Physicists Frame Life as Vehicle Driving Cosmic Entropy Increase

The Most Misunderstood Concept in Physics — Veritasium

Veritasium

Earth sustains biological order not by hoarding solar energy, but by converting each concentrated solar photon into 20 diffuse infrared photons to dissipate universal energy.

AI-assisted summary of The Most Misunderstood Concept in Physics — original source. Check the original for full context.

Source published:

Key ideas

  • Earth maintains thermodynamic equilibrium by radiating into space the exact same quantity of energy it absorbs from the sun, but it trades a single high-energy solar photon for roughly 20 dispersed, lower-energy infrared photons. (14:50)
  • Sadi Carnot demonstrated that heat engine efficiency is fundamentally capped by the ratio of absolute temperatures between hot and cold reservoirs, not engine design or material choice. (6:22)
  • Entropy is not merely disorder but a statistical measure of energy dispersal; backward heat flow is not physically forbidden, but exponentially improbable as particle count scales. (11:24)
  • Black holes represent the primary entropy sinks of the modern cosmos, with Sagittarius A* holding roughly 10^91 Boltzmann constants—thousands of times more entropy than the entire post-Big Bang observable universe. (21:36)
  • Complex structures like galaxies, weather systems, and biological life emerge exclusively during the intermediate transition between low and high entropy rather than at either extreme. (25:00)

Detailed notes

## Background Published on YouTube by the educational channel Veritasium (hosted by Derek Muller) on July 1, 2023, this video explores entropy, described in the show notes as one of the most important yet least understood concepts in all of physics. The description highlights the collaborative physics research behind the presentation alongside references to foundational historical and modern papers.

## TL;DR The video deconstructs common misconceptions about entropy, explaining that it represents the natural statistical tendency of concentrated energy to disperse rather than mere macroscopic disorder. Tracing the concept from Sadi Carnot's 19th-century analysis of heat engine efficiency through Rudolf Clausius and Ludwig Boltzmann's statistical mechanics, Derek Muller shows how the Second Law of Thermodynamics governs everyday phenomena. Crucially, the Earth does not retain net energy from the Sun but rather absorbs low-entropy, high-energy photons and reradiates them as diffuse infrared photons. This cosmic gradient drives biological complexity, explains why time flows in only one direction, and charts the eventual thermodynamic path toward the heat death of the universe.

## Atmosphere & style The presentation blends street-level curiosity with cinematic, mind-expanding scientific storytelling. Derek Muller acts as an approachable, inquisitive guide whose casual man-on-the-street interviews smoothly transition into rigorous historical narratives and elegant theoretical physics. The tone evolves from playful confusion to an awe-inspiring, almost poetic appreciation of existence, framing human life and cosmic structure as brief, magnificent eddies flourishing in the intermediate zone between absolute order and ultimate thermodynamic dissolution.

## Bulletin - [14:50](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=890s) Earth maintains thermodynamic equilibrium by radiating into space the exact same quantity of energy it absorbs from the sun, but it trades a single high-energy solar photon for roughly 20 dispersed, lower-energy infrared photons. - [06:22](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=382s) Sadi Carnot demonstrated that heat engine efficiency is fundamentally capped by the ratio of absolute temperatures between hot and cold reservoirs, not engine design or material choice. - [11:24](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=684s) Entropy is not merely disorder but a statistical measure of energy dispersal; backward heat flow is not physically forbidden, but exponentially improbable as particle count scales. - [21:36](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=1296s) Black holes represent the primary entropy sinks of the modern cosmos, with Sagittarius A* holding roughly 10^91 Boltzmann constants—thousands of times more entropy than the entire post-Big Bang observable universe. - [25:00](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=1500s) Complex structures like galaxies, weather systems, and biological life emerge exclusively during the intermediate transition between low and high entropy rather than at either extreme.

## Key takeaways - [00:00](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=0s) Energy conservation dictates that over long timescales, the Earth must radiate into space the exact amount of energy it receives from the sun to prevent runaway warming. - [04:12](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=252s) Sadi Carnot formulated an idealized reversible engine operating between two thermal reservoirs to determine the absolute theoretical limit of thermodynamic efficiency. - [06:22](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=382s) Carnot efficiency depends solely on reservoir temperatures (eta = 1 - T_cold / T_hot), meaning 100% efficiency requires an impossible cold reservoir at absolute zero. - [09:05](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=545s) Rudolf Clausius formulated entropy to quantify irreversible energy dispersal, establishing that universal entropy relentlessly tends toward a maximum. - [11:24](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=684s) Ludwig Boltzmann showed that heat flows spontaneously from hot to cold due to statistical counting: thermal equilibrium corresponds to overwhelmingly numerous microstates compared to non-equilibrium configurations. - [14:50](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=890s) Local entropy decreases, such as domestic air conditioning or cellular organization, must always be compensated by a greater net entropy increase in the wider environment. - [14:50](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=890s) Life functions as an open thermodynamic engine, driving metabolic processes by transforming concentrated solar radiation into diffuse planetary infrared waste. - [18:10](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=1090s) Dissipative adaptation suggests that matter driven by external energy naturally self-organizes into structures that accelerate entropy production. - [19:11](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=1151s) Under the Past Hypothesis, the early universe occupied an exceptionally low-entropy state because matter was smoothly distributed rather than clumped by gravity. - [21:36](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=1296s) Jacob Bekenstein and Stephen Hawking revealed that black hole entropy is proportional to event horizon surface area, with black holes dominating cosmic entropy today. - [23:31](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=1411s) The thermodynamic arrow of time is defined by the irreversible progression from statistically improbable low-entropy states toward high-probability equilibrium. - [23:31](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=1411s) The heat death of the universe represents an ultimate statistical equilibrium where no work can be extracted and the macroscopic arrow of time disappears. - [25:00](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=1500s) Structural and biological complexity peaks during the dynamic intermediate phase between low and high entropy, while both pure order and maximum entropy lack complexity.

## Detailed walkthrough ### [00:00](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=0s) The Street Interview: What Does Earth Get from the Sun?

A fundamental concept in physics governs everything from microscopic molecular collisions to colossal atmospheric storms, shaping the evolution of the universe from the Big Bang to its eventual demise. It may determine why time flows only forward and even explain why life itself emerged. Yet, despite its sweeping importance, it remains deeply misunderstood. To highlight the widespread confusion, Derek Muller takes to the street to pose a deceptively basic question to passersby: what does the Earth get from the sun?

People readily reply with heat, light, warmth, ultraviolet rays that produce vitamin D, and, broadly, energy. Derek agrees, but then follows up with a crucial second question: relative to the amount of energy the Earth absorbs from the sun, how much does it radiate back out into space? The responses immediately reveal intuitive misconceptions. People guess that Earth retains most of it, offering estimates like 20% or 70% radiated back. One person reasons, “Probably not as much, I, you know, I don't believe it's just radiating right back,” while another explains, “Because we use some of the energy.” But energy cannot simply be used up or disappear.

As Derek explains, if the Earth absorbed more energy than it released, it would continuously heat up, creating an unlivable catastrophe. As Derek puts it: “For most of the earth's history, it should be exactly the same amount of energy in from the sun as earth radiates into space.” This leaves the interviewees visibly puzzled. If the total energy entering the Earth is equal to the energy radiated back out into space, what is Earth actually receiving? While one bystander jokes, “It gives us a nice tan,” the true answer requires examining an intellectual breakthrough that began two centuries ago.

- Passersby intuitively assume Earth absorbs and holds onto most solar energy to power civilization and natural processes. - Because energy is conserved, the net incoming solar radiation must balance the outgoing terrestrial thermal radiation over the long term to maintain a steady planetary temperature.

---

[02:15](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=135s) Sadi Carnot and the Search for Engine Efficiency

In the winter of 1813, France faced invasion from the allied armies of Austria, Prussia, and Russia. Sadi Carnot, a 17-year-old student and the son of one of Napoleon Bonaparte’s generals, wrote directly to Napoleon on December 29 requesting permission to fight on the front lines. Preoccupied with the war, the Emperor never replied. A few months later, when enemy forces marched on Paris, Carnot and his fellow students took up arms to defend a chateau east of the capital. They were hopelessly outmatched, Paris fell within a single day, and the young Carnot was left devastated by the defeat.

Seven years later, Carnot traveled to Prussia to visit his exiled father, who was not only a military leader but also an accomplished physicist with published work on mechanical energy transfer. Their conversations gravitated toward the defining technology of the industrial age: the steam engine. Steam engines were rapidly reshaping global infrastructure by propelling ships, draining mine shafts, and dredging harbors. Carnot realized that the economic and military supremacy of nations would belong to whoever possessed the most powerful, fuel-efficient engines.

At the time, French engineering lagged noticeably behind British industry. The most advanced steam engines in existence managed to convert only about 3% of their thermal energy into usable mechanical work. Determined to restore France’s national standing, Carnot spent the next three years investigating the fundamental physics of heat engines to determine whether there was an absolute upper limit to their efficiency.

- On December 29, 1813, 17-year-old Carnot wrote to Napoleon asking to serve, but Paris fell shortly after in a one-day battle. - Contemporary early-19th-century steam engines operated at an abysmal conversion efficiency of approximately 3%.

---

[04:12](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=252s) The Ideal Carnot Engine and Reversibility

To analyze heat engines without the messiness of real-world flaws, Carnot conceived a theoretical, frictionless engine that suffered no heat losses to the ambient surroundings. This idealized machine operates between two large thermal reservoirs: a hot metal bar and a cold metal bar. The working mechanism consists of an insulated cylinder filled with air, sealed by a movable piston connected to a heavy flywheel, with a conductive base allowing heat transfer exclusively through the bottom.

```mermaid flowchart TD accTitle: The Four Stages of the Carnot Cycle accDescr: Step by step progression of the Carnot cycle through isothermal and adiabatic expansion followed by isothermal and adiabatic compression. subgraph HotStage ["Hot Reservoir Phase"] A["1. Isothermal Expansion<br>(Chamber on hot bar; gas absorbs heat and expands, pushing piston)"] B["2. Adiabatic Expansion<br>(Hot bar removed; gas continues expanding and cools to cold bar temperature)"] end subgraph ColdStage ["Cold Reservoir Phase"] C["3. Isothermal Compression<br>(Chamber on cold bar; flywheel pushes piston, dumping heat into cold bar)"] D["4. Adiabatic Compression<br>(Cold bar removed; gas compressed further, heating to initial hot state)"] end A --> B B --> C C --> D D --> A ```

The cycle begins with the air inside the cylinder slightly below the hot bar's temperature. Bringing the hot bar against the base causes the gas to expand at constant temperature as heat enters, driving the piston upward and turning the flywheel. The hot bar is then pulled away, allowing the air to continue expanding adiabatically; because no new heat enters, the gas cools until it matches the temperature of the cold bar. Next, the cold bar is brought into contact with the base while the momentum of the flywheel forces the piston back down, compressing the gas and discharging heat into the cold sink. Finally, the cold bar is detached, and the flywheel compresses the gas further, warming it back up to just below the hot bar's temperature so the cycle can repeat.

Carnot recognized that this idealized engine is entirely reversible. If the flywheel's momentum is driven in the opposite direction, the gas expands to draw heat out of the cold bar, gets compressed, and deposits that exact amount of heat back into the hot bar. As Derek notes: “However many cycles were run in the forward direction, you could run the same number in reverse, and at the end, everything would return to its original state with no additional input of energy required.” Because an ideal engine can run forward and backward indefinitely without altering the total state of the system or its surroundings, running it produces no permanent change in the universe.

---

[06:22](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=382s) Efficiency Limits and the Kelvin Scale

A fully reversible engine might seem capable of 100% efficiency, but Carnot proved this is mathematically impossible. The net mechanical energy added to the flywheel during each cycle equals the heat absorbed from the hot reservoir minus the heat rejected into the cold reservoir. The efficiency is simply this net work divided by the heat input from the hot side. Because the gas is hotter and under higher pressure during the expansion stroke than during the cold compression stroke, the work done by the expanding gas exceeds the work required to compress it back down. Still, some heat must always be dumped into the cold reservoir to reset the piston.

Decades later, Lord Kelvin realized that Carnot's ideal engine established an absolute temperature scale independent of any specific material. If a gas could expand until it cooled to a state where molecular motion ceased entirely, the particles would exert zero pressure on the piston. Compressing the gas on the cold side would then require zero mechanical work, meaning no waste heat would be rejected, yielding an engine of 100% efficiency. This theoretical zero-point defines absolute zero on the Kelvin scale.

Because the heat absorbed and rejected in an ideal cycle is directly proportional to absolute temperature, the efficiency of a Carnot engine simplifies to $\eta = 1 - (T_{\text{cold}} / T_{\text{hot}})$. As Derek emphasizes: “What we have learned is that the efficiency of an ideal heat engine doesn't depend on the materials or the design of the engine, but fundamentally on the temperatures of the hot and cold sides.” Achieving 100% thermal efficiency would require either an infinitely hot heat source or a cold sink at absolute zero—both physically unreachable. In Carnot's era, high-pressure steam engines operated at temperatures around 160 °C, establishing a theoretical Carnot efficiency ceiling of 32%, far above their real-world 3% performance.

- Maximum thermal efficiency is strictly determined by the ratio of the absolute temperatures of the heat reservoirs: $\eta = 1 - \frac{T_C}{T_H}$. - 19th-century steam engines operating at 160 °C (around 433 K) and exhausting near ambient temperatures had a maximum theoretical limit of 32%.

---

[09:05](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=545s) Rudolf Clausius, Entropy, and the Second Law

Real heat engines fall far short of theoretical limits because they are plagued by internal friction, imperfect heat transfer across fluctuating temperatures, and uncontained thermal dissipation into the surrounding air, cylinder walls, and flywheel axles. Once heat leaks into the wider environment, it disperses randomly into microscopic vibrations that cannot be gathered back together. The mechanical cycle becomes strictly irreversible. Even though the first law of thermodynamics guarantees that total energy is conserved, the energy undergoes a qualitative decline. As Derek observes: “Energy is most usable when it is concentrated and less usable when it's spread out.”

In the 1850s and 1860s, German physicist Rudolf Clausius analyzed Carnot's thermodynamic framework and created a mathematical formulation to track this dissipation of energy, naming the property *entropy*. When energy is tightly concentrated within a hot source, entropy is low and the capacity to do useful work is high. When that energy disperses into the cooler environment, cylinder casing, and mechanical bearings, entropy increases, rendering that energy unavailable for work.

In 1865, Clausius distilled thermodynamics down to two fundamental tenets: “First, the energy of the universe is constant. And second, the entropy of the universe tends to a maximum.” This second law dictates why heat flows spontaneously from hot objects to cold ones, why an unconstrained gas expands to fill its container, and why perpetual motion machines are an impossibility. Derek points out that while popular science typically equates entropy with "disorder"—reflecting how systems naturally evolve toward mixed and random states—the most accurate physical intuition is simply “the tendency of energy to spread out.”

- Real-world engines are irreversible because friction and heat dissipation disperse concentrated energy into the surrounding environment. - Clausius coined the term *entropy* to quantify how dispersed energy becomes relative to absolute temperature. - The two fundamental laws of thermodynamics formulated by Clausius (1865): 1. The total energy of the universe remains constant. 2. The total entropy of the universe tends toward a maximum.

[11:24](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=684s) Boltzmann’s Statistical Insight and the Improbability of Heat Reversal

The fundamental laws of physics operate identically whether running forwards or backwards in time, raising the question of where the observed arrow of time actually originates. To demonstrate this on a microscopic scale, Derek Muller introduces a simplified model of two small metal bars brought together, one hot and one cold, containing just eight atoms each. The thermal energy of each atom corresponds to the number of discrete energy packets it holds; the more packets it contains, the more vigorously it vibrates. Starting with seven packets in the left bar and three in the right, these packets continuously hop at random between atoms across the lattice while the total energy remains constant. When the two bars touch, packets hop across the boundary, creating countless possible energy arrangements where every single microscopic configuration is equally likely.

