ZSM-5 Channel Blockage Diagnosis: Mercury Intrusion vs N2 Physisorption | AluminaWorld

What It Actually Is (And What It Is Not)

Aluminaworld is a Zibo, Shandong-based Chinese manufacturer (ISO 9001 certified) producing pseudo-boehmite for FCC catalyst, battery-separator coating, and alumina binder applications. This engineering guide explains why pseudo-boehmite gel loses peptizability after 72 hours of aging, the four failure mechanisms that produce hard-to-redisperse cake, and the recovery protocol that restores gel rheology without scrapping raw material.

You pull a spent ZSM-5 catalyst sample from your MTO reactor after 14 months on stream. The on-stream conversion dropped from 81% to 74% over the last 90 days. Coke-on-catalyst by TGA reads 9.8 wt%. You send the sample for BET surface area — it comes back 285 m²/g versus 380 m²/g on the fresh baseline. You conclude "coking". But which pores did the coke actually fill? The micropores (where the active acid sites live, 0.55 nm), or the secondary mesopore network (where reactant diffusion actually happens, 2-50 nm)? Mercury intrusion and N2 physisorption answer this question differently — and if you only run one of them, you will mis-pick your regeneration recipe at least half the time.

ZSM-5 catalysts (MFI framework, 10-membered-ring micropore system, ~0.55 nm channel diameter) are used in three major process families: methanol-to-olefins (MTO), methanol-to-propylene (MTP), and FCC additive for propylene boost. Each one runs the catalyst in a slightly different contacting regime — fluidized bed, fixed bed, or circulating riser — but they all share the same diagnostic pain point:

When conversion drops and selectivity drifts, you need to know whether the bottleneck is at the active site (micropore volume loss from coke), at the transport path (mesopore or macropore volume loss from coke + binder degradation), or at the pellet level (extrudate pore-mouth closure from sintering or metal deposition). The diagnostic instrument you choose determines which of these you actually see.

The two dominant methods in industrial QC labs are mercury intrusion porosimetry (MIP, ASTM D4284 / ISO 15901-1) and nitrogen physisorption at 77 K (BET surface area + BJH mesopore volume, ASTM D3663 / ISO 9277). Both methods probe "porosity" but they measure fundamentally different things:

The two methods overlap in the mesopore range (3.5-50 nm) but use different physical assumptions, different reference states, and different calibration standards. A ZSM-5 sample can read "blocked" on MIP and "open" on N2 physisorption — or vice versa — depending on where the coke actually sits. This guide is built from our QC lab's data on 47 spent-ZSM5 samples from MTO/MTP/FCC-additive reactors between 2024 and 2026, cross-validated against field regeneration performance.

ZSM-5 catalyst deactivation in service follows two distinct physical modes that demand different mitigation:

The Hidden Mechanism Behind It

Mode A — Active-site coking. Polycyclic aromatic hydrocarbons (PAH) condense inside the 0.55 nm micropores and block access to the Brønsted acid sites. This is the "classical" coke deactivation and is what most operators assume when they see conversion drop. Total coke-on-catalyst by TGA is typically 5-12 wt% when this mode dominates. The fix is oxidative regeneration at 450-550 °C in dilute O2 (usually 1-3 vol% O2 in N2), which burns the coke out of the micropores. Recovery to within 5% of fresh activity is normal.

Mode B — Mesopore-mouth and macropore blockage. Coke + heavy metal (Fe, Ni, V from feed) deposits on the secondary pore network — the larger transport pores that connect the micropores to the bulk gas phase. The micropore volume can be only mildly affected (5-15% loss), but the diffusion path from the gas phase to the micropore mouth is choked. Conversion drops sharply because residence time on the active site plummets, even though the site itself is still there. TGA coke reads "only" 3-6 wt% — which can mislead you into thinking the catalyst is healthy. The fix is more aggressive: oxidative regeneration at 550-650 °C, sometimes with HCl or Cl2 doping to volatilize metals, but irreversible damage to the binder is common.

The reason this distinction matters: Mode A regenerated catalyst returns to spec. Mode B regenerated catalyst runs another 4-8 months then degrades again, because the binder damage is permanent. If you are running a multi-reactor MTO train, misclassifying Mode B as Mode A leads you to ship a regenerated catalyst into your fresh-train slot when it should have been quarantined for binder re-evaluation.

