Cryogenic liquid propellants deliver the highest specific impulse of any routinely used chemical propulsion. The three combinations that dominate modern space launch, liquid hydrogen with liquid oxygen, liquid methane with liquid oxygen, and refined kerosene with liquid oxygen, span a performance range from approximately $300$ seconds to more than $450$ seconds vacuum specific impulse. Each combination trades a different set of properties against absolute performance. Liquid hydrogen with liquid oxygen delivers the highest specific impulse at the cost of hydrogen’s low density and demanding storage requirements. Liquid methane with liquid oxygen delivers intermediate specific impulse with clean combustion and modest density. Refined kerosene with liquid oxygen delivers the lowest specific impulse of the three but the highest density and the most tractable ground handling. This article treats the specific chemistries, delivered performance, and engine cycles that make each combination viable at the level the opening article of this series establishes and by the same taxonomy that the previous article on solid propellants applied to composite and double-base chemistries.

The distinguishing feature of these combinations is that at least one propellant is stored at cryogenic temperature. Liquid oxygen boils at $90$ kelvin. Liquid methane boils at $111$ kelvin. Liquid hydrogen boils at $20$ kelvin. Refined kerosene is a room-temperature fluid, but its combination with liquid oxygen is treated as cryogenic because the oxidizer sets the storage boundary. Cryogenic storage introduces boil-off losses, insulation requirements, and tanking operations that storable combinations avoid. It also permits the highest specific impulse chemistry because low-molecular-weight combustion products expand most efficiently through a nozzle at high chamber temperatures.

Cryogenic Liquid Rocket Engine Anatomy

A cryogenic liquid rocket engine consists of two propellant tanks, turbopumps that deliver propellant to the combustion chamber at chamber pressure, an injector plate that atomizes and mixes the propellants against the head of the chamber, a combustion chamber where reactions proceed to equilibrium, a nozzle throat that accelerates the flow to sonic velocity, and a divergent nozzle section that expands the flow supersonically to produce thrust. Cooling passages carry propellant through the chamber and nozzle walls to protect the wall metal from the combustion gas. A power cycle drives the turbopumps by burning a small fraction of the propellant flow through a preburner or gas generator and extracting shaft work from the resulting hot gas.

The mechanical complexity of a cryogenic liquid engine is substantially higher than that of a solid rocket motor. The engine contains rotating turbomachinery operating at tens of thousands of revolutions per minute, high-pressure combustion at hundreds of bars, thin-walled cooling passages that must not fail under thermal and pressure loads, and a mixture of subsystems that interlock through the power balance. Turbopumps for a high-performance engine deliver hundreds of megawatts of shaft power at mass flow rates of hundreds of kilograms per second. The engine’s power cycle determines how efficiently propellant enthalpy can be converted to useful work at the turbines that drive the pumps, and this conversion efficiency directly limits achievable chamber pressure.

The propellant tanks and their pressurization are engineering problems in their own right. Cryogenic tanks require internal insulation against boil-off, external insulation against solar heating in flight, thermal shrinkage accommodation during chill-down, and pressurization by ullage gas that is usually gaseous helium for hydrogen and either gaseous helium or autogenously heated propellant for oxygen and methane. Tank chill-down is the process of cooling the empty tank to cryogenic temperature by flowing propellant vapor through it before liquid loading begins. Chill-down consumes several percent of the propellant load and takes minutes to hours depending on the tank size and design.

