Decision in brief
For a first cycloalkane-oriented phenolic-HDO screen, Ni on a characterized ZnO–Al₂O₃ composite is the most defensible practical starting point; Ni/ZnFe₂O₄ is the higher-risk, mechanistically interesting challenger. This is a research-priority recommendation—not a proven performance ranking. Ni–Zn–Al mixed oxides have been tested on a phenolic substrate, vanillin, but that study principally removed an aldehyde oxygen and retained aromatic methoxy and hydroxyl groups. It does not prove deep HDO to hydrocarbons. Ni–Fe catalysts have direct phenol-HDO precedent, but that precedent is not a demonstration of Ni supported on zinc ferrite. [1] [2]
Four-way answer: (1) Ni/ZnO–Al₂O₃: first practical composite screen; (2) Ni/ZnAl₂O₄: spinel-rich comparator, not proven better or worse; (3) Ni/ZnFe₂O₄: potentially valuable Ni–Fe–Zn chemistry, with greater uncertainty about reduction and working phase; (4) Ni/ZnO–iron oxide: valuable phase-mixture challenger, but its exact identity must be specified. No retrieved phase-isolated, head-to-head phenolic-HDO experiment ranks these four. An observed ranking must be established experimentally.
Formula clarification. “ZnO–Fe₂O₄” does not specify a conventional charge-balanced binary iron oxide: Fe₂O₄ requires a counter-cation in a spinel formula. This review treats it provisionally as a proposed ZnO + ZnFe₂O₄ composite, while also considering ZnO + Fe₂O₃ as an alternative synthesis. Those are not equivalent. Identify the intended oxide and molar proportions before preparing or naming the sample.
Evidence boundary. Twelve distinct research papers are assessed individually below. Two directly address phenolic conversion with Ni–Zn–Al or Ni–Fe systems, but neither is a matched four-support comparison. Several other papers are substrate-matched with different support, or support-matched with different chemistry. Direct evidence for Ni/ZnFe₂O₄ phenolic HDO was not located in this focused search; this is not a proof of absence. Some items were available only as abstracts/records. No confidential experimental results are published here.
1 · Define what “good” means
For cyclohexane and alkylcyclohexanes: favor high carbon-molar yield of oxygen-free products at matched conversion, adequate H₂ activation, sufficient C–O cleavage and acceptable carbon balance and stability. Cyclohexanol and methoxycyclohexane remain oxygenated even when their aromatic rings are fully hydrogenated. The relevant quantity is product yield, not cyclohexane selectivity in isolation. For aromatics, minimize ring hydrogenation and C–C cleavage while promoting oxygen removal; a different Ni–support environment may be better. [6] [7]
For selective phenolic chemicals, retaining one or more oxygen-containing groups can be desirable. Vanillin conversion to 2-methoxy-4-methylphenol is a useful side-chain oxygen-removal reaction, but the product still contains phenolic OH and methoxy oxygen. It must not be interpreted as complete deoxygenation or cycloalkane yield. [1]
2 · Materials and comparative assessment
A · Ni/ZnAl₂O₄
What it is: Ni added to a zinc-aluminate spinel support. The nominal stoichiometry alone does not determine surface termination, inversion, residual ZnO, Ni location or reduction behavior.
Why test: a phase-defined spinel-rich reference helps ask whether a separate ZnO-rich phase matters. Surface acidity varies with synthesis and thermal history, and Ni–support interaction can stabilize small particles but also alter Ni reducibility. Zn-containing environments may change Ni chemistry. [10] [11]
Limitation: no sufficiently matched Ni/ZnAl₂O₄ phenolic-HDO result was verified. Do not assume it is inert or that more acidity is always better.
B · Ni/ZnO–Al₂O₃
What it is: Ni on a composite; calcination can create ZnO, ZnAl₂O₄ and residual alumina in proportions that vary with preparation. A label does not establish that all three are present.
