Open and closed systems describe how matter and energy move (or don’t move) between a system and its surroundings. An open system exchanges both energy and matter with its environment, a closed system exchanges energy but not matter, and an isolated system exchanges neither. These categories show up across physics, chemistry, biology, and earth science, and understanding them is essential because the rules governing what can happen inside a system depend fundamentally on what crosses its boundary.
The Three Categories and What Separates Them
Science recognizes three types of thermodynamic system, and the differences hinge entirely on what is allowed to pass through the system’s boundary. An open system lets both energy and matter flow in and out. A pot of boiling water on a stove is open: heat enters from the burner, steam escapes into the air, and you can add more water. A closed system allows energy transfer but blocks the flow of matter. A sealed pressure cooker on the same stove is closer to closed: heat enters through the metal walls, but the water and steam stay trapped inside. An isolated system permits neither energy nor matter to cross. A perfect thermos bottle, if one existed, would be isolated: nothing gets in or out.
In practice, perfectly closed or perfectly isolated systems are idealizations. Every real container leaks a little heat; every seal is imperfect over long enough timescales. Scientists use these labels to describe which exchanges are significant enough to matter for the question at hand. A well-insulated calorimeter is treated as isolated for the few minutes of an experiment, even though it will eventually equilibrate with the room. The categories are tools for thinking clearly about what drives change inside a boundary.
Earth as the Textbook Example
Earth is the example that makes the distinction click for most people, because it sits right on the boundary between open and closed. Sunlight pours in, and Earth radiates heat back into space, so energy flows freely across the boundary. But matter is a different story. Apart from the occasional meteorite falling in and a tiny trickle of atmospheric gas escaping to space, Earth holds on to essentially the same atoms it has always had. This makes Earth approximately closed with respect to matter but open with respect to energy.1arXiv. Earth as a Closed Life-Support System: Finite Buffers, Dynamical Thresholds, and Planetary Sustainability
That energy openness is what makes life possible. Solar energy drives photosynthesis, weather patterns, and ocean currents. The material closure is why recycling of elements matters so much: the nitrogen, carbon, and phosphorus that living things need have to cycle through ecosystems rather than being replenished from outside. Many smaller ecosystems on Earth mirror this arrangement. Microbial mats, lakes, and open ocean gyres are roughly closed to matter and sustain themselves by self-organizing stable nutrient cycles powered by external light energy.2PubMed Central. Closed ecosystems extract energy through self-organized nutrient cycles
Why Living Things Must Be Open Systems
Every living organism is an open system, and it cannot be otherwise. You eat food (matter in), breathe oxygen (matter in), exhale carbon dioxide (matter out), radiate body heat (energy out), and excrete waste (matter out). This constant exchange is not incidental to being alive; it is what keeps you alive. From a thermodynamic standpoint, organisms maintain their internal order by exporting the disorder they produce, primarily as heat. The metabolically produced entropy gets shipped outward, and the organism stays organized.3MDPI / International Journal of Molecular Sciences. Entropy Perspectives of Molecular and Evolutionary Biology
This is sometimes misunderstood as a violation of the second law of thermodynamics, which says that disorder in an isolated system always increases. But a living cell is not isolated. It continuously pulls in low-entropy raw materials (food, oxygen) and pushes out high-entropy waste products (heat, carbon dioxide). The total entropy of the organism plus its surroundings still increases, as the second law requires. The organism just manages to keep its own local patch of the universe orderly by making the surroundings slightly more disordered. The compartmentation of chemicals inside cells and the fact that metabolic reactions sit far from equilibrium account for measurable reductions in entropy, on the order of 40 to 50 joules per kelvin per liter, that distinguish living tissue from a non-living chemical mixture.3MDPI / International Journal of Molecular Sciences. Entropy Perspectives of Molecular and Evolutionary Biology
How Openness Changes the Rules in Chemistry
The distinction between open and closed matters enormously in chemistry because it determines what a system settles into. A closed chemical system, cut off from exchange of matter and energy with its surroundings, will eventually reach equilibrium: the forward and reverse reactions balance out, energy stops being consumed or released on net, and nothing interesting happens anymore. The free energy of the system decreases until it hits its minimum, and the system parks there.4ACS Publications. Open-System Nonequilibrium Steady State: Statistical Thermodynamics, Fluctuations, and Chemical Oscillations
An open chemical system behaves very differently. When fresh reactants are continuously supplied and products are continuously removed, the system never reaches equilibrium. Instead, it can settle into a nonequilibrium steady state, a condition where concentrations stay constant not because reactions have stopped but because inflow and outflow balance each other. These steady states have nonzero flows of material and energy moving through them, and they convert chemical energy into heat continuously.4ACS Publications. Open-System Nonequilibrium Steady State: Statistical Thermodynamics, Fluctuations, and Chemical Oscillations This is a crucial difference. Equilibrium is static and featureless. A nonequilibrium steady state can sustain patterns, oscillations, and complexity, which is exactly what living cells do.
