IntroductionBASF has introduced its latest FCC catalyst manufacturing innovation, the MSRC platform. This breakthrough manufacturing development takes advantage of staged reactions with different catalytic attributes much the same way that staged hydrotreating catalyst loading permits different reaction zones in a fixed bed reactor vessel. The concept of staged reactions is not new to the refining industry, but its application to a circulating system like FCC is a true step forward in catalyst technology.

The MSRC platform uses existing catalyst technologies such as BASF’s proven Distributed Matrix Structures (DMS)1, 2 and Proximal Stable Matrix and Zeolite (Prox-SMZ)3, 4 materials, but through a novel manufacturing technology combines two or more existing functionalities within the same catalyst particle. The location of the various stages can be specifically engineered to achieve maximum value for the refiner. This staging approach can be applied to allow processing of heavier feedstocks or to maximize specific product yields.

The manufacturing process for this unique approach is based on BASF’s in situ technology and involves several key

sequential manufacturing steps. One of the important attributes in the development of MSRC technology is the binding process. Unique to the in situ FCC manufacturing process, Y zeolite grows across the boundary between catalyst stages, acting as a binder and giving the catalyst particle excellent attrition resistance properties.

The first product from this new platform, Fortress™, is designed for resid feed applications, where contaminant metals passivation is critical. In particular, traces of nickel in the resid feed have a detrimental effect on the catalyst performance due to dehydrogenation reactions leading to coke and hydrogen production. In state-of-the-art resid catalysts, a nickel trap based on specialty alumina is used to trap the nickel and render it less detrimental 2. By using the MSRC concept, the spatial distribution of this specialty alumina within the particle is adjusted to maximize its efficiency in nickel trapping and lead to improved catalyst performance.

The MSRC manufacturing technology was successfully scaled up and demonstrated in 2010 and two refinery trials were initiated. This paper discusses the results of this effort.

Technical Note

Multi-Stage Reaction Catalysts (MSRC): A Breakthrough Innovation in Fluid Catalytic Cracking (FCC) Technology

MSRC Concept The approach of the new BASF MSRC platform is to combine two or more existing FCC catalyst functionalities within a single catalyst particle. The location of the various stages can be specifically engineered to achieve maximum value for the customer. This can be related to processing of heavier feedstocks, or to maximize specific product yields in the FCC unit. The MSRC platform can include existing catalyst technologies like DMS (for gasoline maximization) or Prox-SMZ (for diesel maximization), but is based on a novel and unique manufacturing concept.

The first product from this new platform, Fortress, is designed for resid feed applications, where contaminant feed metal passivation is crucial. In particular, traces of nickel in the resid feed have a detrimental effect on the catalyst performance. In state-of-the-art resid FCC catalysts like Flex-Tec®, a specialty alumina is integrated in the catalyst formulation to trap the nickel and form nickel aluminate which is less deleterious for dehydrogenation reactions in the FCC riser. By examining spent FCC catalysts from refineries with electron microscopy, it was generally observed that while vanadium is distributed homogenously through the particles, nickel mainly deposits and accumulates on the outer surface of the catalyst, as depicted in Figure 1. It would thus be advantageous to

concentrate the nickel trapping alumina at the outer layer of the catalyst to make it more effective. With current catalyst technology the specialty alumina is uniformly distributed throughout the catalyst microsphere. This results in a large portion of the alumina, located in the interior of the particle, being unavailable to react with the nickel and, in essence, wasted.

BASF has addressed this situation using the MSRC approach. The inner stage of the catalyst has the DMS structure to allow enhanced diffusion of heavy molecules and selective pre-cracking on the exposed zeolite surface, maximizing gasoline yields1. The outer-stage is also based on DMS technology, but is enriched with specialty alumina to trap the nickel where it enters and deposits on the catalyst surface as depicted in Figure 2. The nickel trapping technology is analogous to that employed with Flex-Tec catalyst2 and Stamina™ catalyst4, but the improved spatial distribution of the trapping alumina offers more efficient material utilization and better performance potential.