Hydrogen and Oxygen

The combination of liquid hydrogen with liquid oxygen produces water as the sole combustion product at stoichiometric conditions.

\[2 H_2 + O_2 \rightarrow 2 H_2O\]

The stoichiometric mixture ratio by mass is $8$ to $1$ oxidizer to fuel, because oxygen’s molecular weight of $32$ combines with two hydrogen molecules totaling $4$ mass units. Rocket engines using this combination operate at a mixture ratio of approximately $5$ to $6$ to $1$ rather than at stoichiometric, because the low molecular weight of hydrogen-rich exhaust more than compensates for the lower chamber temperature that running fuel-rich produces. The oxidizer-to-fuel mass ratio is denoted $O/F$ and equals the mass flow rate of oxidizer divided by the mass flow rate of fuel.

\[O/F = \frac{\dot{m}_{ox}}{\dot{m}_f}\]

Rocket engine designers select the operating $O/F$ to maximize delivered specific impulse rather than to match stoichiometry. The ideal specific impulse depends on the chamber temperature $T_c$ divided by the average exhaust molecular weight $M$ through the square root of that ratio, as the opening article of this series establishes. The $O/F$ that maximizes ideal specific impulse satisfies a stationary condition on $T_c/M$ at fixed chamber pressure.

\[\left. \frac{\partial}{\partial (O/F)} \left( \frac{T_c}{M} \right) \right|_{P_c} = 0\]

Fuel-rich operation reduces $T_c$ but reduces $M$ more rapidly over a range of mixture ratios below stoichiometric. The maximum of $T_c/M$ therefore falls below the stoichiometric $O/F$ for all three cryogenic combinations covered in this article.

An $O/F$ ratio of $6.0$ produces a chamber temperature of approximately $3550$ kelvin and an average exhaust molecular weight of approximately $13.5$ grams per mole. An $O/F$ ratio of $5.0$ produces a chamber temperature of approximately $3400$ kelvin and an average molecular weight of approximately $12$ grams per mole. The lower molecular weight at $O/F = 5.0$ raises the ideal specific impulse by approximately $5$ seconds despite the lower chamber temperature. Engines are typically operated at $O/F$ ratios chosen to balance specific impulse against propellant tank volume, because hydrogen’s low density penalizes vehicles heavily when extra hydrogen mass is carried.

Liquid Hydrogen

Liquid hydrogen boils at $20.28$ kelvin at atmospheric pressure and has a liquid density of approximately $71$ kilograms per cubic meter at its normal boiling point. Its low density is the principal engineering challenge for its use as a rocket propellant. A vehicle stage that carries $100$ tonnes of liquid hydrogen requires a tank volume of approximately $1400$ cubic meters, roughly nineteen times the volume the same mass of refined kerosene occupies. Vehicle diameter, structural mass, and aerodynamic drag all scale with tank size, and these penalties partially offset the specific impulse advantage of hydrogen chemistry.

Liquid hydrogen exists as two nuclear spin isomers. Para-hydrogen has the two proton spins antiparallel and is the low-temperature equilibrium form. Ortho-hydrogen has the two spins parallel and is the higher-energy form. At room temperature the equilibrium composition is approximately $75$ percent ortho and $25$ percent para. At $20$ kelvin the equilibrium is essentially pure para. The ortho-to-para transition is slow at low temperature and exothermic, releasing $703$ joules per mole. If hydrogen is liquefied without catalytic ortho-para conversion, boil-off from the residual ortho-to-para conversion in storage can consume a substantial fraction of the liquid over days. Industrial hydrogen liquefaction therefore catalyzes the conversion during liquefaction using iron oxide, chromium oxide, or paramagnetic-salt catalysts.

Liquid hydrogen requires vacuum-jacketed or foam-insulated storage. The Space Launch System core stage uses spray-on foam insulation on the tank exterior at a thickness of approximately $25$ millimeters. Ground storage uses vacuum-jacketed tanks with multilayer insulation between the inner and outer walls. Boil-off losses from a well-insulated ground storage tank are typically $0.05$ to $0.5$ percent per day depending on tank size, with smaller tanks losing a larger fraction because of their higher surface-to-volume ratio.