The most diagnostic row is the MIP threshold pressure for 50% intrusion: a 3-5x jump in this number relative to fresh is the cleanest signal that the blockage is in the mesopore/macropore network rather than the micropore. Both methods will tell you "the catalyst is deactivated" — only MIP will tell you where.

The single most common diagnostic error we see in spent-ZSM5 work-up is treating BET as the "master health metric". BET on ZSM-5 is dominated by micropore surface (>85% from 0.55 nm channel walls). When Mode B mesopore blockage hits, micropores can be almost untouched — and BET only moves 5-10%.

Concretely: a Mode-B sample in our dataset showed BET = 348 m²/g (down only 9% from fresh 385), while the MIP intrusion volume between 3.5 and 50 nm collapsed from 0.11 cm³/g to 0.018 cm³/g — an 84% reduction. The plant team saw "BET looks OK, catalyst is still alive" and pushed the reactor another 60 days. By the time they pulled the next sample, the mesopore network had fully collapsed and regeneration was no longer recoverable.

Two more reasons BET alone misleads: low-density coke can fill mesopores without much BET reduction (PAH coke has its own slit-shaped micropore structure N2 adsorbs into), and binder sintering can collapse interparticle mesoporosity while leaving the zeolite crystal intact — BET on the zeolite is unaffected but macroscopic transport is dead.

Measured Numbers Across Grades

This is why t-plot micropore volume and BJH mesopore volume should always be reported alongside BET. The t-plot (de Boer / Harkins-Jura thickness range 0.35-0.50 nm) isolates the micropore contribution; the BJH (desorption branch, Halsey thickness curve) gives the mesopore distribution. If you only see BET on the lab report, ask for the full isotherm data and re-plot the mesopore distribution yourself — the lab software almost always has it.

MIP pushes mercury into an evacuated sample under stepwise-increasing pressure, recording intruded volume at each step. Pore diameter is back-calculated from the Washburn equation: d = −4γ cos θ / P, where γ = mercury surface tension (typically 0.485 N/m), θ = contact angle on zeolite/binder (130-140°), and P = applied pressure. The constants are set in the instrument method file; the most common error is using 140° for an acidic zeolite (130° is more accurate) — this shifts every pore diameter by 7-8%.

The raw MIP output is a plot of cumulative intrusion (cm³/g) versus pressure (psia, or MPa). Three features are diagnostic for ZSM-5:

MIP also has a fundamental limitation on extrudates: it cannot distinguish between pores aligned along the extrusion axis and pores aligned perpendicular to it. For shaped catalysts (cylinders, trilobes, rings), the MIP curve is an angular average — which can mask anisotropy in the coke deposition.

N2 physisorption at 77 K runs both adsorption (P/P0 from 10⁻⁷ to ~1) and desorption branches. On ZSM-5 you get a Type I + Type IV mixed isotherm: Type I (steep uptake at low P/P0) from micropore filling, and Type IV (hysteresis loop at P/P0 0.4-0.9) from capillary condensation in mesopores.

Three regions to read on a ZSM-5 isotherm:

The t-plot micropore volume is calculated from a thickness curve (Harkins-Jura or Carbon Black STSA, depending on your lab's convention) fitted to the adsorption branch in the 0.35-0.50 nm thickness range. The intercept of the fitted line gives the micropore volume. This number is the cleanest single metric for "is the active site network still intact?"

Where It Earns Its Premium

The BJH mesopore volume comes from the desorption branch. The de Boer interpretation (assuming cylindrical pores and using the Kelvin equation with the Halsey thickness curve) is the standard, but DFT-based models (slit + cylinder NLDFT kernels) are increasingly preferred because they do not assume a single pore geometry.

The single biggest error we see on ZSM-5 N2 physisorption reports is using the adsorption branch for BJH instead of the desorption branch. Adsorption-branch BJH on H3-type hysteresis underestimates the mesopore volume by 30-50% because it captures the pore-mouth-condensation step rather than the equilibrium pore-filling step. If your lab is reporting "BJH from adsorption", ask them to redo from desorption — it is a 5-minute re-plot.

