March 2024

Catalysts

Decarbonize the FCCU through maximizing low-carbon propylene

As the global energy system continues to be threatened by geopolitical tensions, the refining industry continues to demonstrate an outstanding capacity for adaptation.

IP: 3.141.100.120

As the global energy system continues to be threatened by geopolitical tensions, the refining industry continues to demonstrate an outstanding capacity for adaptation. Oil flows swiftly, rerouted after new sanctions have been implemented, with the refined products trade shifting accordingly, ensuring the critical supply of energy to global markets. For this global energy system, energy security at an affordable cost is vital to continue the energy transition, which will reinforce the system with a more diversified, sustainable energy network.

Commitments in carbon dioxide (CO₂) reduction from many of the main refining industry players continue, with goals to reach net-zero emissions by 2050 or earlier in some cases. These targets require reductions in Scope 1 and Scope 2 emissions from daily operations, as well as increased deployment of renewable energy, the production of biofuels, or the building of a hydrogen (H₂) infrastructure. Anticipating the peak and decline in the demand for traditional fossil transportation fuels, many refineries are flexing their operations toward an increased petrochemicals yield, improving their resilience.

Among all petrochemical precursors produced in a refinery, propylene is possibly the most important product. Even though current market conditions for propylene are challenging, the demand for propylene is projected to increase by 45 MMtpy through the remainder of this decade (or 4% annualized), from which refinery contributions are expected to be constant at around 27%. In other words, more propylene from refineries is needed to satisfy the growing demand. One of the key units in the refinery is the fluid catalytic cracking unit (FCCU), which converts a variety of heavy hydrocarbon feedstock types to lighter, higher-value products, including transportation fuels and vital petrochemical precursors like naphtha and propylene. The FCCU is expected to play an expanded role in satisfying the growing demand for petrochemicals, while also providing clean, lower-carbon fuels. The feedstock and operational flexibility of the FCCU helps to minimize the product carbon intensity profile and to adjust the product slate to meet ever-changing market needs.

The co-authors’ company^a has been developing and implementing strategies around decreasing carbon intensity in the FCCU¹ with partners, while also leading the way on co-processing alternative non-fossil feedstocks in the FCCU.^2,3 Additionally, the co-authors’ company has introduced the concept of low-carbon propylene,⁴ with a focus on FCCUs operating to maximize propylene yield, while minimizing carbon emissions.

This article describes the value and sustainability journey that Cepsa has been on with respect to the FCCU, and how the co-authors’ company has supported the company’s efforts at every step along the way. Part 1 describes how the two companies worked together on a catalyst reformulation that allowed expanding the feed slate to include resid feedstock. As a result of the reformulation, the refinery was able to take on the heavier feed slate, which led to increased profit, but this also led to an improved coke selectivity profile, allowing fuels to be produced in a more sustainable manner. Part 2 describes how the co-processing of bio-based feedstocks resulted in a higher propylene yield, and, since the incremental propylene yield is of biogenic origin, it has a lower carbon intensity compared to propylene produced from conventional feedstocks. The authors believe that this is a tremendous opportunity for propylene from the FCCU. While there is the additional challenge of oxygenate removal, the authors discuss available solutions to meet the polymer-grade specifications. Finally, Part 3 describes the value of lab testing to reduce risks, and the successful trial of coprocessing bio-based feedstock in the commercial FCCU. 

Positive Motion: The transformation from traditional refineries to energy parks

Cepsa launched its Positive Motion strategy with the aim to lead in sustainable mobility and energy to create value and a brighter future for society. It includes, among other initiatives, building an e-mobility infrastructure on the Iberian Peninsula; fostering H₂ commercial road transport demand, including the production of green H₂; and supporting the production of biofuels.

Biofuels are typically classified as first-generation or advanced biofuels, depending on whether they come from raw materials competing with the food market or organic waste, respectively. Due to their renewable nature, they generate significantly lower net CO₂ emissions during their lifecycle than traditional fossil fuels (up to 90% less). They come from biomass that has absorbed CO₂ from the atmosphere during its growth. Biofuels’ capacity for direct substitution/drop-in contributes to rapid decarbonization in all sectors where they are applied, especially in those with complex electrification, such as aviation, heavy land transport or maritime transport. Also, they help to increase energy independence by expanding the diversity of energy sources. They can accelerate the energy transition, since they do not require changes in existing vehicles, logistics and/or distribution, enabling immediate implementation. Second-generation biofuels are generally a preferable option, as they do not compete with agricultural crops and minimize waste ending up in landfills or in the environment. 

As part of Cepsa’s Positive Motion strategy, the company is transforming its refineries into diversified energy parks, where Cepsa applies innovation and sustainability to develop new green products that allow Cepsa to decarbonize the productive processes, helping other sectors in their decarbonization goals. These energy parks are strategically located in Southern Europe next to key ports favoring trade. They use digitalization developments like the Internet of Things (IoT) combined with advanced analytics to optimize production processes and build on excellence in logistics. The increased production of biofuels and integration of green H₂ and renewable energies are also key objectives of these energy parks.