Liquid Oxygen

Liquid oxygen boils at $90.19$ kelvin at atmospheric pressure and has a liquid density of approximately $1141$ kilograms per cubic meter at its normal boiling point. Its higher density and higher boiling point relative to hydrogen make it substantially easier to handle. Boil-off losses from insulated ground storage are typically $0.02$ to $0.2$ percent per day. Liquid oxygen is chemically compatible with a broad range of structural metals including stainless steels and aluminum alloys, subject to cleanliness requirements that exclude hydrocarbon contamination that could serve as ignition fuel.

Liquid oxygen is manufactured by cryogenic air separation using distillation columns that separate atmospheric nitrogen, oxygen, and argon. The industrial base for liquid oxygen production is mature and distributed, with major merchant producers including Air Liquide, Linde, Air Products, and Praxair supplying refineries, steel mills, medical facilities, and aerospace ranges from the same production plants. A kilogram of liquid oxygen at the launch site costs approximately fifty United States cents to one United States dollar depending on volume and delivery distance, orders of magnitude less than the cost of the propellants it burns with.

Hydrolox Engines

The RS-25, formerly the Space Shuttle Main Engine, is the archetypal staged-combustion hydrolox engine. It burns liquid hydrogen with liquid oxygen at a chamber pressure of approximately $206$ bar, delivering $366$ seconds sea-level specific impulse and $452$ seconds vacuum specific impulse. Its two-preburner staged combustion cycle sends fuel-rich preburner exhaust through the turbines that drive the fuel and oxidizer turbopumps, then routes that preburner exhaust into the main combustion chamber to complete combustion. The RS-25 first flew on the Space Shuttle in $1981$ and continues in service on the Space Launch System core stage.

The RL10 is the archetypal expander-cycle hydrolox upper-stage engine. Variants of the RL10 have flown on the Centaur upper stage and its derivatives since $1963$, making the RL10 among the longest-serving production rocket engines. The RL10B-2 variant used on the Delta III and Delta IV upper stages delivers approximately $465$ seconds vacuum specific impulse at a chamber pressure of approximately $44$ bar, aided by an extendible carbon-carbon nozzle skirt that increases the expansion ratio from approximately $84$ to approximately $285$. The expander cycle harvests shaft power for the turbopumps from hydrogen that has been heated by absorbing chamber and nozzle heat during regenerative cooling. The absence of a preburner or gas generator eliminates the specific impulse loss those cycles impose but limits the achievable chamber pressure because heat available for pump work is bounded by nozzle area.

The Vulcain 2 engine on the Ariane 5 first stage and its Vulcain 2.1 variant on the Ariane 6 first stage deliver approximately $431$ seconds vacuum specific impulse at a chamber pressure of approximately $117$ bar using a gas-generator cycle. The gas-generator cycle burns a small fraction of the propellant flow in an auxiliary combustor and discharges the low-temperature exhaust from the turbines overboard rather than back into the main chamber. This overboard discharge represents a small specific-impulse loss compared to staged combustion but simplifies the engine and permits lower chamber pressure.

The Vinci upper-stage engine on Ariane 6 uses an expander cycle and delivers approximately $457$ seconds vacuum specific impulse at a chamber pressure of approximately $61$ bar. The Japanese LE-9 engine on the H3 first stage uses an expander-bleed cycle and delivers approximately $425$ seconds vacuum specific impulse.

Methane and Oxygen

The combination of liquid methane with liquid oxygen produces carbon dioxide and water as the principal combustion products.

\[CH_4 + 2 O_2 \rightarrow CO_2 + 2 H_2O\]

The stoichiometric mixture ratio by mass is $4$ to $1$ oxidizer to fuel, because methane’s molecular weight of $16$ combines with two oxygen molecules totaling $64$ mass units. Rocket engines using this combination operate at $O/F$ ratios of $3.4$ to $3.7$, slightly fuel-rich for the same reason hydrolox engines run fuel-rich. Chamber temperature at $O/F = 3.6$ is approximately $3550$ kelvin, comparable to hydrolox at $O/F = 6$, but the exhaust molecular weight of approximately $23$ grams per mole is substantially higher than the hydrolox value of $13$. This higher molecular weight is the principal reason methalox specific impulse is approximately $80$ to $100$ seconds lower than hydrolox specific impulse at comparable expansion ratios.