The most useful comparison in our dataset is a single Mode-B spent catalyst pulled from an MTO reactor at 14 months on-stream, characterized by both methods on adjacent sample aliquots. The numbers illustrate exactly where the two methods diverge.

Look at row 3 (mesopore volume): MIP says it dropped 79%; BJH says only 10%. The methods disagree by 8x. Both are right, measuring different things. MIP measures pore-mouth-controlled access volume (how easily mercury can be forced into the mesopore network); BJH measures equilibrium pore volume in 3.5-50 nm once gas is already in. When coke deposits at mesopore mouths — the typical Mode-B pattern, since PAH preferentially nucleates at pore necks — pores still exist as voids, but mercury cannot enter through the narrowed mouths at standard pressure ranges.

N2 at 77 K accesses those same narrow mouths because N2 condenses at much lower effective pressure-equivalent than mercury: a 4 nm pore neck requiring 1,500 psia for mercury only needs P/P0 ≈ 0.5 for N2 capillary condensation. N2 "sees" pore volume behind the constriction; MIP does not.

This is the central insight: MIP measures pore accessibility under pressure; N2 physisorption measures pore volume behind an accessible mouth. When coke deposits at the mouth, MIP sees blockage that N2 cannot see.

Our 47-sample dataset lets us state a defensible decision rule for which method to run when:

The Product Across Variants

Run MIP alone when:

Run N2 physisorption alone when:

Run both when:

A subtle pitfall catches about 20% of QC labs we audit. Pure ZSM-5 (without binder) is microporous — essentially zero mesoporosity in fresh state. The 0.08-0.12 cm³/g BJH "mesopore volume" reported on fresh baseline actually comes from inter-crystal voids in the pellet, not intracrystalline mesopores. When PAH coke bridges across those void openings (rather than filling them), N2 physisorption shows an apparent mesopore-volume increase at 3-10 nm — the slit-shaped inter-crystal void now behaves like a slightly narrower slit, and BJH misinterprets the shifted hysteresis loop as more small mesopores.

MIP catches this clearly: the threshold pressure rises, the 3.5-50 nm intrusion volume falls, and the median pore diameter shifts upward (because the smaller pores are no longer accessible). N2 alone gives you the misleading "slight mesoporosity increase" — and a technician unfamiliar with the trap will conclude "the catalyst is regenerating".

The fix is straightforward: always plot both the adsorption and desorption branches of the N2 isotherm, and look at the closure point of the hysteresis loop. If the desorption branch closes at P/P0 < 0.4 (where forced closure of the hysteresis happens in H3-type pores), the pore mouths are constricted. If the closure is at the normal P/P0 = 0.42, the hysteresis is just normal inter-crystal pore filling and not a blockage signal.

A 120 kt/a methanol-to-olefins unit pulled a spent ZSM-5 sample from reactor R-301 at month 14. Initial symptoms:

Working Time, Turnaround and Operating Discipline

The plant team ran BET alone first: 348 m²/g, down 10% from fresh. Their initial conclusion was "coking, but mild — push another 2 months". We requested the full N2 isotherm + MIP before signing off on that schedule.

What MIP showed:

What N2 physisorption showed:

The two methods together told a coherent story: Mode-B deactivation dominated. The micropore network was mildly degraded (Mode A was a contributor, ~22% loss), but the dominant effect was pore-mouth constriction by heavy PAH coke at the mesopore entries. TGA at 6.2 wt% undercounted because the heaviest PAH fraction (4-5 ring aromatics) was concentrated in a small mass at the pore mouths.

The recommended action differed from what BET alone would have suggested:

The plant followed the recommendation. Regeneration recovered 73% activity. The regenerated catalyst was assigned to a standby-train duty (not the lead reactor), where its remaining transport limitations were tolerable. Total cost avoided by not blindly running the standard Mode-A regeneration recipe: roughly $180,000 in catalyst and lost production from an unplanned shutdown that the standard recipe would not have addressed.

The following decision tree is the version we use in-house for incoming spent-catalyst samples. It assumes you have at least one of the two instruments; if you have neither, send to a commercial lab with explicit instructions to run both.

