Clathrate Hydrates

A molecular cage, not a chemical bond

Freeze water with a little methane dissolved in it, raise the pressure, and something strange happens: the water does not freeze into ordinary hexagonal ice. Instead, the water molecules arrange themselves into hollow polyhedral cages, and each cage closes around a single methane molecule like a frozen fist. The result is a clathrate hydrate — an ice that is, by mass, mostly water, but which carries a passenger gas locked inside its crystal structure.

The key word is clathrate, from the Latin clathratus, "latticed" or "barred." The guest molecule is not chemically bonded to anything. There is no shared electron pair, no ionic attraction — only the weak van der Waals force between the guest and the wall of water molecules surrounding it, plus the simple fact that the guest is physically too big to slip out through the pentagonal and hexagonal faces of its cage. It is imprisonment by geometry. Remove the surrounding pressure, warm the ice a few degrees, and the cages collapse; the gas escapes in a rush and the water relaxes into a small puddle.

This makes a clathrate an inclusion compound and, crucially, a non-stoichiometric one. Because no chemical bond fixes the ratio of guest to host, the cages can be partly empty. A real methane hydrate might have only 90 percent of its cages occupied. The composition floats with temperature and pressure rather than being pinned by valence, which is why you write the formula with a tilde — CH₄·~5.75 H₂O — instead of a clean integer ratio.

The three crystal structures: sI, sII, sH

Nature builds clathrate cages from a small alphabet of polyhedra, named by the Jeffrey notation that counts faces. The fundamental building block is the 5¹² cage — a pentagonal dodecahedron with twelve five-sided faces, made of 20 water molecules. Larger cages add a few hexagonal faces. From these blocks, three repeating structures dominate:

Structure	Water / cell	Cage types	Typical guests	Where it forms
sI (cubic)	46	2 × 5¹² (small) + 6 × 5¹²6² (large)	CH₄, CO₂, H₂S, Xe	Marine seep gas, Mars CO₂, comets
sII (cubic)	136	16 × 5¹² (small) + 8 × 5¹²6⁴ (large)	N₂, O₂, propane, mixtures, H₂	Natural-gas hydrate, air clathrate in ice cores
sH (hexagonal)	34	3 × 5¹² + 2 × 4³5⁶6³ + 1 × 5¹²6⁸	Large + small guest pair	Deep petroleum systems

Which structure forms is dictated mostly by the size of the guest relative to the cage diameter. Small molecules like methane (diameter ≈ 0.43 nm) fit comfortably in the sI large cage (free diameter ≈ 0.59 nm). Slightly larger guests strain sI and push the system into sII, whose 5¹²6⁴ cage is roomier. Molecules too large for any single cage — like isopentane — can only be trapped in sH, and only if a small "help gas" simultaneously fills the small cages to stabilise the lattice. The trapped fraction matters: a cage occupancy below a critical threshold (around 70–95 percent depending on guest) is too low to hold the open lattice together, and the structure simply will not nucleate.

Why an open ice cage is more stable than dense ice

Ordinary ice Ih is a relatively dense, fully hydrogen-bonded network. Clathrate ice is far more open — about 20 percent of its volume is empty cage interior. Building such an open framework costs hydrogen-bond energy compared with collapsing into ordinary ice, so empty clathrate is thermodynamically unstable and has never been found unsupported in nature. The guest molecules pay the difference. Each enclosed gas molecule adds a few kJ/mol of van der Waals stabilisation by pressing outward on the cage walls, and once enough cages are filled, the filled clathrate becomes the lowest-free-energy phase.

The classic statistical-thermodynamic description is the van der Waals–Platteeuw model (1959). It treats each cage as an independent adsorption site and writes the fractional occupancy θ of a cage of type i by guest j as a Langmuir isotherm in the guest's fugacity f:

θ_ij = C_ij f_j / ( 1 + Σ_k C_kj f_k )

where C_ij is a temperature-dependent Langmuir constant encoding the guest–cage interaction. The chemical potential of water in the hydrate is then lowered relative to empty-lattice water by

Δμ_w = R T Σ_i ν_i ln( 1 − Σ_j θ_ij )

where ν_i is the number of cages of type i per water molecule. Setting this equal to the empty-lattice-to-real-water chemical-potential gap gives the dissociation pressure as a function of temperature — the phase boundary you actually measure in the lab. The takeaway: the more cages you can fill (higher fugacity, i.e. higher gas pressure), the more stable the hydrate, which is exactly why hydrates are creatures of high pressure and low temperature.