Liquid Methane

Liquid methane boils at $111.65$ kelvin at atmospheric pressure and has a liquid density of approximately $422$ kilograms per cubic meter at its normal boiling point. Its intermediate density between liquid hydrogen and refined kerosene, combined with its higher specific impulse than kerosene and lower cost than hydrogen, places it in a favorable middle ground for reusable launch vehicles where propellant density affects first-stage sizing and combustion cleanliness affects turnaround time.

Methane’s normal boiling point differs from that of liquid oxygen by only $21$ kelvin. This proximity permits shared insulation and shared cooling paths for combined propellant tanks in some vehicle architectures. The narrow temperature margin also permits densified propellant operation, in which propellants are subcooled below their normal boiling points to increase density. Densified methane at $95$ kelvin has a density of approximately $464$ kilograms per cubic meter, approximately $10$ percent higher than at normal boiling point. Densified oxygen at $80$ kelvin has a density of approximately $1195$ kilograms per cubic meter, approximately $5$ percent higher than at normal boiling point. Combined densification provides significant vehicle-level performance gains at the cost of additional ground-handling equipment.

Liquid methane is manufactured from natural gas by cryogenic liquefaction. Industrial-grade liquefied natural gas is approximately $85$ to $95$ percent methane by mass with the balance ethane, propane, nitrogen, and trace higher hydrocarbons. Propellant-grade methane requires purification to greater than $99$ percent methane by mass to avoid combustion instabilities and injector fouling from higher hydrocarbon impurities. Merchant liquid methane at the launch site costs approximately fifty United States cents to two United States dollars per kilogram depending on volume and purity, comparable to liquid oxygen and substantially less than refined kerosene.

Methalox Engines

The SpaceX Raptor engine on the Starship vehicle is the first flown full-flow staged-combustion methalox engine. It burns liquid methane with liquid oxygen at a chamber pressure of approximately $300$ bar for the V2 variant, delivering approximately $330$ seconds sea-level specific impulse and approximately $350$ seconds vacuum specific impulse in its sea-level configuration, or approximately $380$ seconds vacuum specific impulse in its vacuum configuration with expanded nozzle. Full-flow staged combustion routes both fuel-rich and oxidizer-rich preburner exhaust into the main chamber, eliminating the propellant flow imbalance between preburners that constrains other staged-combustion cycles.

The Blue Origin BE-4 engine on the New Glenn and Vulcan Centaur first stages burns liquid methane with liquid oxygen at a chamber pressure of approximately $135$ bar using an oxidizer-rich staged-combustion cycle. It delivers approximately $310$ seconds sea-level specific impulse. Its oxidizer-rich cycle burns methane with excess oxygen in a preburner and routes the oxidizer-rich exhaust to the main chamber, contrasting with the RS-25’s fuel-rich configuration and the Raptor’s dual-preburner full- flow configuration.

The Rocketdyne SE-7 methalox engine and various Chinese and Russian methalox engines under development represent the growing methalox engine population.

Kerosene and Oxygen

The combination of refined kerosene with liquid oxygen produces carbon dioxide and water as the principal combustion products. Kerosene is a mixture of aliphatic and aromatic hydrocarbons with an average composition approximately $C_{12}H_{26}$. Its stoichiometric combustion proceeds approximately according to the following equation.

\[C_{12} H_{26} + \tfrac{37}{2} O_2 \rightarrow 12 CO_2 + 13 H_2O\]

The stoichiometric mixture ratio by mass is approximately $3.4$ to $1$ oxidizer to fuel. Rocket engines using this combination operate at $O/F$ ratios of $2.2$ to $2.7$, well fuel-rich for maximum specific impulse. Chamber temperature at $O/F = 2.3$ is approximately $3500$ kelvin. Exhaust molecular weight at that mixture ratio is approximately $23$ grams per mole, similar to methalox. Kerolox delivered specific impulse is consequently a few seconds lower than methalox at comparable operating conditions because kerolox chemistry produces slightly more incomplete combustion products that reduce the effective average molecular weight increment.