Why first: it permits systematic variation of Zn/Al and Ni–Zn–alumina contact. The NiZnAl-derived vanillin study is phenolic-feed evidence for that broader composition family, although it does not isolate a spinel or deep oxygen removal. Zn/Al controls Ni reducibility and dispersion in an oleic-acid study, which is useful characterization context, not a phenolic optimum. [1] [8]
Risk: strong Ni–Al bonding can reduce accessible Ni; excessive Zn or Ni–Zn alloy/oxide decoration can change the number and type of accessible metal sites. Higher reducibility alone is not proof of higher HDO activity. [11]
C · Ni/ZnFe₂O₄
What it is: Ni deposited on zinc ferrite, nominally Zn²⁺Fe³⁺₂O₄ before reduction. Ni can interact with Fe-bearing domains as well as Zn species.
Why test: Ni–Fe alloy systems can promote phenol conversion and subsequent cyclohexanol transformation on carbon; Fe-modified Ni also has guaiacol-HDO precedent on BEA. These establish a Ni–Fe hypothesis, not the behavior of zinc ferrite as a support. [2] [3]
Risk: the reduced/working material may differ from the initial spinel because of reduction, segregation, interdiffusion or possible alloy formation. Such changes must be measured under the actual activation conditions; none is inevitable merely from the starting formula.
D · Ni/ZnO–iron oxide
What it is: specify whether ZnO + Fe₂O₃, ZnO + Fe₃O₄ or ZnO + ZnFe₂O₄ is intended. A physical mixture and a co-precipitated composite can have different interfacial contact and evolve differently on reduction.
Why test: a controlled mixed-phase reference can reveal whether nominal ferrite, separate oxide surfaces or Ni–Fe interface formation matters. Ni–Fe/Al₂O₃ evidence shows Fe effects depend on temperature, composition and time on stream—not just the presence of Fe. [9]
Risk: Fe redox chemistry, changes in Ni particle size, leaching and carbon deposition may alter long-term behavior. Check them rather than attributing every effect to acidity.
3 · What the chemistry suggests—and what it cannot prove
Ni hydrogenation and the oxygen-removal bottleneck
Ni can dissociate H₂ and hydrogenate phenolic rings. On Ni/Al₂O₃, Dutta and colleagues report parallel ring-hydrogenation and acid-associated adsorption/demethylation pathways for anisole. A hydrogenated oxygenate is not the same as a deoxygenated hydrocarbon. Product distribution reflects formation, subsequent conversion and desorption, and varies with pressure, temperature, solvent and time. [6]
What Zn can plausibly do
Zn can alter support acidity and Ni speciation, particle dispersion and adsorption; metallic Ni–Zn alloy, oxidic ZnO and zinc aluminate are distinct states. In the NiZnAl-derived vanillin study the authors discuss Ni–Zn alloy and oxide acidity, but the experiment does not isolate which proposed factor dominates. Ni–Zn/SiO₂ research on anisole is a relevant lead, yet the accessible record does not justify transferring a specific product trend to Zn–Al supports. [1] [5]
What Fe can plausibly do
On carbon-supported Ni–Fe nanoparticles, Han and colleagues associated improved phenol-to-cyclohexane performance with Ni–Fe sites and enhanced cyclohexanol adsorption, proposing cooperative hydrogen activation and oxygenate interaction. This is direct evidence for that alloy/carbon system; a Ni/ZnFe₂O₄ catalyst may form different phases and cannot inherit this mechanism by assumption. [2]
Acidity is a balance, not a single optimum
Acid sites can participate in adsorption, dehydration or ether conversion, but may also alter side reactions. Compare NH₃-TPD at matched pretreatment, and use pyridine-IR or another discriminating probe to assess Brønsted versus Lewis acidity; NH₃-TPD alone does not distinguish them. ZnO–ZnAl₂O₄ acidity changes with composition, and the shape and integrated quantity depend on preparation. [10] [12]
4 · Paper-by-paper evidence and contribution
1. Yue et al. · NiZnAl-derived oxides and vanillin [1]
What they did: prepared Ni–Zn–Al mixed-oxide catalysts from layered double hydroxide precursors and investigated conversion of lignin derivatives. For the reported Ni₀.₅Zn₁.₅Al₁-derived catalyst, vanillin conversion and selectivity to 2-methoxy-4-methylphenol were both reported as 99% at 130 °C, 1 MPa H₂ and 2 h.
Major contribution: direct phenolic-substrate precedent for the Ni–Zn–Al composition family; links preparation, interfacial properties and selective side-chain oxygen removal. Limit: product retains two oxygen-containing functional groups. This is not proof of hydrocarbon fuel production, not a phase-isolated ZnAl₂O₄ study, and not a match to an anisole experiment. Evidence inspected: publisher article excerpts and abstract; solvent and pressure basis not verified here.