```mermaid flowchart LR accTitle: Energy Packet Redistribution Between Two Contacting Solids accDescr: Microscopic hopping of energy packets leads to macroscopic thermal equilibrium based on high-probability microstates.

subgraph Initial["Initial Non-Equilibrium State"] Hot["Hot Bar: 7 Packets"] Cold["Cold Bar: 3 Packets"] Hot <-->|"Energy Exchange"| Cold end

subgraph States["Statistical Microstates"] Improbable["Cold to Hot: 9 on Left / 1 on Right<br/>(91,520 configurations)"] Probable["Equilibrium: 5 on Left / 5 on Right<br/>(627,264 configurations)"] end

Initial --> States ```

If one takes a snapshot after contact, it is entirely possible to find nine packets in the left bar and only one in the right. This represents heat flowing from cold to hot—an apparent decrease in entropy. Ludwig Boltzmann resolved this apparent contradiction by recognizing that heat flowing backwards is not fundamentally forbidden, but merely statistically improbable: “Heat flowing from cold to hot is not impossible, it’s just improbable.” While there are 91,520 configurations where nine packets sit in the left bar, there are 627,264 configurations where energy divides evenly with five packets in each bar, making thermal equilibrium almost seven times more probable.

Even in this eight-atom system, there is still an overall 10.5% chance that the hot bar ends up with more energy than it started with. However, scaling the model up to 80 atoms per bar with 100 energy packets (starting 70 on the left and 30 on the right) drops the probability of the left solid getting hotter to just 0.05%. In everyday macroscopic objects containing roughly 100 trillion trillion ($10^{26}$) atoms and vastly more energy packets, the chance of spontaneous heat concentration approaches zero so closely that it never occurs. Muller illustrates this using a Rubik's cube: starting from a solved state and applying blind, random turns inevitably scrambles it. Because there is only one solved configuration, a handful of nearly solved states, and quintillions of disordered arrangements, random actions overwhelmingly drive systems from rare, ordered configurations into overwhelmingly abundant messy ones.

- Eight atoms per bar with a 7:3 packet split: 91,520 microstates for a 9:1 split vs. 627,264 microstates for a 5:5 split. - Chance of left bar ending hotter in the 8-atom toy model: 10.5%. - Chance of left solid ending hotter in an 80-atom system (70:30 starting split): 0.05%. - Real-world macroscopic solids contain on the order of 100 trillion trillion atoms.

---

[14:50](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=890s) Local Order Paid by Global Entropy: Air Conditioners and the Sun

If the natural law of the universe is for energy to disperse and disorder to grow, localized processes that concentrate energy seem puzzling. An air conditioner cools an already cool room by dumping heat into an already hotter outdoors, clearly decreasing the entropy inside the house. However, this local decrease is only achieved by generating a far larger quantity of entropy elsewhere in the broader system. Running the air conditioner requires electricity from a power plant where concentrated chemical energy stored in coal is liberated, creating vast amounts of thermal waste at the generation site, dispersing heat through miles of transmission lines, and venting friction and electrical losses from fans and compressors. The localized drop in domestic entropy is heavily outmatched by the total entropy generated to power the device: “Whatever decrease in entropy is achieved at the house is more than paid for by an increase in entropy required to make that happen.”

This mechanism explains how complex order and life persist on Earth without violating thermodynamics. If Earth were an isolated, closed system, all internal temperature differences would flatten out, chemical complexity would rot and homogenize, and all life would perish. Earth avoids this equilibrium because it is an open system bathed in continuous radiation from the sun. The sun delivers a continuous inflow of low-entropy, concentrated energy. Terrestrial life captures this bundled energy—plants synthesize sugars via photosynthesis, herbivores consume plants, and carnivores consume herbivores. At each link in the food chain, energy cascades outward and disperses further.

Muller discusses the planetary balance sheet with an off-camera colleague, jokingly reminding him: “I'm a PhD physicist,” to which the colleague replies, “I trust you.” In steady state, Earth emits the exact same total quantity of energy back into space as it absorbs from solar radiation. The entropy generation of this planetary exchange is revealed by photon counting: for every single high-energy, low-entropy solar photon Earth absorbs, it radiates roughly 20 lower-energy infrared photons back into the cosmic void. Every terrestrial phenomenon—winds, oceanic currents, biological metabolism, sleeping, and ecosystems—is powered entirely along this gradient of converting few energetic photons into 20 times as many diffuse ones.

```mermaid flowchart TD accTitle: Earth Thermodynamic Budget and Photon Multiplication accDescr: Earth absorbs concentrated solar photons and re-emits dispersed infrared photons, driving biological and physical processes.

Sun["Sun: Emits 1 High-Energy Photon<br/>(Concentrated Low Entropy)"] -->|"Absorbed"| Earth["Earth System<br/>(Photosynthesis, Weather, Metabolism)"] Earth -->|"Dissipated as Waste Heat"| Space["Space: Emits 20 Low-Energy Photons<br/>(Dispersed High Entropy)"] ```

---

[18:10](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=1090s) Dissipative Structures, Jeremy England, and the Origin of Life

Because life dramatically accelerates the dispersion of energy, several physicists suggest that life is not an improbable accident opposing thermodynamics, but rather a direct vehicle of the Second Law. If the cosmos drives toward maximizing entropy, self-organizing structures emerge because they convert low entropy into high entropy far more efficiently than inanimate matter. In marine environments, for instance, surface waters enriched with cyanobacteria and organic material generate between 30% and 680% more entropy than identical barren seawater.

Physicist Jeremy England expanded this concept into the physics of self-replication and driven assembly. When an arbitrary collection of matter is subjected to an external driving stream of concentrated energy, physical forces naturally select and stabilize configurations capable of absorbing and dissipating that energy into the environment. Over time, matter undergoes dissipative adaptation, evolving increasingly sophisticated dissipators until complex lifelike structures emerge. Muller quotes Jeremy England's formulation directly: “You start with a random clump of atoms, and if you shine light on it for long enough, it should not be so surprising that you get a plant.”

- Surface seawater containing cyanobacteria and organic matter increases entropy production by 30% to 680% compared to abiotic seawater. - Jeremy England’s dissipative adaptation framework models living systems as optimized structures that unpack and discard concentrated free energy into surrounding heat baths.

---

[19:11](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=1151s) Gravity, the Big Bang, and the Cosmic Past Hypothesis

Tracking the chain of low entropy backwards reveals a cosmic puzzle. If Earth relies on the sun's low entropy, the sun owes its concentrated energy to the universe at large. Because cosmic entropy increases monotonically over time, entropy must have been lower yesterday, lower still millions of years ago, and minimal at the moment of the Big Bang. This requirement is known in cosmology as the *past hypothesis*. While it does not explain why the initial conditions were so ordered, it states that an exceptionally low-entropy starting condition was essential for galaxies, stars, and life to form.

This poses a major paradox: the infant universe was hot, dense, and nearly uniform, with temperature variations across space not exceeding 0.001% (one part in 100,000). In a standard gas, complete uniformity denotes maximum entropy. The missing variable in that assessment is gravity. In a self-gravitating system, a smooth, uniform distribution of mass is extraordinarily unnatural and improbable; gravity's natural drive is to pull matter into dense clumps. Thus, an ultra-smooth early universe represents an extremely low-entropy state:

```mermaid flowchart LR accTitle: Gravitational Entropy Production Across Cosmic Time accDescr: The early smooth matter distribution collapses under gravity, releasing potential energy and driving star formation and entropy growth.

Uniform["Uniform Gas Post-Big Bang<br/>(Low Gravitational Entropy)"] -->|"Gravitational Collapse"| Clumping["Clumping into Stars & Galaxies<br/>(Potential Energy to Kinetic Energy)"] Clumping -->|"Collisions & Radiation"| Dissipation["Thermal Dissipation & Black Holes<br/>(Massive Entropy Increase)"] ```

As the cosmos expanded and cooled, gravitational instability drew matter into denser pockets. This collapse converted enormous reserves of gravitational potential energy into kinetic energy. When particles collided during collapse, kinetic energy degraded into heat, dissipating useful energy and increasing overall entropy while forging stars, planets, and galaxies.

At the beginning of the universe, total entropy stood at roughly $10^{88}$ Boltzmann constants ($k_\text{B}$). In the present epoch, the observable stellar mass accounts for about $9.5 \times 10^{80}\ k_\text{B}$. The combined interstellar and intergalactic medium contains nearly ten times that amount, though still a tiny fraction of the initial universal value. Considerably larger pools of ordinary entropy reside in cosmic relic neutrinos and the photons of the Cosmic Microwave Background (CMB).

---

[21:36](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=1296s) Black Hole Thermodynamics: Bekenstein and Hawking

In 1972, Jacob Bekenstein proposed that black holes possess entropy and that this entropy is directly proportional to the surface area of their event horizons, meaning a black hole's entropy increases whenever it swallows matter and expands. Mainstream physicists initially dismissed the proposal as absurd. Classical thermodynamics dictates that anything endowed with entropy must also possess a non-zero temperature, and any body with a finite temperature must emit blackbody radiation—meaning a black hole could not be entirely black.

Stephen Hawking set out to disprove Bekenstein's conjecture using quantum field theory in curved spacetime. Instead of debunking it, Hawking discovered that quantum effects near the event horizon cause black holes to emit a steady stream of thermal radiation, now celebrated as Hawking radiation:

```mermaid flowchart TD accTitle: Development of Black Hole Thermodynamics accDescr: Bekenstein's controversial area-entropy conjecture confirmed by Hawking's calculation of black hole temperature and radiation.

Bekenstein["1972: Jacob Bekenstein proposes<br/>Entropy ∝ Horizon Surface Area"] Skepticism["Physics Community Objects:<br/>Finite Entropy implies Temperature & Radiation"] Hawking["Stephen Hawking attempts disproof<br/>Discovers Hawking Radiation & validates Bekenstein"]

Bekenstein --> Skepticism Skepticism --> Hawking ```

Hawking's calculations confirmed that black holes have both a definite temperature and enormous entropy, refining Bekenstein's model into the Bekenstein-Hawking entropy formula. For stellar and supermassive black holes, this temperature is minuscule. The supermassive black hole at the core of the Milky Way (Sagittarius A*) has a temperature of approximately one hundred trillionth of a Kelvin ($10^{-14}\text{ K}$), radiating far too weakly to be detected against the cosmic background.

Hawking's equations revealed that black holes are the dominant repositories of entropy in the cosmos: - Sagittarius A* contains roughly $10^{91}$ Boltzmann constants of entropy—1,000 times more entropy than the entire observable universe possessed shortly after the Big Bang, and ten times more than all other cosmic particles combined. - All black holes across the observable universe hold a combined entropy of approximately $3 \times 10^{104}\ k_\text{B}$. - Because black holes concentrate virtually all cosmic entropy, the early post-Big Bang universe held a mere $0.000000000000003\%$ ($3 \times 10^{-15}\%$) of the entropy present in the universe today.

[23:31](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=1411s) The Arrow of Time and the Heat Death of the Universe

Every dynamic event in cosmic history—from the formation of planetary systems and the collision of galaxies to crashing asteroids, dying stars, and the emergence of life itself—is driven by a single fundamental reality: the universe began in a state of remarkably low entropy, and that entropy has been climbing ever since. Crucially, physical transformations happen in only one direction. We observe asteroids shattering upon impact, but we never witness an asteroid spontaneously "uncrash," just as we never see a mature solar system unmix itself back into a pristine, dispersed cloud of primordial gas and dust. This fundamental asymmetry is what separates the past from the future. As Derek Muller explains, “The fact that we are going from unlikely to more likely states is why there is an arrow of time.”

```mermaid flowchart TD accTitle: The Thermodynamic Arrow of Time accDescr: Progression from the Big Bang's low-entropy state through cosmic evolution to the ultimate heat death of the universe.

subgraph Past ["Low Entropy (Past)"] A["Big Bang: highly improbable, low-entropy state"] end

subgraph Present ["Increasing Entropy (Present)"] B["Cosmic Evolution: stars, galaxies, and life flourish"] end

subgraph Future ["Maximum Entropy (Distant Future)"] C["Evaporation of last black holes (>10¹⁰⁰ years)"] D["Heat Death: energy uniformly spread, arrow of time vanishes"] end

A --> B B --> C C --> D ```

This thermodynamic trajectory is projected to continue until energy is distributed so completely and evenly that no thermodynamic work can ever be extracted again. This ultimate state of stagnation is known as the heat death of the universe. In the inconceivably distant future—more than $10^{100}$ years from now, once the final black holes have completely evaporated through Hawking radiation—the universe will reach its most probable statistical state. At that point, even across vast cosmic distances, moving forward in time will be physically indistinguishable from moving backward, and the macroscopic arrow of time will cease to exist.