The 160-to-1 number, worked out

The single most quoted fact about methane hydrate is that one volume of the solid yields about 160 volumes of methane gas at the surface. Here is where that comes from. A structure-I unit cell contains 46 water molecules and, fully occupied, 8 methane molecules (2 small + 6 large cages). So the molar ratio is 46/8 = 5.75 water per methane — the famous CH₄·5.75 H₂O.

Per 1 mole CH₄ in fully-occupied sI hydrate:
  water carried     = 5.75 mol H₂O
  mass of hydrate   = 16.0 + 5.75 × 18.0 = 119.5 g
  density of hydrate ≈ 0.91 g/cm³
  volume of hydrate = 119.5 / 0.91 ≈ 131 cm³

Volume of 1 mol CH₄ as ideal gas at STP (0 °C, 1 atm):
  V_gas = 22,414 cm³

Expansion ratio = 22,414 / 131 ≈ 171  (≈160 with real-gas + partial cage filling)

The ~160 figure simply accounts for cages that are not 100 percent occupied and for real-gas departures from ideality. Either way, the lesson is dramatic: a clathrate is a way of storing gas at near-liquid density inside a solid that floats on water. It is the reason marine methane hydrate is both a potential energy resource and a climate concern, and the reason a small, cold icy moon can hide an enormous volatile inventory in a thin clathrate crust.

The stability field: pressure versus temperature

A clathrate exists only inside a wedge of pressure–temperature space bounded by its dissociation curve. The curve rises steeply: warmer conditions require dramatically higher pressure to keep the cages closed. A few anchor points for the methane–water system:

Temperature	Min. CH₄ pressure to stay solid	Setting
−10 °C (263 K)	~1.5 MPa (~15 bar)	Permafrost
0 °C (273 K)	~2.6 MPa (~26 bar)	Seafloor near freezing
+10 °C (283 K)	~7.5 MPa (~75 bar)	~700 m water column
+20 °C (293 K)	~22 MPa (~220 bar)	Deep ocean / drill pipe

On a planetary surface, that translates into a gas hydrate stability zone: a depth band where it is both cold enough and deep enough. In a terrestrial ocean margin the zone starts a few hundred metres below the seafloor and ends where the rising geothermal gradient pushes the temperature back across the dissociation curve, typically a few hundred metres further down. On a cold body the relevant pressure can come from the weight of overlying ice rather than water. CO₂ hydrate, with stronger guest–cage interaction, is stable to lower pressures than methane hydrate at the same temperature — a fact that matters for Mars, where the atmosphere is mostly CO₂.

Comets: clathrate or amorphous ice?

Comet nuclei are the Solar System's deep-freeze archives, and how they store their volatiles is a live debate. Two trapping mechanisms compete. Clathrate hydrate requires crystalline water ice that formed or annealed warm enough to build cages around gas. Amorphous water ice, deposited below roughly 100 K, traps gas in a disordered glassy matrix and releases it in bursts as it crystallises near 130–140 K, an exothermic transition that can run away and power outbursts.

The distinction has observable consequences. A comet still far from the Sun — beyond about 6 AU, where surface temperatures are below the water-sublimation point — can nonetheless show vigorous CO and CO₂ activity. Amorphous-ice crystallisation naturally explains that distant outgassing and the spectacular outbursts seen in objects like 29P/Schwassmann–Wachmann 1. Pure clathrates, which need a crystallisation event in the presence of gas, are harder to make in the cold protosolar nebula. The modern consensus is that primordial comet volatiles are trapped mainly in amorphous ice, with crystalline clathrate forming as a secondary, processed phase once a nucleus has been warmed by repeated perihelion passages. The Rosetta mission's measurements of comet 67P/Churyumov–Gerasimenko — its abundant CO, CO₂, N₂, and noble gases — are most consistent with amorphous-ice trapping, although clathrate layers near the surface remain plausible.

