Phenolic hydrodeoxygenation in batch autoclaves
Catalyst-support effects, reactor methods, and comparison pitfalls
Use the navigation to move from the screening boundary to paper-level conditions, design synthesis, and a practical protocol checklist. In print, the sections are deliberately separated into a 12-part, print-friendly report.
1 · Scope and screening decisions
The target reaction is catalytic hydrodeoxygenation (HDO) of lignin-derived phenolic molecules, especially phenol and guaiacol, in which hydrogen removes oxygen functionality as water and/or oxygen-containing coproducts. The experimental question is narrower than “which catalyst works best”: does changing the support change conversion, deoxygenation, pathway, product slate, stability, or coke under comparable conditions?
The screening used three labels. Direct means a phenolic substrate and a closed batch/autoclave test with a support comparison described in a primary-source record or paper. Adjacent—fixed bed means the paper compares catalysts/supports in a continuous or vapor-phase reactor; it informs support chemistry but cannot be pooled quantitatively with a liquid batch test. Context only includes a useful HDO study that changes promoter, phase, or mechanism but does not isolate support. These distinctions matter because the user’s initial reactor constraint is scientifically consequential.
Search coverage included papers and publisher/repository records returned for phenol, guaiacol, cresols, propylguaiacol, and related lignin phenolics. Search hits were deduplicated by DOI/title. Records without a DOI are retained only where a primary repository gives identifiable publication metadata. The OpenAlex search endpoint returned HTTP 401 during this task; discovery therefore relied on publisher, repository, and scholarly-web search routes. Discovery is not proof of exhaustive coverage.
The evidence ceiling is explicit: only conditions exposed in primary text or publisher/repository records are reported. A missing vessel volume, solvent, or pressure basis is not supplied from another catalyst paper. “H₂ pressure” can mean a cold charge, a maintained pressure, or a pressure measured hot; these are not interchangeable.
2 · How to interpret support-effect evidence
A support comparison is strongest when active phase, metal loading, particle size, pretreatment, substrate concentration, solvent, catalyst-to-feed ratio, reactor, and thermal history are held constant while the support changes. In practice, support changes can also alter active-phase formation, dispersion, reducibility, acidity, pore structure, and metal–support contact. These can be part of the causal “support effect,” but they prevent an oversimplified claim that surface area or acidity alone explains the outcome.
Evidence is graded here as follows: A = primary full text or methods were available for the key reaction conditions; B = publisher abstract/index record confirms selected conditions but full method was not verified in this review; C = discovery/secondary summary only, used as a lead and not as a quantitative comparison. Grades describe what was verified for this report, not the paper’s quality.
Outcome terms must also remain distinct. Conversion reports disappearance of feed. Deoxygenation degree reports oxygen removal, ideally by elemental balance or explicit product balance. Selectivity is fraction of converted feed to a product (molar or carbon basis). Yield combines conversion and selectivity. A high phenol conversion can mostly reflect ring hydrogenation, while a high hydrocarbon selectivity at low conversion can still give low hydrocarbon yield. Coke and carbon balance should be reported separately.