A practical illustration comes from food biotechnology. Batch fermentation operates as a closed system: all the nutrients go in at the start, and the culture progresses through its growth phases until there is nothing left to consume. Continuous fermentation, by contrast, is an open system: fresh medium is fed in at a steady rate while an equal volume of culture is drawn off. Properly designed continuous systems can sustain stable metabolism for weeks or months, often achieving five to ten times higher productivity than batch processes and producing more consistent product quality.5Elsevier. Continuous fermentation in food biotechnology: bridging the gap between batch and continuous processing The difference in system type is the difference between a one-shot process that runs down and a sustained operation that keeps going.
The Universe as an Isolated System
If Earth is the go-to example of a closed system and a living cell is the go-to example of an open system, the universe itself is the standard example of an isolated system. There is, by definition, nothing outside the universe for it to exchange energy or matter with. Its total mass-energy is constant. Entropy is created as matter converts to energy inside stars and as that energy radiates irreversibly outward into colder regions of space.6Geochimica et Cosmochimica Acta. Thermodynamic cosmology
This is one reason the second law of thermodynamics applies so cleanly at the cosmic scale. In an isolated system, entropy can only increase (or, at equilibrium, stay the same). The universe has no surroundings to dump waste heat into, no external energy source to draw from. Everything that happens inside it has to work with the fixed budget of mass-energy and the ever-increasing total entropy. The long-term implication, sometimes called the “heat death” of the universe, is that the universe gradually approaches a state where temperature differences even out and no further work can be extracted from anything. That is the endgame for any truly isolated system, though the timescale is so vast it is more of a philosophical observation than a practical concern.
Some cosmological analyses treat large comoving volumes of the universe as isolated subsystems, since the large-scale homogeneity means there are no net flows of matter or energy across neighboring volumes.7Monthly Notices of the Royal Astronomical Society: Letters. Does information entropy play a role in the expansion and acceleration of the Universe? This allows researchers to apply thermodynamic reasoning to cosmological questions, including the interplay between structure formation (which creates local order) and the overall entropy budget.
Self-Organization and Dissipative Structures
One of the most counterintuitive findings in twentieth-century science is that open systems, far from equilibrium, can spontaneously organize themselves into complex patterns. The physicist Ilya Prigogine coined the term “dissipative structures” for these organized states, recognizing that the very process of dissipating energy (increasing entropy in the surroundings) could create and sustain order within the system itself.8AIP Publishing (Chaos: An Interdisciplinary Journal of Nonlinear Science). Dissipative structures and irreversibility in nature: Celebrating 100th birth anniversary of Ilya Prigogine (1917–2003)
Think of a hurricane. It is an open system that draws energy from warm ocean water and expels heat at higher altitude. As long as the energy supply continues, the hurricane maintains its organized spiral structure. Cut off the warm water (when the hurricane moves over land, for instance), and the structure falls apart. The organization was not built into the system from the start; it emerged from the flow of energy through the system. Chemical oscillations, convection cells in heated fluids, and biological morphogenesis all fall into this category. The key insight is that none of them could occur in a closed system at equilibrium. Openness to energy and matter flow is a prerequisite for self-organization.
This realization had broad implications. For the first class of systems where an equilibrium state is possible, information is zero and entropy is maximal at equilibrium. But in self-organizing systems driven away from equilibrium, information increases as the system moves further from its equilibrium state.9Physics-Uspekhi. Entropy and information of open systems In plain terms, an open system can become more structured and information-rich the harder it is driven, as long as energy keeps flowing through it.
