Supercritical Water as Hydrogen Source for the Hydrothermal Conversion of Fatty Acids to Alkanes: A Comprehensive Review of Experimental Studies

The conversion of fatty acids to fuel-range alkanes using supercritical water (SCW) as both the reaction medium and in-situ hydrogen source has emerged as a promising route for renewable fuel production. Below is a detailed synthesis of the experimental literature on this topic, organized as a comparative table.

Comparative Table of Experimental Studies

Study	Catalyst System	Hydrogen Source	Feedstock	Conditions	Main Products	Byproducts/Side Products	Performance	Key Mechanism	Role of Metal Oxide (if used)
Fu et al. (2011) [1]	Activated carbon (Darco AC-1; Norit AC-2); no noble metals	Supercritical water (no H₂ added); water serves as H-donor via gasification (C + H₂O → CO + H₂; WGS: CO + H₂O → CO₂ + H₂)	Palmitic acid (C16:0); Oleic acid (C18:1)	330–385 °C; near-/supercritical water (density 0.15–0.50 g/cm³); 3 h batch; no H₂ added	Palmitic → C₈–C₁₅ n-alkanes (mainly C₁₅ pentadecane). Oleic → C₁₂–C₁₇ n-alkanes (mainly C₁₇ heptadecane) + some C₁₇ olefins	Stearic acid (from hydrogenation of oleic); C₈–C₁₄ alkanes (cracking); 2-heptanone; minor cracking products	~20–24% conversion (palmitic over AC-1); 9–10% pentadecane yield (AC-1); 19% pentadecane yield (AC-2); 77% mass balance for oleic; Ea = 125 kJ/mol	Decarboxylation is dominant (no C16 from palmitic, ruling out HDO). AC catalyzes both decarboxylation and hydrogenation. Water provides H atoms. At supercritical conditions, higher water density inhibited decarboxylation rate.	No metal oxide; SiO₂ impurity in AC-1 was tested and showed no activity
Fu et al. (2010) [3]	5% Pt/C; 5% Pd/C; also tested: metal salts (NaCl, MnCl₂, ZnCl₂, CoCl₂, CuSO₄, MgSO₄), bases (NaOH, KOH), Raney® Ni, activated carbon	Supercritical/near-critical water (no H₂ added); in situ H₂ detected only from Pt/C runs	Palmitic acid	290–380 °C; water (95% reactor volume filled); batch; no H₂ added	Pentadecane (C₁₅) — >90% selectivity	1-tetracosanol, decylcyclopentane, 4-methyltetradecane, C₈–C₁₆ alkanes	Pt/C: 76% pentadecane yield in 1 h at 370 °C; Pd/C: 63% yield in 3 h at 370 °C; Pt/C reusable ≥3× without loss; Ea = 79 ± 5 kJ/mol; reaction first-order in palmitic acid	Decarboxylation (removal of O as CO₂); Pt/C more active than Pd/C (opposite trend to dodecane solvent, showing solvent effect). Metal dispersion dropped after use but activity maintained.	Not applicable
Fu et al. (2011) [2]	5% Pt/C	Supercritical water (no H₂ added); 30% H₂ molar yield detected in gas phase (proved in situ H₂ formation)	Stearic (C18:0), palmitic (C16:0), lauric (C12:0), oleic (C18:1), linoleic (C18:2) acids	330 °C; water; batch; no H₂ added	Saturated FAs → corresponding C_(N–1) n-alkanes (>90% selectivity); Unsaturated FAs → hydrogenated first to saturated FA, then decarboxylated	Ketones (e.g., tetradecylfuranone); C₁₄–C₁₆ alkanes; aromatics; from linoleic: heavy coupling products, oleic acid intermediate	Saturated FA decarboxylation rates independent of carbon number; oleic → stearic acid yield ~40% after 2 h; heptadecane yield increased steadily; linoleic conversion 97% at shortest time	Sequential hydrogenation → decarboxylation for unsaturated FAs. In situ H₂ via aqueous-phase reforming (Pt/C catalyzes reforming of FA to H₂, CO₂, CH₄, C₂H₆). Water-gas shift consumes CO.	Not applicable
Watanabe et al. (2006) [4]	Alkali hydroxides (NaOH, KOH); metal oxides (CeO₂, Y₂O₃, ZrO₂)	Supercritical water (SCW); SCW stabilizes stearic acid; additives provide catalytic decarboxylation	Stearic acid (C17H₃₅COOH)	400 °C (673 K); 0.17 g/cm³ water (~25 MPa); 30 min batch; no H₂ added	KOH → C₁₇ alkane + CO₂ (monomolecular decarboxylation). Metal oxides → C₁₆ alkene + CO₂	2-Nonadecanone (2-ND, from bimolecular decarboxylation with ZrO₂); carbonyl compounds; CH₄; trace CO	Conversion: SCW alone: 2%; +NaOH: 13%; +KOH: 32%; +CeO₂: 30%; +Y₂O₃: 62%; +ZrO₂: 68%	KOH: monomolecular decarboxylation via carboxylate anion (RCOO⁻ → R⁻ + CO₂); ZrO₂/CeO₂/Y₂O₃: bimolecular decarboxylation producing ketone + CO₂; then ketone decomposes to alkene + shorter acid. SCW stabilizes the FA molecule.	ZrO₂ (most active): provides acid–base bifunctional sites; promotes bimolecular decarboxylation between two FA molecules (or FA + CH₃COOH) to form ketone (2-ND). CeO₂ and Y₂O₃ also have acid–base sites. Y₂O₃ converted to YOOH during reaction.