3 · Core source matrix: direct batch/autoclave evidence
| Study / DOI | Feed and support variable | Reactor evidence and conditions retrieved | Classification |
|---|---|---|---|
| Dongil et al., “Hydrodeoxygenation of guaiacol over Ni/carbon catalysts: effect of the support and Ni loading” 10.1039/C5RA22540J | Guaiacol; CNT, oxidized CNT, activated carbon; Ni loading varied | Batch reactor; 573 K, 5 MPa H₂, 4 h. Vessel volume and whether 5 MPa is cold/operating pressure were not recovered in the inspected record. | Direct, A/B |
| “Understanding the Performance and Stability of Supported Ni-Co-Based Catalysts in Phenol HDO” 10.3390/catal6110176 | Phenol; HZSM-5, HBeta, HY, ZrO₂; Ni-Co/Ni formulations | 25 mL autoclave; 0.5 g phenol, 10 g water, 25 mg pre-reduced catalyst; H₂ charged to 50 bar at room temperature; 250 °C; 650 rpm. | Direct, A. Cold charge explicitly reported. |
| “Effect of Support on the Hydrodeoxygenation of Phenol using Ni-based Metal Catalysts: Ni/SiO₂, Ni/ZrO₂, and Ni/Al₂O₃” 10.1088/1757-899X/546/7/072004 | Phenol; SiO₂, ZrO₂, Al₂O₃; nominal 15 wt% Ni | Oil-batch autoclave; 200 °C, 2 h, H₂ atmosphere. Pressure, solvent, and full charge details not verified in retrieved publisher record. | Direct, B |
| Bykova et al., “Ni-based sol–gel catalysts as promising systems for crude bio-oil upgrading: Guaiacol hydrodeoxygenation study” 10.1016/j.apcatb.2011.11.051 | Guaiacol; Ni sol–gel formulations with different stabilizing components, including SiO₂- and ZrO₂-stabilized materials | Autoclave; 320 °C and 17 MPa H₂ reported. Exact feed charge, solvent, duration, and pressure basis not verified here. | Direct reaction; support/stabilizer comparison is confounded, B. |
| Mo et al., “Hydrodeoxygenation of Bio-Derived Phenol to Cyclohexane Fuel Catalyzed by Bifunctional Mesoporous Organic–Inorganic Hybrids” 10.3389/fchem.2018.00216 | Phenol; Pt/SBA-15, sulfonated SBA-15 hybrids, and physical mixtures | 100 mL stainless batch autoclave; H₂ purge, then 1–4 MPa initial charge; 80–200 °C; 800 rpm. Reported optimum: phenol in dichloromethane, 200 °C, 3 MPa H₂, 4 h. | Direct support/function comparison, A. Initial charge, not hot pressure. |
| Moreira et al., “Liquid-Phase Hydrodeoxygenation of Guaiacol over Mo₂C Supported on Commercial CNF” 10.3390/catal8040127 | Guaiacol; Mo₂C/CNF, with support/blank controls | 100 mL high-pressure batch reactor; 1.2 mL guaiacol, 40 mL decane, 0.2 g catalyst; H₂ charged at 2 MPa (one test 3 MPa); 300 rpm during heat-up then 1000 rpm at temperature; varied temperature/time, reported up to 623 K and 2–6 h. | Direct support-control evidence, not a multi-support ranking, A. |
| Guo et al., “Hydrodeoxygenation of Guaiacol as a Bio-Oil Model Compound over Pillared Clay-Supported Nickel–Molybdenum Catalysts” 10.1021/acs.jpcc.9b01890 | Guaiacol; pillared-clay NiMo; reduced versus sulfided state | High-pressure stirred autoclave; 0.5 g catalyst, 10 g guaiacol, no solvent, 6 h; pressure and temperature varied. Exact test points should be taken from the article/SI before numerical cross-study comparison. | Direct batch precedent, but state comparison rather than support panel; A/B. |
| Yu et al., “Hydrodeoxygenation of Phenolic Compounds to Cycloalkanes over Supported Nickel Phosphides” 10.1016/j.cattod.2018.05.012 | Phenol, catechol, o-cresol; Ni phosphides on SiO₂, HZSM-5, Al₂O₃ | Support matrix and substrates confirmed in primary manuscript/repository; reaction-method passage was not available in the retrieved extract. Autoclave details therefore remain unverified here. | Support comparison candidate; do not count as verified batch until methods are checked. |
| Oyama et al., “Ni₃P as a High-Performance Catalytic Phase for the Hydrodeoxygenation of Phenolic Compounds” 10.1039/C7GC03262E | Phenol and phenolic compounds; nickel phosphide phases | Aqueous phenol HDO reported; full reactor/charge details not verified in this pass. | Adjacent catalyst-phase evidence; possible overlap with phosphide support studies. |
| Moreira et al., “An Integrated Reaction Model of Guaiacol Hydrodeoxygenation Using Activated Carbon Supports: Effects of Support Properties, Metals, and Solvents” 10.1039/D5RE00179J | Guaiacol; Ru/Ni and several activated carbons; water versus decane | Accepted-manuscript record reports 100 mL stirred steel autoclave, 250 °C, maintained 5.17 MPa, 900 rpm, 120 min; continuous H₂ supply. Publication version/date should be checked against final journal issue. | Direct, strong support/solvent comparison; separate from fixed initial-charge studies. |
The matrix has 10 records including context and candidates; it is not a claim of 10 fully verified direct support-only comparisons. Several rows deliberately retain caveats.