Titan, Enceladus, Mars, and the outer worlds

• Titan. Saturn's giant moon has a thick nitrogen–methane atmosphere whose methane is photochemically destroyed on a ~10–30 Myr timescale. Something must resupply it. A leading idea is a clathrate-rich crust over Titan's subsurface ocean: as the interior cools and the ocean partly freezes, methane clathrate at the ice–ocean boundary can dissociate and outgas, buffering the atmosphere over geological time. Episodic cryovolcanism may tap the same reservoir.
• Enceladus. The south-polar plume of this small Saturnian moon vents water vapour laced with CO₂, methane, and other species. One model feeds those gases from clathrates that dissociate as ocean water rises and depressurises through the tiger-stripe fractures — effectively decanting a clathrate reservoir into space.
• Mars. The polar caps and regolith may host CO₂ clathrate. Because CO₂ hydrate is more stable than CH₄ hydrate at low pressure, it has been proposed as a storage form for the planet's CO₂ and as a possible source of episodic gas release; it also has implications for any martian methane, whose origin remains debated.
• The cold outer Solar System. Pluto, Triton, the Kuiper-belt bodies, and the icy satellites all formed cold enough to trap their birth volatiles. Clathration of N₂, CO, and CH₄ in their interiors influences how — and whether — those gases later vent to the surface, and clathrate's low thermal conductivity can act as an insulating lid that helps keep a subsurface ocean liquid.

Clathrate hydrate versus ordinary ice and amorphous-ice trapping

Property	Ice Ih	Clathrate hydrate	Amorphous ice (gas-laden)
Water arrangement	Dense ordered lattice	Open cage lattice	Disordered glass
Gas content	Negligible	Up to ~15 mol % guest	Variable, trapped in voids
Bond to guest	—	van der Waals only	van der Waals only
Forms at	0 °C, 1 atm	Low T + high gas pressure	Vapour deposit below ~100 K
Releases gas by	—	Dissociation along P–T curve	Crystallisation near 130–140 K
Gas release style	—	Gradual, pressure-driven	Burst / runaway
Density	0.92 g/cm³	~0.91 g/cm³ (CH₄)	~0.94 g/cm³ (low-density form)
Astrophysical role	Bulk of icy bodies	Buffered volatile store	Primary comet volatile store

The two trapping phases are not mutually exclusive — a real icy body can hold amorphous ice deep down, crystalline ice and clathrate where it has warmed, and ordinary ice as bulk filler. The interplay sets the timing and style of outgassing, which is what we actually observe as cometary activity or plume composition.

Discovery and the people behind it

Clathrate hydrates have a long pedigree. Humphry Davy first reported a "chlorine hydrate" in 1810, and Michael Faraday characterised it in 1823, fixing the water-to-chlorine ratio. For a century they were a laboratory curiosity. That changed in 1934 when Hammerschmidt showed that gas hydrates — not ordinary ice — were the cause of plugged natural-gas pipelines, which turned hydrates into a serious engineering problem and funded decades of phase-equilibrium measurement. The molecular picture arrived with X-ray work by von Stackelberg and others in the early 1950s, establishing the sI and sII structures; the predictive statistical theory came from J. H. van der Waals and J. C. Platteeuw in 1959. The planetary connection was drawn in the 1950s–80s as Delsemme, Lewis, and others proposed clathrates to explain cometary gas release and the volatile budgets of the outer planets and their moons. Today, hydrate science spans energy (a global methane-in-hydrate inventory variously estimated at hundreds to thousands of gigatonnes of carbon), climate, carbon sequestration (CO₂-for-CH₄ swap in reservoirs), and planetary science.

Common misconceptions and edge cases

• "The gas is chemically bonded to the water." No. The only thing holding the guest in is van der Waals attraction and the geometric fact that it cannot fit through the cage faces. Break the cage and the gas leaves chemically unchanged.
• "Empty clathrate ice exists on its own." It does not, in nature. The open lattice is unstable without enough guest molecules propping it open; below a critical cage occupancy the structure will not form. (Hyper-stable empty hydrate frameworks like ice XVI have been made in the lab only by carefully evacuating a pre-formed clathrate at very low temperature.)
• "Methane hydrate is just methane frozen in ice." Frozen methane (solid CH₄) and methane dissolved in ordinary ice are different things. The hydrate is a distinct crystal phase with its own structure, density, and dissociation curve — and a vastly higher gas-storage capacity than gas trapped in ice Ih.
• "Comets are basically clathrate balls." The evidence, especially from Rosetta at 67P, points to amorphous ice as the primary volatile reservoir, with clathrate as a secondary phase. Treating a comet as pure clathrate over-predicts how tightly its gases should be bound and mis-times its outgassing.
• "Warming always dissociates hydrate immediately." Dissociation can be kinetically inhibited by self-preservation (anomalous metastability): between roughly 240 K and 270 K at low pressure, a thin skin of ice Ih can armour a methane-hydrate grain and stall its decomposition for hours to days, which matters for how hydrate behaves when an icy body is gently warmed.