Why People Confuse Closed and Isolated
One of the most persistent sources of confusion, even among students who have taken a thermodynamics course, is mixing up “closed” and “isolated.” Research on student understanding has found that some students assume that a gas heated inside a sealed container cannot absorb heat, because they equate a closed container with an isolated system. Part of the problem is linguistic: the word “closed” in everyday English implies “sealed off from everything,” which sounds a lot like isolated. In thermodynamics, though, “closed” only means matter cannot cross the boundary. Energy absolutely can.10European Journal of Physics. Student understanding of thermodynamic processes, variables and systems
The confusion is sometimes reinforced by instructors themselves, who occasionally use “closed” when they mean “isolated.” Once students absorb this conflation, they overgeneralize it and struggle to analyze problems correctly. A sealed container sitting on a hot plate is closed (matter stays inside) but not isolated (heat enters through the walls). A thermos that was truly perfect would be isolated. The distinction matters because the equations governing what happens inside the system are different in each case. If you misclassify the system, you predict the wrong outcome.
Open and Closed Systems in Hydrology
The open-versus-closed framing shows up in fields you might not expect. In hydrology, researchers studying river catchments (the land areas that drain into a particular stream) have traditionally modeled them as closed water balance systems. The assumption is tidy: rain falls, some evaporates, the rest runs off into the stream, and no water sneaks in or out underground. But many catchments are not actually closed in this sense. Groundwater can flow between neighboring catchments beneath the surface, which means the water balance does not add up unless you account for that underground exchange.11Water Resources Research. The Impact of an Open Water Balance Assumption on Understanding the Factors Controlling the Long‐Term Streamflow Components
When researchers studied over 700 catchments in Brazil under an open water balance assumption, allowing for inter-catchment groundwater flow, they found that the factors controlling long-term streamflow components looked quite different than the closed-system models predicted.11Water Resources Research. The Impact of an Open Water Balance Assumption on Understanding the Factors Controlling the Long‐Term Streamflow Components This is a good example of how the choice of system type is not just a philosophical label; it determines what your model gets right and what it gets wrong. Treating an open system as closed can lead to systematic errors, whether the system is a chemical reactor, a watershed, or a living cell.
How the Open-Closed Framework Shaped Modern Science
The distinction between open and closed systems was not always central to science. Classical physics, for most of its history, focused on closed or isolated systems. The equations of Newtonian mechanics describe objects that exchange forces but not matter. Classical thermodynamics was built around closed systems approaching equilibrium. It was the biologist Ludwig von Bertalanffy who, in the mid-twentieth century, argued forcefully that biology could not be understood through closed-system models. Living organisms, he insisted, are fundamentally open systems defined by their material and energetic exchanges with their surroundings. His “General System Theory” contrasted this organismic approach with the cybernetic approach of the time, which focused on information transfer and feedback loops but did not deal with material and energetic exchanges.12PubMed Central. On the history of Ludwig von Bertalanffy’s “General Systemology”, and on its relationship to cybernetics – part III: convergences and divergences
Bertalanffy’s push to take open systems seriously influenced fields well beyond biology, including ecology, sociology, and management theory. The later work of Prigogine on dissipative structures reinforced the point from the physics side: the most interesting phenomena in nature, from weather patterns to the origin of life, occur in open systems far from equilibrium. The closed-system framework remains indispensable for problems where the simplification is justified (combustion in an engine cylinder, for instance), but the broader trend in science over the past century has been to recognize that most real-world systems of interest are open, and to develop tools for analyzing them accordingly.
When a System Changes Category
Systems do not always stay in one category. A living cell that dies stops exchanging matter and energy with its environment and gradually approaches equilibrium, effectively transitioning from an open system to a closed one on its way to becoming something like an isolated system (at least locally, for a while). A sealed terrarium starts as an open system while you are building it and becomes approximately closed the moment you seal the lid. From that point on, matter recycles internally while light energy passes through the glass.
In engineering, deliberately shifting a system’s category is a design choice. A batch reactor is loaded, sealed, and allowed to react as a closed system. A continuous-flow reactor is plumbed to receive fresh feed and discharge products, operating as an open system. Choosing between the two depends on whether you need tight control over reaction conditions (easier in a closed batch) or high throughput and consistency (easier in a continuous open system). The fermentation example mentioned earlier illustrates the tradeoff: batch is simpler to set up but limited in productivity, while continuous operation demands more sophisticated engineering but can sustain output for much longer.5Elsevier. Continuous fermentation in food biotechnology: bridging the gap between batch and continuous processing
Even the Earth’s system classification is not perfectly stable over geological time. Major asteroid impacts add significant matter. The slow escape of hydrogen from the upper atmosphere removes it. These fluxes are small enough that “approximately closed to matter” remains a useful description, but the approximation is better in some eras than others. The lesson is that open, closed, and isolated are descriptions of what is happening at a given scale and timescale, not permanent properties stamped onto a system. Choosing the right label is a judgment call, and getting it right is often the first step to understanding what the system will do next.