The Fischer-Tropsch Process

What the Fischer-Tropsch process does

Start with the cheapest carbon you have — coal, natural gas, biomass, even municipal waste — and you cannot pour it into a fuel tank. The Fischer-Tropsch (FT) process is the industrial bridge that turns that carbon into liquid hydrocarbons that pour, pump, and burn like diesel. It does this in two stages. First you shatter the feedstock down to the smallest possible carbon fragment, carbon monoxide, mixed with hydrogen: this mixture is called synthesis gas or syngas. Then Fischer-Tropsch reassembles those one-carbon fragments into long hydrocarbon chains.

The net reaction, ignoring the branching and the alkene side-products, is a surface-catalyzed polymerization:

    n CO  +  (2n+1) H₂   ──Fe or Co, 200–350 °C, 20–40 bar──→   CₙH₂ₙ₊₂  +  n H₂O

    e.g. for n = 12 (a diesel-range molecule):
    12 CO  +  25 H₂  →  C₁₂H₂₆  +  12 H₂O

Every carbon in the product came from one CO molecule; every CO gave up its oxygen as water (on cobalt) or as CO₂ (on iron, via the water-gas shift). Fischer-Tropsch is best understood as a surface polymerization: the metal is a workbench that grips one-carbon monomers, strips off their oxygen, and links them into a chain — the same statistical logic that governs any step-growth polymer, which is why the products come out as a spread of chain lengths rather than one clean compound.

The mechanism: surface carbide chain growth

The dominant modern picture is the carbide (or alkyl) mechanism. Everything happens on the metal surface, on adjacent atoms, and the electron bookkeeping is the bookkeeping of chemisorption: bonds to the metal form and break, and hydrogen atoms are transferred one at a time.

1. Dissociative CO adsorption. A CO molecule lands on the metal and its already-weak C≡O bond breaks entirely. You are left with a surface carbon atom (C*) and a surface oxygen atom (O*), each σ-bonded to the metal. This C–O scission is the committed step — it is why FT needs a metal (Fe, Co, Ru) that binds CO strongly enough to break it, and why it does not happen on a metal like copper that only makes methanol.
2. Oxygen removal. Adsorbed hydrogen (from dissociated H₂) attacks the surface oxygen, carrying it off as water: O* + 2 H* → H₂O. On iron the oxygen can instead be swept away by CO through the water-gas shift, leaving CO₂. Either way, the oxygen is gone and a bare surface carbon remains.
3. Build the monomer. That surface carbon is hydrogenated stepwise: C* → CH* → CH₂*. The methylene unit CH₂* is the true building block — the "monomer" of Fischer-Tropsch. Fully hydrogenating it to CH₃* and then CH₄ is the unwanted shortcut that produces methane.
4. Initiate a chain. One methylene picks up a third hydrogen to become a surface methyl, CH₃* — the seed of a growing alkyl chain.
5. Propagate (chain growth). A neighboring CH₂* inserts into the metal–alkyl bond: the growing chain migrates onto the new methylene, extending it by exactly one carbon. R–CH₂* + CH₂* → R–CH₂–CH₂*. This migratory insertion repeats — one carbon per step — and it is the heart of the whole process.
6. Terminate. The chain leaves the surface one of two ways. β-hydride elimination hands a hydrogen from the second carbon back to the metal and releases a terminal 1-alkene (R–CH=CH₂). Alternatively, the chain grabs a surface hydrogen and desorbs as a saturated n-alkane (R–CH₂–CH₃). This is why FT output is a mix of paraffins and α-olefins.

    CO(g) ──► C≡O adsorbs ──► C* + O*        (dissociative adsorption)
    O* + 2 H* ──► H₂O                        (oxygen rejection)
    C* + 2 H* ──► CH₂*                        (make the monomer)
    CH₂* + H* ──► CH₃*                        (initiate: surface methyl)

    propagation (repeat n times):
      R-CH₂*  +  CH₂*  ──►  R-CH₂-CH₂*        (grow the chain by one C)

    termination:
      R-CH₂-CH₂*  ──►  R-CH=CH₂  +  H*        (β-H elimination → 1-alkene)
      R-CH₂-CH₂*  +  H*  ──►  R-CH₂-CH₃       (hydrogenation → n-alkane)

The elegance of this scheme is that a single microscopic coin-flip — grow again versus terminate now — is happening at every carbon, and that single probability controls the entire product slate.