The manufacturing process for this two-stage reaction catalyst is based on BASF’s unique DMS manufacturing technology and involves multiple consecutive steps (Figure 3). First a precursor microsphere based on proven DMS technology with a reduced particle size is formed. This microsphere is then added to a second slurry, which has a formulation similar to Flex-Tec but includes elevated concentrations of the nickel passivating specialty alumina. The slurry is then spray-dried to yield a two-stage microsphere with the required final FCC particle size.

V o

r N

i Co

nce

ntr

atio

n (

wt%

)

Distance from the Center (microns)

0.6

0.4

0.2

0-50 -25 0 25 50

V Ni

Figure 1: Elemental line scans across equilibrium catalyst particles show distribution of contaminants nickel and vanadium

Alumina

Flex-Tec Fortress (Resid MSRC)

Figure 2: Simplistic representation of alumina distribution in current resid catalyst Flex-Tec (left) versus novel MSRC resid catalyst technology (right)

2

The FCC catalyst is then manufactured by growing zeolite from the kaolin nutrients in the multi-stage microsphere. The zeolite grows in both stages and across the interface and will act as a binder to hold the stages together, preventing selective attrition of the outer stage. Zeolite growth and catalyst activity are in line with current DMS platform products, as are all physical properties.

The novel MSRC and Ni passivation enhancement concepts result in cost-effective resid FCC performance that is a breakthrough innovation and offers significant potential for future MSRC applications.

Development of FortressThe basis for the development of the Fortress catalyst was the observation that Ni accumulates mainly in the outer 5–10 microns of the equilibrium FCC catalyst5. This is depicted in Figure 1. Geometric calculations show that the weight ratios of the two stages for an ~8 micron outer stage thickness and a normal FCC particle size distribution are about 1:1. The addition of a solid inner-stage particle to the recipe makes the production of MSRC more complex than existing FCC catalysts. Scanning electron microscopy of the early

H2O H2ODryer

Calciner

Dryer

Calciner

BaseExchange

Rare-earthExchange Calciner

Step 1: Inner stage microsphere (MS) Step 2: Outer stage microsphere

clay slurry

Alumina freeInner stage MS(~65 µm)

clay slurry+ alumina+ Inner stage MS

Two-Stage MS(~80 µm)

Step 3: Zeolite crystallization and after-treatment

MicrosphereCausticSilicate

Multi-Stage ReactionCatalyst

Na-Y

Figure 3: Manufacturing sequence of MSRC technology

Figure 4: SEM pictures of MSRC samples demonstrate the multi-stage concept

samples clearly demonstrates the multi-stage configuration of the particles (Figure 4). By modifying the microsphere formulation, the densities of inner and outer stage components

3

were optimized in order to allow maximum diffusion rates, without impacting the staged configuration. After successful demonstration of manufacturing MSRC catalysts on pilot scale and encouraging testing data, the project was quickly transferred to commercial manufacturing scale. The plant production trials required the capabilities of multiple manufacturing plants to first make the inner stage microspheres, followed by the construction of the outer stage material, and finally zeolite crystallization and finishing to make the catalyst (see Figure 2).

The particle size distribution of the resulting MSRC microspheres is shown in Figure 6. The graph shows that the inner stage particle has an average particle size (d50) of 64 μm and the final particle (inner + outer stage) of 80 μm, resulting in an average outer stage thickness of 8 μm. The thickness is fairly constant throughout the particle size distribution. This is important since the expected Ni penetration depth would be fairly constant across the entire particle size range, so that all particle size ranges should have comparable outer stage thickness.

The attrition resistance of the commercially manufactured catalyst was measured with two standard attrition testing techniques: Roller and Airjet (Table 1). The table shows that attrition properties are very similar to Flex-Tec, indicating that the two stages are very well connected and form a mechanically stable particle. Since the zeolite grows across the interface and chemically bonds to the DMS matrix support materials in both stages, attrition resistance and other physical properties such as density are not any different than existing DMS products such as Flex-Tec. This was also validated in the commercial performance, which is described later in this paper.