2. Han et al. · Ni–Fe particles and phenol [2]
What they did: compared Ni–Fe nanoparticles on mesoporous carbon spheres against monometallic Ni and Fe for phenol HDO. The tested Ni/Fe 3:1 formulation gave the highest cyclohexane yield among their ratios; they investigated the sequence phenol → cyclohexanol → cyclohexane.
Major contribution: establishes a concrete Ni–Fe cooperative mechanism and demonstrates the importance of downstream alcohol removal. Limit: carbon-supported alloy, not zinc ferrite; 3:1 is not a predicted optimum for a ZnFe₂O₄-based catalyst. Evidence: institutional primary abstract/record.
3. Fe/Mo/W-modified Ni/BEA and guaiacol [3]
What they did: studied promoter effects on Ni/BEA structure and catalytic guaiacol hydrodeoxygenation, including Fe. Major contribution: a substrate-matched Ni–Fe promoter comparison relevant to oxygenated aromatic substrates. Limit: zeolite acid sites and formulation differ from ferrite, and the retrieved record does not support transferring numerical selectivity or operating conditions. Evidence: publisher record and supporting-information record; full text not inspected.
4. Loricera et al. · ZnNi catalysts and phenol [4]
What they did: examined supported ZnNi catalysts for oxygen removal from bio-liquids and aromatics hydrogenation, including phenol-HDO support comparisons. Major contribution: a direct phenol and Ni–Zn precedent with support dependence. Limit: not a Zn–Al versus Zn–Fe support comparison; a ZnNi phase cannot be assumed in an uncharacterized Ni/ZnO sample. Evidence: bibliographic record and abstract-level source; detailed conditions not inspected.
5. Wang et al. · Zn modification of Ni/SiO₂ for anisole [5]
What they did: investigated the effect of Zn on Ni/silica in anisole HDO. Major contribution: a particularly relevant substrate–promoter precedent to retrieve for details. Limit: silica is not alumina or ferrite; the accessible record here verifies the study but not its detailed results. Evidence: bibliographic record only.
6. Dutta et al. · anisole pathways on Ni/Al₂O₃ [6]
What they did: studied 10 wt% Ni/Al₂O₃ and anisole adsorption/reaction, including a methoxycyclohexane pathway probe. The abstract reports reactive adsorption on alumina acid sites and demethylation between 200 and 300 °C.
Major contribution: separates support-mediated and Ni-mediated pathway questions for a practical reference catalyst. Limit: no Zn or Fe; different conditions cannot be assumed equivalent. Evidence: publisher abstract and excerpts.
7. Zhang et al. · Ni support comparison for phenolics [7]
What they did: compared Ni/γ-Al₂O₃ and Ni/SiO₂ for conversion of phenolic compounds and mixtures, examining product distribution and carbon deposition. Major contribution: shows that the support can redirect reaction pathways and stability even with the same nominal metal. Limit: neither support is Zn–Al or Zn–Fe; reaction details and mixed-feed outcomes cannot be reduced to one universal support ranking. Evidence: publisher abstract/record.
8. Chen et al. · Zn/Al ratio, Ni and oleic acid [8]
What they did: varied the Zn/Al ratio in Ni/ZnO–Al₂O₃ for oleic-acid deoxygenation. The authors report improved Ni reducibility at higher Zn/Al but a dispersion penalty at excessive Zn; Zn/Al 2:1 performed best in their series.
Major contribution: shows why accessible Ni and reducibility must be measured together. Limit: oleic acid is not a phenolic ether, and the cited study reports substantial decarbonylation chemistry. Do not transfer its composition optimum or call all alkane formation HDO. Evidence: publisher/accepted manuscript excerpts.
9. Vajglova et al. · Fe–Ni/Al₂O₃ and dihydroeugenol [9]
What they did: compared Ni-, Fe- and Fe–Ni-containing catalysts for solventless continuous dihydroeugenol upgrading at 275–300 °C and reported 30 bar H₂. An FeNi/Al₂O₃ formulation and a NiCeAl catalyst gave substantial oxygen-free product at 300 °C.