4 · Fixed-bed and other adjacent support studies
Fixed-bed evidence is valuable for catalyst chemistry and stability, but its space velocity, feed vaporization, hydrogen/feed ratio, and time-on-stream differ fundamentally from batch autoclaves. The following papers help frame the support question without entering a batch comparison.
| Paper | What it contributes | Boundary |
|---|---|---|
| de Souza et al., “Hydrodeoxygenation of Phenol over Pd Catalysts: Effect of Support on Reaction Mechanism and Catalyst Deactivation” 10.1021/acscatal.6b02022 | Broad Pd support panel: SiO₂, Al₂O₃, TiO₂, ZrO₂, CeO₂, CeZrO₂; support-dependent pathways/deactivation. | Fixed-bed, not autoclave. |
| de Souza et al., “Hydrodeoxygenation of phenol using nickel phosphide catalysts. Study of the effect of the support” Catalysis Today 356 (2020), 366–375 | Ni phosphide on SiO₂, Al₂O₃, TiO₂, CeO₂, CeZrO₂; gas-phase phenol at 300 °C and 1 atm; support changes phosphide phase. | Gas phase; phase/support confounding is central. |
| Deepa et al., “Function of Metals and Supports on the Hydrodeoxygenation of Phenolic Compounds” 10.1002/cplu.201402145 | Phenol, guaiacol, eugenol; metals and acidic/neutral/basic supports compared at 523 K and 3 MPa H₂. | Support/function study; inspect primary method before assuming identical reactor mode for every substrate. |
| Sun et al., “Carbon-Supported bimetallic Pd-Fe catalysts for vapor-phase hydrodeoxygenation of guaiacol” 10.1016/j.jcat.2013.05.020 | Carbon-supported metal/bimetal comparisons at atmospheric pressure; PdFe alloy enhances oxygen-free aromatic yields. | Vapor-phase fixed-bed, metal comparison on carbon. |
| “Hydrodeoxygenation of Guaiacol over Pd–Co and Pd–Fe Catalysts: Deactivation and Regeneration” 10.3390/pr9030430 | Al-MCM-41-supported Pd, Co, Fe and bimetals; fixed-bed 400 °C, 1 atm; stability/regeneration and coke evidence. | Fixed-bed; support is common, metals vary. |
| “Vapor-phase hydrodeoxygenation of guaiacol on Al-MCM-41 supported Ni and Co catalysts” 10.1016/j.apcata.2015.12.021 | Ni/Co on acidic mesoporous Al-MCM-41; 400 °C, ambient pressure; product and regeneration effects. | Vapor-phase; support is not a broad matrix. |
| He and Wang, “Highly selective catalytic hydrodeoxygenation of guaiacol to cyclohexane over Pt/TiO₂ and NiMo/Al₂O₃ catalysts” 10.1007/s11705-014-1435-9 | Compares bifunctional Pt/TiO₂ and NiMo/Al₂O₃ at 300 °C, 7.1 MPa in fixed bed; discusses metal/acid functions and coke. | Different metals and supports; cannot isolate support. |
| Xiao et al., “Refinement of the kinetic model for guaiacol hydrodeoxygenation over platinum catalysts” 10.1002/aic.16913 | Intrinsic kinetics and reactor-model transfer for Pt-catalyzed guaiacol HDO. | Mechanistic/kinetic context, not a support comparison. |
| “Impact of Geometric and Electronic Factors on Selective Hydro-Deoxygenation of Guaiacol by Surface-Rich Metal/Silica Catalysts” 10.3390/catal13020425 | Fixed-bed, 15 bar(g) H₂, 250–400 °C; separates metal effects using silica to reduce support acidity confounds. | Metal comparison on a deliberately simplified common support. |