Selectivity: the Anderson-Schulz-Flory distribution

Because chain growth is a repeated probabilistic step, you cannot make a single product. Define the chain-growth probability α as the chance that a given surface chain grows by one more carbon rather than terminating. The chance a chain reaches exactly n carbons and then stops is the geometric outcome of flipping that same weighted coin — this is the Anderson-Schulz-Flory (ASF) distribution:

    xₙ  =  (1 − α) · αⁿ⁻¹                 (mole fraction of n-carbon chains)
    Wₙ  =  n · (1 − α)² · αⁿ⁻¹            (weight fraction of n-carbon chains)

The consequence is unavoidable: the yield of any single carbon number is capped. The gasoline cut (C₅–C₁₁) peaks at only ~48 wt% around α ≈ 0.76; the diesel cut (C₁₀–C₂₀) peaks near ~40 wt% around α ≈ 0.87. Methane (n = 1) is always over-produced relative to the ideal curve because full hydrogenation of the monomer is a competing, non-selective escape. The practical lever is α, and you set it with the catalyst and the conditions:

• High α (0.90–0.95). Cobalt at low temperature (~200–240 °C). Makes mostly long-chain wax (C₂₀+). Wax is not the end product — it is deliberately over-grown so it can be hydrocracked cleanly into diesel and jet fuel with almost no light-ends waste. This is the modern gas-to-liquids strategy.
• Low α (0.70–0.80). Iron at high temperature (~320–350 °C). Skews toward gasoline-range molecules and light olefins (ethylene, propylene) used as chemical feedstock — the classic South-African high-temperature route.

Reagents, catalysts, and conditions

Fischer-Tropsch is really a three-part industrial train: make the syngas, condition it, then run the synthesis.

• The feed. Syngas, CO + H₂, with the ratio tuned toward the stoichiometric ~2.0–2.15 H₂/CO that chain growth consumes. From methane (steam reforming, CH₄ + H₂O → CO + 3H₂) you start H₂-rich; from coal you start H₂-lean and shift it up with water-gas shift.
• The killer poison: sulfur. H₂S and other sulfur species chemisorb irreversibly on the metal and poison it at the parts-per-billion level. The syngas must be desulfurized to below ~50–100 ppb before it ever touches the catalyst — sulfur removal is one of the largest capital costs of a plant.
• Cobalt catalyst. ~15–25 wt% Co dispersed on a high-surface-area oxide (Al₂O₃, SiO₂, TiO₂), promoted with a little Pt, Re, or Ru to help reduce the cobalt and keep it metallic. Very active, low CO₂, long-lived — the choice for clean methane-derived syngas.
• Iron catalyst. Fused or precipitated iron promoted with potassium (which boosts CO adsorption and chain growth) and a structural promoter like SiO₂ or Al₂O₃. Under reaction it converts to iron carbides (the active phase) plus magnetite. Cheap, sulfur-tolerant, water-gas-shift active — the choice for dirty, H₂-lean coal syngas.
• Conditions. Low-temperature FT (LTFT) runs ~200–250 °C, 20–30 bar, for waxes and diesel; high-temperature FT (HTFT) runs ~320–350 °C, ~25 bar, for gasoline and olefins. Pressure raises rate and chain growth; higher temperature raises rate but lowers α and pushes toward methane.
• The reactor. Modern plants favor the slurry bubble-column reactor: fine catalyst suspended in molten product wax, syngas bubbled up through it. The liquid wax is a near-isothermal heat bath that carries away the huge reaction heat into internal steam tubes, avoiding the hot spots that plague fixed beds.

Fischer-Tropsch vs related C1 conversions

	Fischer-Tropsch (Co)	Fischer-Tropsch (Fe)	Methanol synthesis
Feedstock	Syngas (H₂-rich, ~2:1)	Syngas (H₂-lean OK)	Syngas (~2:1) + CO₂
Catalyst	Co / Al₂O₃ + Pt/Re	Fe carbide + K promoter	Cu / ZnO / Al₂O₃
C–O bond broken?	Yes (dissociative)	Yes (dissociative)	No (associative)
Water-gas shift	Negligible	Strong (makes CO₂)	Runs in tandem
Oxygen leaves as	H₂O	CO₂	H₂O
Product	Waxes → diesel/jet	Gasoline + olefins	CH₃OH (one carbon)
Product spread	ASF, α ≈ 0.90–0.95	ASF, α ≈ 0.70–0.80	Single compound
Temperature	200–250 °C	320–350 °C	200–300 °C
Sulfur tolerance	Very low (ppb)	Moderate	Very low (ppb)

The deep contrast is in step 1 of the mechanism. Fischer-Tropsch dissociates CO — it breaks the C–O bond and grows carbon chains. Methanol synthesis on copper keeps the C–O bond intact and simply hydrogenates CO to CH₃OH. That one difference — break the bond or not — is the fork in the road between building long fuels and building a single oxygenate.

Worked example: Pearl GTL, Qatar

The Pearl gas-to-liquids plant in Ras Laffan, Qatar (Shell, on stream 2011) is the largest Fischer-Tropsch plant in the world and a clean illustration of the modern high-α strategy end to end.