Hossain et al. (2018) [6]	Activated carbon (Darco G-60); no noble metal	Subcritical water reacting with C catalyst (C + H₂O → CO + H₂; WGS: CO + H₂O → CO₂ + H₂); also FA thermal cracking yields H₂	Oleic acid	300–400 °C; continuous fixed-bed; <500 psi; water-to-OA ratio 2:1–5:1; space time 0.25–2.5 h; no H₂ added	Heptadecane (89.3% selectivity at max conditions)	Heptadecene (0.5% at max); stearic acid (9.4% unconverted); C₈–C₁₆ and C₁₈–C₂₀ alkanes; CH₄; CO; CO₂; H₂	Max decarboxylation: 91% at 400 °C, 2 h, water:OA = 4:1; liquid yield 63.5%; HHV = 45.0 MJ/kg (close to diesel); density comparable to kerosene/diesel	Dual role of AC: (1) catalyzes decarboxylation of stearic acid → C₁₇ + CO₂; (2) reacts with water (C + H₂O → CO + H₂) producing in situ H₂ that hydrogenates oleic → stearic. AC is partially consumed (~4.5 wt% loss).	Not applicable (no metal oxide)
Zhang et al. (2019) [5]	5% Ru/C; also tested Pt/C, Pd/C, Rh/C	Supercritical water; in situ H₂ via aqueous-phase reforming of glycerol (from triglyceride hydrolysis) and alkane reforming	Stearic acid; 1,2-distearoyl-3-palmitoyl-sn-glycerol (lipid)	330 °C; 1–5 MPa N₂ or H₂ headspace; batch; no H₂ added	Stearic → C₇–C₁₇ n-alkanes (mixture); Lipid → C₇–C₁₇ alkanes (faster conversion)	CH₄, CO₂, H₂, C₂H₆, C₃H₈; trace C₁₈ (HDO product)	Activity: Pt/C > Ru/C > Pd/C ≈ Rh/C. Ru/C: complete conversion in 12.5 h (1 MPa N₂); TOF₀ = 17.5 h⁻¹. Under H₂: complete conversion in 1 h (TOF₀ = 131.5 h⁻¹). Lipid: full conversion in 2.5 h (2 MPa N₂). Ru/C deactivates after reuse.	Decarboxylation (DOX) followed by sequential cracking of C₁₇ → shorter alkanes + CH₄. Ru/C promotes parallel reactions: DOX, decarbonylation, cracking, methanation, WGS. For lipids: glycerol reforming provides extra H₂. Higher H₂ pressure enhances both DOX and cracking.	Not applicable
Konwar & Mikkola (2022) [7]	Ru, Pt, Pd on carbon supports with varying acid–base properties (ACPcomm, NC, NAC, AC)	Subcritical water; in situ H₂ via glycerol aqueous-phase reforming and WGS	Tristearin (saturated C18 triglyceride)	256–326 °C; subcritical water; batch reactors; no H₂ added	Ru-based → C₅–C₂₀ n-alkanes + iso-alkanes (7–8% iso); Pt/Pd-based → mainly n-C₁₇ (85–92% selectivity)	CO₂, H₂, CH₄, C₁–C₅ hydrocarbons; cracking products; isomerized products (with Ru)	TOF order: Ru > Pt > Pd (all supports). Ea: Ru (99 kJ/mol) < Pt (121 kJ/mol) < Pd (146 kJ/mol). Pt/Pd activity positively correlated with support basicity; Ru activity independent of support acid–base properties.	Pt/Pd: decarboxylation dominant pathway; Ru: parallel DOX + decarbonylation + cracking + isomerization + methanation. H₂ is co-produced. Basic supports (N-doped) enhance Pt/Pd stability and activity; acidic supports promote coking/deactivation.	Not applicable (carbon supports only, but N-doping of carbon provides basic sites that stabilize metal nanoparticles and reduce leaching)
Li et al. (2015) [8]	B(C₆F₅)₃ (homogeneous Lewis acid); no heterogeneous catalyst; uses PMHS (poly(methylhydrosiloxane)) as reductant	Silane (PMHS), not supercritical water	Fatty acids (lauric, myristic, palmitic, stearic, oleic), their methyl esters, triglycerides	Room temperature; CH₂Cl₂ or cyclohexane solvent; 6 h; mild conditions	Saturated FAs → corresponding n-alkanes (dodecane to octadecane, 81–93% yields); Oleic acid → 9-octadecene (80%); Triglycerides → alkanes	Siloxane residues (from PMHS conversion)	93% dodecane from lauric acid; 84% octodecane from stearic acid; 95% octodecane from tristearin (10 mol% catalyst); real oils (colza, olive) successfully converted	Hydrosilylative reduction — B(C₆F₅)₃ activates Si–H bond of silane, which reduces carboxylic acid/ester sequentially to silyl ethers, then to alkane. No decarboxylation; complete deoxygenation via C–O bond cleavage.	Not applicable (homogeneous catalysis; no metal oxide)
Yang et al. (2017) [9]	Ni-MOFs (Ni-BTC, Ni-BM65, Ni-BM73) grown on zeolite 5A beads; no noble metal	CO₂ atmosphere (20 bar); not supercritical water; H-source not supercritical water	Oleic acid	340 °C; 20 bar CO₂; 2 h batch; no H₂ added	Heptadecane (up to ~77% selectivity with Ni-BM65/zeolite 5A)	Branched paraffins (isomerization); C₇–C₁₆ alkanes (cracking); octadecane	Conversion ~90%; heptadecane selectivity: Ni-BM65/5A (77%) > Ni-BM73/5A (65%) > Ni-BTC/5A (63%). Zeolite 5A alone gave only ~14% heptadecane. Catalyst recyclable with some loss.	Decarboxylation at Ni sites in MOF; acid sites on zeolite 5A support enhance selectivity. Acid site density correlates with activity. Micropore volume of MOF enables mass transport. CO₂ atmosphere may suppress side reactions.	Not applicable (Ni-MOF with zeolite 5A support; Ni²⁺ is active metal center; CO₂ atmosphere used instead of supercritical water)