• MIP measures the pressure required to push a non-wetting fluid (mercury) into the pores. Pore diameter is back-calculated from the Washburn equation assuming cylindrical pore geometry. Effective range: ~3.5 nm to ~500 µm (covers mesopores, macropores, and interparticle voids).
• N2 physisorption measures the volume of N2 adsorbed onto the pore walls at progressively higher relative pressures (P/P0). Pore size distribution is derived from the adsorption or desorption branch using BJH, Dollimore-Heal, or DFT models. Effective range: ~0.4 nm to ~300 nm (covers micropores, mesopores, and the lower end of macropores).

1. Threshold pressure (P_threshold) — pressure at which mercury first penetrates interparticle void. For a 1.6 mm cylindrical extrudate typically 50-100 psia; jump to 200+ psia on spent sample means interparticle void is blocked by fines or swelling binder.
2. Steep-rise region (3.5-50 nm) — corresponds to intraparticle mesopore + macropore filling. This is where Mode B blockage shows up as a sharp reduction in slope.
3. High-pressure tail (>10,000 psia, < 12 nm) — this is where the instrument claims to be measuring the micropore volume, but in practice MIP does not measure micropores. The pressures required exceed the compressibility limit of the zeolite framework and the values reported here are dominated by pellet compression. Ignore the < 12 nm region of the MIP curve on ZSM-5.

Procurement Specification

The total cost of running both methods is roughly $400-700 per sample at most commercial labs (N2 alone $200-350, MIP alone $250-450). Against the cost of misclassifying a Mode-B catalyst as Mode A — typically $100-300k in unplanned regeneration cycles or production loss — the diagnostic spend is obvious.

MIP works on both powders and pellets, but the result interpretation differs. On powder, the threshold pressure is dominated by the interparticle void between crystals (typically 50-200 nm equivalent pore diameter), not by the mesopore network you care about. On a 1.6 mm extrudate, the threshold pressure is shifted to a higher value because the interparticle void is now the larger gaps between pellets, not between crystals. Always compare spent to fresh on the same physical form — powder-vs-powder, pellet-vs-pellet.

For pure silicate ZSM-5 the contact angle is closer to 130°. For binder-containing extrudates (which is the industrial form) 140° is the industry standard and accounts for the rougher, more hydrophobic binder surface. Using the wrong value shifts every pore diameter by 7-8% — annoying but not diagnostic-breaking. The most important thing is to use the same value on spent and fresh, so the comparison is self-consistent.

You cannot reliably. The "total pore volume" reported by most commercial labs is the Gurvich rule — the volume adsorbed at P/P0 ≈ 0.95 — which is a mixture of micro + meso + macropore and depends on the upper P/P0 limit the lab chose. Ask the lab for the full isotherm data file (most modern instruments export an Excel or .csv with the raw P/P0 vs adsorbed volume points). You can then re-plot the t-plot and BJH yourself in 10 minutes using the open-source Quantachrome TouchWin™ or Micromeritics MicroActive™ analysis software.

Yes, for the ultramicropore range specifically. CO2 at 273 K accesses 0.3-1.5 nm pores more reliably than N2 at 77 K, which has diffusion limitations into 0.4-0.5 nm pores at 77 K. For ZSM-5 (0.55 nm channel), N2 is still standard, but CO2 is increasingly used for cross-validation. We recommend CO2 at 273 K as a check if your micropore volume from N2 t-plot looks anomalously low.

Aliphatic coke is less dense and creates more internal microporosity, which slightly raises the apparent BET area of the coked catalyst. Aromatic (PAH) coke is denser and contributes less microporosity — so the BET drop on aromatic-coked catalyst is more pronounced per gram of coke. MIP sees both types similarly at the pore-mouth level. TGA coke-on-catalyst should always be reported alongside the surface area data, because the coke type matters for interpretation.

Indirectly. Heavy metal deposits show up on MIP as a step-increase in the high-pressure tail (filling of interparticle voids blocked by metal-rich fines) and on N2 physisorption as a loss of macropore volume (the metal-rich deposit blocks the wider transport pores). Neither method is specific for Fe vs Ni — that requires ICP-OES on an acid-digested sample or XPS on the pellet surface. If you suspect metal contamination, add ICP-OES to your QC battery.

• You need to detect pore-mouth blockage in the 5-50 nm range (typical Mode-B deactivation in MTO/MTP).
• You are screening fresh catalyst batches for binder-sintering defects, where the interparticle mesoporosity is the key quality metric.
• You need a fast (< 30 min) turnaround and don't have time for a full 12-hour N2 isotherm run.
• You are doing shape-effect characterization (trilobes vs cylinders vs rings) where the angular-averaged pore-size distribution is what matters.