- Everything dynamic in the universe relies on the gradient created by its initial low-entropy state. - Irreversible macroscopic processes establish the directionality of time. - The heat death represents the ultimate equilibrium state of maximum probability and maximum entropy. - After black holes evaporate past $10^{100}$ years, macroscopic temporal directionality disappears entirely.

---

[25:00](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=1500s) The Emergence of Complexity Between Extremes

While it may seem that entropy is simply a destructive force dragging reality toward the dullest conceivable fate, high entropy should not be mistaken as the sole enemy of form and beauty. Just as maximum entropy corresponds to minimal complexity, a state of absolute minimum entropy is equally devoid of complex behavior. Derek illustrates this using the everyday example of tea and milk. Held completely separate in a mug and pitcher, the liquids have low entropy but are static and uninteresting. However, the instant the milk is poured into the hot tea, intricate, swirling plumes and delicate fractal patterns bloom across the liquid before eventually fading away into a bland, uniform mixture.

```mermaid flowchart LR accTitle: Complexity Curve Relative to Entropy accDescr: Charting complexity across low, intermediate, and maximum entropy states using the milk and tea analogy.

A["Separate Tea & Milk<br>(Low Entropy / Low Complexity)"] --> B["Active Mixing & Eddies<br>(Medium Entropy / Peak Complexity)"] B --> C["Uniform Mixture<br>(Maximum Entropy / Low Complexity)"] ```

Both extremes—pure order and pure disorder—are fundamentally poor in structure. Complex, organized phenomena do not exist at the absolute extremes; they emerge and thrive exclusively in the intermediate transition zone while entropy is on the rise. Derek highlights our place within this ongoing thermodynamic cascade: “Both low and high entropy are low in complexity. It's in the middle where complex structures appear and thrive.” Because human existence and the biosphere are situated right in this sweet spot, he urges viewers to take advantage of the universe's remaining low entropy while it lasts.

- Maximum entropy features low complexity because everything is uniformly mixed and homogeneous. - Pure low entropy also features low complexity due to its static, unmixed separation. - Rich, intricate structures (including living organisms) emerge exclusively during the intermediate transition from low to high entropy.

---

[25:47](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=1547s) Equipping Your Mind: Brilliant.org

Reflecting on the overarching lessons of thermodynamics, Derek notes that having the proper conceptual frameworks empowers people to demystify complex systems, remarking: “With the right tools, we can understand just about anything, from a cup of tea cooling down to the evolution of the entire universe.” He transitions to highlighting the video's sponsor, Brilliant.org, as a practical platform for building those analytical capabilities across STEM fields, including mathematics, physics, data science, and programming.

Rather than relying on passive lectures, Brilliant emphasizes active, problem-based learning. Derek highlights their interactive course, "Thinking in Code," which uses a drag-and-drop environment to teach foundational programming principles such as nesting and conditional logic without getting bogged down in syntax errors. The course guides learners through tangible challenges—ranging from programming a virtual robot to designing real-world automation scripts, such as setting phone reminders or filtering prospective matches on a dating app. He concludes by offering a 30-day free trial along with a 20% discount on an annual premium subscription for the first 200 subscribers who visit the dedicated link.

- Brilliant provides guided interactive pathways across physics, programming, math, and data science. - The "Thinking in Code" curriculum introduces essential programming constructs, including conditionals and nested logic, via intuitive drag-and-drop interfaces. - The learning methodology focuses on practical applications and immediate feedback to develop real intuition. - Viewers are offered a 30-day free trial and a 20% discount on an annual premium tier at `brilliant.org/veritasium`.

## Notable quotes > [11:24](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=684s) Heat flowing from cold to hot is not impossible, it’s just improbable. — Ludwig Boltzmann

> [18:10](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=1090s) You start with a random clump of atoms, and if you shine light on it for long enough, it should not be so surprising that you get a plant. — Jeremy England

> [25:00](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=1500s) Both low and high entropy are low in complexity. It's in the middle where complex structures appear and thrive. — Derek Muller

> [23:31](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=1411s) The fact that we are going from unlikely to more likely states is why there is an arrow of time. — Derek Muller

> [09:05](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=545s) Energy is most usable when it is concentrated and less usable when it's spread out. — Derek Muller

> [06:22](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=382s) What we have learned is that the efficiency of an ideal heat engine doesn't depend on the materials or the design of the engine, but fundamentally on the temperatures of the hot and cold sides. — Derek Muller

## Mentioned Sadi Carnot, Napoleon Bonaparte, Lord Kelvin, Rudolf Clausius, Ludwig Boltzmann, Jeremy England, Jacob Bekenstein, Stephen Hawking, Derek Muller, Brilliant.org, Sagittarius A*, Elements of Truth, Snatoms

## Video description One of the most important, yet least understood, concepts in all of physics. Head to https://brilliant.org/veritasium to start your free 30-day trial, and the first 200 people get 20% off an annual premium subscription.

In 2025, we launched the Elements of Truth game on Kickstarter and were thrilled by the response. We’re now in the final stages of production, with Kickstarter rewards shipping soon. If you missed the campaign, you can now pre-order the retail edition through our new website here - https://ve42.co/EntropyDes

If you're looking for a molecular modeling kit, try Snatoms - a kit I invented where the atoms snap together magnetically: https://snatoms.com

We have a tabletop game launching! Pledge your support now - https://ve42.co/KickStarterD

▀▀▀ A huge thank you to those who helped us understand different aspects of this complicated topic - Dr. Ashmeet Singh, Supriya Krishnamurthy, Dr. Jos Thijssen, Dr. Bijoy Bera, Dr. Timon Idema, Álvaro Bermejillo Seco and Dr. Misha Titov.

▀▀▀ References: Carnot, S. (1824). Reflections on the motive power of heat: and on machines fitted to develop that power. - https://ve42.co/Carnot1890

Harnessing The True Power Of Atoms | Order And Disorder Documentaries, Spark via YouTube - https://ve42.co/OrderDisorder

A better description of entropy, Steve Mould via YouTube - https://ve42.co/Mould2016

Dugdale, J. S. (1996). Entropy and its physical meaning. CRC Press. - https://ve42.co/Dugdale1996

Schroeder, D. V. (1999). An introduction to thermal physics. - https://ve42.co/Schroeder2021

Fowler, M. Heat Engines: the Carnot Cycle, University of Virginia. - https://ve42.co/Fowler2023

Chandler, D.L. (2010). Explained: The Carnot Limit, MIT News - https://ve42.co/Chandler2010

Entropy, Wikipedia - https://ve42.co/EntropyWiki

Clausius, R. (1867). The mechanical theory of heat. Van Voorst. - https://ve42.co/Clausius1867

What is entropy? TED-Ed via YouTube - https://ve42.co/Phillips2017

Thijssen, J. (2018) Lecture Notes Statistical Physics, TU Delft.

Schneider, E. D., & Kay, J. J. (1994). Life as a manifestation of the second law of thermodynamics. Mathematical and computer modelling, 19(6-8), 25-48. - https://ve42.co/Schneider1994

Lineweaver, C. H., & Egan, C. A. (2008). Life, gravity and the second law of thermodynamics. Physics of Life Reviews, 5(4), 225-242. - https://ve42.co/Lineweaver2008

Michaelian, K. (2012). HESS Opinions" Biological catalysis of the hydrological cycle: life's thermodynamic function". Hydrology and Earth System Sciences, 16(8), 2629-2645. - https://ve42.co/Michaelian2012

England, J. L. (2013). Statistical physics of self-replication. The Journal of chemical physics, 139(12), 09B623_1. - https://ve42.co/England2013

England, J. L. (2015). Dissipative adaptation in driven self-assembly. Nature nanotechnology, 10(11), 919-923. - https://ve42.co/England2015

Wolchover, N. (2014). A New Physics Theory of Life, Quantamagazine - https://ve42.co/Wolchover2014

Additional References: https://ve42.co/EntropyRefs

▀▀▀ Special thanks to our Patreon supporters: Emil Abu Milad, Tj Steyn, meg noah, Bernard McGee, KeyWestr, Amadeo Bee, TTST, Balkrishna Heroor, John H. Austin, Jr., john kiehl, Anton Ragin, Diffbot, Gnare, Dave Kircher, Burt Humburg, Blake Byers, Evgeny Skvortsov, Meekay, Bill Linder, Paul Peijzel, Josh Hibschman, Mac Malkawi, Juan Benet, Ubiquity Ventures, Richard Sundvall, Lee Redden, Stephen Wilcox, Marinus Kuivenhoven, Michael Krugman, Sam Lutfi.

▀▀▀ Written by Casper Mebius, Derek Muller & Petr Lebedev Edited by Trenton Oliver & Jamie MacLeod Animated by Mike Radjabov, Ivy Tello, Fabio Albertelli and Jakub Misiek Filmed by Derek Muller, Albert Leung & Raquel Nuno Molecular collisions video by CSIRO's Data61 via YouTube: Simulation of air Additional video/photos supplied by Getty Images, Pond5 and by courtesy of NASA, NASA's Goddard Space Flight Center, NASA Goddard Flight Lab/ CI Lab, NASA/SDO and the AIA, EVE, HMI, and WMAP science teams. As well as the Advanced Visualization Laboratory at the National Center for Supercomputing Applications, B. Robertson, L. Hernquist Music from Epidemic Sound & Jonny Hyman Produced by Derek Muller, Petr Lebedev, Emily Zhang, & Casper Mebius

[Source](https://www.youtube.com/watch?v=DxL2HoqLbyA)

中文简报

物理学家将生命视为推动宇宙熵增的载体 地球维系生物秩序并非依靠储存太阳能,而是通过将每个集中的太阳光子转化为20个漫射的红外光子来耗散宇宙能量。 该视频破除了关于熵的常见误解,解释了熵所代表的是集中能量自发耗散的统计学趋势,而不仅仅是宏观上的无序。从萨迪·卡诺19世纪对热机效率的分析,到鲁道夫·克劳修斯和路德维希·玻尔兹曼的统计力学,德里克·穆勒展示了热力学第二定律如何支配日常现象。至关重要的是,地球并没有净保留来自太阳的能量,而是吸收低熵、高能量的光子,并以漫射红外光子的形式重新辐射出去。这种宇宙梯级驱动了生物的复杂性,解释了时间为何只朝一个方向流逝,并描绘了宇宙最终走向热寂的热力学路径。 地球通过向太空辐射与吸收自太阳完全等量的能量来维持热平衡,但它将单个高能太阳光子转化为大约20个分散的、能量较低的红外光子。 (14:50) 萨迪·卡诺证明,热机效率在根本上受限于冷热热源之间的绝对温度比,而非发动机设计或材料选择。 (6:22) 熵不仅是无序,更是能量耗散的统计学量度;热量反向流动在物理上并非绝不可能,但随着粒子数量的增加,其概率呈指数级降低。 (11:24) 黑洞是现代宇宙主要的熵汇,人马座A*拥有约10^91个玻尔兹曼常数的熵——是大爆炸后整个可观测宇宙熵的数千倍。 (21:36) 星系、天气系统和生物等复杂结构,仅出现在低熵向高熵演变的中间过渡阶段,而非出现于任何一个极端状态。 (25:00)

Reading summary