Key Findings and Mechanistic Insights

1. Supercritical Water as a Source of Hydrogen

A central finding across the hydrothermal decarboxylation literature is that water itself serves as the hydrogen source without requiring external H₂ addition. Multiple pathways have been proposed:

Carbon gasification (with carbon-based catalysts): For activated carbon catalysts, water reacts with the carbon surface via C + H₂O → CO + H₂, followed by water–gas shift (CO + H₂O → CO₂ + H₂). Hossain et al. [6] directly demonstrated this by passing only water over activated carbon at 400 °C, producing measurable H₂, accompanied by ~4.5 wt% consumption of the carbon catalyst.
Aqueous-phase reforming (with metal catalysts): Over Pt, Pd, and Ru catalysts, a small fraction of the fatty acid or (in the case of triglycerides) the glycerol co-product undergoes reforming to produce H₂, CO₂, and light hydrocarbons. Fu et al. [2] directly measured 30% H₂ molar yield from oleic acid over Pt/C at 370 °C. Zhang et al. [5] showed that when processing a triglyceride (lipid), the glycerol released by hydrolysis produces nearly double the H₂ compared to stearic acid alone, accelerating decarboxylation.
Water as H-donor: In the case of activated carbons, Fu et al. [1] noted that hydrogenation of the C=C bond in oleic acid occurred without any added H₂, implying that water molecules (or H atoms derived from water) donate hydrogen to the unsaturated substrate.

2. Dominant Deoxygenation Pathway: Decarboxylation vs. Hydrodeoxygenation

Across all studies using supercritical water without added H₂, decarboxylation (DOX) — removal of oxygen as CO₂, producing a hydrocarbon with one fewer carbon (Cₙ→Cₙ₋₁) — is the dominant pathway. Evidence includes:

No observation of C₁₆ (hexadecane) from palmitic acid (C₁₆) [1,3], confirming absence of hydrodeoxygenation (HDO, which would yield C₁₆).
High selectivity (>85–90%) to the Cₙ₋₁ alkane from saturated fatty acids over Pt/C [2,3].
Detection of CO₂ (not CO or H₂O) as the main oxygen-containing gaseous product.