0

20

40

60

80

100

0 30 60 90 120 150

% F

iner

Th

an

Particle Size (micron)

Inner stage

+ Outer stage

0.20

0.25

0.30

0.35

0.40

0.45

0.50

Hyd

rog

en (

wt%

)

65.0 70.0 75.0 80.0

Conversion (wt%)

Fortress

Flex-Tec

Figure 5: Hydrogen yields in ACE testing of Fortress and Flex-Tec. The samples were deactivated with CMDU to 3000 ppm Ni and 1000 ppm V

Figure 6: Particle size distribution of inner stage and final plant produced microspheres

Table 1: Attrition properties

Flex-Tec Fortress

Roller (wt% loss/h)

18 16

Airjet Rate* (wt% loss/h)

2.7 2.9

*Losses in first hour of operation are disregarded

4

Flex-Tec Fortress

Hydrogen 0.53 0.44

Total C4- 15.9 15.5

Gasoline 49.4 50.4

LCO 20.1 20.1

HCO 9.9 9.9

Coke 4.7 4.1

Table 3: Product yields (wt%) at 70% conversion

The product from the first manufacturing campaign was evaluated in the lab using an Advanced Catalytic Evaluation (ACE) bench scale testing unit. ACE testing was done with a commercial resid feed at 960°F and variable cat/oil ratios to generate yield response curves. The performance of Fortress was compared to the Flex-Tec catalyst that was being supplied to the target trial refinery FCC unit. Prior to ACE testing the catalyst samples were deactivated and metallated using BASF’s Cyclic Metallation and Deactivation Unit (CMDU) with intermediate steaming to target approximately 2800 ppm Ni and 1100 ppm V, as well as commercially representative equilibrium catalyst activity. Table 2 shows the properties of the pseudo-equilibrium catalysts. As stated earlier, the chemical and physical properties of Fortress are very similar to Flex-Tec; the difference is the distribution of alumina which is concentrated in the outer-stage (Fortress) versus uniform (Flex-Tec).

It must be noted that the proper laboratory preparation of a pseudo-equilibrium catalyst for a resid cracking performance evaluation is not trivial. This is because conventional (Mitchell) pore volume metals impregnation methods result in Ni being distributed uniformly throughout the catalyst microsphere, thus penetrating much deeper into the microsphere than observed on a FCC unit’s Equilibrium Catalyst (E-Cat). Hydrogen production from impregnated metals will not show the benefit of the improved distribution of the alumina in the Fortress catalyst. Resid FCC units however will result in Ni deposition mostly on the alumina-enriched outer stage, so the correct answer will be for Fortress to outperform catalyst technologies employing uniformly dispersed active ingredients. This correct metals tolerance ranking cannot be obtained by pore volume impregnation techniques.

Table 2: Properties of catalysts for ACE testing

Flex-Tec Fortress

Fresh

TSA (m2/g) 300 301

MSA (m2/g) 72 69

REO (wt%) 3.1 3.7

APS (μm) 76 76

0–40 μm (wt%) 11 10

0–80 μm (wt%) 54 55

Deactivated (CMDU)

Ni (ppm) 2797 2718

V (ppm) 1133 1050

Steamed TSA (m2/g) 115 123

Steamed MSA (m2/g) 36 38

ZSA/MSA 2.2 2.2

5

Commercial Performance of FortressFortress entered into the first of two commercial trials in the second half of 2010. The trial that is furthest along is a short contact time resid unit located in the United States. The unit operates in deep partial burn and low reactor severity, with primary constraints of wet gas compressor and air blower capacity. The refiner’s economic objectives are to maximize residue rates while minimizing bottoms yields. With the excellent metals tolerance of Fortress, the expected reduction in hydrogen yield can alleviate the wet gas compressor limit and the improvement in coke selectivity can allow more operating flexibility within these constraints.

The unit was using Flex-Tec together with purchased E-cat for metals flushing. Flex-Tec had replaced a competitive catalyst in early 2010 based on the results of a catalyst evaluation. For a clear comparison, the purchased E-cat was removed prior to the start of the trial. As such there are three different periods of operation compared: Flex-Tec with E-cat, Flex-Tec without E-cat, and Fortress without E-cat.