PART 1: EXPANDING THE FEED DIET WINDOW THROUGH CATALYST REFORMULATIONS

The FCCU sits at the heart of the refinery as a key conversion unit cracking heavy, low-value hydrocarbon streams into lighter, high-value liquid fuels and petrochemical precursors. The FCCU has proven to be one of the most versatile processing units in a refinery due to its ability to process a wide variety of feedstocks with different properties and contaminant levels. Cepsa Gibraltar has a long history of adding value to its operations through reformulations in the company’s max-propylene mode FCCU,⁵ usually targeting to increase propylene, butylene and gasoline production, while minimizing slurry, dry gas and coke. The unit^b has a side-by-side design, with a high-efficiency regenerator. It was also recently revamped to a packed stripper and a vortex separation system (VSS) separator design.

Cepsa Gibraltar’s FCCU historically processed blends of pre-treated and straight-run vacuum gasoil (VGO), while adding some amounts of other process streams, when needed. Evaluations of optimal economics at the refinery concluded that processing atmospheric residue was profitable, which pushed the team to increase its content in the feed to concentrations never explored before. Residual feedstock is heavier by nature, and it contains more polyaromatics and has a higher level of contaminants—including vanadium (V), nickel (Ni) and iron (Fe), among others—compared to traditional VGO feedstock types. Broadly speaking, a residue-to-propylene catalyst was designed to withstand these harsher conditions by incorporating metal traps for V and Ni, while adjusting the zeolite hydrogen transfer and Z/M ratio to enhance activity retention while still delivering high amounts of olefins and preserving optimal bottoms cracking.

Cepsa asked the co-authors’ company to help mitigate the negative impacts of processing heavier feedstocks through catalyst reformulation. Several options were tested, leveraging vast experience in residue-to-propylene units⁶ and implementing catalysts formulated to withstand harsher unit conditions and contaminant metal levels. A crucial performance driver was to improve coke and dry gas selectivity. To better understand the enhanced metals passivation, tests were conducted in a pilot plant^c at two different contaminant metal levels to reflect historical operation with light feeds (800 mg/kg V + 1,800 mg/kg Ni), as well as future operation with elevated contaminants (1,200 mg/kg V + 2,800 mg/kg Ni).

FIGS. 1 and 2 show step-out improvements in coke and gas selectivity, with the new reformulation at the high metals condition. The improved Ni trapping was sufficient to cover all the metal expected in the new conditions, as seen in the constant dry gas observed at the two metals levels. Coke selectivity was improved, as seen in TABLE 1. In addition, the improved activity retention was clear when examining yield deltas at a constant cat-to-oil ratio (TABLE 2), which helped minimize catalyst additions (kg catalyst/thousand tons of feed) for the residual feedstock case. These promising results in the laboratory provided Cepsa with the confidence to proceed with a commercial trial of the catalyst.

 

During the commercial trial, close monitoring of both the circulating equilibrium catalyst (Ecat) inventory’s physical and chemical properties and the unit data was conducted. The primary objective of the new catalyst design was to improve catalyst stability to maintain optimal levels of activity and in-unit conversion at higher Ecat metals content. FIG. 3 shows the Ecat microactivity test (MAT) remained at 2–4 times higher activity at equivalent Ecat V levels compared to the previous catalyst. The higher metals tolerance and hydrothermal stability enabled constant specific catalyst additions (kg catalyst/thousand tons of feed), while maximizing the conversion to liquid products and improved coke selectivity. The catalyst reformulation better positioned the unit to improve profitability while processing heavier feeds, and also reduced the carbon intensity of FCC products through coprocessing non-fossil-derived feedstocks.

 

PART 2: LOOKING AHEAD—PRODUCING LOW-CARBON PROPYLENE IN THE FCCU VIA COPROCESSING

The propylene produced by the FCCU already has a relatively low carbon footprint compared to other commercial-scale propylene production processes due to the FCCU’s catalytic conversion via β-scission and the relatively moderate reaction conditions (FIG. 4).⁶ The end member product of β-scission is C₃=.

 

There are routes to further improve upon the carbon intensity of C₃= product from the FCCU, including addressing Scope 1 emissions by reducing the severity (riser temperature, feed temperature, catalyst cooler duty or degree of partial burn) of the FCC process, or by addressing Scope 3 emissions by reducing the fossil carbon content of FCC propylene by renewables coprocessing.

The FCCU is unique in its operation in heat balance. The unit generates sufficient coke yield to satisfy the heat requirements for feedstock heating and vaporization, endothermic catalytic cracking and heat losses. Therefore, reducing the coke yield requires a reduction in one of the heat requirements. Lowering the riser outlet temperature (ROT) of the FCCU will reduce the production of coke and, therefore, CO₂ emissions from the FCCU regenerator. Commercial data has revealed that coke yield and corresponding CO₂ emissions could be reduced by 5%–10%.