3. Catalyst Performance Comparison

Catalyst	Typical Conditions	Activity/Conversion	Selectivity	Key Advantage	Key Limitation
Activated carbon (alone)	370–400 °C, SCW	20–24% (palmitic, 3 h) [1]; 91% (oleic, continuous, 2 h) [6]	40–50% (batch); 89% C₁₇ (continuous)	Low cost, no noble metal	Lower activity; some carbon is consumed; deactivates after ~30 h on stream
5% Pt/C	290–380 °C, SCW	76% in 1 h (370 °C) [3]; near-complete in 6 h at 290 °C	>90% to Cₙ₋₁ alkane	Highest activity; excellent reusability (≥3 cycles)	High cost of Pt; metal dispersion decreases in hydrothermal conditions
5% Pd/C	330–370 °C, SCW	63% in 3 h (370 °C) [3]	>90% to Cₙ₋₁ alkane	Lower cost than Pt	Less active than Pt in water (opposite trend to organic solvents)
5% Ru/C	256–330 °C, SCW	Complete conversion in 12.5 h (330 °C, 1 MPa N₂) [5]; TOF₀ = 17.5 h⁻¹	C₇–C₁₇ mixture (cracking)	Low cost; produces fuel-range mixture (diesel-like); active at lower temperatures	Deactivates significantly; produces more gaseous products
Metal oxides (ZrO₂, CeO₂, Y₂O₃)	400 °C, SCW	68% (ZrO₂, 30 min) [4]	C₁₆ alkene (not alkane)	Low cost; reusable oxides	Produces unsaturated alkenes, not alkanes; requires further hydrogenation
KOH/NaOH	400 °C, SCW	32% (KOH) [4]	C₁₇ alkane (monomolecular)	Simple homogeneous base	Homogeneous, non-recyclable; corrosive

4. Role of Metal Oxides (in SCW Systems)

The study by Watanabe et al. [4] specifically investigated metal oxides in supercritical water. Key findings:

ZrO₂ was the most active (68% conversion) and produced predominantly C₁₆ alkene (not alkane), via a proposed bimolecular decarboxylation mechanism where two fatty acid molecules react to form a long-chain ketone (2-nonadecanone) and CO₂. The ketone further decomposes to C₁₆ alkene and a shorter acid.
The acid–base bifunctional nature of ZrO₂ (both Lewis acid and base sites on its surface) was critical for this bimolecular pathway.
CeO₂ and Y₂O₃ behaved similarly but with lower activity.
Notably, these metal oxides produced alkenes rather than alkanes — they did not provide hydrogenation activity — unlike carbon-based catalysts (activated carbon, Pt/C, Ru/C) which produced fully saturated alkanes.

5. Effect of Support Surface Chemistry

Konwar and Mikkola [7] systematically demonstrated that the acid–base properties of carbon supports strongly influence catalyst performance in hydrothermal decarboxylation:

For Pt and Pd catalysts, activity positively correlated with basic surface sites (e.g., N-containing functional groups) and negatively with acidic sites (e.g., phosphate groups). Basic supports stabilized metal nanoparticles, reduced leaching, and suppressed coking.
For Ru catalysts, activity was independent of support acid–base properties, but stability (reusability) improved on basic supports.
Operating at lower temperatures (≤256 °C) markedly improved catalyst lifetime by reducing hydrothermal degradation of the support and metal leaching.

6. Reaction Pathways: Saturated vs. Unsaturated Fatty Acids

A consistent finding across studies is that unsaturated fatty acids follow a sequential hydrogenation–decarboxylation pathway [1,2,5]:

Hydrogenation of the C=C bond (using in situ H₂ derived from water) occurs first, producing the saturated fatty acid.
The saturated fatty acid then undergoes decarboxylation to the corresponding Cₙ₋₁ n-alkane.

This explains why oleic acid (C18:1) ultimately produces heptadecane (C₁₇), not heptadecene. The hydrogenation step is faster than direct decarboxylation of the unsaturated acid.

7. Product Distribution Tuning

An important advantage of the hydrothermal approach is the ability to tune product distribution:

Pt/C and Pd/C produce a narrow product range (nearly pure Cₙ₋₁ alkane) through selective decarboxylation [3].
Ru/C produces a broader C₇–C₁₇ mixture through coupled decarboxylation + cracking, more closely resembling petroleum-derived diesel and jet fuel [5].
Reaction parameters such as temperature, water density, and headspace pressure significantly influence product yields. For instance, Fu et al. [1] showed that higher water density in supercritical conditions (0.31–0.50 g/cm³) inhibited decarboxylation, while lower density (0.15 g/cm³) gave higher pentadecane yields.