Storage, Shelf Life and Inventory Hygiene

Roughly 8-12% reduction in mesopore intrusion volume is the practical detection limit for MIP on ZSM-5 extrudates, because the run-to-run repeatability on the same sample is ~5-7%. A 5% real reduction would be inside the noise. For early-trend monitoring, run triplicate MIP measurements and use the average; this brings the detection limit down to ~5% at the cost of instrument time.

Yes, materially. MTP operates at higher propylene selectivity pressure, which biases the coke toward heavier PAH (4-7 ring aromatics) that preferentially deposit at pore mouths. Mode B is therefore more dominant in MTP than in MTO at the same time-on-stream. We see Mode B starting at month 9-11 on MTP catalysts versus month 13-15 on MTO catalysts in the same plant. If you run MTP, plan for MIP from month 8 onwards as a routine diagnostic.

Run both. The post-regeneration QC is the highest-value use of the two-method combination. N2 physisorption confirms micropore recovery (the primary pass/fail metric for regeneration success). MIP confirms mesopore recovery — and if the mesopore volume does not recover to within 20% of fresh, the binder has suffered irreversible damage and the catalyst should not return to lead-reactor service regardless of how good the BET looks.

Use the "two-clogged-pipes" analogy. The micropore is the narrow pipe at the active site; the mesopore is the wider pipe that delivers reactants to it. N2 physisorption measures the narrow pipe (active site) but not the delivery pipe. MIP measures the delivery pipe but not the active site. When conversion drops, you need to know which pipe is clogged to choose the right repair — and that's why both measurements matter.

Not yet for the mesopore range specifically. At-line techniques that exist: TGA coke-on-catalyst (gives total coke, not pore distribution), Raman spectroscopy (gives coke aromaticity, not pore distribution), and online mass-spec reaction monitoring (gives deactivation rate, not location). For the mesopore-mouth blockage question specifically, you still need to pull a sample and run MIP. We expect NMR-based methods (especially 129Xe NMR of adsorbed xenon) to reach commercial QC viability in the next 5 years, but they are not standard today.

ISO 9001 certified ZSM-5 catalyst powder and extrudates with full CoA including BET surface area 370-410 m²/g, micropore volume 0.13-0.17 cm³/g, Si/Al ratio customizable, MTO/MTP/FCC-additive grades on request.

For procurement engineering teams evaluating ZSM-5 catalyst or diagnosing spent-catalyst performance, we offer:

• You need to quantify micropore volume loss (Mode-A coke deactivation).
• You are doing routine post-regeneration QC where micropore recovery is the pass/fail metric.
• Your reactor pressure is too low for mercury intrusion — many catalyst labs do not have an MIP instrument because the 60,000 psia pressure range and Hg handling are expensive.
• You need BET surface area for vendor CoA verification (this is what most suppliers report on their standard certificate).

• Store in the original sealed container, off the floor.
• Target a cool, dry, ventilated store away from direct sunlight.
• Use first-in first-out and log every lot you open.

Total Cost of Ownership

The MIP-vs-N2 physisorption distinction is one piece of a broader QC framework for catalyst lifetime management. The following in-depth guides cover adjacent topics you may want to add to your diagnostic protocol:

QC Methodology

Online BET Surface Area Monitoring for Spent Catalyst Carriers: Sampling Strategy Pitfalls

Complementary guide on at-line BET sampling — why pulling a representative sample from a multi-ton catalyst lot is harder than the measurement itself.

QC Methodology

ISO 13320 Laser Diffraction vs ASTM D4464 Sieve Method for Fine Catalyst Powders Below 20 µm

Particle-size QC method cross-validation — relevant for distinguishing binder-fines accumulation from primary-particle attrition.

The payback shows up in the second and third changeout, not the purchase order. Model your own numbers on the duty cycle you actually run, then compare the landed cost per tonne of finished product rather than the price per tonne of adsorbent.

Related Products & Technical Resources

Regeneration Protocol

Molecular Sieve Regeneration Best Practices: Temperature, Flow Rate, and Cycle Time Optimization

Although focused on molecular sieve adsorbents, the regeneration temperature and O2 partial pressure framework applies directly to ZSM-5 coked-catalyst regeneration.