Refined Kerosene RP-1

Rocket-grade kerosene, designated RP-1 in the United States and by various national designations elsewhere, is a narrow-boiling-range distillate cut from petroleum with tightly controlled sulfur, aromatic, and unsaturated hydrocarbon content. The military specification MIL-DTL-25576 defines RP-1 as a distillate of density $0.799$ to $0.815$ grams per cubic centimeter, aromatic content less than $5$ percent, olefin content less than $2$ percent, and sulfur content less than $0.003$ percent. These constraints exclude the reactive impurities that would otherwise coke or foul the engine’s regenerative cooling passages.

RP-1 has a room-temperature density of approximately $810$ kilograms per cubic meter and remains liquid from approximately $220$ kelvin to approximately $520$ kelvin at atmospheric pressure. It is a room-temperature-storable fluid that requires only rudimentary ground handling. Its cost at the launch site is approximately three to seven United States dollars per kilogram depending on volume and specification, substantially higher than liquid oxygen or liquid methane but still small compared to vehicle costs.

RP-2 is a modernized specification that further reduces sulfur content and adds stricter thermal stability requirements. RG-1 and Naftil are the Russian equivalents. Chinese and Indian kerosene grades follow similar compositional constraints.

Kerolox Engines

The Rocketdyne F-1 engine, developed for the Saturn V first stage and flown on all Saturn V missions from $1967$ to $1973$, remains the highest- thrust liquid rocket engine ever flown. It delivered approximately $263$ seconds sea-level specific impulse and $304$ seconds vacuum specific impulse at a chamber pressure of approximately $70$ bar using a gas- generator cycle. Its low chamber pressure by modern standards reflected the technology of the early nineteen sixties and its design mass flow rate of approximately $2600$ kilograms per second per engine.

The SpaceX Merlin 1D engine on the Falcon 9 first stage delivers approximately $282$ seconds sea-level specific impulse and $348$ seconds vacuum specific impulse at a chamber pressure of approximately $97$ bar using a gas-generator cycle. The Merlin Vacuum variant on the Falcon 9 second stage delivers approximately $348$ seconds vacuum specific impulse at a chamber pressure of approximately $108$ bar with a large expansion ratio nozzle. The Merlin is a modern kerolox engine that exemplifies efficient gas-generator design and has demonstrated the highest thrust-to- weight ratio of any flown kerolox engine.

The Russian RD-180 engine, used on the Atlas III and Atlas V first stages, burns refined kerosene with liquid oxygen at a chamber pressure of approximately $262$ bar using a dual-preburner oxidizer-rich staged- combustion cycle. It delivers $311$ seconds sea-level specific impulse and $338$ seconds vacuum specific impulse. Its oxidizer-rich cycle achieves substantially higher chamber pressure than any American kerolox engine because oxidizer-rich preburner exhaust delivers more turbine work per unit specific impulse loss than fuel-rich exhaust does. The Russian RD-170 and RD-171 four-chamber engines used similar cycles.

The Chinese YF-100 and YF-100K kerolox engines use oxidizer-rich staged combustion at chamber pressures of approximately $118$ bar and deliver approximately $335$ seconds vacuum specific impulse.

Coking and Cooling Constraints

Refined kerosene undergoes thermal decomposition to solid carbon deposits at wall temperatures above approximately $560$ kelvin. This coking process is the principal cooling constraint for kerolox engines because regenerative cooling passages heat the kerosene above its own coking temperature over long-duration or high-heat-flux operation. Coke deposition reduces heat transfer at the wall and increases pressure drop through the coolant channels. Both effects compound and can lead to wall failure if operation continues without margin.