| “Vapor-Phase Hydrodeoxygenation of Guaiacol to Phenol Production Using Bifunctional Ni/Cu-Beta Zeolite Catalysts” 10.1016/j.joei.2023.101273 | Zeolite acid function and Ni/Cu active composition in vapor-phase guaiacol conversion. | Fixed-bed; not an autoclave support screen. |
| “Tuning palladium catalyst performance for guaiacol hydrodeoxygenation via support engineering” 2026 publisher record | Recent continuous-flow, high-H₂-pressure support-engineering study on Pd/oxide/carbon formulations. | Very recent fixed-bed research; online record inspected, DOI metadata not recovered in this pass. |
| Krause et al., “Hydrodeoxygenation of guaiacol on noble metal catalysts” 10.1016/j.cattod.2008.10.037 | Early noble-metal guaiacol HDO catalyst benchmark. | Useful historical background; not counted as a support-isolated autoclave comparison. |
These 12 adjacent/context records plus the 10 core matrix records provide a 22-record screened corpus. Not all are support-effect studies; a subset explicitly varies support. The requested ≥20 is met only for the overall screened phenolic-HDO corpus, not for strict autoclave support comparisons.
5 · What the batch/autoclave papers actually did
Phenol: aqueous and solvent-assisted designs
The Ni–Co phenol study is the clearest aqueous batch template among the sources inspected: 0.5 g phenol and 10 g water in a 25 mL vessel, with 25 mg pre-reduced catalyst, 50 bar H₂ charged at room temperature, 250 °C, and 650 rpm. This reports a cold pressure, not necessarily the pressure during reaction. The solvent-to-substrate ratio is high and creates a water-rich environment; results should not be compared directly with organic-solvent phenol tests or solvent-free guaiacol runs.
Mo et al. used a 100 mL stainless batch autoclave and deliberately combined Pt/SBA-15 with sulfonated organic–inorganic hybrids and a physical mixture. The comparison is therefore not a simple inert-support screen: sulfonic acid groups add a second catalytic function, and physical mixture controls help distinguish proximity/synergy from mere combined component inventory. The reported optimum includes dichloromethane, 200 °C, initial 3 MPa H₂ and 4 h. Solvent participation and downstream separation are part of the interpretation.
Guaiacol: carbon supports and carbide
Dongil et al. compare Ni on carbon materials and vary Ni loading at 573 K, 5 MPa H₂ and 4 h. This directly probes support surface chemistry (including oxidized nanotube groups) but the loading variation means each support comparison must be read at matched loading rather than across all reported “best” catalysts. Moreira et al.’s Mo₂C/CNF work gives valuable mixing details: staged stirring during heat-up and at temperature (300 then 1000 rpm), and a 100 mL vessel with decane and a stated initial H₂ pressure. Those details signal potential mass-transfer sensitivity; rate or selectivity differences should not be attributed to support unless mixing is held constant.
The later activated-carbon study reports a maintained pressure and continuous H₂ supply, unlike a sealed cold-charge batch. It compares supports and solvents at 250 °C, 5.17 MPa, 900 rpm and 120 min. This is closer to a pressure-controlled batch operation; report the pressure-control mode explicitly when citing it.