With Purchased E-cat

FACT 71 wt%

Nickel 7400 ppm

Vanadium 2600 ppm

Delta Iron 0.23 wt%

Antimony 2000 ppm

Lead 100 ppm

Carbon on Catalyst 0.2–0.5 wt%

Equivalent Nickel 5600 ppm

Without Purchased E-cat

FACT 71 wt%

Nickel 8900 ppm

Vanadium 2700 ppm

Delta Iron 0.30 wt%

Antimony 2300 ppm

Lead 100 ppm

Carbon on Catalyst 0.2–0.6 wt%

Equivalent Nickel 6800 ppm

Table 4: E-cat evaluation during first trial

The trial was supported by analysis of the refinery’s E-cat and fines data. The refinery sent twice weekly E-cat samples and weekly fines samples. Samples enter BASF’s Refinery Returns system and are systematically analyzed for chemical properties, physical properties, activity, and selectivities. The activity and selectivities are measured on an ACE unit. This is a standardized testing method using a fixed fluid bed, which simulates Short Contact Time (SCT) riser performance. All the E-cats are evaluated under the same conditions (feed, temperature, and cat/oil ratio) providing a good basis for comparison. Below are the typical E-cat values for before and after the purchased E-cat was removed.

The ACE results are listed in Table 3. Fortress catalyst hydrogen and coke yields were reduced about 15% at similar metals content as Flex-Tec, with a corresponding increase in gasoline yield. Testing in the absence of metals showed no differences in coke and bottoms yields, demonstrating that no diffusion limitations occur in lab testing and that the effect observed in metals testing is a passivation effect. Since H2 production is a direct measure of metals dehydrogenation activity, it is clear that Fortress is achieving improved passivation. This is in line with the laboratory prototypes shown in Figure 5.

6

0.20

0.25

0.30

0.35

0.40

0.45

0.50

0.55

5000 5500 6000 6500 7000 7500 8000Equivalent Nickel (ppm)

AC

E H

2 Y

ield

(w

t%)

Flex-Tec with E-cat

Flex-Tec without E-cat

Fortress without E-cat

0.2

0.3

0.4

0.5

0.6

0.7

-200 -150 -100 -50 0 50 100Day

AC

E H

2 Y

ield

3000

3500

4000

4500

50005500

60006500

7000

7500

8000

Fortress without E-cat

Flex-Tec without E-cat

Flex-Tec with E-cat

Eq

uivalen

t Nickel (p

pm

)

ACE H2 Yield E-cat Equivalent Nickel

Figure 7: Figure ACE E-cat data - ACE hydrogen yield and E-cat equivalent nickel

Figure 8: ACE E-cat data - ACE hydrogen yield versus E-cat equivalent nickel

With the removal of purchased E-cat, the metals increased due to lower catalyst addition rates. Nickel alone increased from ~7,500 ppm to ~9,500 ppm. Despite the increase in metals, the hydrogen make off the ACE unit did not increase as shown in Figure 7. Equivalent nickel, calculated by Ni + V/4 + Fe/10 + 5 Cu – 4/3 Sb, represents the equivalent dehydrogenation activity of the metals on the E-cat and the antimony passivation effect. Transitioning from Flex-Tec without purchased E-cat to Fortress without purchased E-cat, the hydrogen selectivity measured by the ACE improved by 12%. The plot in Figure 8 shows how the average hydrogen yield decreased from 0.41 wt% to 0.36 wt% with Fortress, demonstrating the improved metals passivation.

7

1.0

1.3

1.6

1.9

2.2

2.5

-200 -150 -100 -50 0 50 1003000

3500

4000

4500

5000

5500

6000

6500

7000

7500

8000

Day

AC

E C

oke

Fac

tor

Eq

uivalen

t Nickel (p

pm

)

ACE Coke Factor Equivalent Nickel

Flex-Tec with E-cat

Flex-Tec without E-cat

Fortress without E-cat

2

2.5

3

3.5

4

4.5

5

5.5

6

AC

E C

oke

Yie

ld (

wt%

)

66 68 70 72 74 76ACE Conversion (wt%)

Figure 9: ACE E-cat data - ACE coke factor and equivalent nickel

Figure 10: ACE E-cat data - ACE coke yield versus conversion

The improved metals tolerance of Fortress can also be seen by the improvement in coke selectivity. Figure 9 shows the ACE coke factor (coke yield adjusted for activity) as it corresponds to the equivalent nickel. As with the ACE hydrogen yield, the removal of purchased E-cat did not increase the coke factor despite the increase in equivalent nickel of over 1000 ppm. With Fortress, the coke factor has significantly decreased. Comparing the ACE coke yield from the transition of Fortress to Flex-Tec (Figure 10), the coke yield decreased by 17%. This dramatic decrease in coke of ~0.7 wt% for the same conversion is unexpected. A general rule of thumb is a 7:1 ratio of hydrogen reduction to coke reduction. With a hydrogen yield reduction of 0.05 wt% this corresponds to an expected 0.35 wt% reduction in coke. A reduction in coke of ~0.7 wt% is twice what can be explained from the hydrogen reduction, and in a positive direction.