The coking constraint is one of the reasons kerolox engines historically used gas-generator cycles rather than staged combustion. The Russian oxidizer-rich staged-combustion approach circumvents the coking problem by burning kerosene with a large excess of oxygen in the preburner, producing an oxygen-rich exhaust that cannot coke the turbine and manifold surfaces. Fuel-rich staged combustion with kerolox is possible but requires either very short-duration operation or advanced cooling schemes that are unattractive at production scale.

Methane does not exhibit this coking behavior because it lacks the polynuclear aromatic and long-chain compounds whose thermal decomposition products drive coking. This makes methalox amenable to fuel-rich staged combustion and full-flow staged combustion at higher chamber pressures than kerolox permits. The methalox coking advantage is one of the principal engineering motivations for the industry-wide shift toward methalox in reusable launch vehicles.

Historical Ethanol and Oxygen

Ethanol, chemical formula $C_2H_5OH$, was the first hydrocarbon rocket fuel used at flight scale. The German V-2 rocket used a mixture of approximately $75$ percent ethanol and $25$ percent water burned with liquid oxygen at a chamber pressure of approximately $15$ bar. The water dilution reduced flame temperature to protect the uncooled and regeneratively cooled steel chamber and nozzle from thermal failure. The V-2 delivered approximately $239$ seconds sea-level specific impulse at mass flow rates and chamber pressures far below modern practice. The combustion equation for anhydrous ethanol with oxygen is written below.

\[C_2 H_5 OH + 3 O_2 \rightarrow 2 CO_2 + 3 H_2O\]

The Redstone missile and Mercury-Redstone launcher continued the ethanol- oxygen tradition through the late nineteen fifties. American practice shifted to kerosene by the early nineteen sixties because refined kerosene delivers approximately $10$ to $15$ seconds higher specific impulse at similar chamber pressures and has a substantially higher density than ethanol. Ethanol-oxygen engines have appeared in some sounding rockets and student-built launch vehicles because ethanol is nontoxic and inexpensive, but ethanol is no longer used at production scale.

Cryogenic Handling and Boil-off

Cryogenic propellant operations impose ground infrastructure that storable-propellant operations do not require. Insulated storage tanks, cryogenic transfer piping, vacuum-jacketed valves, low-pressure ullage gas conditioning, and boil-off recovery systems all add to launch complex capital cost. A modern launch complex serving a hydrolox vehicle carries approximately fifty to two hundred million United States dollars in cryogenic infrastructure across ground storage, tanking, and vent systems.

Boil-off during ground hold and during pre-launch is a load-planning concern. A vehicle stage tanked several hours before launch loses several percent of its propellant to boil-off before ignition. Boil-off losses are absorbed by continuous replenishment from ground storage during the countdown. Once the vehicle disconnects from ground support at ignition minus a few seconds, the residual boil-off from tank warming and pump chill-down consumes propellant that has already been paid for in performance. Cryogenic upper stages that must coast in orbit for hours before their second burn face additional boil-off losses that constrain mission profiles.

Power Cycles

The power cycle of a liquid rocket engine determines how efficiently propellant enthalpy can be converted to shaft work at the turbopumps and how much of the propellant flow is available to enter the main chamber at full combustion enthalpy. Five power cycles are used in production cryogenic engines.

Gas-generator cycle. A small fraction of the propellant flow, typically $2$ to $5$ percent, is burned in an auxiliary combustor that produces hot gas at temperatures below the turbine’s material limit. The turbine extracts shaft work from this gas and dumps the exhaust overboard or through a small nozzle for a small thrust contribution. The specific impulse penalty for this discarded flow is typically $3$ to $10$ seconds depending on engine sizing. The F-1, Merlin, Vulcain 2, and RS-27 all use this cycle.