Derek Muller explains that the true gift the Sun provides Earth is not net energy—since incoming and outgoing energy must balance—but low entropy, the fundamental gradient of spreading energy that drives cosmic evolution, the arrow of time, and the emergence of life. The Planetary Energy Balance and Entropy Gradient (0:00) A common intuition is that Earth absorbs and stores most solar radiation to fuel natural and human processes. However, as Derek Muller demonstrates, because energy is conserved, the planet must radiate back the exact amount it absorbs to avoid continuous, catastrophic heating. Earth operates in a steady state, emitting approximately 20 diffuse, low-energy infrared photons into space for every single concentrated, high-energy solar photon absorbed. This thermodynamic degradation powers all planetary phenomena, from weather patterns to biological metabolism. From Carnot Efficiency to Statistical Irreversibility (4:12) The conceptual origin of entropy traces back to Sadi Carnot's 1820s analysis of heat engines, which revealed that maximum thermal efficiency is strictly capped by the temperature ratio of hot and cold reservoirs, $\eta = 1 - (T_C / T_H)$. While theoretical engines operating without friction are reversible, real engines leak heat, dispersing concentrated energy into microscopic vibrations. Rudolf Clausius formalized this loss as entropy, positing in 1865 that the energy of the universe remains constant while its total entropy continuously rises toward a maximum. Ludwig Boltzmann resolved why these macroscopic processes never run backward. Microscopic laws are time-reversible, but thermal states are governed by overwhelming statistical probabilities. In an eight-atom model, there is a 10.5% chance for heat to spontaneously move from cold to hot, but scaling to 80 atoms drops that probability to 0.05%. In macroscopic objects containing roughly $10^{26}$ atoms, the statistical likelihood of dispersed energy spontaneously concentrating approaches zero, giving rise to an irreversible arrow of time. Local Order and Dissipative Living Systems (14:50) Localized reductions in entropy, such as a refrigerator cooling a room or organisms synthesizing complex tissues, do not violate the second law because they are offset by far larger entropy increases in their surrounding environments. In open systems, living matter accelerates overall thermodynamic dissipation. Physicist Jeremy England proposes that matter under an external energy stream undergoes dissipative adaptation, naturally evolving into configurations that disperse concentrated free energy into surrounding heat baths. Supporting this, surface seawater containing cyanobacteria and organic material generates 30% to 680% more entropy than barren seawater. Cosmic Origins, Gravitational Clumping, and Black Holes (19:11) Tracking universal entropy backward requires the past hypothesis: the early cosmos must have begun in an exceptionally low-entropy state. Although the post-Big Bang universe was nearly uniform—with spatial temperature fluctuations under 0.001%—this uniformity represents minimum entropy only in a self-gravitating system, where matter naturally clumps. Gravitational collapse converted immense gravitational potential energy into kinetic energy and heat, forging stars and galaxies. Following Jacob Bekenstein's area-entropy proposal and Stephen Hawking's discovery of Hawking radiation, black holes were confirmed as the primary reservoirs of cosmic entropy. The early post-Big Bang universe contained only about $10^{88}$ Boltzmann constants ($k_B$) of entropy, or roughly $3 \times 10^{-15}\%$ of the current cosmos. Today, Sagittarius A* alone holds about $10^{91}\ k_B$, and all black holes combined account for roughly $3 \times 10^{104}\ k_B$. Complexity in the Intermediate Zone (23:31) Neither extreme order nor maximum disorder fosters structural complexity. Complete separation of states possesses low entropy but remains static, while a thoroughly mixed system at maximum entropy is uniform and devoid of dynamic organization. Muller notes that complex forms, including life, emerge and thrive solely in the intermediate transitional phase where entropy is actively increasing. In the projected heat death of the universe—more than $10^{100}$ years in the future, after all black holes have evaporated—energy will be uniformly distributed, and the macroscopic arrow of time will cease. The video highlights that living systems and complex cosmic structures depend strictly on an unrepeatable initial condition of low entropy established at the Big Bang. While the universal trajectory toward maximum thermodynamic equilibrium and eventual heat death is considered irreversible, human existence and biological organization occupy the transient, highly structured intermediate zone.

精炼阅读摘要

Veritasium 的 Derek Muller 指出,“熵”是物理学中最常被误解的概念;它并非通俗语境下的“混乱度”,而是能量自发扩散的客观趋势。整个宇宙的演变、复杂生命的维系以及不可逆的时间之箭,均建立在宇宙初始罕见的低引力熵向高概率平衡态扩散的热力学梯度之上。 地球能量收支与低熵输入的本质 (0:00) 公众在街头访谈中普遍误认为地球会长期吸收并滞留大部分太阳能以驱动自然与文明进程。然而根据能量守恒定律,为了维持长期的行星温度稳定,地球辐射回太空的热量与吸收的太阳能量必须完全相等;若吸收多于释放,地球将持续变热直至彻底毁灭。地球真正从太阳获取的核心财富并非能量的总量,而是能量的“品质”或低熵状态。 在稳态下,太阳向地球注入的是高度集中的低熵高能光子。地球上的大气流动、洋流、光合作用以及动植物代谢,本质上都在利用这一高度集中的能量梯度。最终,地球每吸收 1 个高能太阳光子,就会向宇宙深空释放大约 20 个低能量、弥散的红外光子。生命与地球生态圈正是依靠将少量集中光子降解转化为 20 倍弥散光子所产生的耗散过程,才得以维持高度复杂的组织秩序。 热机极限与克劳修斯对熵的定义 (2:15) 19 世纪初,法国物理学家 Sadi Carnot 为了提升法国工业与军事实力,着手探寻热机效率的绝对理论极限。当时早期的蒸汽机热效率仅约为 3%。Carnot 构想了一种无摩擦、无散热损失的理想可逆热机,证明其最大热效率完全由高低温热源的绝对温度比值决定,公式为 η = 1 - (T_cold / T_hot);以 19 世纪蒸汽机在 160 °C(约 433 K)运行的工况估算,其理论极限效率仅为 32%。Lord Kelvin 随后据此创立了绝对温标,并指出只有当冷库温度降至绝对零度(0 K)或热源处于无限高温时,理想效率才可能达到 100%,而这在现实物理中皆不可能实现。 现实中的机器由于内部摩擦与不可避免的热传导,必然是不可逆的。高温热源中高度集中的能量一旦散逸至周围冷环境,便退化为无法重新聚合的微观随机振动。德国物理学家 Rudolf Clausius 于 1865 年提出“熵”这一概念以量化能量的耗散程度,并总结出热力学两大基石:宇宙的总能量保持恒定,而宇宙的总熵趋向极大值。能量总量虽不会减少,但其能够转化为有效机械功的做功能力却随熵增而永久衰减。 统计概率与时间之箭的不可逆性 (11:24) 微观物理规律在时间正向与反向上完全对称,但宏观热量却绝不会自发由低温传向高温。Ludwig Boltzmann 揭示了这背后的统计力学真相:热量反向流动并非被物理定律严格禁止,而是概率极低。通过左右各 8 个原子的玩具模型分析,在接触后系统出现热侧更热(9:1 能量包分布)的微观排列有 91,520 种,而两边均分(5:5 分布)的微观排列有 627,264 种;尽管 8 原子系统中热量倒流的概率仍有 10.5%,但当比例扩大到两侧各 80 个原子时,热侧变得更热的概率便骤降至 0.05%。 对于包含约 100 万亿亿(10^26)个原子的宏观真实物体而言,能量自发聚集逆流的概率趋近于零,因而在实际宇宙时间尺度内绝不可能发生。正如盲目旋转打乱的魔方只可能从唯一的还原态走向数以百亿亿计的混乱构型,宏观系统压倒性地由稀有的低熵构型演化为极其丰富的多微观态构型。这种由统计概率决定的不可逆性,在物理世界中塑造了单向流动的时间之箭。 耗散结构、黑洞熵与宇宙终极演化 (14:50) 局部的熵减与生命秩序并未违背热力学第二定律。家用空调在降低室内熵的同时,必须消耗由发电厂提供的电能,燃烧过程、输电线路及压缩机摩擦所产生的全局熵增,远远超过了室内降低的熵。物理学家 Jeremy England 提出的耗散适应理论表明,生命系统本质上是高效解开并耗散自由能的物理结构;在外部能量驱动下,富含蓝细菌与有机质的表层海水产生的熵比无生命的纯净海水高出 30% 到 680%,这表明生物自组织是宇宙最大化熵产生的有力载体。 追溯这股低熵流的源头,天体物理学面临着所谓的“过去假说”:大爆炸初期近乎完全均匀的炽热气体,由于引力具有自发聚集成团的特性,实际上构成了极低的引力引力熵状态。随后的引力坍缩释放了大量重力势能,转化为动能并最终耗散为恒星发光与黑洞。Jacob Bekenstein 与 Stephen Hawking 的研究证实黑洞具有正比于视界面积的巨大熵;当前可观测宇宙所有黑洞的熵总和约为 3 × 10^104 k_B,占据宇宙总熵的绝大部分,而大爆炸初期可观测宇宙的熵仅占当今宇宙的 3 × 10^-15%。 Muller 强调,完全的低熵与极端的最大熵实际上都缺乏结构复杂度,正如泾渭分明的茶与牛奶,或是彻底混合均质的液体都缺乏活力;唯有在系统从低熵迈向高熵的过渡阶段,才会孕育出精细的旋涡、复杂的生物圈与文明。在超过 10^100 年后,当宇宙中最后的黑洞完全通过霍金辐射蒸发殆尽,能量将被彻底均分且无法再提取任何有用功,宇宙将迎来不可挽回的热寂,宏观的时间之箭也将就此终结。

Detailed notes — zh

## Background 该视频由知名科普频道 Veritasium(主持人 Derek Muller)于 2023 年 7 月 1 日发布在 YouTube 平台。视频简介将其标榜为整个物理学中最重要却又最常被误解的概念之一,并附上了大量关于热力学、统计物理与生命起源的经典学术文献参考。 ## TL;DR 视频从“地球从太阳究竟获得了什么”这一街头反常识提问切入,指出地球净能量吸收基本为零,真正索取的是“低熵”。随后,节目回顾了卡诺理想热机与开尔文温标对效率极限的奠基,梳理了克劳修斯对熵作为“能量扩散倾向”的定义,以及玻尔兹曼通过微观统计排布揭示的时间单向性本质。最后,视频将视野拓展至引力主导的早期极低熵宇宙、吞噬海量熵的黑洞终局,并用茶融入牛奶的比喻解释了为何高度复杂的生命结构只在宇宙走向热寂的中间过渡期得以绽放。 ## Atmosphere & style 全片洋溢着深邃、引人入胜且充满哲理探索的叙事张力。Derek 从轻松幽默的街头快问快答起步,逐步引导观众进入精密严谨的历史思想实验与硬核微观概率模拟,最终将视角推向百亿年尺度的壮阔宇宙史诗。观看体验既像是一场拆解直觉陷阱的智力过山车,又带有一种在浩瀚熵增荒原中见证生命与秩序奇迹的敬畏感与诗意。 ## Bulletin - [00:00](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=0s) 地球从太阳获得的真正资产并非净能量,而是低熵流:地球向太空辐射的能量总量与接收的总量几乎相等,但每接收1个高能光子就会向外辐射约20个低能红外光子,正是这股低熵流驱动了地表的一切活动与生命演化。 - [06:22](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=382s) 卡诺热机理论揭示热机效率极限根本上仅取决于高低温源的绝对温度(η = 1 - T_cold / T_hot),哪怕消除所有机械摩擦和散热损失,任何现实热机也绝不可能达到100%的能量转换效率。 - [11:24](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=684s) 玻尔兹曼指出微观上热量从低温流向高温并非绝对不可能,只是极度不概率;宏观系统自发表现出的单向演化与时间之箭,纯粹源自系统从极低概率构型不可阻挡地滑向极大概率构型的统计必然性。 - [19:11](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=1151s) 早期宇宙在大爆炸后极其均匀且处于极致低熵状态,原因是引力聚团使均匀分布成为极低概率构型;现代宇宙绝大多数的熵已被锁在黑洞之中(总量达约 3 × 10^104 k_B)。 ## Key takeaways - [00:00](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=0s) 街头大众普遍误以为地球消耗了太阳输入的能量,但实际上地球处于热平衡状态,向太空辐射的辐射能量总量与接收量基本持平。 - [02:15](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=135s) 萨迪·卡诺构建了完全可逆且无摩擦耗散的理想热机模型,论证了热机做功必须依赖高温源向低温源转移热量。 - [06:22](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=382s) 开尔文勋爵基于卡诺热机构建了绝对温标,证明了热机理论效率极限 η = 1 - (T_cold / T_hot),且因无法实现绝对零度或无穷高温,100% 效率在物理上不可达成。 - [09:18](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=558s) 克劳修斯将衡量能量在系统中扩散程度的物理量定义为“熵”,并确立了热力学第二定律:孤立系统的熵永远趋向于极大值,能量的自发耗散不可逆转。 - [09:18](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=558s) 相比于将熵理解为模糊的“混乱度”或“无序度”,更严谨的物理直觉是将其视为能量向外自发扩散分散的内在倾向。 - [11:24](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=684s) 玻尔兹曼的统计力学揭示,微观状态的可能构型数量悬殊导致宏观熵增;在微观尺度下局部逆向传热概率客观存在,但在包含 10^26 个原子的宏观固体中逆向概率趋近于零。 - [14:50](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=890s) 空调等局部制冷设备产生的局部熵减并未违背物理法则,其运转过程中在发电厂、输电线和机械压缩机处产生的高额熵增远超室内的熵减总量。 - [14:50](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=890s) 生命不仅没有违背热力学第二定律,反而扮演了高效耗散低熵的催化机器;英格兰的耗散驱动自组织理论提出物质在持续能量流冲刷下会自发演化出利于散热的复杂结构。 - [19:11](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=1151s) 在引入长程吸引力引力后,早期宇宙炽热均匀的分布实际上对应着极其罕见的低熵状态,引力坍缩点燃恒星与星系的过程释放了海量引力势能并导致宇宙熵剧增。 - [21:36](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=1296s) 贝肯斯坦提出黑洞具有与视界面积成正比的熵,霍金通过推导霍金辐射证实黑洞具有微弱温度与极高熵值,现今宇宙几乎所有熵(3 × 10^104 k_B)都集中在黑洞内部。 - [23:31](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=1411s) 宇宙宏观时间之箭的存在完全由大爆炸的极低初始熵(过去假说)决定,直到超过 10^100 年后最后一颗黑洞蒸发陷入宇宙热寂,时间之箭才会彻底消亡。 - [24:52](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=1492s) 系统的复杂度与熵并非线性反比,而是呈现“倒 U 型”曲线:正如牛奶刚倒入茶水时的复杂涡流,唯有在低熵走向高熵的中间过渡阶段,复杂结构与生命才能繁盛生长。 ## Detailed walkthrough ### [00:00](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=0s) 地球究竟从太阳得到了什么? 物理学中有一个极其重要却又最常被误解的概念,它支配着从分子碰撞到狂暴飓风的一切现象,横跨宇宙自创生、演化直至终局的漫长历程,甚至可能决定了时间箭头的方向以及生命诞生的根源。然而,公众对这一概念的普遍认知却充满盲区。面对“地球从太阳得到了什么”这样一个简单的问题,街头受访者的第一反应往往是光线、温暖、热量、紫外线合成的维生素 D,或者最直截了当的答案——能量。 ```mermaid flowchart LR accTitle: 地球与太阳的能量进出平衡 accDescr: 展示太阳光照射地球后,辐射能量进出相等的平衡关系。 Sun["太阳辐射能量 (光照)"] -->|"能量流入"| Earth["地球系统 (处于热平衡)"] Earth -->|"能量辐射回太空 (红外长波)"| Space["外太空 (能量收支平衡)"] ``` 然而,当被进一步追问“地球向太空辐射回去的能量占太阳输入能量的多少”时,大多数人都会陷入直觉陷阱。有人猜测辐射回去的能量“明显更少”,估算仅为输入量的 70% 甚至 20%,理由是人类或地球生命“消耗利用了一部分能量”。这种直觉违背了物理规律:能量既不能被创造也不能被消灭,更无法被真正“用光”。事实上,在地球漫长的历史中,地球向太空辐射的能量总量与从太阳接收到的能量几乎完全相等。正如主持人向受访者解释的那样:“因为如果我们不这样做,地球就会变得热得多,那将是一个大麻烦。”既然净能量流入基本为零,那么地球从太阳那里真正获得的、不可或缺的独特资产,显然并非单纯的“能量”本身。 - 街头随机受访者对能量辐射比例的直觉猜测多在 20% 至 70% 之间,普遍误以为能量在地球内部被生物或工业“消耗掉”了。 - 若地球长期吸收大于辐射,热量持续累积将导致地表急剧升温而无法维持宜居环境。 - 解开这一悖论的关键,必须回溯到两个世纪前工业革命时期的一项重大物理发现。 --- ### [02:15](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=135s) 卡诺与理想热机的诞生 1813 年严冬,拿破仑帝国面临奥地利、普鲁士和俄国联军的入侵。拿破仑麾下将军之子、年仅 17 岁的萨迪·卡诺(Sadi Carnot)写信给拿破仑请求参军保卫祖国,但拿破仑因战事焦头烂额未予回复。数月后巴黎遇袭,卡诺与同学们在城东一座城堡参与防守,却难敌联军攻势,巴黎仅抵抗一天便告陷落。战败迫使卡诺撤退,年轻的他遭受了巨大打击。 拿破仑倒台七年后,卡诺前往普鲁士探望流亡的父亲。其父不仅是军界将领,也是一名物理学家,曾撰写过关于机械系统中最有效能量传递方式的论著。父子二人深入探讨了当时最前沿的技术突破——蒸汽机。蒸汽机在当时已广泛用于船舶驱动、矿石开采和港口疏浚,国力的工业与军事竞争很大程度上取决于蒸汽机的技术水准。然而法国的蒸汽机设计明显落后于英国,卡诺决心找出背后的核心原因。当时顶尖的蒸汽机也只能将约 3% 的热能转化为有用机械功。卡诺意识到,如果能大幅提升这一效率,便能赋予法国巨大优势,恢复其大国地位。 ```mermaid flowchart TD accTitle: 卡诺理想热机循环工作流程 accDescr: 描绘卡诺热机在热源、绝热膨胀、冷源和绝热压缩四个冲程之间的循环。 subgraph Expansion["膨胀做功阶段"] A["1. 气缸接触高温热源:等温吸热,气体膨胀推动活塞做功"] B["2. 移开高温热源:气体绝热膨胀,温度降至冷源温度"] end subgraph Compression["压缩复位阶段"] C["3. 气缸接触低温冷源:飞轮带动活塞等温压缩,废热排入冷源"] D["4. 移开低温冷源:气体绝热压缩,温度回升至接近热源温度"] end A --> B B --> C C --> D D -->|"循环往复"| A ``` 卡诺潜心研究热机三年,提出了完全消除摩擦力和环境热散失的“理想热机”构想: - **热源与气缸构造**:配备一个高温金属块和一个低温金属块;气缸内部充满空气,仅底部允许热量进出,内部活塞连接至一个储能飞轮。 - **等温膨胀**:气体初始温度略低于高温块。高温块接触气缸底部,热量流入,气体在恒温下膨胀并推动活塞向上做功,驱动飞轮旋转。 - **绝热膨胀**:移开高温块,气体在无外界热量注入的情况下继续膨胀降温,直至温度降至与低温块一致。 - **等温压缩**:将低温块与气缸底部接触,借助飞轮的惯性带动活塞向下压缩气体,在此过程中多余热量被排入低温块。 - **绝热压缩**:移开低温块,飞轮继续压缩气体使其进一步升温,直至温度恢复到略低于高温块的初始状态,随即重新接入高温块开启下一个循环。 这一循环的关键特性在于其完全可逆(Reversible)。若将飞轮逆向旋转,气体会先膨胀降温,接触冷源吸收热量,再被压缩升温,并在接触热源时借助飞轮能量将热量反向泵回高温块。无论正向运行多少个循环,只需以相同周期反向运行,整个系统与外界环境就能毫无残留地恢复至原初状态,无需额外能量输入。“因此,运行理想热机并没有真正改变什么,你总能撤销所做的一切。” --- ### [06:22](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=382s) 开尔文绝对温标与热机效率极限 由于卡诺理想热机完全可逆且没有任何耗散,直觉上可能认为其效率能达到 100%,但现实并非如此。在每一次做功循环中,飞轮获得的净能量等于从高温端流入的气体吸热量减去流向低温端的放热量。计算热机效率,即是用这部分转化为机械功的净能量除以来自高温端吸入的总热量。 高温侧气体推动活塞所做的功,必然大于冷端活塞压缩气体时外界所做的功,因为相同质量的气体在高温时对活塞施加的压强远大于低温时的压强。因此,要提高理想热机的效率,手段只有两种:提高高温端的温度,或者降低低温端的温度(或两者并行)。 威廉·汤姆森(开尔文勋爵)在了解卡诺理想热机后意识到,这套理论为建立绝对温标奠定了基石。设想气体膨胀到极限程度,剧烈冷却直至所有气体粒子完全停止运动,此时粒子对活塞的压强为零,在冷端进行压缩时完全不需要做功,也就不会向外耗散热量。这一状态即为“绝对零度”(Absolute Zero),在此极限下热机效率将达到 100%。 ```mermaid flowchart LR accTitle: 卡诺热机效率计算逻辑 accDescr: 说明理想热机效率公式由吸放热量比转化为基于开尔文绝对温标的温差比。 Qin["高温端吸热量 (Q_in) ∝ T_hot"] Qout["低温端放热量 (Q_out) ∝ T_cold"] Work["净机械功 (W = Q_in - Q_out)"] Qin --> Work Qout --> Work Work --> Efficiency["理想效率 η = 1 - (T_cold / T_hot)"] ``` 借助开尔文绝对温标,吸热量与放热量可直接用高温侧绝对温度 $T_{\text{hot}}$ 与低温侧绝对温度 $T_{\text{cold}}$ 替换,热机效率由此表达为: $$\eta = 1 - \frac{T_{\text{cold}}}{T_{\text{hot}}}$$ 这揭示了一个颠覆性的物理事实:“理想热机的效率并不取决于发动机的材质或机械构造,而是从根本上取决于高温端和低温端的温度。”要实现 100% 的效率,要么在高温端拥有无限高的温度,要么在低温端达到绝对零度,而这两者在现实物理世界中都是无法实现的。哪怕排除了所有的机械摩擦阻力和对外部环境的散热损失,任何热机也不可能达到 100% 的转换效率,原因在于活塞若想返回初始位置重新做功,就必须向冷源排出热量,绝不可能将吸入的热能全额锁留在飞轮中。 - 卡诺所处时代的高压蒸汽机运行温度最高仅能达到约 160 ℃。 - 依据卡诺公式换算,当时蒸汽机的理论极限效率仅约为 32%,而实际测量效率更是低至 3% 左右。 - 实际效率与理论极限差距悬殊的原因在于不可逆损失:机械摩擦生热、向周围环境散热,以及非恒温热传递。 --- ### [09:18](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=558s) 克劳修斯、能量耗散与熵的定义 在实际运行的真实热机中,同样的输入热量最终转化为飞轮动能的比例要低得多。其余的能量并没有凭空消失,而是耗散散落到了气缸壁上、飞轮的转轴轴承摩擦处,并以辐射形式向周围环境扩散。这种能量的扩散过程是单向不可逆的:能量一旦分散开来,就再也无法自发收集重拢。“总能量虽然保持不变,但它的可用性降低了。当能量高度集中时,它是最可用的;而当能量分散开来时,它就变得不再那么可用了。” 几十年后,德国物理学家鲁道夫·克劳修斯(Rudolf Clausius)深入钻研了卡诺的研究成果。他创造性地提出了一种用于衡量“能量在系统中扩散分散程度”的物理量,并将其命名为——**熵(Entropy)**。 | 状态维度 | 低熵状态 (Low Entropy) | 高熵状态 (High Entropy) | |:--- |:--- |:--- | | **能量分布特征** | 能量高度集中(如聚焦于卡诺循环的高温金属块) | 能量广泛分散(如耗散至气缸壁、转轴与周围空气中) | | **对外做功能力** | 能量具备高可用性,能高效转化为机械外功 | 能量分散为微观杂乱热运动,对外做功能力严重衰减 | | **系统演化方向** | 自然孤立演化的起点形态 | 孤立系统经历不可逆过程后自发趋向的最终状态 | 1865 年,克劳修斯对热力学第一定律与第二定律做出了流传百世的终极概括: 1. **热力学第一定律**:“宇宙的能量是恒定的。”(First, the energy of the universe is constant.) 2. **热力学第二定律**:“宇宙的熵趋向于极大值。”(And second, the entropy of the universe tends to a maximum.) 简而言之,热力学第二定律的精髓就是“能量随着时间的推移必然会自发地扩散开来”。第二定律构成了世间大量不可逆自然现象的底层支柱:它是热水变凉、冷饮升温的根本原因;它解释了为何容器中的气体总是自发膨胀以充满整个空间;它也宣判了永动机在物理上的绝对死刑,因为闭合系统中的可用能量始终在不可逆转地衰减减少。 尽管在大多数物理科普与教科书中,人们最习惯将熵描述为“无序度”(Disorder)——这与事物混合、随机化及秩序丧失的表象高度吻合——但要深刻理解物理世界的运作机理,主持人指出:“我认为理解熵的最佳方式,是将其视为能量向外扩散的内在倾向。”这就引出了一个更为根本的物理学谜题:为什么能量随着时间的推移一定会向外扩散?要知道,大多数微观层面的物理定律本身,其实都具有完全的时间反演对称性。 ### [11:24](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=684s) 玻尔兹曼的统计洞见:热量从冷到热并非不可能,只是极度不概率 物理定律在微观层面上不论时间向前还是向后运行都是完全相同的,那这种明确的时间单向性(时间之箭)究竟从何而来?为了直观展示这一点,可以设想两根微小的金属棒:一根较热,一根较冷。在这个简化模型中,假设每根棒各只有 8 个原子。每个原子根据自身拥有的能量包(energy packets)数量振动,能量包越多,振动越剧烈。初始状态下,左边的热棒拥有 7 个能量包,右边的冷棒拥有 3 个能量包。棒内能量包的数量即代表系统的“状态”。 在单独观察左棒时,7 个能量包在晶格内永不停歇地在各个原子间随机跳跃,不断形成不同的微观能量排布(configurations),而总能量始终保持不变。当冷棒(带 3 个能量包)与其接触时,能量包开始在两根棒之间自由穿梭,每个独特的微观构型出现的概率完全均等。如果此时随机拍一张系统快照,甚至可能抓拍到极其违背直觉的瞬间:左棒聚集了 9 个能量包,而右棒只剩 1 个。这意味着热量居然从冷物体流向了热物体,表面上看似熵减了。 物理学家路德维希·玻尔兹曼(Ludwig Boltzmann)正是由此做出了突破性洞见:“热量从冷流向热并不是不可能的,它只是极度不可能(Heat flowing from cold to hot is not impossible, it's just improbable.)”。通过统计组合数可以发现: - 左棒拥有 9 个能量包的微观构型数为 91,520 种; - 能量完全均分(两棒各 5 个能量包)的微观构型数则高达 627,264 种(能量均分的概率是前者的 6 倍以上)。 即便如此,在 16 个原子的极小尺度下,左棒最终能量包比初始状态更多的概率仍高达 10.5%。然而,一旦将尺度扩展到每棒 80 个原子、共 100 个能量包(左 70、右 30),左棒最终变得更热的概率便断崖式下跌至仅 0.05%。在现实日常固体中,原子数高达约 $10^{26}$(100 万亿亿)个,伴随更多的能量包,热量自发逆向流动的概率趋近于绝对零。 魔方也是极佳的类比:一个完全复原的魔方如果闭上眼睛随意转动,必然会越来越混乱。因为复原状态只有 1 种,接近复原的状态也极少,而彻底混乱的随机状态有数十百亿亿(quintillions)种。没有任何思考和外力干预,每一次随意的转动都必然驱使系统从罕见的有序态滑向极大概率的混乱态。 --- ### [14:50](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=890s) 空调悖论与耗散结构:地球为何没有陷入均一死亡 如果宇宙的自然倾向是能量扩散、事物变乱,那人类是如何制造出空调的?空调使本来就凉爽的室内变得更冷,使炎热的室外变得更热,室内熵明显减少了。这种局部熵减绝未违背热力学第二定律,因为局部的熵减必须以其他地方产生更大规模的熵增为代价。从发电厂燃烧煤炭释放化学能、加热环境与发电机、经过高压电线产生电阻热,到空调风机与压缩机的摩擦废热,为了让室内减少一点熵,整个环境付出了多得多的熵增代价。 这也引出了更核心的生命之谜:如果整个宇宙的熵永远在增加,地球为什么能维持复杂的结构、保留冷热差异,甚至孕育繁衍出生命?答案在于:地球不是一个孤立系统,地球拥有太阳。 ```mermaid flowchart TD accTitle: Earth Low Entropy Engine accDescr: Sun provides concentrated low-entropy energy that drives Earth processes and reradiates as dispersed high-entropy heat. subgraph SpaceInput["来自太空的输入"] Sun["太阳高能光子 (低熵)"] end subgraph EarthSystem["地球开放系统"] Life["光合作用与食物链转化"] Activity["地质演变、气象与生命活动"] end subgraph SpaceOutput["排向太空的废热"] Waste["20倍数量的低能红外光子 (高熵)"] end Sun --> Life Life --> Activity Activity --> Waste ``` 太阳赋予地球的本质上是一股稳定的“低熵流”——即高度聚集、致密的能量。地球接收到的能量在总量上与辐射回太空的热辐射能量完全相等(正如视频中 Derek 作为物理学博士向采访对象特意强调并确认的事实)。区别在于熵的形式:地球每接收 1 个来自太阳的高能光子,就会向深空辐射出 20 个低能量的光子。从植物生长、树木倒塌、兽群奔腾,到飓风咆哮以及人类的呼吸睡眠,地球上的一切活动,本质上都是将少量的聚集高能光子转化为 20 倍数量的耗散低能光子的过程。 生命不仅没有违背热力学第二定律,甚至很可能是该定律的直接产物。因为生命是宇宙中极其高效的“将低熵转化为高熵的机器”。例如,含有蓝细菌和其他有机物的海水表层,其产熵效率比无生物的海水高出 30% 到 680%。物理学家杰里米·英格兰(Jeremy England)进一步提出“耗散驱动的自组织理论”,认为在持续聚集的能量流冲刷下,物质会自发演化出更善于耗散能量的结构,最终孕育生命。正如他精辟指出的那样: > “你从一团随意的原子开始,如果用光照它足够长的时间,产生一株植物不应该是什么令人惊讶的事。”(“You start with a random clump of atoms, and if you shine light on it for long enough, it should not be so surprising that you get a plant.”) --- ### [19:11](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=1151s) 引力、大爆炸与过去假说:早期宇宙为何是低熵的 既然地球依赖太阳的低熵生存,太阳的低熵又从何而来?答案必须追溯至宇宙大爆炸本身。宇宙总熵随时间递增,意味着昨天比今天熵低,最终推导出一个必然结论:宇宙在大爆炸刚发生时处于熵最低的状态。这一前提在宇宙学中被称为“过去假说”(past hypothesis)。它并不解释为何初始熵如此之低,只说明要形成今天观测到的宇宙,初始状态必须如此。 但这带来了一个巨大的反常识悖论:早期宇宙炽热、致密且极其均匀,各处的温度差异最多不超过 0.001%,宛如一锅搅拌均匀的热汤。这种“均匀混合”在普通热力学中恰恰是最大熵的特征,为何反而被定义为极度低熵? 关键在于此前忽略了引力的存在: 1. **引力的聚团本性**:引力是一种只具有吸引力的长程力,它的自然演化方向是将弥散的物质拉扯聚集在一起。 2. **均匀态是极低概率态**:在包含引力的系统中,物质漫无目的地均匀平铺反而是最不可思议、概率极低的构型(即引力低熵)。 3. **势能向动能与热能的转化**:随着宇宙膨胀冷却,物质在引力作用下向密集区域坍缩,海量引力势能转化为动能,随后在粒子撞击中转化为无序热能,导致有效能量耗散、熵大幅攀升,进而逐步点燃恒星、塑造星系并孕育生命。 - 宇宙诞生之初的熵大约只有 $10^{88} k_B$(玻尔兹曼常数); - 当前可观测宇宙中所有恒星的熵约为 $9.5 \times 10^{80} k_B$; - 星际介质与星系际介质的熵约为恒星的近 10 倍; - 中微子和宇宙微波背景辐射(CMB)的光子储存了更高的熵。 --- ### [21:36](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=1296s) 黑洞:宇宙绝大部分熵的终极藏身地 1972 年,雅各布·贝肯斯坦(Jacob Bekenstein)提出了一个颠覆性的假说:黑洞也具有熵,而且黑洞的熵与其事件视界的表面积成正比,随着黑洞吞噬物质体积增大,其熵必然增加。当时主流物理学家大多认为这是天方夜谭,因为根据经典热力学,如果黑洞具有熵,就必须拥有温度;而有温度的物体必然对外辐射,这与黑洞“只进不出、全黑”的定义完全相悖。 史蒂芬·霍金(Stephen Hawking)起初着手计算是为了证伪贝肯斯坦的理论,然而出乎他意料的是,量子场论与广义相对论的结合推导显示:黑洞确实具有温度,并且确实会向外缓慢释放热辐射(即霍金辐射)。以银河系中心的超大质量黑洞人马座 A* 为例,其温度大约只有 100 万亿分之一开尔文($10^{-14}\text{ K}$),发出的辐射极其微弱以至于完全无法探测,因此在观测上依然“漆黑一片”。 霍金不仅证实了贝肯斯坦是正确的,还精确计算出了黑洞熵的数值。其结果令人震惊: - 仅银河系中心的超大质量黑洞,其熵就高达约 $10^{91} k_B$; - 单单这一个黑洞的熵,就是早期可观测宇宙总熵的 1,000 倍,且是宇宙中所有其他常规粒子加起来的 10 倍以上; - 全宇宙所有黑洞汇总起来,所占据的熵高达 $3 \times 10^{104} k_B$。 这意味着,宇宙中绝大部分的熵实际上都被锁在了黑洞之中。反观早期宇宙,其拥有的熵仅相当于现代宇宙的 0.000000000000003%(约为 $10^{-14}\%$ 甚至更少),正因为起点处于如此不可思议的极致低熵状态,宇宙才拥有了长达百亿年演化出恒星、星系与生命繁荣的宏大舞台。 ### [23:31](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=1411s) 低熵的起源、时间之箭与宇宙的热寂 宇宙能够发生的一切宏大与微小的事件——无论是行星系统的形成、星系的并合、小行星的撞击、恒星的死亡,还是生命本身的繁衍演化——全都有赖于一个根本的前提:宇宙在初始状态下的熵非常低,并且一直在持续增加。这一过程具有绝对的单向性。在现实世界中,我们绝不可能看到小行星撞击后自动复原并退回太空,也绝不可能看到行星系统解体逆向变回最初由尘埃和气体构成的星云。过去与未来之间存在着泾渭分明的区别,而这一界限完全是由熵来界定的。正如德里克·穆勒(Derek Muller)所言:“The fact that we are going from unlikely to more likely states is why there is an arrow of time.”(我们正在从极不可能的状态走向更可能的状态,这正是时间之箭存在的原因。) 这种熵增的过程预计将一直持续下去,直到宇宙中的能量被彻底且极其均匀地分散开来,届时将不再有任何有趣的事情发生,这便是所谓的“宇宙热寂”(heat death of the universe)。在极为遥远的未来——距今超过 $10^{100}$ 年之后,当最后一颗黑洞也彻底蒸发殆尽时,宇宙将步入其最可能出现的高熵状态。到了那个阶段,即便在极其宏观的尺度上,人们也无法分辨时间是在向前流逝还是向后倒退,时间之箭本身将彻底消失。 ```mermaid flowchart LR accTitle: 宇宙从低熵演化至热寂的过程 accDescr: 展示宇宙从大爆炸初期的极低熵状态,演变到当前的复杂结构活跃期,最终走向时间之箭消失的热寂终局。 subgraph Past["过去 (早期宇宙)"] A["极低熵状态<br/>(高度不可能发生的微观构型)"] end subgraph Present["现在 (结构与演化)"] B["熵持续单向增加<br/>(产生明确的时间之箭)"] C["恒星、星系并合、行星形成与生命繁衍"] end subgraph Future["遥远未来 (>10^100 年)"] D["最后黑洞蒸发完毕"] E["热寂状态<br/>(最可能状态,时间之箭消失)"] end A --> B B --> C C --> D D --> E ``` - **不可逆的物理过程**:行星系统不会倒退分离为气体与尘埃云,小行星撞击绝不会逆向“未撞击化”,宏观历史只朝一个方向推进。 - **时间之箭的本质**:宏观时间单向流动的直觉,来源于物理系统从低概率宏观态向高概率宏观态的自发跃迁。 - **热寂的时间跨度**:超过 $10^{100}$ 年后,黑洞蒸发殆尽,能量分布完全平坦化,宏观状态不再发生任何变化。 --- ### [24:52](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=1492s) 复杂度与熵的倒 U 型关系:茶与牛奶的比喻 虽然听上去熵增是一个不可逆转的诅咒,正不可避免地将整个宇宙推向能想象到的最枯燥结局,但这并不代表熵与复杂度成简单的反比。最大熵状态固然意味着极低的复杂度(完全均匀单调),但这绝不意味着低熵状态就具有最高的复杂度。 德里克·穆勒用向茶水中倒入牛奶的过程做了一个极为生动的实验与比喻:分别端着一杯未混合的纯茶和一小罐纯牛奶,这种状态尽管处于低熵,但其实毫无生趣;而当把牛奶倒入茶中时,两者开始剧烈混合,交界面上瞬间涌现出绚丽且精细的涡流与复杂图案。然而,这些美丽的花纹稍纵即逝,很快牛奶和茶水就彻底混为一体,重新变回毫无特征的单调液体。德里克·穆勒总结道:“Both low and high entropy are low in complexity. It's in the middle where complex structures appear and thrive.”(低熵和高熵的复杂度都很低。唯有在二者之间的过渡阶段,复杂的结构才会涌现并蓬勃发展。)既然人类文明恰好置身于这个熵增的中间过渡期,我们理应充分利用身边现有的低熵资源,创造属于我们的精彩。 | 演化阶段 | 对应状态(以茶与牛奶为例) | 熵水平 | 复杂度特征 | |:--- |:--- |:--- |:--- | | **初始低熵期** | 未混合的纯茶与纯牛奶分离放置 | 极低 | 结构单调且乏味,缺乏交织的宏观结构 | | **混合过渡期** | 牛奶倒入茶中,产生丰富翻滚的涡流纹理 | 中等 | **复杂度达到巅峰**,精致复杂结构(如星系、生命)蓬勃涌现 | | **高熵平衡期** | 牛奶与茶水彻底混合均匀,成为均质奶茶 | 极大 | 失去一切动态特征与分层,趋于死寂 | --- ### [25:47](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=1547s) 赞助商介绍与结语:Brilliant.org 的思维训练 在节目的结尾,德里克·穆勒强调,只要掌握了正确的思维工具,我们就能理解世间的一切事物——从一杯放凉的茶水,到整个宇宙的演化终局。如果你想为自己的思维武器库增添强大的工具,赞助商 Brilliant.org 便是一个绝佳的学习平台。在 Brilliant 上,学习者可以系统掌握从数学、数据科学到编程以及物理学的核心概念。用户只需设定自己的学习目标,平台便会量身定制一条学习路径,提供达成目标所需的一切知识与工具。 针对想要学习像程序员一样思考的学习者,Brilliant 推出了最新课程《Thinking in Code》(代码思维)。该课程采用直观的拖拽式编辑器,讲授代码嵌套(nesting)和条件分支(conditionals)等核心逻辑。学习者可以从直观的机器人编程任务入手,进而将这些技能应用到日常生活中,比如在手机上自动设置提醒,或者编写自动化机器人来筛选约会软件上的匹配对象。德里克·穆勒表示,Brilliant 最打动他的一点是它能将抽象知识与真实世界紧密相连,并通过动手实践培养扎实的直觉。他在片尾向观众致谢,并邀请大家通过专属链接获取为期 30 天的免费试用及年度订阅 20% 的优惠。 - **课程特色与实用案例**:课程通过直观的可视化拖拽降低学习门槛,涵盖自动化手机提醒与构建社交筛选脚本等实际应用。 - **优惠信息**:观众可通过访问 `brilliant.org/veritasium` 获得 30 天免费完整试用,前 200 名注册用户可享受 Brilliant 年度高级订阅(Annual Premium Subscription)20% 的折扣。 ## Notable quotes > [09:18](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=558s) 我认为理解熵的最佳方式,是将其视为能量向外扩散的内在倾向。 — Derek Muller > [23:31](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=1411s) The fact that we are going from unlikely to more likely states is why there is an arrow of time. — Derek Muller > [24:52](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=1492s) Both low and high entropy are low in complexity. It's in the middle where complex structures appear and thrive. — Derek Muller > [14:50](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=890s) You start with a random clump of atoms, and if you shine light on it for long enough, it should not be so surprising that you get a plant. — Jeremy England > [11:24](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=684s) Heat flowing from cold to hot is not impossible, it's just improbable. — Ludwig Boltzmann > [09:18](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=558s) 总能量虽然保持不变,但它的可用性降低了。当能量高度集中时,它是最可用的;而当能量分散开来时,它就变得不再那么可用了。 — Derek Muller > [06:22](https://www.youtube.com/watch?v=DxL2HoqLbyA&t=382s) 理想热机的效率并不取决于发动机的材质或机械构造,而是从根本上取决于高温端和低温端的温度。 — Derek Muller ## Mentioned Derek Muller, Sadi Carnot, William Thomson (Lord Kelvin), Rudolf Clausius, Ludwig Boltzmann, Jacob Bekenstein, Stephen Hawking, Jeremy England, Brilliant.org, Elements of Truth, Snatoms, 人马座 A* ## Video description One of the most important, yet least understood, concepts in all of physics. Head to https://brilliant.org/veritasium to start your free 30-day trial, and the first 200 people get 20% off an annual premium subscription. In 2025, we launched the Elements of Truth game on Kickstarter and were thrilled by the response. We’re now in the final stages of production, with Kickstarter rewards shipping soon. If you missed the campaign, you can now pre-order the retail edition through our new website here - https://ve42.co/EntropyDes If you're looking for a molecular modeling kit, try Snatoms - a kit I invented where the atoms snap together magnetically: https://snatoms.com We have a tabletop game launching! Pledge your support now - https://ve42.co/KickStarterD ▀▀▀ A huge thank you to those who helped us understand different aspects of this complicated topic - Dr. Ashmeet Singh, Supriya Krishnamurthy, Dr. Jos Thijssen, Dr. Bijoy Bera, Dr. Timon Idema, Álvaro Bermejillo Seco and Dr. Misha Titov. ▀▀▀ References: Carnot, S. (1824). Reflections on the motive power of heat: and on machines fitted to develop that power. - https://ve42.co/Carnot1890 Harnessing The True Power Of Atoms | Order And Disorder Documentaries, Spark via YouTube - https://ve42.co/OrderDisorder A better description of entropy, Steve Mould via YouTube - https://ve42.co/Mould2016 Dugdale, J. S. (1996). Entropy and its physical meaning. CRC Press. - https://ve42.co/Dugdale1996 Schroeder, D. V. (1999). An introduction to thermal physics. - https://ve42.co/Schroeder2021 Fowler, M. Heat Engines: the Carnot Cycle, University of Virginia. - https://ve42.co/Fowler2023 Chandler, D.L. (2010). Explained: The Carnot Limit, MIT News - https://ve42.co/Chandler2010 Entropy, Wikipedia - https://ve42.co/EntropyWiki Clausius, R. (1867). The mechanical theory of heat. Van Voorst. - https://ve42.co/Clausius1867 What is entropy? TED-Ed via YouTube - https://ve42.co/Phillips2017 Thijssen, J. (2018) Lecture Notes Statistical Physics, TU Delft. Schneider, E. D., & Kay, J. J. (1994). Life as a manifestation of the second law of thermodynamics. Mathematical and computer modelling, 19(6-8), 25-48. - https://ve42.co/Schneider1994 Lineweaver, C. H., & Egan, C. A. (2008). Life, gravity and the second law of thermodynamics. Physics of Life Reviews, 5(4), 225-242. - https://ve42.co/Lineweaver2008 Michaelian, K. (2012). HESS Opinions" Biological catalysis of the hydrological cycle: life's thermodynamic function". Hydrology and Earth System Sciences, 16(8), 2629-2645. - https://ve42.co/Michaelian2012 England, J. L. (2013). Statistical physics of self-replication. The Journal of chemical physics, 139(12), 09B623_1. - https://ve42.co/England2013 England, J. L. (2015). Dissipative adaptation in driven self-assembly. Nature nanotechnology, 10(11), 919-923. - https://ve42.co/England2015 Wolchover, N. (2014). A New Physics Theory of Life, Quantamagazine - https://ve42.co/Wolchover2014 Additional References: https://ve42.co/EntropyRefs ▀▀▀ Special thanks to our Patreon supporters: Emil Abu Milad, Tj Steyn, meg noah, Bernard McGee, KeyWestr, Amadeo Bee, TTST, Balkrishna Heroor, John H. Austin, Jr., john kiehl, Anton Ragin, Diffbot, Gnare, Dave Kircher, Burt Humburg, Blake Byers, Evgeny Skvortsov, Meekay, Bill Linder, Paul Peijzel, Josh Hibschman, Mac Malkawi, Juan Benet, Ubiquity Ventures, Richard Sundvall, Lee Redden, Stephen Wilcox, Marinus Kuivenhoven, Michael Krugman, Sam Lutfi. ▀▀▀ Written by Casper Mebius, Derek Muller & Petr Lebedev Edited by Trenton Oliver & Jamie MacLeod Animated by Mike Radjabov, Ivy Tello, Fabio Albertelli and Jakub Misiek Filmed by Derek Muller, Albert Leung & Raquel Nuno Molecular collisions video by CSIRO's Data61 via YouTube: Simulation of air Additional video/photos supplied by Getty Images, Pond5 and by courtesy of NASA, NASA's Goddard Space Flight Center, NASA Goddard Flight Lab/ CI Lab, NASA/SDO and the AIA, EVE, HMI, and WMAP science teams. As well as the Advanced Visualization Laboratory at the National Center for Supercomputing Applications, B. Robertson, L. Hernquist Music from Epidemic Sound & Jonny Hyman Produced by Derek Muller, Petr Lebedev, Emily Zhang, & Casper Mebius [Source](https://www.youtube.com/watch?v=DxL2HoqLbyA)

Source quotes

[9:18] Derek Muller: 我认为理解熵的最佳方式,是将其视为能量向外扩散的内在倾向。 [23:31] Derek Muller: The fact that we are going from unlikely to more likely states is why there is an arrow of time. [24:52] Derek Muller: Both low and high entropy are low in complexity. It's in the middle where complex structures appear and thrive. [14:50] Jeremy England: You start with a random clump of atoms, and if you shine light on it for long enough, it should not be so surprising that you get a plant. [11:24] Ludwig Boltzmann: Heat flowing from cold to hot is not impossible, it's just improbable. [9:18] Derek Muller: 总能量虽然保持不变,但它的可用性降低了。当能量高度集中时,它是最可用的;而当能量分散开来时,它就变得不再那么可用了。 [6:22] Derek Muller: 理想热机的效率并不取决于发动机的材质或机械构造,而是从根本上取决于高温端和低温端的温度。

Source quotes

[11:24] Ludwig Boltzmann: Heat flowing from cold to hot is not impossible, it’s just improbable. [18:10] Jeremy England: You start with a random clump of atoms, and if you shine light on it for long enough, it should not be so surprising that you get a plant. [25:00] Derek Muller: Both low and high entropy are low in complexity. It's in the middle where complex structures appear and thrive. [23:31] Derek Muller: The fact that we are going from unlikely to more likely states is why there is an arrow of time. [9:05] Derek Muller: Energy is most usable when it is concentrated and less usable when it's spread out. [6:22] Derek Muller: What we have learned is that the efficiency of an ideal heat engine doesn't depend on the materials or the design of the engine, but fundamentally on the temperatures of the hot and cold sides.

Original transcript

[0:00] - This is a video about one of the most important, [0:02] yet least understood concepts in all of physics. [0:07] It governs everything from molecular collisions [0:09] to humongous storms. [0:12] From the beginning of the universe [0:14] through its entire evolution, to its inevitable end. [0:19] It may, in fact, determine the direction of time [0:22] and even be the reason that life exists. [0:25] To see the confusion around this topic, [0:27] you need to ask only one simple question. [0:31] What does the Earth get from the sun? [0:35] - What does the earth get from sun? [0:39] - Well, it's light rays? [0:41] - What do we get from the sun? - Heat. [0:43] - Warmth. [0:43] - Warmth, light. [0:45] - Vitamin D, we get vitamin D from- [0:47] - We do get vitamin D from the ultraviolet rays. [0:49] - Well, a lot of energy. [0:51] - What does the earth get from this, energy? [0:53] - Yeah, energy. [0:54] - Energy. [0:56] - Nailed it. [0:57] Every day, the earth gets a certain amount of energy [0:59] from the sun. [1:00] And then how much energy does the earth radiate back [1:04] into space relative to that amount [1:06] that it gets from the sun? [1:07] - Probably not as much, I, you know, [1:09] I don't believe it's just radiating right back. [1:12] - I'd say less. [1:14] - Less. [1:14] - Less. - I say less. [1:16] - I guess about 70%? [1:17] - It is a fraction. [1:18] - I'd say 20%. [1:19] - Because... - Because we use some of it. [1:22] - We use some of the energy. - Mm-hmm. [1:23] - We consume a lot, right? [1:24] - But the thing about energy is it never really goes away. [1:28] You can't really use it up. [1:30] - It would have to break even, wouldn't it? [1:32] Same amount, yeah. [1:33] - You know, cause and effect. [1:35] It'd be equal in some ways, right? [1:37] - For most of the earth's history, [1:39] it should be exactly the same amount of energy [1:41] in from the sun as earth radiates into space. [1:45] - Wow. [1:46] - Because if we didn't do that, [1:48] then the earth would get a lot hotter, that'd be a problem. [1:50] - That'd be a big problem. [1:51] - So, if that is the case... [1:53] - Yeah. [1:54] - Then what are we really getting from the sun? [1:57] - That's a good question. [1:58] - Hmm. [1:59] - It gives us a nice tan. [2:01] - It gives us a nice tan, I love it. [2:03] We're getting something special from the sun. [2:05] - I don't know, what do we get without the energy? [2:07] - But nobody talks about it. [2:09] To answer that, we have to go back to a discovery [2:12] made two centuries ago. [2:15] In the winter of 1813, France was being invaded [2:18] by the armies of Austria, Prussia, and Russia. [2:22] The son of one of Napoleon's generals [2:24] was Sadi Carnot, a 17-year-old student. [2:27] On December 29th, he writes a letter to Napoleon [2:30] to request to join in the fight. [2:32] Napoleon preoccupied in battle, never replies. [2:36] but Carnot gets his wish a few months later [2:39] when Paris is attacked. [2:41] The students defend a chateau just east of the city, [2:44] but there're no match for the advancing armies, [2:47] and Paris falls after only a day of fighting. [2:50] Forced to retreat, Carnot is devastated. [2:56] Seven years later, he goes to visit his father [2:59] who's fled to Prussia after Napoleon's downfall. [3:02] His father was not only a general, but also a physicist. [3:06] He wrote an essay on how energy [3:07] is most efficiently transferred in mechanical systems. [3:11] When his son comes to visit, they talk at length [3:14] about the big breakthrough of the time, steam engines. [3:18] Steam engines were already being used to power ships, [3:21] mine ore, and excavate ports. [3:24] And it was clear that the future industrial [3:26] and military might of nations [3:28] depended on having the best steam engines. [3:33] But French designs were falling behind [3:35] those of other countries like Britain. [3:37] So, Sadi Carnot took it upon himself to figure out why. [3:45] At the time, even the best steam engines [3:47] only converted around 3% of thermal energy [3:50] into useful mechanical work. [3:52] If he could improve on that, he could give France [3:55] a huge advantage and restore its place in the world. [3:59] So he spends the next three years studying heat engines, [4:03] and one of his key insights [4:05] involves how an ideal heat engine would work, [4:08] one with no friction and no losses to the environment. [4:12] It looks something like this. [4:16] Take two really big metal bars, one hot and one cold. [4:20] The engine consists of a chamber filled with air, [4:24] where heat can only flow in or out through the bottom. [4:28] Inside the chamber is a piston, [4:30] which is connected to a flywheel. [4:32] The air starts at a temperature just below that [4:35] of the hot bar. [4:36] So first, the hot bar is brought into contact [4:38] with the chamber. [4:39] The air inside expands with heat flowing into it [4:42] to maintain its temperature. [4:44] This pushes the piston up, turning the flywheel. [4:48] Next, the hot bar is removed, [4:50] but the air in the chamber continues to expand, [4:53] except now without heat entering, the temperature decreases. [4:57] In the ideal case, [4:58] until it is the temperature of the cold bar. [5:01] The cold bar is brought into