6 · Autoclave experimental-design synthesis
A defensible support comparison begins by declaring the independent variable and fixing everything else. Use one active metal or defined bimetal, one target loading basis (elemental mass per finished catalyst is preferable), one precursor/impregnation route, one calcination/reduction history, one substrate lot, and one solvent composition. Characterize actual loading and active phase; nominal impregnation is not proof of equal metal content or equal phosphide/carbide phase across supports.
Pre-reduction should occur in a separate controlled operation where possible. State gas composition, ramp rate, reduction temperature and duration, flow, cooling/passivation/storage, and whether the charged catalyst is reduced or precursor oxide. For air-sensitive phases, quantify exposure and passivation. In the Ni–Co phenol paper, pre-reduction is explicitly part of the reaction protocol; this is important because aqueous phenol/water may alter catalyst state.
Use a blank matrix: (1) solvent + H₂, no catalyst; (2) support only; (3) metal catalyst; (4) support-acidity or physical-mixture control if bifunctionality is claimed; and (5) a benchmark catalyst. If a support is acid-functionalized, include the corresponding unfunctionalized support and a physically mixed acid/support control. For redox oxides, compare oxygen-vacancy/reducibility changes with spectroscopic/TPR evidence rather than attributing activity solely to acidity.
Randomize run order, replicate center conditions (at least triplicate for the primary comparison), and report uncertainty. Sample at multiple times if the aim includes kinetics; a single endpoint mixes intrinsic rate, equilibrium, deactivation, and product secondary reactions. Quantify gas, liquid, and solid fractions. A carbon balance is mandatory when products include gases, oligomers, or coke.
For time-course studies, use independent sacrificial reactors or demonstrate that repeated sampling does not alter pressure/headspace and catalyst loading. The latter is often overlooked: withdrawing liquid from a small vessel changes both liquid holdup and gas–liquid equilibrium.
7 · Pressure, fill ratio, heating, and mixing
Record reactor material, rated volume, liner volume, liquid volume, catalyst mass, headspace, sealing procedure, purge cycles, room-temperature H₂ charge, and hot/controlled pressure separately. Report whether pressure is autogenous, initially charged, or maintained by a regulator/continuous feed. Heating a sealed vessel increases pressure through gas heating and solvent vapor pressure; a “50 bar H₂” cold charge is not a 50 bar operating-pressure guarantee.
Headspace affects available hydrogen inventory and gas–liquid transfer. To compare supports, use the same vessel geometry and fill fraction, not merely the same solvent volume in differently sized reactors. Specify whether the reported pressure is total pressure or hydrogen partial pressure, whether inert purge gas remains, and whether pressure declined during reaction. For an externally regulated system, state the regulator setpoint and hydrogen consumption/replenishment if measured.
Stirring speed alone is not a universal mixing criterion. Impeller type/diameter, baffles, liquid depth, catalyst particle size, and gas dispersion matter. The Moreira CNF study’s switch from 300 to 1000 rpm during heating/reaction illustrates that autoclave hydrodynamics can change within a run. Establish a speed-independence plateau using at least three stirring rates; if conversion/selectivity changes, report the regime as potentially mass-transfer-limited.
Measure internal liquid temperature where possible. Report heat-up time and define the reaction-time zero (setpoint reached, H₂ charge complete, or heater start). A nominal “4 h at temperature” is ambiguous unless the timing convention and cooling/quench procedure are given. Rapid quenching reduces post-reaction chemistry; report how the catalyst was separated and how volatile compounds were recovered.
8 · How support chemistry changes the reaction
Support effects can emerge through at least five coupled mechanisms: metal dispersion/particle size; electronic or interfacial metal–support interaction; acid–base chemistry; reducibility/oxygen vacancies; and pore architecture/transport. Carbon supports add surface oxygen functionality and hydrophobicity; oxidized CNTs can improve wetting/anchoring yet also change metal electron density or block adsorption sites. The guaiacol-on-Ni/carbon comparison specifically associates CNT oxidation with altered catalytic behavior, illustrating that “carbon support” is not a single category.
