How Rockets Work: Thrust, Propulsion Systems and Orbital Mechanics Explained

Every time a rocket lifts off — whether it is a crewed Falcon 9, a European Ariane launcher or a small Earth-observation satellite booster — it is applying the same handful of physics principles that engineers have used since the early 20th century. Public interest in these launches stays high year-round, yet the underlying mechanics of thrust, propulsion and orbit are often skipped over. Agencies including NASA and the European Space Agency (ESA) publish detailed explanations of how rockets actually work; this piece draws on that material.

Producing thrust: Newton's third law in action

According to NASA's Glenn Research Center, a rocket engine burns stored fuel together with a stored oxidizer inside a combustion chamber, producing hot, high-pressure exhaust gas. That gas is then forced through a nozzle, which accelerates it further before it leaves the vehicle. NASA states plainly that "thrust is produced according to Newton's third law of motion": the gas is pushed one way out of the nozzle, and the rocket is pushed the opposite way with equal force. ESA describes the same mechanism for younger readers using a balloon analogy — air rushing out of a balloon's neck in one direction pushes the balloon the other way.

NASA's thrust equation expresses this numerically as F = ṁVe + Ae(pe − p0), where ṁ is the mass flow rate of exhaust through the engine, Ve is the exhaust's exit velocity, Ae is the nozzle's exit area, pe is the pressure at the nozzle exit, and p0 is the surrounding (free-stream) pressure. In practice, the amount of thrust a rocket produces depends chiefly on how much exhaust mass leaves per second and how fast that exhaust is moving when it exits the nozzle.

Why rockets work in the vacuum of space

Jet engines and propellers depend on drawing in surrounding air, which means they cannot operate once a vehicle leaves the atmosphere. Rockets avoid this limitation because, as NASA notes, they carry their own oxidizer on board rather than relying on atmospheric oxygen. Since there is no outside air involved in the reaction at all, the thrust equation has no "free-stream" air-intake term — which is also why, as NASA puts it, a rocket can generate thrust in a vacuum where there is no other source of oxygen.

Specific impulse: measuring an engine's efficiency

Engineers compare rocket engines using a figure called specific impulse, or Isp. NASA's Glenn Research Center defines it as thrust divided by the weight flow rate of propellant consumed, expressed through the formula Isp = F/(ṁ·g0), where g0 is standard gravitational acceleration (9.8 m/s², or 32.2 ft/s²). NASA explains that an engine with a higher specific impulse is more efficient because it produces more thrust for the same amount of propellant burned — making Isp the rocket-propulsion equivalent of fuel economy.

Multistage rockets: shedding dead weight

Most launch vehicles are built in two or three stages. ESA explains that once a stage has burned through all of its propellant, it is separated and discarded, because hauling an empty tank and spent engine hardware any further would waste fuel needed to accelerate the rest of the vehicle. Large launchers also attach solid- or liquid-fuelled booster rockets to the first stage for extra thrust at liftoff; these boosters, too, are jettisoned once spent. ESA notes that spacecraft need to reach roughly 28,000 km/h to enter orbit, a speed that staging helps achieve by continually lightening the vehicle as it climbs.

Chemical vs. electric propulsion

Not every propulsion system is designed to launch a vehicle off the ground. Chemical engines — burning a fuel and oxidizer together, as described above — remain the only practical way to lift a rocket out of Earth's atmosphere. For maneuvering once already in space, however, agencies increasingly use electric propulsion. NASA describes systems that use electricity, generated by solar arrays or a nuclear reactor, to ionize a gas such as xenon or krypton and accelerate the resulting ions out of a thruster using electric and magnetic fields.

NASA's Dawn mission illustrates the trade-off: it used electric propulsion to spend roughly five years reaching the asteroid belt before maneuvering between the bodies Vesta and Ceres — a slow, steady acceleration impossible to achieve with the brief, powerful burns of chemical engines, but efficient over long missions. ESA makes the same point about its SMART-1 lunar probe and the ongoing BepiColombo mission to Mercury, both of which rely on ion engines that, in ESA's words, "can go on pushing gently for months or even years" using only a small supply of propellant and a few kilowatts of power from solar panels.

Reaching and staying in orbit

Getting to space and staying there are two different problems. The U.S. National Oceanic and Atmospheric Administration (NOAA) explains that a satellite stays in orbit when its forward momentum is in balance with Earth's gravitational pull: gravity keeps bending the satellite's path toward Earth, while its sideways speed keeps carrying it onward, so the two effects combine into a circular or elliptical path rather than a straight fall. If a satellite moves too slowly for its altitude, gravity dominates and it descends; if it moves too fast, it breaks out of that balanced path instead of maintaining its orbit.

The speed needed for that balance changes with altitude. NOAA gives the example of its NOAA-20 satellite, which orbits a few hundred miles up at about 17,000 mph, compared with its GOES-East satellite, which orbits roughly 22,000 miles high at only about 6,700 mph — illustrating that the farther a satellite sits from Earth, the weaker gravity's pull and the lower the speed required to stay in orbit. NASA's Earth Observatory catalog gives concrete figures for common orbit classes: its Aqua satellite takes about 99 minutes to circle Earth from roughly 705 kilometers up (low Earth orbit); GPS satellites fly in a semi-synchronous medium Earth orbit about 20,200 kilometers above the surface with a 12-hour period; and geostationary satellites sit about 36,000 kilometers up, completing one orbit every 23 hours, 56 minutes and 4 seconds so that they appear fixed over one point on Earth.