contact with the chamber [5:04] and the flywheel pushes the piston down. [5:06] And as the air is compressed, [5:08] heat is transferred into the cold bar. [5:11] The cold bar is removed. [5:12] The flywheel compresses the gas further [5:15] increasing its temperature until it is just below that [5:17] of the hot bar. [5:19] Then the hot bar is connected again and the cycle repeats. [5:23] Through this process, heat from the hot bar is converted [5:26] into the energy of the flywheel. [5:30] And what's interesting to note about Carnot's ideal engine [5:33] is that it is completely reversible. [5:35] If you ran the engine in reverse, [5:38] first the air expands lowering the temperature, [5:41] then the chamber is brought into contact with the cold bar, [5:44] the air expands more, drawing in heat from the cold bar. [5:48] Next, the air is compressed, increasing its temperature. [5:52] The chamber is placed on top of the hot bar [5:54] and the energy of the flywheel is used to return the heat [5:58] back into the hot bar. [6:00] However many cycles were run in the forward direction, [6:03] you could run the same number in reverse, and at the end, [6:07] everything would return to its original state [6:10] with no additional input of energy required. [6:14] So by running an ideal engine, nothing really changes. [6:17] You can always undo what you did. [6:22] So what is the efficiency of this engine? [6:25] Since it's fully reversible, [6:26] you might expect the efficiency to be 100%, [6:30] but that is not the case. [6:32] Each cycle, the energy of the flywheel increases [6:35] by the amount of heat flowing [6:36] into the chamber from the hot bar, [6:38] minus the heat flowing out of the chamber at the cold bar. [6:42] So to calculate the efficiency, we divide this energy [6:46] by the heat input from the hot bar. [6:50] Now the heat in on the hot side is equal to the work done [6:53] by the gas on the piston, [6:55] and this will always be greater than the work done [6:58] by the piston on the gas on the cold side, [7:01] which equals the heat out. [7:03] And this is because on the hot side, [7:05] the hot gas exerts a greater pressure [7:07] on the piston than that same gas when cold. [7:11] To increase the efficiency of the engine, [7:13] you could increase the temperature of the hot side, [7:16] or decrease the temperature of the cold side, or both. [7:21] Lord Kelvin learns of Carnot's ideal heat engine [7:24] and realizes it could form the basis [7:27] for an absolute temperature scale. [7:30] Imagine that the gas is allowed to expand an extreme amount, [7:34] so much that it cools to the point [7:36] where all the gas particles effectively stop moving. [7:39] Then they would exert no pressure on the piston, [7:42] and it would take no work to compress it [7:44] on the cold side, so no heat would be lost. [7:48] This is the idea of absolute zero, [7:52] and it would make for a 100% efficient engine. [7:56] Using this absolute temperature scale, the Kelvin scale, [8:00] we can replace the amount of heat in and out [8:02] with the temperature of the hot and cold side respectively, [8:06] because they are directly proportional. [8:09] So we can express efficiency like this, [8:11] which we can rewrite like this. [8:15] What we have learned is that the efficiency [8:17] of an ideal heat engine doesn't depend on the materials [8:20] or the design of the engine, [8:21] but fundamentally on the temperatures [8:23] of the hot and cold sides. [8:26] To reach 100% efficiency, you'd need infinite temperature [8:29] on the hot side or absolute zero on the cold side, [8:32] both of which are impossible in practice. [8:35] So even with no friction or losses to the environment, [8:37] it's impossible to make a heat engine 100% efficient. [8:41] And that's because to return the piston [8:43] to its original position, [8:44] you need to dump heat into the cold bar. [8:47] So not all the energy stays in the flywheel. [8:51] Now, in Carnot's time, high pressure steam engines [8:54] could only reach temperatures up to 160 degrees Celsius. [8:57] So their theoretical maximum efficiency was 32%, [9:02] but their real efficiency was more like 3%. [9:05] That's because real engines experience friction, [9:08] dissipate heat to the environment, [9:10] and they don't transfer heat at constant temperatures. [9:13] So for just as much heat going in, [9:15] less energy ends up in the flywheel. [9:18] The rest is spread out over the walls of the cylinder, [9:21] the axle of the flywheel, [9:23] and is radiated out into the environment. [9:26] When energy spreads out like this, [9:29] it is impossible to get it back. [9:31] So this process is irreversible. [9:34] The total amount of energy didn't change, [9:37] but it became less usable. [9:39] Energy is most usable when it is concentrated [9:43] and less usable when it's spread out. [9:48] Decades later, German physicist, Rudolf Clausius, [9:51] studies Carnot's engine, and he comes up with a way [9:54] to measure how spread out the energy is. [9:58] He calls this quantity, entropy. [10:02] When all the energy is concentrated in the hot bar, [10:05] that is low entropy, [10:07] but as the energy spreads to the surroundings, [10:10] the walls of the chamber and the axle [10:12] will entropy increases. [10:15] This means the same amount of energy is present, [10:17] but in this more dispersed form, [10:19] it is less available to do work. [10:23] In 1865, Clausius summarizes the first two laws [10:27] of thermodynamics like this. [10:29] First, the energy of the universe is constant. [10:33] And second, the entropy of the universe tends to a maximum. [10:38] In other words, energy spreads out over time. [10:44] The second law is core to so many phenomena in the world. [10:48] It's why hot things cool down and cool things heat up, [10:51] why gas expands to fill a container, [10:54] why you can't have a perpetual motion machine, [10:57] because the amount of usable energy in a closed system [11:00] is always decreasing. [11:04] The most common way to describe entropy is as disorder, [11:08] which makes sense because it is associated with things [11:10] becoming more mixed, random, and less ordered. [11:14] But I think the best way to think about entropy [11:16] is as the tendency of energy to spread out. [11:19] So why does energy spread out over time? [11:22] I mean, most of the laws of physics [11:24] work exactly the same way forwards or backwards in time. [11:28] So how does this clear time dependence arise? [11:34] Well, let's consider two small metal bars, [11:36] one hot and one cold. [11:38] For this simple model, [11:40] we'll consider only eight atoms per bar. [11:43] Each atom vibrates according to the number [11:45] of energy packets it has. [11:46] The more packets, the more it vibrates. [11:49] So let's start with seven packets of energy [11:51] in the left bar and three in the right. [11:54] The number of energy packets in each bar [11:55] is what we'll call a state. [11:58] First, let's consider just the left bar. [12:01] It has seven energy packets, [12:02] which are free to move around the lattice. [12:05] This happens nonstop. [12:06] The energy packets hop randomly from atom to atom [12:09] giving different configurations of energy, [12:12] but the total energy stays the same the whole time. [12:16] Now, let's bring the cold bar back in [12:18] with only three packets and touch them together. [12:21] The energy packets can now hop around [12:22] between both bars creating different configurations. [12:26] Each unique configuration is equally likely. [12:30] So what happens if we take a snapshot at one instant in time [12:33] and see where all the energy packets are? [12:36] So stop, look at this. [12:39] Now there are nine energy packets in the left bar, [12:43] and only one in the right bar. [12:45] So heat has flowed from cold to hot. [12:49] Shouldn't that be impossible because it decreases entropy? [12:53] Well, this is where Ludwig Boltzmann [12:55] made an important insight. [12:57] Heat flowing from cold to hot is not impossible, [13:01] it's just improbable. [13:04] There are 91,520 configurations [13:07] with nine energy packets in the left bar, [13:10] but 627,264 with five energy packets in each bar. [13:16] That is the energy is more than six times [13:19] as likely to be evenly spread between the bars. [13:22] But if you add up all the possibilities, [13:24] you find there's still a 10.5% chance that the left bar [13:28] ends up with more energy packets than it started. [13:32] So, why don't we observe this happening around us? [13:36] Well, watch what happens as we increase the number of atoms [13:38] to 80 per bar and the energy packets to 100, [13:42] with 70 in the left bar and 30 in the right. [13:45] There is now only a 0.05% chance that the left solid [13:49] ends up hotter than it started. [13:52] And this trend continues as we keep scaling up the system. [13:55] In everyday solids, there are around 100 trillion, [13:59] trillion atoms and even more energy packets. [14:02] So heat flowing from cold to hot is just so unlikely [14:06] that it never happens. [14:10] Think of it like this Rubik's cube. [14:12] Right now, it is completely solved, [14:14] but I'm gonna close my eyes and make some turns at random. [14:18] If I keep doing this, it will get further and further [14:21] from being solved. [14:22] But how can I be confident that I'm really messing [14:26] this cube up? [14:27] Well, because there's only one way for it to be solved, [14:30] a few ways for it to be almost solved, [14:32] and quintillions of ways [14:34] for it to be almost entirely random. [14:36] Without thought and effort, [14:38] every turn moves the Rubik's cube [14:40] from a highly unlikely state that of it being solved [14:43] to a more likely state, a total mess. [14:50] So if the natural tendency of energy is to spread out [14:53] and for things to get messier, [14:55] then how is it possible to have something [14:57] like air conditioning where the cold interior of a house [15:01] gets cooler and the hot exterior gets hotter? [15:06] Energy is going from cold to hot, [15:08] decreasing the entropy of the house. [15:12] Well, this decrease in entropy is only possible [15:16] by increasing the entropy a greater amount somewhere else. [15:20] In this case, at a power plant, [15:21] the concentrated chemical energy and coal is being released, [15:26] heating up the power plant in its environment, [15:28] spreading to the turbine the electric generators, [15:31] heating the wires all the way to the house, [15:33] and producing waste heat in the fans and compressor. [15:37] Whatever decrease in entropy is achieved at the house [15:40] is more than paid for by an increase in entropy required [15:44] to make that happen. [15:47] But if total entropy is constantly increasing [15:50] and anything we do only accelerates that increase, [15:54] then how is there any structure left on earth? [15:58] How are there hot parts separate from cold parts? [16:01] How does life exist? [16:05] Well, if the earth were a closed system, [16:07] the energy would spread out completely, [16:09] meaning, all