logo


Feb-2004

FCC catalyst evaluation

A post audit at a Japanese refinery using a proprietary FCC model and catalyst-evaluation methodology determines a real improvement in profitability

Sanjay Bhargava, KBC Advanced Technologies

Viewed : 7279


Article Summary

After 10 years, a Japanese refinery decided to embark on an FCC unit catalyst change in order to increase profitability. The change was a tricky proposition, as the most important product from the FCC unit was decanted oil (used as needle coker feedstock), and the product specifications for this were very tight on sulphur, viscosity and aromatic content. The key to increasing profitability was to maximise the production of decanted oil while meeting its specifications by either maximising the feed and/or lowering the conversion. Yield estimates were received from different catalyst vendors and, based on a detailed evaluation, a single vendor was selected and their proposed catalyst introduced.

A post audit was then performed by KBC using its proprietary FCC model (CATOP) and catalyst-evaluation methodology to determine the real improvement in profitability.

As is typical in an FCC catalyst change, several operating variables changed during the two to three-month change-out. CATOP was employed to isolate the effect of catalyst alone on FCC yields using a unit factor (UF) methodology.

The actual benefit to the refinery as a result of the catalyst change alone was estimated to be $2.76 MM/yr. This included feed increase benefits of $2.0 MM/yr, yield credits of $0.8–1.0 MM/yr and catalyst cost savings of $0.16 MM/yr. The catalyst change allowed the refiner to run the FCC at a lower riser top temperature (by 5–6°C) at the same decanted oil specifications. This in turn resulted in a higher FCC feed rate, lower gas yields and higher liquid yields (better selectivity).

Refinery history
The refinery had used catalyst “A” for 10 years, occasionally making minor reformulation changes. It decided to test a new catalyst “B” for its FCC from a different vendor. The selection was based on a detailed evaluation of yield estimates from different vendors. The vendors were provided with the refinery’s typical FCC feedstock, economics, strategies and constraints. Catalyst B additions were started during the first week of September 1999. The percentage change-out was determined from the rare earth (RE) content of the equilibrium catalysts (E-Cat).
The catalyst B reformulation involved upping the zeolite on the catalyst significantly (increases liquid yield selectivity – less coke and gas) and increasing RE from 1.6–2.8 wt%. The total surface area was held constant.

Test runs used
The refinery had been performing weekly monitoring test runs using a monitoring calibration, target-setting application from mid-July 1999. This allowed it to benchmark the FCC operation prior to the introduction of catalyst B. The model is a Microsoft Excel-driven interface, with the FCC model running on the proprietary Petrofine platform. As part of the monitoring exercise, the application also generated catalyst-specific UFs used to determine the effect of the new catalyst on yields at the same constraints.

Catalyst A UFs and catalyst B UFs at different levels of change-out were determined by using the average of multiple test runs described as follows:
— Base case (Cat A) Average of three test runs 8/17, 8/20, 9/1/99
— 41% Cat B/59% Cat A Average of 10/20, 10/27/99
— 61% Cat B/49% Cat A Average of 1/13, 1/19, 1/26/00
— 73% Cat B/27%Cat A Average of 2/2, 2/9, 2/16/00

All good test runs were material balanced by adjusting the feed rate for a 100 per cent material balance closure (most test runs had 99 per cent closure on raw data). They were then used to generate UFs to isolate the catalyst effects. The test runs were grouped in the four sets previously described, and the average data was then extrapolated using statistical regression to obtain the effect of a 100 per cent change-out.
Some of the important parameters that also changed during the catalyst evaluation were:
— Feed rate increased by 15 m3/h
— Stripping steam rate increased by 0.8 tph
— Dispersion steam decreased by 0.7 tph
— LCO/slurry cut-point was changed by 10°C.

To determine the effect of the catalyst change on yields, the results of all of these changes were backed out of the evaluation using the model and UF methodology.

Unit factor analysis
Vendor B had performed pilot plant studies to determine shifts in yields at constant conversion. The results were as follows:
— 11 per cent reduction in coke make selectivity
— 7 per cent increase in (C1 + C2) selectivity
— 7 per cent reduction in LPG selectivity
— 1 per cent increase in bottoms yield selectivity.

A comparison of the UFs from the 11 test runs showed the following:
— 11 per cent increase in coke make selectivity
— 32 per cent increase in (H2 + C1 + C2) selectivity
— 10 per cent increase in LPG yield selectivity
— No change in bottoms yield selectivity.


Add your rating:

Current Rating: 3


Your rate: