Catalysis 2026 Issue

Yield shifts (Denali Action versus incumbent)

Process variable shifts (Denali Action versus incumbent)

Delta (Denali Action-Action)

Feed rate, MBPD

-4.5

Corrected conversion Corr. dry gas yield

4.3

Feed API, º

0.2 1.0

-0.53 -1.37

Vol. avg. riser T, F

Corr. vol% C₃ Corr. vol% C₃= Corr. vol% C₄=

Regen. dense bed temp., F Cat cooler duty, MMBTU/hr

-5.0 -1.4

0.2

-0.3 -0.8

Delta coke, wt/wt

-0.04

Corr. vol% total C₄s Corr. vol% gasoline Corr. vol% LCO Corr. vol% slurry

Unit flue gas CO/CO₂ ratio Fresh cat dry add. rate, TPD Purch. E-cat add. rate, TPD ZSM-5 additive add., Ib/day

-0.1

2.1

0.1

-3.4

-1.6

-0.79

-567

Corr. wt% coke

0.64

E-cat FST, wt% conv.

1.2

E-cat Ni, wppm E-cat V, wppm E-cat REO, wt% E-cat P205, wt%

-417 -961

Table 1

-0.1 -0.3

measurement accuracy, process hysteresis effects, and other transient conditions that may not be completely captured by routine analysis. When reviewing calibration factors, there are four things to watch for: u Excessive variability: Fluctuations in calibration factor values by more than 20% from a nominal value may indicate a noisy process variable. v Flyers: A single calibration factor that deviates signifi - cantly from all preceding factors’ values usually indicates a measurement error. It may be possible to keep the case by correcting the bad data. Many factors on the same day taking on notably different values often indicate a process upset. In that situation, discarding the case may be prudent. w Trends: Gradual shifts are often observed during cata- lyst transition periods. Confirm if the trend correlates with catalyst changeout. If not, investigate other variables to understand the reason before using the calibration factors for predictions. x Correlation with independent variables: When correla- tion between calibration factors and process variables exists, this is an indication that model tuning may be necessary. Tuning is a different process from calibration, and due to its complexity, it is recommended to refer to model documenta- tion or developer assistance prior to attempting. Catalyst post-auditing As previously discussed, the model captures the fingerprint of catalyst formulation effects in the calibration factors. If the FCC unit was neither revamped nor subject to mechanical failures during the two time periods with different catalysts, the differences in calibration factors can reflect the effects of the catalyst change.

Purch. E-cat FST, wt% conv. Purch. E-cat Ni, wppm Purch. E-cat V, wppm Purch. E-cat REO, wt% Purch. E-cat P205, wt

2.0

-106 -172

0.1

-0.3

Table 2

The following example demonstrates a catalyst post- audit performed with a kinetic FCC model. The Marathon Anacortes Refinery FCC unit had a goal to improve liquefied petroleum gas (LPG) olefinicity, maintain bottoms cracking and activity, and lower delta coke. Ketjen developed the pro- prietary Denali Action catalyst to replace its incumbent Action catalyst. It incorporates ZT-600 zeolite technology, providing improved catalyst stability and better coke selectivity. Increased stability allows a lower RE content in the cata- lyst, reducing hydrogen transfer while maintaining activity. Enhanced coke selectivity allows for a higher cat-to-oil ratio; combined with lower RE, it allows for improved LPG yield and olefinicity. The bottoms cracking capability was held constant by maintaining constant inputs of MT-20 (matrix cracking functionality) and MT-60 (nickel-trapping alumina) matrix technologies. The Anacortes FCC unit is a challenging FCC unit com- prised of two risers, two main fractionators, a single par- tial-combustion regenerator, and two downstream CO boilers. Each riser is wholly independent, having unique feed composition, riser outlet temperature setpoints, and cat- to-oil ratios. Complicating the situation, purchased E-cat is added to the unit along with fresh catalyst. If it were possible to hold feed and independent varia- bles constant across catalyst periods, a catalyst evalua- tion would be straightforward: simply compare the yields. However, effective catalyst reformulations are designed to relieve operational constraints, enabling the unit to operate in formerly inaccessible and more profitable regimes. As the unit transitioned to Denali Action, feed rate, feed quality, riser temperatures, regenerator operation, catalyst addition rates, E-cat metals, and ZSM-5 additive consumption rates all changed. The only way to confidently separate catalyst effects from the feed and operational variable changes is to use a kinetic FCC model.

FST yield shifts (Denali Action versus incumbent)

E-cat (fresh + purchased)

Purchased only

Propylene C₄ olefins Isobutane Naphtha

-1.2 -0.5 0.05

-4.2 -1.9

-0.14

4.3

7.6

LCO

-0.2 -1.1

-0.3 -1.7 0.26

Bottoms

Coke

-0.56

Table 3

Catalysis 2026