Major contribution: phenolic-substrate, continuous-flow evidence that metal combination, support and time on stream matter. Limit: it is not a zinc-ferrite study; the 275 °C comparisons also show monometallic Ni can outperform a particular FeNi formulation, so Fe is not automatically beneficial. Evidence: open publisher article and institutional record; pressure basis beyond reported wording not checked.
10. Matveyeva et al. · ZnO–ZnAl₂O₄ support acidity [10]
What they did: studied ZnO–zinc aluminate nanocomposites with NH₃-TPD and other characterization for dehydrogenation catalyst supports. Major contribution: direct support-family evidence that ZnO fraction changes measured acidity; their Fig. 6 displays NH₃-TPD profiles. Limit: no phenolic HDO ranking; a ZnO–spinel mixture is not automatically identical to a ZnO–Al₂O₃ preparation. Evidence: open article text and figure caption.
11. Sokolov et al. · Ni on ZnAl oxides in methane reforming [11]
What they did: examined Ni supported on Zn–Al mixed oxides for CO₂ reforming. The authors reported easier reduction on Zn-rich materials and excess uptake relative to NiO-only stoichiometry, attributed in part to Zn chemistry and alloy formation.
Major contribution: warns that H₂-TPR uptake is not necessarily Ni-only and that nominal supports can affect the reduced metal state. Limit: methane reforming conditions are not phenolic HDO and TPR alone does not prove an alloy. Evidence: article record/excerpts.
12. Taghvaei et al. · Ni on modified alumina–silica for anisole [12]
What they did: examined anisole conversion over Ni on alumina–silica supports with and without plasma treatment, using NH₃-TPD and other characterization. Major contribution: illustrates concurrent changes in support acidity, texture and Ni dispersion that complicate a support comparison. Limit: different support and low-pressure reaction context, with phenol and benzene among major products. Evidence: open article abstract and text excerpts.
5 · A practical experiment to choose, rather than guess
Stage A: define the materials. Prepare (i) Ni/γ-Al₂O₃ baseline, (ii) Ni/ZnO–Al₂O₃, (iii) Ni/ZnAl₂O₄, (iv) Ni/ZnFe₂O₄ and (v) the intended Ni/ZnO–iron oxide composite, plus bare-support blanks and Ni/ZnO and Ni/Fe₂O₃ controls where feasible. Specify oxide formula, support preparation, calcination, actual Zn:Al or Zn:Fe, and whether the metal is impregnated after support synthesis.
Stage B: normalize fairly. As a proposed screening point—not a literature optimum—use the same nominal elemental Ni loading, catalyst mass and particle size range. Report actual elemental compositions and surface area. Run both identical reduction history and, if justified, separately optimized activation; otherwise a single pretreatment can systematically disadvantage one material. Measure accessible Ni sites and report rate per gram and per accessible Ni where meaningful. Metal dispersion and phase may change, so equal loading is not equal site count.
Stage C: benchmark the reaction. Start with anisole for methoxy chemistry and phenol or guaiacol as an additional challenge. Use the same temperature, reactor, solvent and reaction duration for all samples; specify whether H₂ pressure is a cold initial charge, hot operating pressure, total pressure or H₂ partial pressure. Keep mixing and transport conditions comparable. Collect repeated time points and compare both equal time and matched conversion.
Stage D: measure the correct outputs. Report conversion, each product selectivity and carbon-molar yield, oxygen-removal degree with a defined basis, carbon balance including gas, hydrogen consumption where possible, and stability/reuse. Ring hydrogenation is not oxygen removal. Analyze methoxycyclohexane and cyclohexanol as separate feeds to test downstream steps. Compare replicates and error bars.
Stage E: identify the working state. Combine fresh/reduced/spent XRD, elemental analysis, BET, H₂-TPR with blanks, NH₃-TPD plus pyridine-IR, electron microscopy/EDS and appropriate Ni/Fe/Zn XAS or XPS; Mössbauer spectroscopy can help distinguish Fe environments where available. Test leaching if liquid-phase operation is used. Use operando methods where feasible. A catalyst starting as spinel may not retain that phase after activation; neither XRD non-detection of nanoparticles nor TPR alone establishes absence or identity.
