Effects of Size and Location of Nanoparticles in Catalysis - Krijn de Jong

Scientific Concepts, Discoveries, and Nature/Physical Phenomena

Climate/energy context for catalysis

• Climate change and carbon budget
  • Framing via cumulative CO₂ emissions vs. temperature increase, consistent with Paris Agreement constraints.
  • “Carbon budget” urgency motivates rapid emissions reductions.
• Role of catalysis in climate mitigation
  • Catalysis enables more efficient conversion processes, reducing CO₂ emissions through higher efficiency.
  • Distinction between:
    • CCS (carbon capture and storage)
    • CCU (carbon capture and utilization)
  • Thermodynamic critique: converting CO₂ → fuels is constrained by first-law thermodynamics.
    • More efficient pathways may instead use renewable H₂ directly, rather than “make CO₂ then reduce it.”
• Synthesis gas (syngas) as a platform
  • Syngas = CO + H₂, produced from varied carbon feeds: coal, natural gas, biomass, and potentially CO₂.
  • Product classes tied to catalyst chemistry:
    • Fischer–Tropsch: long-chain alkanes (diesel-range)
    • CO/olefin routes: short-chain alkenes (“olefins route”)

---

Nanoparticles: size effects (surface structure and catalytic sites)

• Key idea: nanoparticle size controls surface structure
  • As particle diameter changes, the fraction of specific surface sites changes strongly.
  • Example concept: fraction of step-edge (B₅-type) sites can increase dramatically with size over certain ranges (from near-zero to several percent), with largest changes around ~1–6 nm.
• Catalytic performance is non-monotonic with size
  • For cobalt in syngas-to-diesel-like chemistry:
    • There is a maximum activity around ~6 nm.
  • Above that size:
    • Activity per surface atom decreases approximately like 1/diameter.
  • Below that size:
    • Turnover frequency drops sharply.
• Mechanistic explanation: site-specific “overbinding” on small particles
  • On very small particles, edge/corner atoms bind reactants too strongly, becoming catalytically ineffective.
  • As particle size increases, the contribution shifts more toward terrace sites.
  • Net effect: activity can drop by about an order of magnitude in the cobalt case.

---

Syngas-to-olefins: particle size + promoters govern selectivity

• Iron-based syngas conversion to lower olefins (C2–C4)
  • Catalysts involve promoters (Na and sulfur) and iron carbide surface chemistry.
• Size trends
  • Smaller particles may show higher turnover, but:
    • Promoter-driven tuning of selectivity is strongest for larger sizes.
  • Promoters suppress undesired methane formation (“methane-like pathways”) and change barriers toward olefin formation.
• Promoter mechanism (electronic/chemical effects)
  • A Na–S (sodium sulfide) species modifies electronic density at the iron carbide surface:
    • Weakens C–H bonds
    • Strengthens Fe–H interactions
  • Consequences described:
    • Higher barriers for unwanted “methanation”
    • Lower barriers toward ethylene/olefin formation
    • → improved selectivity

---

Controlling nanoparticle size using supports/precursors

• Support pore confinement
  • Ordered mesoporous silica with ~10 nm pores acts as a nanoreactor.
  • Adjusting precursor thermal decomposition conditions yields different metal oxide/particle sizes and can promote embedding vs. external growth.
• Thermal treatment chemistry impacts precursor decomposition
  • Air calcination vs. NO-containing gas streams can drastically change decomposition behavior:
    • With NO: particles remain more confined and embedded in pores.
    • Without NO: growth can generate rods/particles outside pores (less confinement).

---

Catalyst stability and nanoparticle growth (sintering / Ostwald-like transport)

• Time-on-stream size evolution can invert initial size advantages
  • For nickel methanation catalysts:
    • Initially smallest particles show highest activity but deactivate faster.
• Proposed growth driver
  • Nickel carbonyl transport: smaller particles may have higher monomer concentration due to higher thermodynamic potential (Kelvin effect).
  • Net transport from small → large particles promotes coarsening.
• Growth regime dependence
  • Two growth modes:
    • Unconfined growth: may crush/break support pores
    • Confined growth: growth stops upon contacting pore walls
  • Starting with larger particles can yield smaller growth ratios and improved stability.

---

Copper nanoparticles: structure sensitivity and promoter/support roles

• Relevance to methanol synthesis
  • Focus on Cu-based methanol synthesis, including the role of Cu–Zn systems with ZnO as a promoter.
• Size sensitivity below ~10 nm
  • Much literature centers on >10 nm; the talk emphasizes <10 nm behavior.
• In situ TEM observation of nanoparticle formation
  • A co-precipitated copper “felo-silicate” precursor on silica is reduced under controlled conditions.
  • Real-time imaging shows:
    • Cu species convert and form nanoparticles early
    • Growth accelerates during the first ~20 minutes
    • Particles become effectively immobile (no “walking vacuum cleaners” behavior)
  • Mechanistic implication: autocatalysis is crucial to the growth behavior (as argued in the talk).
• Turnover frequency trends
  • Highest maxima for Cu/Zn silicate relative to Cu alone and relative to more conventional Cu/Zn oxide systems (Cu–ZnO/Al₂O₃).
  • Reported enhancement: ~10× higher maximum turnover frequency vs. copper alone (consistent with improvements for Cu–ZnO).