Since less mass is being converted into hydrogen and coke, the delta mass is showing up as 0.7 wt% more LPG + Gasoline Yield (Figure 11). A commercial unit can realize additional benefits from the improved coke selectivity by being able to lower the regenerator temperature and increasing cat/oil ratio, or by lowering the CO2/CO ratio, resulting in increased liquid yield.

8

61.0

61.5

62.0

62.5

63.0

63.5

64.0

64.5

65.0

65.5

66.0

67 68 69 70 71 72 73 74ACE Conversion (wt%)

AC

E L

PG

+ G

aso

line

(wt%

)Flex-Tec without E-cat

Fortress without E-cat

0

2

4

6

8

10

12

14

16

0.1 1 10 100 1000Particle Size (microns)

Fra

ctio

nal

(%

)

Day 103 Day 68 Day 40 Day -3 Day -72

Figure 11: ACE E-cat data - LPG + Gasoline vs. Conversion

Figure 12: PSD of ESP fines shows no changes in moving from Flex-Tec to Fortress

As expected from the fresh catalyst properties, there were no changes observed in attrition or unit losses following the transition from Flex-Tec to Fortress. The particle size distribution (PSD) of the fines from the Electrostatic Precipitator (ESP) shows no change across the transition (Figure 12).

A post-audit study of the trial was completed on the ACE unit using the refiner’s feed and mimicking the refiner’s operating conditions comparing the fully changed out E-cat samples of Flex-Tec and Fortress. The ACE comparison shows a decrease in hydrogen of 15%, decreasing from 0.65 wt% to 0.55 wt%. The coke yield decreased by 0.5 wt% for the same conversion. This is in alignment with both expectations from the development program and the data already discussed from the routine E-cat monitoring. Post audit ACE results are shown in Figure 13.

9

10

0.40

0.45

0.50

0.55

0.60

0.65

0.70

0.75

0.80

60 65 70Conversion (wt%) Conversion (wt%)

Conversion (wt%)Conversion (wt%)

Conversion (wt%)

Hyd

roge

n (w

t%)

75 80

Flex-Tec Fortress

50

52

54

56

58

60

62

64

66

60 65 70

LPG

+ G

asol

ine

(wt%

)

75 80

Flex-Tec Fortress

2.0

2.2

2.4

2.6

2.8

3.0

3.2

3.4

3.6

60 65 70 75 80

Dry

Gas

(wt%

)

3.8

Flex-Tec Fortress

6

8

10

12

14

16

18

60 65 70 75 80

Bot

tom

s (w

t%)

Flex-Tec Fortress

5

6

7

8

9

10

11

12

13

14

60 65 70

Cok

e (w

t%)

6

8

10

12

14

16

18

6 8 10 12 14Coke (wt%)

Bot

tom

s (w

t%)

Flex-Tec FortressFlex-Tec Fortress

75 80

Figure 13: ACE E-cat post-audit testing using the refinery feed and mimicking the refiner’s operating conditions

11

With Purchased E-cat

FACT 71 wt%

Copper 70 ppm

Nickel 2700 ppm

Vanadium 2200 ppm

Delta Iron 0.9 wt%

Antimony 600 ppm

Lead 200 ppm

Equivalent Nickel 3400 ppm

Without Purchased E-cat

FACT 69 wt%

Copper 70 ppm

Nickel 3500 ppm

Vanadium 3200 ppm

Delta Iron 1.1 wt%

Antimony 900 ppm

Lead 200 ppm

Equivalent Nickel 4200 ppm

Table 5: E-cat evaluation during second trial

There is a second trial of Fortress in another US refinery. The benefit of the improved nickel passivation in this unit is expected to be slightly less than the prior unit because the nickel is lower at ~3000 ppm. Like the previous refinery, this is a comparison to Flex-Tec, and purchased E-cat was removed from the inventory. The typical E-cat properties before and after the removal of purchased E-cat are shown below. Another feature of this unit is a very high level of contaminant iron (~1 wt%). At this time the unit is operating well with Fortress but it is too early to summarize the results of the trial.