6 · Scope, search and confidence
This is a relevance-selected, non-exhaustive search of publisher records, institutional repositories, open article text and DOI/Crossref metadata on 5 November 2026. Query families combined Ni, ZnAl₂O₄, ZnO–Al₂O₃, ZnFe₂O₄, iron oxides, phenol, anisole, guaiacol and hydrodeoxygenation. References are distinct original studies, not a citation-ranked top list. Full texts or supporting information were not accessible or independently inspected for all entries; evidence levels are stated per paper. The specific four-way Ni/support ranking remains an untested recommendation, and not finding a paper is not a novelty claim.
References · 12 original papers
- [1] Yue, Zhang, Sun, Gao, Gao, Cheng, Shang, Gao, Wang (2021). Highly efficient hydrodeoxygenation of lignin-derivatives over Ni-based catalyst. Applied Catalysis B: Environmental. DOI: 10.1016/j.apcatb.2021.120243
- [2] Han, Rehman, Wang, Rykov, Gutiérrez, Zhao, Wang, Ma, Lercher (2019). The synergistic effect between Ni sites and Ni-Fe alloy sites on hydrodeoxygenation of lignin-derived phenols. Applied Catalysis B: Environmental. DOI: 10.1016/j.apcatb.2019.04.065
- [3] Yan, Tian, Kennedy, Tkachenko, Stockenhuber (2021). Influence of Promoters (Fe, Mo, W) on the Structural and Catalytic Properties of Ni/BEA for Guaiacol Hydrodeoxygenation. ACS Sustainable Chemistry & Engineering. DOI: 10.1021/acssuschemeng.1c06266
- [4] Loricera, Castaño, Infantes-Molina, Hita, Gutiérrez, Arandes, Fierro, Pawelec (2012). Designing supported ZnNi catalysts for the removal of oxygen from bio-liquids and aromatics from diesel. Green Chemistry. DOI: 10.1039/C2GC35901D
- [5] WANG, GU, SHI, CHEN (2022). Effect of Zn on performance of Ni/SiO2 for hydrodeoxygenation of anisole. Journal of Fuel Chemistry and Technology. DOI: 10.1016/S1872-5813(22)60029-X
- [6] Dutta, Shumeiko, Aubrecht, Karásková, Fridrichová, Pacultová, Hlinčík, Kubička (2024). Evaluation of anisole hydrodeoxygenation reaction pathways over a Ni/Al2O3 catalyst. Journal of Catalysis. DOI: 10.1016/j.jcat.2024.115553
- [7] Zhang, Tang, Zhang, Wang, Ma (2018). Hydrodeoxygenation of lignin-derived phenoic compounds to hydrocarbon fuel over supported Ni-based catalysts. Applied Energy. DOI: 10.1016/j.apenergy.2017.08.078
- [8] Chen, Zhang, Li, Li (2017). Effect of Zn/Al ratio of Ni/ZnO-Al2O3 catalysts on the catalytic deoxygenation of oleic acid into alkane. Applied Catalysis A: General. DOI: 10.1016/j.apcata.2016.11.012
- [9] Vajglová, Yevdokimova, Gallego-Villada, Vondrová, Martinez-Klimov, Simakova, Eränen, Tirri, Peuronen, Lastusaari, Manabayeva, Tungatarova, Yergaziyeva, Dossumov, Lindén, Mäki-Arvela, Murzin (2026). Solventless hydrodeoxygenation of dihydroeugenol in a continuous reactor over Ni catalysts modified by Fe and Ce. Reaction Kinetics, Mechanisms and Catalysis. DOI: 10.1007/s11144-026-03188-9
- [10] Matveyeva, Omarov, Nashchekin, Popkov, Murzin (2022). Catalyst supports based on ZnO–ZnAl 2 O 4 nanocomposites with enhanced selectivity and coking resistance in isobutane dehydrogenation. Dalton Transactions. DOI: 10.1039/D2DT02088B
- [11] Sokolov, Radnik, Schneider, Rodemerck (2017). Low-temperature CO2 reforming of methane over Ni supported on ZnAl mixed metal oxides. International Journal of Hydrogen Energy. DOI: 10.1016/j.ijhydene.2017.01.013
- [12] Taghvaei, Rahimpour, Bruggeman (2017). Catalytic hydrodeoxygenation of anisole over nickel supported on plasma treated alumina–silica mixed oxides. RSC Advances. DOI: 10.1039/C7RA02594G