• Feedstock. ~1.6 billion cubic feet per day of natural gas from the North Field. Methane is converted to syngas by partial oxidation with pure oxygen (from a giant air-separation unit), giving a clean, sulfur-free, H₂-rich feed — ideal for cobalt.
• Synthesis. Cobalt catalyst in Shell's SMDS (Shell Middle Distillate Synthesis) reactors, run at high α to deliberately over-produce long-chain paraffin wax rather than aiming for diesel directly.
• Upgrading. The heavy wax is then hydrocracked and isomerized — long straight chains are cut to the diesel/jet range and branched to fix cold-flow behavior. Because the wax has no sulfur, no aromatics, and no branching to start with, the hydrocracker runs clean and the yield of on-spec middle distillate is very high.
• Output. ~140,000 barrels per day of GTL liquids — ultra-clean diesel, jet fuel, naphtha, lubricant base oils, and normal paraffins for detergents — plus ~120,000 barrels per day of natural-gas liquids and ethane.

The logic is worth restating: rather than fight the ASF distribution to hit the diesel cut directly (capped at ~40%), Pearl pushes α as high as possible to make wax at high selectivity, then cracks the wax down into diesel — turning the ASF ceiling from an obstacle into a design choice.

Limitations and side reactions

• The ASF ceiling. No single carbon number can dominate. You either accept a broad slate or over-grow to wax and crack back down. This statistical spread is the single defining constraint of the whole technology.
• Methane over-production. Fully hydrogenating the CH₂* monomer to CH₄ is always competing with chain growth, and it worsens with temperature. Methane is the least valuable FT product, so keeping α high and temperature moderate is partly a fight to suppress it.
• Heat management. At ≈ −165 kJ per mole of CH₂, the reaction dumps enormous heat. Poor heat removal creates hot spots that spike methane selectivity and sinter or coke the catalyst — the reason slurry reactors displaced fixed beds.
• Sulfur poisoning. Sub-ppm sulfur permanently kills the metal sites. Feed cleanup to the ppb level is mandatory and expensive.
• Carbon deposition (coking) and fouling. Too little hydrogen or too much temperature lets surface carbon polymerize into inactive graphitic coke or bulk carbide that encapsulates the metal, deactivating iron catalysts especially.
• Carbon and water penalty. Coal-to-liquids via iron FT rejects a large fraction of the feed carbon as CO₂ and consumes a great deal of water in shift and cooling — which is why FT fuels have a heavy carbon and water footprint unless the CO₂ is captured or the carbon source is biomass.
• Capital cost. An FT plant is really three plants stacked (gasification/reforming, gas cleanup, synthesis + upgrading). It is only economic when the feedstock is very cheap (stranded gas, mine-mouth coal) or a fuel-security premium applies.

History: from Mülheim coal to Sasol

The chemistry was born of a scarcity. Germany in the 1920s had abundant coal and almost no oil. At the Kaiser Wilhelm Institute for Coal Research in Mülheim an der Ruhr, Franz Fischer (the institute's director) and Hans Tropsch showed in 1925–1926 that syngas passed over iron and cobalt catalysts at moderate pressure gave a mixture of liquid hydrocarbons — the "synthol"/"kogasin" process. It built directly on the earlier work of Sabatier (metal-catalyzed CO hydrogenation to methane) and ran alongside the higher-pressure Bergius coal-liquefaction route.

Germany industrialized it in the 1930s and operated nine Fischer-Tropsch plants during World War II, producing synthetic fuel and lubricants from coal to feed a war machine cut off from crude oil. After the war the intellectual property spread. In apartheid-era South Africa — coal-rich and, under international oil embargo, oil-starved — the state company Sasol (South African Coal, Oil and Gas Corporation) commercialized coal-based Fischer-Tropsch from 1955, eventually making a large share of the country's transport fuel from coal. Sasol's Secunda complex remains the largest single FT operation on Earth. The modern renaissance came from natural gas: Shell's Bintulu plant in Malaysia (1993) and then Pearl GTL in Qatar (2011) proved that gas-to-liquids, on cobalt, could monetize stranded natural gas as clean liquid fuel.

Industrial and environmental notes

• Why anyone bothers. Fischer-Tropsch is the only mature way to turn solid coal or stranded/flared gas into a drop-in liquid transport fuel. Its niche is energy security (a country with coal but no oil) and monetizing gas too remote to pipeline.
• Fuel quality is genuinely superior. Zero sulfur, zero aromatics, high cetane (70–80). FT jet fuel (synthetic paraffinic kerosene) is certified for blending in commercial aviation, and FT diesel is used to sweeten dirty crude-oil diesel to meet emissions specs.
• The carbon question. Coal-to-liquids FT roughly doubles the well-to-wheel CO₂ of the resulting fuel versus crude oil, because so much feed carbon is rejected as CO₂. The same reactor, fed biomass-derived or CO₂-derived syngas (with green hydrogen), becomes a route to carbon-neutral synthetic e-fuels — an active area for sustainable aviation fuel.
• Safety. The process handles large volumes of carbon monoxide (acutely toxic, odorless) and hydrogen (wide flammability range) at pressure and temperature, plus hot molten wax and high-pressure steam. CO detection, inerting, and rigorous heat-removal control dominate plant safety design.