ConclusionBASF has introduced an innovative new FCC catalyst concept, MSRC. The first product adaptation of this concept, Fortress, has been successfully developed and scaled up to commercial production and is demonstrating positive results for improved metals passivation performance in early refinery trials. BASF intends to both expand the number of Fortress applications and to introduce exciting new technologies using the MSRC concept.

References1) NaphthaMax® - Breakthrough FCC Catalyst Technology for

Short Contact Time Applications; J. B. McLean and D. M. Stockwell, NPRA AM-01-58.

2) Distributed Matrix Structures – a Technology Platform for Advanced FCC Catalytic Solutions; J. B. McLean, W. A. Weber, and D. H. Harris, NPRA AM-03-38.

3) The Role of FCC Catalyst Technology in Maximizing Diesel Production, J. B. McLean, NPRA AM-09-34.

4) Stamina - New FCC Catalyst for Maximum Distillate Yield Demonstrated in Big West’s Salt Lake City Refinery; M. Kraus, N. Kiser, Q. Fu, J. Yang, O. Thornton, and J. Finch, NPRA AM-10-171.

5) Kugler, E.L., and Leta, D.P., Journal of Catalysis 109 (1988), 387.

AuthorsJ. B. McLean Global Technology Manager BASF Catalysts Division Houston, TX

B. W. Hoffer Senior Chemist BASF Catalysts Division Houston, TX

G. M. Smith Research Chemist BASF Catalysts Division Iselin, NJ

D. M. Stockwell Senior Research AssociateBASF Catalysts DivisionIselin, NJ

A. S. Shackleford Technical Services Engineer BASF Catalysts Division Houston, TX

About Us

BASF’s Catalysts division is the world’s leading supplier of environmental and process catalysts. The group offers exceptional expertise in the development of technologies that protect the air we breathe, produce the fuels that power our world and ensure efficient production of a wide variety of chemicals, plastics and other products. By leveraging our industry-leading R&D platforms, passion for innovation and deep knowledge of precious and base metals, BASF’s Catalysts division develops unique, proprietary catalyst and adsorbent solutions that drive customer success.

BASF - The Chemical Company

Americas Sales Office BASF Corporation 25 Middlesex/Essex Turnpike P.O. Box 770 Iselin, NJ 08830-0770 Tel: +1-732-205-7188 Fax: +1-732-205-7725 Email: refining-catalysts@basf.com

Asia Sales Office BASF South East Asia Pte Ltd 7 Temasek Boulevard #35-01 Suntec Tower One Singapore 038987 Tel: +65-6337-0330 Fax: +65-6398-5104 Email: refining-catalysts@basf.com

Europe, Middle East, Africa Sales Office BASF plc Earl Road, Cheadle Hulme P.O. Box 4 Cheadle, CHESHIRE, SK8 6QG, UK Tel: +44-(0)-161-488-5156 Fax: +44-(0)-161-485-5487 Email: refining-catalysts@basf.com

Flex-Tec, Fortress, NaphthaMax, and Stamina are trademarks of BASF.

Although all statements and information in this publication are believed to be accurate and reliable, they are presented gratis and for guidance only, and risks and liability for results obtained by use of the products or application of the suggestions described are assumed by the user. NO WARRANTIES OF ANY KIND, EITHER EXPRESS OR IMPLIED, INCLUDING WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE, ARE MADE REGARDING PRODUCTS DESCRIBED OR DESIGNS, DATA OR INFORMATION SET FORTH. Statements or suggestions concerning possible use of the products are made without representation or warranty that any such use is free of patent infringement and are not recommendations to infringe any patent. The user should not assume that toxicity data and safety measures are indicated or that other measures may not be required. © 2011 BASF

BF-9722 USL 10/11www.catalysts.basf.com/refining