Staged-combustion cycle. Preburner exhaust drives the turbines and then flows into the main chamber to complete combustion. The main chamber receives the entire propellant flow at full combustion enthalpy. This eliminates the specific-impulse penalty of the discarded gas- generator flow. Higher chamber pressures are achievable because turbine work is not bounded by the overboard discharge conditions. The RS-25 uses fuel-rich staged combustion. The RD-180 and RD-170 use oxidizer-rich staged combustion.

Full-flow staged combustion. Both a fuel-rich preburner and an oxidizer-rich preburner drive dedicated turbines, and both preburner exhaust streams flow into the main chamber. The two preburners are sized independently so that turbine flow matches shaft work demand without the propellant flow imbalance that constrains single-preburner staged-combustion. The SpaceX Raptor is the first flown production full-flow staged combustion engine.

Expander cycle. Chamber and nozzle heat vaporize and heat the fuel, typically hydrogen, in the regenerative cooling passages, and the heated fuel drives the turbines before entering the injectors. No preburner or gas generator is used. This cycle eliminates all combustion-related specific-impulse loss but limits chamber pressure because turbine enthalpy is bounded by the nozzle area available for heat pickup. The RL10 and Vinci use this cycle.

Expander-bleed cycle. A variant of the expander cycle that bleeds some heated fuel overboard after the turbine rather than routing it back to the injectors. This variant relaxes the chamber-pressure limit of the pure expander cycle at the cost of a small specific-impulse penalty. The Japanese LE-9 uses this cycle.

Regenerative Cooling

Regenerative cooling routes propellant through channels in the chamber and nozzle walls before it enters the injectors. The propellant absorbs heat from the wall metal, protecting the metal from thermal failure and returning the absorbed enthalpy to the combustion process. Regenerative cooling is the standard cooling scheme for high-performance liquid rocket engines and is the essential feature that permits sustained operation at chamber temperatures well above the melting points of the wall materials.

The typical wall construction is a milled-channel copper alloy inner liner backed by a nickel electroformed outer jacket. The channels run axially from the injector face to the nozzle exit with cross-sections sized to keep wall temperature below the copper alloy’s yield-strength limit at the local heat flux. Wall heat flux peaks at the throat where gas velocities and heat transfer coefficients are highest, typically $50$ to $150$ megawatts per square meter for staged-combustion engines. Cooling channel pressure drops of $30$ to $80$ bar are typical, which directly reduces the pump discharge pressure available at the injectors.

Performance Comparison

Modern hydrolox upper-stage engines deliver $450$ to $465$ seconds vacuum specific impulse. Modern hydrolox first-stage engines deliver $425$ to $452$ seconds vacuum specific impulse and $315$ to $370$ seconds sea- level specific impulse, with the range’s upper end reflecting the RS-25’s staged-combustion architecture and its lower end reflecting the Vulcain 2 gas-generator cycle. Modern methalox engines deliver $340$ to $380$ seconds vacuum specific impulse and $300$ to $330$ seconds sea-level specific impulse. Modern kerolox engines deliver $330$ to $360$ seconds vacuum specific impulse and $260$ to $315$ seconds sea-level specific impulse depending on cycle and chamber pressure.

Density specific impulse, the product of bulk propellant density and specific impulse defined in the previous article on solid propellants, provides a different ranking. Hydrolox delivers approximately $190000$ seconds times kilograms per cubic meter, methalox approximately $310000$, kerolox approximately $310000$, and V-2- era ethanol-oxygen approximately $220000$. Modern methalox and kerolox tie approximately at density specific impulse despite kerolox’s higher propellant density, because methalox specific impulse advantage compensates for kerolox density advantage.

Tradeoffs

Cryogenic liquid propellants win over solid propellants on absolute specific impulse, throttleability, and restart capability. They lose on storability, mechanical complexity, and density specific impulse.