life would cease, [16:12] everything would decay and mix, [16:14] and eventually, reach the same temperature. [16:17] But luckily, earth is not a closed system, [16:21] because we have the sun. [16:28] What the sun really gives us is a steady stream [16:31] of low entropy that is concentrated bundled up energy. [16:35] The energy that we get from the sun is more useful [16:39] than the energy we give back. [16:41] It's more compact, it's more clumped together. [16:44] Plants capture this energy and use it to grow [16:46] and create sugars. [16:48] Then animals eat plants and use that energy [16:51] to maintain their bodies and move around. [16:54] Bigger animals get their energy [16:55] by eating smaller animals and so on. [16:58] And each step of the way, [17:00] the energy becomes more spread out. [17:03] - Okay, interesting. [17:04] - Yeah. [17:05] - Oh wow, I did not know that. [17:06] - There you go. [17:08] Ultimately, all the energy that reaches earth from the sun [17:11] is converted into thermal energy, [17:13] and then it's radiated back into space. [17:17] But in fact, it's the same amount. [17:19] I know this is a- - You do know this is... [17:21] - I'm a PhD physicist. [17:23] - Oh, okay, but anyway, so... - I trust you. [17:27] The increase in entropy can be seen [17:29] in the relative number of photons arriving at [17:32] and leaving the earth. [17:34] For each photon received from the sun, [17:38] 20 photons are emitted, [17:40] and everything that happens on earth, [17:42] plants growing, trees falling, herds stampeding, [17:46] hurricanes and tornadoes, people eating, [17:49] sleeping, and breathing. [17:51] All of it happens in the process of converting fewer, [17:55] higher energy photons [17:57] into 20 times as many lower energy photons. [18:01] Without a source of concentrated energy [18:03] and a way to discard the spread out energy, [18:06] life on earth would not be possible. [18:10] It has even been suggested that life itself [18:12] may be a consequence of the second law of thermodynamics. [18:16] If the universe tends toward maximum entropy, [18:19] then life offers a way to accelerate that natural tendency, [18:23] because life is spectacularly good [18:26] at converting low entropy into high entropy. [18:29] For example, the surface layer of seawater produces [18:32] between 30 to 680% more entropy when cyanobacteria [18:38] and other organic matter is present than when it's not. [18:42] Jeremy England takes this one step further. [18:45] He's proposed that if there is a constant stream [18:48] of clumped up energy, this could favor structures [18:51] that dissipate that energy. [18:53] And over time, this results [18:55] in better and better energy dissipators, [18:57] eventually resulting in life. [19:00] Or in his own words, [19:01] "You start with a random clump of atoms, [19:03] and if you shine light on it for long enough, [19:06] it should not be so surprising that you get a plant." [19:11] So life on earth survives on the low entropy from the sun, [19:15] but then where did the sun get its low entropy? [19:19] The answer is the universe. [19:22] If we know that the total entropy of the universe [19:25] is increasing with time, then it was lower entropy yesterday [19:29] and even lower entropy the day before that, [19:31] and so on, all the way back to the Big Bang. [19:37] So right after the Big Bang, [19:38] that is when the entropy was lowest. [19:41] This is known as the past hypothesis. [19:44] It doesn't explain why the entropy was low, [19:47] just that it must have been that way for the universe [19:49] to unfold as it has. [19:51] But the early universe was hot, dense, [19:54] and almost completely uniform. [19:56] I mean, everything was mixed and the temperature [19:59] was basically the same everywhere, [20:01] varying by at most 0.001%. [20:05] So how is this low entropy? [20:09] Well, the thing we've left out is gravity. [20:11] Gravity tends to clump matter together. [20:15] So taking gravity into account, [20:16] having matter all spread out like this, [20:19] would be an extremely unlikely state, [20:21] and that is why it's low entropy. [20:26] Over time, as the universe expanded and cooled, [20:29] matter started to clump together in more dense regions. [20:33] And in doing so, enormous amounts of potential energy [20:36] were turned into kinetic energy. [20:38] And this energy could also be used [20:41] like how water flowing downhill can power a turbine. [20:45] But as bits of matter started hitting each other, [20:48] some of their kinetic energy was converted into heat. [20:51] So the amount of useful energy decreased. [20:55] Thereby, increasing entropy. [20:58] Over time, the useful energy was used. [21:01] In doing so, stars, planets, galaxies, and life were formed, [21:06] increasing entropy all along. [21:09] The universe started with around 10 [21:11] to the 88 Boltzmann constants worth of entropy. [21:15] Nowadays, all the stars in the observable universe [21:17] have about 9.5 times 10 to the 80. [21:20] The interstellar and intergalactic medium combined [21:23] have almost 10 times more, [21:25] but still only a fraction of the early universe. [21:29] A lot more is contained in neutrinos [21:31] and in photons of the cosmic microwave background. [21:36] In 1972, Jacob Bekenstein proposed [21:39] another source of entropy, black holes. [21:46] He suggested that the entropy of a black hole [21:48] should be proportional to its surface area. [21:51] So as a black hole grows, its entropy increases. [21:55] Famous physicists thought the idea was nonsense [21:58] and for good reason. [21:59] According to classical thermodynamics, [22:01] if black holes have entropy, [22:03] then they should also have a temperature. [22:05] But if they have temperatures, they should emit radiation [22:09] and not be black after all. [22:12] The person who set out to prove Bekenstein wrong [22:15] was Stephen Hawking. [22:17] But to his surprise, his results showed that black holes [22:20] do emit radiation, now known as Hawking radiation, [22:24] and they do have a temperature. [22:27] The black hole at the center of the Milky Way [22:29] has a temperature of about a hundred trillionth of a Kelvin, [22:33] emitting radiation that is far too weak to detect. [22:36] So still pretty black. [22:38] But Hawking confirmed that black holes have entropy [22:40] and Bekenstein was right. [22:43] Hawking was able to refine Bekenstein's proposal [22:46] and determine just how much entropy they have. [22:49] The super massive black hole at the center of the Milky Way [22:51] has about 10 to the 91 Boltzmann constants of entropy. [22:56] That is 1,000 times as much [22:58] as the early observable universe, [23:00] and 10 times more than all the other particles combined. [23:04] And that is just one black hole. [23:06] All black holes together account for 3 times 10 [23:09] to the 104 Boltzmann constants worth of entropy. [23:14] So almost all the entropy of the universe [23:17] is tied up in black holes. [23:19] That means, the early universe [23:21] only had about 0.000000000000003% of the entropy it has now. [23:31] So the entropy was low, and everything that happens [23:35] in the universe like planetary systems forming, [23:37] galaxies merging, asteroids crashing, [23:40] stars dying, to life itself flourishing, [23:44] all of that can happen because the entropy [23:46] of the universe was low and it has been increasing, [23:50] and it all happens only in one direction. [23:52] We never see an asteroid uncrash [23:55] or a planetary system unmix [23:57] into the cloud of dust and gas that made it up. [24:00] There is a clear difference between going to the past [24:03] and the future, and that difference comes from entropy. [24:07] The fact that we are going from unlikely [24:09] to more likely states is why there is an arrow of time. [24:16] This is expected to continue until eventually, [24:20] the energy gets spread out so completely [24:22] that nothing interesting will ever happen again. [24:26] This is the heat death of the universe. [24:29] In the distant future, [24:31] more than 10 to the 100 years from now, [24:33] after the last black hole has evaporated, [24:36] the universe will be in its most probable state. [24:40] Now, even on large scales, you would not be able to tell [24:43] the difference between time moving forwards or backwards, [24:46] and the arrow of time itself would disappear. [24:52] So it sounds like entropy is this awful thing [24:55] that leads us inevitably [24:57] towards the dullest outcome imaginable. [25:00] But just because maximum entropy has low complexity [25:03] does not mean that low entropy has maximum complexity. [25:07] It's actually more like this tea and milk. [25:09] I mean, holding it like this is not very interesting. [25:12] But as I pour the milk in, the two start to mix [25:17] and these beautiful patterns emerge. [25:20] They arise in an instant and before you know it, [25:23] they're gone back to being featureless. [25:26] Both low and high entropy are low in complexity. [25:30] It's in the middle where complex structures [25:33] appear and thrive. [25:35] And since that's where we find ourselves, [25:37] let's make use of the low entropy we've got while we can. [25:47] With the right tools, we can understand just about anything, [25:50] from a cup of tea cooling down [25:52] to the evolution of the entire universe. [25:54] And if you're looking for a free and easy way [25:56] to add powerful tools to your arsenal, [25:59] then you should check out [26:00] this video sponsor, brilliant.org. [26:02] With Brilliant, you can master key concepts in everything [26:05] from math and data science to programming and physics. [26:08] All you need to do is set your goal, [26:10] and Brilliant will design the perfect learning path for you, [26:13] equipping you with all the tools you need to reach it. [26:16] Want to learn how to think like a programmer? [26:18] Then Brilliant's latest course, "Thinking in Code" [26:20] is a fast and easy way to get there. [26:22] Using an intuitive drag and drop editor, [26:25] it teaches you what you really need to know, [26:27] including essential concepts like nesting and conditionals. [26:31] You can start by jumping right in to program a robot [26:33] and then learn how to apply your new tools [26:36] to your everyday life, [26:37] like automating reminders on your phone [26:39] or building a bot that filters your matches on a dating app. [26:42] What I love about Brilliant is that they connect [26:45] what you learn to real world examples. [26:47] And because each lesson is hands-on, [26:49] you'll build real intuition, [26:51] so you can put what you've learned to good use. [26:53] To try everything Brilliant has to offer free [26:56] for a full 30 days, visit brilliant.org/veritasium. [27:00] I will put that link down in the description. [27:02] And through that link, the first 200 of you to sign up [27:05] will get 20% off Brilliant's annual premium subscription. [27:09] So I wanna thank Brilliant for sponsoring this video, [27:12] and I wanna thank you for watching.