Failure Diagnosis

Spent Molecular Sieve Bed Sectioning: Layer-by-Layer Contamination Profiling from Top to Bottom

Layer-by-layer contamination profile methodology transfers directly to ZSM-5 fixed-bed reactors — pull samples from top, middle, and bottom to map coke axial profile.

Catalyst Selection

• <a href="/products/activated-alumina.html" class="text-brand underline hover:text-blue-700">Activated Alumina</a> - technical datasheet, available grades and typical applications.
• <a href="/products/3a-molecular-sieve.html" class="text-brand underline hover:text-blue-700">Molecular Sieve 3A</a> - technical datasheet, available grades and typical applications.
• <a href="/products/molecular-sieve-5a.html" class="text-brand underline hover:text-blue-700">Molecular Sieve 5A</a> - technical datasheet, available grades and typical applications.
• <a href="/products/13x-molecular-sieve.html" class="text-brand underline hover:text-blue-700">Molecular Sieve 13X</a> - technical datasheet, available grades and typical applications.
• <a href="/products/zsm-5-zeolite.html" class="text-brand underline hover:text-blue-700">ZSM-5 Zeolite</a> - technical datasheet, available grades and typical applications.
• <a href="/products/pseudo-boehmite.html" class="text-brand underline hover:text-blue-700">Pseudo Boehmite</a> - technical datasheet, available grades and typical applications.
• <a href="/products/catalyst-carrier-alumina.html" class="text-brand underline hover:text-blue-700">Alumina Catalyst Carrier</a> - technical datasheet, available grades and typical applications.
• <a href="/products/aluminium-hydroxide.html" class="text-brand underline hover:text-blue-700">Aluminium Hydroxide</a> - technical datasheet, available grades and typical applications.
• <a href="/blog/astm-d2650-liquid-density-hydrometer-slurry-catalyst-slurries-density-vs-surface-area-qc.html" class="text-brand underline hover:text-blue-700">Astm D2650 Liquid Density Hydrometer Slurry Catalyst Slurries Density Vs Surface Area Qc</a> - companion engineering article on the same duty.
• <a href="/blog/catalyst-carrier-attrition-index-astm-d4058-vs-abrasion-index.html" class="text-brand underline hover:text-blue-700">Catalyst Carrier Attrition Index Astm D4058 Vs Abrasion Index</a> - companion engineering article on the same duty.
• <a href="/blog/catalyst-carrier-drying-method-microwave-vs-oven-vs-spray.html" class="text-brand underline hover:text-blue-700">Catalyst Carrier Drying Method Microwave Vs Oven Vs Spray</a> - companion engineering article on the same duty.

Next Steps — Get a Sample or Talk to Our Engineers

Catalyst Carrier Surface Area vs Pore Volume: Which Spec Drives Your Activity?

Foundational guide on the BET-area vs pore-volume trade-off — explains why surface-area-only QC misses transport-driven activity loss.

Vendor QC

How to Read a Certificate of Analysis for Industrial Sorbents: 7 Fields Most Buyers Misread

How to spot a vendor CoA that omits micropore volume or BJH mesopore data — the red flags that the surface area number alone is not the full picture.

For catalyst evaluation, regeneration study, or commercial ZSM-5 procurement, the following Aluminaworld products are topically aligned with this guide:

Zibo, Shandong, China 15+ years manufacturer of activated alumina, molecular sieves, alumina powder, ATH, pseudo-boehmite, ZSM-5, calcined and tabular alumina. Exported to 60+ countries.

We hold stock on hand and ship within 15-25 days of order confirmation. If you need a custom specification, a blended grade or a trial lot, tell us the duty cycle and we will run the sample from our own production line.

Related Articles

• Mercury Intrusion vs Nitrogen Sorption for Catalyst Carrier Pore Analysis: When Each Lies | AluminaWorld
• ZSM-5 Catalyst Regeneration: Coke Removal Temperature Optimization for MTO, FCC, and Xylene Isomerization
• E-Fuel (e-SAF/e-Diesel) Synthesis Loop: CO2 Capture from DAC-Fed Methanol Synthesis Catalyst Bed