Absolute specific impulse is the strongest cryogenic-liquid advantage. Hydrolox delivers roughly $150$ seconds higher specific impulse than the best composite solid propellants. Methalox delivers roughly $70$ seconds higher specific impulse than composite solids. Kerolox delivers roughly $60$ seconds higher specific impulse than composite solids. This specific impulse advantage compounds through the rocket equation to produce substantially higher payload fraction for a given vehicle mass.

Throttleability and restart capability are second-order cryogenic-liquid advantages but essential for many applications. Liquid engines can be throttled from full thrust down to approximately $40$ percent of nominal thrust in most designs, permitting max-Q throttling during ascent and soft-landing profiles for reusable vehicles. Solid motors cannot be throttled. Liquid engines can be shut down and restarted, permitting staged burns, orbit circularization, deorbit maneuvers, and rendezvous operations that require multiple thrust events. Solid motors cannot be restarted.

Storability is the strongest cryogenic-liquid disadvantage. Cryogenic propellants boil off in storage, cannot be pre-loaded and left ready to fire, and require ground infrastructure for tanking that adds hours to turn-around cycles. This constraint is why every American ballistic missile in service uses solid propellant for its warhead-delivery propulsion and why cryogenic upper stages face mission-duration limits absent active zero-boil-off refrigeration.

Mechanical complexity is the second-strongest cryogenic-liquid disadvantage. Turbopumps, injectors, cooling passages, and cycle plumbing add production cost, add mass to the engine, and add failure modes that solid motors avoid entirely. Modern reusable engines mitigate this disadvantage by amortizing engine cost across many missions.

Density specific impulse is the third cryogenic-liquid disadvantage, particularly for hydrolox. Hydrogen’s low density penalizes first-stage vehicles because the extra tank volume produces extra structural mass and extra aerodynamic drag. Methalox and kerolox occupy a middle density-Isp regime that has become the preferred choice for reusable first stages.

Applications

Cryogenic liquid propellants dominate three application categories.

Space-launch first stages are one dominant category. The Space Launch System core stage uses hydrolox RS-25 engines. The Ariane 6 core stage uses hydrolox Vulcain 2.1. The Falcon 9 first stage uses kerolox Merlin 1D. The New Glenn first stage uses methalox BE-4. The Starship booster uses methalox Raptor. Trends since approximately $2010$ have shifted first-stage propellant choice from hydrolox and kerolox toward methalox for reusable vehicles, driven by methalox’s combination of clean combustion, high specific impulse relative to kerolox, and density advantages relative to hydrolox.

Space-launch upper stages are the second dominant category. The Centaur family of upper stages has flown hydrolox RL10 variants since $1963$. The Delta IV upper stage used RL10B-2. The Vulcan Centaur upper stage uses RL10C. The Ariane 6 upper stage uses Vinci. The Falcon 9 second stage uses kerolox Merlin Vacuum. The Starship ship uses methalox Raptor.

Deep-space and interplanetary propulsion is the third application category where cryogenic liquid engines appear, though monopropellant thrusters and storable bipropellant thrusters handle the majority of spacecraft propulsion after separation from the launch vehicle upper stage. Cryogenic liquid engines are used for major deep-space maneuvers that require both high specific impulse and substantial delta-v, subject to the boil-off limitations that restrict how long a cryogenic stage can remain in space before its second burn.

Conclusion

Cryogenic liquid propellants deliver the highest specific impulse of any routinely used chemical propulsion at the cost of storability, mechanical complexity, and density specific impulse. Hydrogen-oxygen combinations deliver the highest specific impulse of the three practical combinations but suffer from hydrogen’s low density. Methane-oxygen combinations deliver intermediate specific impulse, clean combustion that permits full-flow staged combustion at high chamber pressures, and density compatible with reusable first-stage vehicles. Kerosene-oxygen combinations deliver the lowest specific impulse of the three but the highest density and most tractable ground handling. Engine cycles trade specific impulse against chamber pressure through gas-generator, staged- combustion, expander, and full-flow-staged-combustion architectures.

The next article, A220, covers storable and hypergolic liquid propellants.

References