---

Nanoparticle location effects (spatial distribution on supports)

• Two distinct “location” concepts
  1. Inter-particle distance: clustering vs uniform dispersion (particle–particle spacing)
  2. Bifunctional catalysis distance: spacing between different functional components (e.g., metal vs acid sites)
• “Coffee-stain” effect
  • During impregnation/drying, precursors can form patches, leading to clustered nanoparticles after calcination.
• Nitric oxide impact on precursor redistribution
  • NO-containing calcination can produce more uniform distributions by altering precursor chemistry/mobility compared with air calcination.
• How location effects were measured
  • Electron tomography: 3D mapping of nanoparticle distribution inside porous supports
  • Small-angle X-ray scattering (SAXS): assesses electron density uniformity and pore filling/emptying
• Catalysis consequences
  • Uniform vs clustered distributions yield different activity/stability:
    • Clustered, high-density domains tend to deactivate more strongly, likely via increased particle diffusion/coalescence within those domains.

---

Bifunctional catalysis: nanoscale intimacy in zeolite + metal systems

• System: Pt on alumina binder with zeolite (acid sites)
  • Zeolite provides Brønsted acid sites → carbenium ion chemistry (isomerization, cracking via acid-catalyzed steps).
  • Metals provide hydrogenation/dehydrogenation functions.
• “Closer is better” expectation vs experimental outcomes
  • A common assumption: putting metal inside zeolite minimizes metal–acid distance for maximal selectivity control.
  • The talk compares:
    • Pt on alumina binder
    • Pt inside zeolite crystals
• Findings
  • Conversion may roughly follow Thiele-modulus-like expectations, but selectivity changes strongly.
  • Pt on binder can increase isomerization selectivity (feed example basis: nC10/nC19).
  • Pt inside zeolite increases secondary hydrocracking/overprocessing, increasing C–C fragmentation and altering the product pattern.
• Proposed reasoning
  • If metals are outside, olefins only require short diffusion distances, limiting exposure to metal-driven secondary reactions.
  • If metals are inside, repeated diffusion and longer residence increase the probability of undesired metal-side reactions.

---

Colloidal/chemical guidance: support acid-base and charge effects

• A Q&A emphasized that precursor deposition and particle size can depend strongly on:
  • Support surface charge / point of zero charge
  • Acid–base properties
• Implications suggested:
  • Near charge neutrality (e.g., silica near unit charge conditions), acid–base effects may be limited.
  • For alumina-containing materials (e.g., MFI zeolites with different Si/Al), acidity and ion-exchange behavior can strongly alter adsorption and resulting nanoparticle formation.

---

Gas-stream chemistry during precursor decomposition (NO effects)

• Nitric oxide influences decomposition via two main pathways
  1. Acid–base/ionic transformation acceleration
     • Metal nitrates can convert to metal hydroxide nitrates and/or hydrogen nitrates, changing decomposition rate and chemistry.
  2. Radical chemistry / decomposition kinetics
     • NO modifies radical pathways differently for different metals (discussed for Ni, Cu, Co), moderating the “explosive” decomposition behavior of nitrates.
• Differential scanning calorimetry observations
  • Without NO: sharper endothermic peaks (more violent decomposition)
  • With NO: more moderated peaks

---

Evidence of special copper behavior (volatilization risk)

• The talk notes an observed phenomenon for copper nitrate precursors:
  • A “student found a black ring” in the reactor, interpreted as volatilization of copper compounds, motivating a literature search.

---

Methodologies / Experimental Approaches Outlined

• Nano site analysis vs particle size
  • Use theoretical/atomistic modeling (DFT + multiscale modeling) to track surface site fractions (e.g., step-edge sites) vs nanoparticle diameter.
• Particle size control via catalyst synthesis variables
  • Vary metal loading (impregnation) on support
  • Use different precursor/impregnation techniques (including examples like carbon nanofiber supports)
  • Use ordered mesoporous supports with controlled pore diameters (nanoconfinement)
  • Change thermal treatment atmosphere (air vs NO-containing vs trace NO vs other gas streams)
• Transient kinetic studies
  • Time-resolved/transient methods to infer whether smaller particles contribute less due to stronger binding at edges/corners.
• In situ / operando electron microscopy
  • In situ TEM under low pressure with differential pumping
  • Real-time imaging of nanoparticle nucleation/growth during reduction (e.g., copper felo-silicate → Cu nanoparticles)
• 3D structural mapping of particle location
  • Electron tomography to reconstruct nanoparticle spatial distributions in porous supports.
• Scattering characterization of distribution uniformity
  • SAXS to infer pore filling/emptying and changes in electron density uniformity.
• Catalyst testing
  • Measure activity/turnover frequency vs particle size and time-on-stream
  • Map selectivity between olefins vs methane/undesired products
• Bifunctional catalysis testing in zeolite systems
  • Compare metal placement:
    • Pt on alumina binder vs Pt inside zeolite
  • Evaluate conversion/selectivity using different feed molecules (examples included nC10, nC19, nC7)

---

Researchers and Sources Featured (As Explicitly Mentioned)

• Krijn de Jong (speaker; referenced by name)
• Professor Peruvian (name appears in opening thanks; exact spelling uncertain from transcript)
• Hardy [felt?] (referenced as source of seminal 1960s work on nanoparticle surface-site changes; full name unclear)
• Peter from Hell / Peter [from] hell.the (Sasol and South Africa; transcript unclear)
• Hollins group (collaborator for transient kinetics; identity not specified)
• Paul (mentioned as having proposed a distance-based selectivity concept decades earlier: “Paul why”)
• Johann Martens / Johann Martens (industry/expert collaborator; hydrocracking experiments)
• Paul Weiss (named source of an expectation regarding Thiele-modulus-like behavior)
• Dr. Petra DeYoung (named near the end: “professor Petra DeYoung”)
• Hari better (transcribed acknowledgement; full name unclear)
• Additional colleagues acknowledged generally for the interface of electron microscopy and catalysis (names not provided beyond those listed)

Note: Several names appear corrupted/auto-transcribed; only those visible in subtitles are listed.