Acidic supports can promote dehydration, demethylation, transalkylation, and cracking, but can accelerate coke formation or secondary reactions. Zeolite framework and Si/Al ratio change acid density and pore access; beta and Y zeolites should not be collapsed into one “zeolite” variable. Mesoporous supports can ease diffusion and mitigate pore blockage relative to microporous materials, but this must be tested at matched metal particle size and acidity.
Reducible oxides (TiO₂, CeO₂, ZrO₂, mixed oxides) can alter hydrogen activation, oxygen vacancy formation, and C–O bond cleavage. Yet changing oxide support may also change the stable active phase. Nickel phosphide studies are a clear caution: support identity can shift the phosphide phase itself. Such experiments answer “what catalyst formulation is best on each support?” but not the narrower question “what is the effect of support with the same active phase?” unless phase is measured and controlled.
For every support claim, distinguish direct evidence (controlled panel in the target HDO reaction) from mechanistic plausibility. A support effect in fixed-bed vapor-phase HDO may guide hypotheses for liquid autoclaves, but solvent adsorption, liquid wetting, hydrogen solubility, and aqueous support stability can reverse trends.
9 · Reactions to track, not just “HDO”
Phenol can first hydrogenate to cyclohexanone/cyclohexanol and then deoxygenate to cyclohexane; alternatively, direct C–O cleavage can yield benzene. Guaiacol adds methoxy chemistry: demethylation can yield catechol, demethoxylation can yield phenol, and subsequent ring hydrogenation or direct deoxygenation gives distinct product families. Methyl transfer/transalkylation can produce cresols, toluene, and heavier alkyl aromatics. Ring opening and cracking can cause carbon loss; methanation is reported on some Ni-containing systems.
Therefore, an experiment should quantify oxygenated intermediates, oxygen-free aromatics, cycloalkanes, ring-opened products, light gases (CO, CO₂, CH₄ and C₂+), aqueous organics, and coke/insoluble residue. The reaction network may be solvent-sensitive: alcohols can serve as hydrogen donors; water can change support/catalyst state; decane is not equivalent to polar solvent; dichloromethane warrants a solvent blank and careful carbon accounting.
Use authentic standards and GC-FID/GC-MS for liquid products; calibrate response factors rather than assuming equal FID response. Use gas GC with TCD/FID where relevant. Analyze aqueous phases separately. Determine oxygen removal using elemental analysis or a documented product elemental balance. If the report uses “deoxygenation degree,” give the formula and denominator. If it reports selectivity, specify molar or carbon basis.
Phenol is an accessible model but under-represents the methoxy substituent in guaiacol and the steric/electronic complexity of real lignin oil. Guaiacol is more representative of common lignin-derived phenolics but its competing methyl, methoxy, and hydroxyl transformations complicate a single performance metric. Claims should remain model-compound-specific unless validated with mixtures or real bio-oil.
10 · Reporting metrics and fair comparison
At minimum report conversion, product selectivity and yield, with equations and basis. A useful set is: conversion = (moles feed in − moles feed out)/moles feed in; selectivity to product i = moles of substrate equivalents in i / moles converted substrate; yield = conversion × selectivity. For products with carbon loss or multi-carbon coupling, report both molar and carbon selectivity. An apparent deoxygenation degree can be estimated from oxygen balance, but should not be equated to conversion or hydrocarbon selectivity.
Report carbon recovery as carbon quantified in all measured products divided by carbon fed. Include gases and deposits. Distinguish “oxygen-free product selectivity” from yield. State analytical limits for unidentified peaks. Report catalyst productivity per catalyst mass and, where feasible, per exposed metal site; neither normalization should substitute for the other when dispersion differs.