8. Comparison with Non-Hydrothermal Approaches

For completeness, two additional papers not using supercritical water as the hydrogen source are included in the table:

Li et al. (2015) [8] demonstrated a metal-free homogeneous system using B(C₆F₅)₃ and silane (PMHS) at room temperature. This system achieves high alkane yields but requires stoichiometric silane as the hydrogen source, not water.
Yang et al. (2017) [9] used Ni-MOF/zeolite 5A catalysts under CO₂ atmosphere for oleic acid decarboxylation, achieving ~77% heptadecane selectivity without noble metals — but again without using supercritical water as the hydrogen source.

These approaches are mechanistically distinct from the hydrothermal (SCW-based) route but are relevant for comparison, particularly regarding the goal of eliminating noble metals and external H₂.

Summary

The body of experimental literature demonstrates that supercritical water can serve simultaneously as the reaction medium and as the hydrogen source for converting fatty acids to alkanes. Activated carbon, Pt/C, Pd/C, and Ru/C all catalyze this transformation without added H₂, with water providing hydrogen through carbon gasification (with carbon catalysts) or aqueous-phase reforming (with metal catalysts). Decarboxylation is the dominant deoxygenation pathway. Metal oxides (ZrO₂, CeO₂, Y₂O₃) in SCW promote decarboxylation but produce alkenes rather than alkanes, as they lack hydrogenation functionality. The choice of catalyst determines the product distribution — from narrow-range (Pt/C, Pd/C) to broad-range hydrocarbon mixtures suitable as drop-in fuels (Ru/C, activated carbon in continuous mode).

References

[1]Fu, J.; Shi, F.; Thompson, L. T., Jr.; Lu, X.; Savage, P. E. Activated Carbons for Hydrothermal Decarboxylation of Fatty Acids. ACS Catalysis 2011, 1 (3), 227–231
DOI: 10.1021/cs1001306 [2]Fu, J.; Lu, X.; Savage, P. E. Hydrothermal Decarboxylation and Hydrogenation of Fatty Acids over Pt/C. ChemSusChem 2011, 4 (4), 481–486
DOI: 10.1002/cssc.201000370 [3]Fu, J.; Lu, X.; Savage, P. E. Catalytic Hydrothermal Deoxygenation of Palmitic Acid. Energy & Environmental Science 2010, 3 (3), 311–317
DOI: 10.1039/b923198f [4]Watanabe, M.; Iida, T.; Inomata, H. Decomposition of a Long Chain Saturated Fatty Acid with Some Additives in Hot Compressed Water. Energy Conversion and Management 2006, 47 (18–19), 3344–3350
DOI: 10.1016/j.enconman.2006.01.009 [5]Zhang, J.; Huo, X.; Li, Y.; Strathmann, T. J. Catalytic Hydrothermal Decarboxylation and Cracking of Fatty Acids and Lipids over Ru/C. ACS Sustainable Chemistry & Engineering 2019, 7 (17), 14400–14410
DOI: 10.1021/acssuschemeng.9b00215 [6]Hossain, Md. Z.; Chowdhury, M. B. I.; Jhawar, A. K.; Xu, W. Z.; Charpentier, P. A. Continuous Low Pressure Decarboxylation of Fatty Acids to Fuel-Range Hydrocarbons with in Situ Hydrogen Production. Fuel 2018, 212, 470–478
DOI: 10.1016/j.fuel.2017.09.092 [7]Konwar, L. J.; Mikkola, J.-P. Carbon Support Effects on Metal (Pd, Pt and Ru) Catalyzed Hydrothermal Decarboxylation/Deoxygenation of Triglycerides. Applied Catalysis A: General 2022, 638, 118611
DOI: 10.1016/j.apcata.2022.118611 [8]Li, X.-Y.; Shang, R.; Fu, M.-C.; Fu, Y. Conversion of Biomass-Derived Fatty Acids and Derivatives into Hydrocarbons Using a Metal-Free Hydrodeoxygenation Process. Green Chemistry 2015, 17 (5), 2790–2793
DOI: 10.1039/c5gc00556f [9]Yang, L.; McNichols, B. W.; Davidson, M.; Schweitzer, B.; Gómez-Gualdrón, D. A.; Trewyn, B. G.; Sellinger, A.; Carreon, M. A. Noble Metal-Free Catalytic Decarboxylation of Oleic Acid to n-Heptadecane on Nickel-Based Metal–Organic Frameworks (MOFs). Catalysis Science & Technology 2017, 7 (14), 3027–3035
DOI: 10.1039/c7cy00564d

Echoes Of Day

Bu Blogda Ara