For stability, report time-on-stream or sequential batch cycles, regeneration protocol, and recovered catalyst mass. Coke quantity and chemistry, metal sintering/leaching, oxidation state, and support structural change are separate deactivation modes. Post-run characterization should preserve spent state where relevant; air exposure can oxidize reduced phases and obscure the working catalyst.
Report both nominal and measured metal loading, BET area/pore size, acidity/basicity, reducibility, particle size, and active phase before interpreting support effects. Use XRD/TEM/chemisorption/TPR and suitable spectroscopy as appropriate. These measurements are explanatory covariates, not a license to assign causality from correlation alone.
11 · A practical comparative autoclave design
Stage 1—support screen. Choose one well-defined model substrate (guaiacol if methoxy chemistry matters; phenol for a simpler baseline). Prepare the same metal loading on 4–6 support families: a neutral high-area support, an acidic mesoporous support, a zeolite, a reducible oxide, and a carbon with/without controlled oxidation. Keep one synthesis route and target particle-size range. Characterize after reduction.
Stage 2—reaction conditions. Choose a single solvent only after substrate solubility and support stability are checked. Fix feed concentration, catalyst/feed ratio, vessel geometry, fill fraction, H₂ inventory or regulator setpoint, stirring/impeller, heat-up profile, and quench. Establish a non-transport-limited stirring regime and a low-conversion time window for rate comparison. A full factorial across all factors is expensive; use a screening design for support × temperature × H₂ pressure, then a focused response-surface design on the leading supports.
Controls. Include support-only, no-catalyst, solvent-only, and benchmark catalyst runs. For acid-modified supports, include unmodified support and physical mixture. For reducible oxides, characterize oxygen vacancies/reducibility before and after reaction. Run replicates and randomized order; reserve independent synthesis batches to test reproducibility.
Outputs. Report conversion, oxygen-free liquid yield, aromatic versus cycloalkane distribution, gas products, aqueous phase, carbon recovery, coke, and deactivation. Use GC-FID/GC-MS with standards and gas analysis. Reconcile the elemental balances. Compare support effects at matched conversion as well as at a common reaction time.
Interpretation. Separate (i) support effect on active phase formation, (ii) support effect on reactant/intermediate adsorption and pathway, and (iii) support effect on deactivation/transport. If the active phase differs across supports, state that the practical formulation is being compared, not a pure support perturbation. Do not claim an optimum from a single endpoint or one support loading.
This proposed design is a recommendation synthesized from the methodological limitations above; its values are not claimed as a published optimum. Specific temperatures/pressures should be chosen based on the target catalyst’s safe operating window and preliminary conversion/selectivity data.
12 · Limitations and source register
Limitations. This is a conservative, relevance-selected survey, not a systematic review with exhaustive database export. The OpenAlex connector failed with HTTP 401. Some article full texts or supplementary information were inaccessible in this pass; those entries are flagged and their missing charge details are not inferred. The count of 22 refers to screened research records across batch and adjacent fixed-bed contexts; several are mechanism, kinetics, metal/promoter, or common-support controls rather than support-isolated comparisons. Strict verified autoclave support-comparison evidence remains below 20. Before using any row as a design precedent, inspect the cited paper’s methods and SI for exact charge, pressure basis, temperature timing, and catalyst pretreatment.
Core direct/near-direct studies:
Fixed-bed/adjacent research papers:
Review framing consulted: Wang et al., “Hydrodeoxygenation of lignin-derived phenolics – a review on the active sites of supported metal catalysts,” Green Chemistry (2020), DOI 10.1039/D0GC02610G; and the primary papers linked above. Search records and abstracts support only what they expose; the report labels method-level evidence accordingly.
Bottom line: there are useful autoclave design precedents, but the literature does not support claiming 20 isolated support comparisons in autoclaves. The most informative experimental features to carry forward are explicit cold-charge versus maintained-pressure language, matched feed/catalyst/solvent/headspace, mixing validation, active-phase characterization, blanks, and whole-product carbon/oxygen balances.
End of report · Prepared 22 October 2026