In a 40,000-bpd FCCU, a ROT reduction of 25°C reduces coke yield and CO₂ emissions by about 0.4 wt% FF and 8%, respectively. The latter results in CO₂ reductions of 30,000 tpy. Such a significant severity reduction affects the overall feedstock conversion and yield pattern from the unit. This ROT reduction causes the commercial unit’s propylene (C₃=) yield to decline significantly (> 2 wt% FF), which negatively impacts the refinery’s profitability. To mitigate such a detrimental impact on light olefin yields and profitability, the operator should consider adding a ZSM-5 light olefins additive to the unit, which selectively increases the yield of propylene, butylene and isobutane by cracking FCC gasoline components. The use of such an additive compensates for the lower FCCU’s impact on light olefin yields. The additive has no significant impact on the coke yield and, therefore, on FCCU Scope 1 CO₂ emissions. To balance the significant decrease in C₃= yield, a high-activity commercial additive^d is recommended to limit the dilution effect of the additional catalytic component⁷ on the base-cracking catalyst.

Another route to reduce the carbon intensity of the FCCU C₃= product is to coprocess some renewable feed streams to the FCCU.^4,7 The higher the coprocessing rate, the greater the impact on the carbon intensity of the related FCC products. Assuming an equal distribution of the renewable carbon among the FCC products, the coprocessing rate (mass-based) can be directly related to a reduction in carbon intensity; consequently, the oxygen content of the renewable feed source must be considered. The oxygen content in the renewable feed is mostly converted to water, carbon monoxide (CO) and CO₂ yields. It can be estimated that, considering a coprocessing rate of 10 wt% renewable feed with an oxygen content of about 10 wt% (in the range of many seed oils), the carbon intensity of the resulting C₃= would reduce by 9%.

The ultimate impact of coprocessing renewable feed components on the yield structure is likely to be different to this mass balance approach. However, this must be determined in FCCU pilot plant testing and commercial applications, as they depend on the fossil feed type, unit conditions and FCC catalyst properties. To assess the amount of C₃= stemming from the renewable feed component, highly sophisticated analytical methods for modern carbon determination might be required.⁸

Data testing for some renewable feed types and test conditions showed that the renewable carbon containing feed might be preferentially converted to C₃= compared to fossil feed components. FIG. 5 shows bench scale pilot plant testing results, which indicated that the C₃= yield in this case (fossil feed type, quality and catalyst dependent) increased by about 0.3 wt% FF by blending 9 wt% palm oil with the VGO. Considering the incremental yield concept,⁹ it is estimated that palm oil yields 6 wt%–7 wt% FF C₃=, nearly double the yield of the fossil-based VGO in this particular case.

 

While FIG. 5 illustrates the potential C₃= increase by renewable coprocessing, challenges with coprocessing should be considered. These potential challenges are often associated with the significantly higher oxygen content of the renewable feed component relative to traditional feedstocks. Despite the absence of added H₂, the FCC process offers a high degree of deoxygenation of renewable feed streams. Most oxygen species are converted to hydrocarbons and water, CO₂ and CO, which will leave the FCCU on the reactor side and could pose challenges downstream. In pilot plant testing of renewable feed coprocessing, the effects of trace oxygenates are often not considered. Nevertheless, these are likely to occur with oxygen-containing feed streams. Trace amounts of oxygenates are commonly found in fossil-feed-based FCC product streams like LPG or cracked naphtha. Increasing the combined FCC feed oxygen content by the coprocessing of renewable feed streams like vegetable oils will increase the amount of these oxygenate species. This might negatively influence the downstream processing of the FCCU products and cause products to exceed specification limits. Pilot plant testing will help operators understand the magnitude of changes in oxygenates and water, CO and CO₂ yields. In addition, a close cooperation with the catalyst supplier will help discover areas of concern and monitoring requirements.

The impact of biomass processing on propylene quality

The coprocessing of oxygen-containing feedstocks like renewable feeds will lead to the increased formation of oxygenate molecules (acetaldehyde, acetone and others) that will fractionate into the C₃ fraction downstream of the FCCU. FCCU operators aiming to produce chemical or polymer-grade propylene must be aware of an increased impurity presence in their propylene grades. Especially with polymer-grade propylene, oxygenates will have a negative poisoning impact on polypropylene (PP) catalysts and must be removed.

Oxygenate removal from propylene can be accomplished using wide-pore zeolite molecular sieves, bringing the oxygenate contaminant levels down to the required specifications (< 1 mg/kg). The adsorption targeted is a thermal swing adsorption process, where the adsorbent is regenerated with a clean, dry regeneration gas at higher temperatures (250°C–260°C). Due to the olefinic nature of propylene, a freshly regenerated bed must be carefully returned to service. Olefins adsorb on wide-pore zeolite molecular sieve adsorbents, and excessive heat of adsorption of the olefinic molecules could lead to a thermal runaway. Therefore, the co-authors’ company recommends using a preloading step where propylene is introduced to the fresh bed at low concentrations, and then slowly ramped up to feed concentration levels while monitoring the adsorbent bed temperatures. This can be done in a vapor phase or in a liquid phase. The co-authors’ technology company offers the adsorbent types needed to help refiners properly design oxygenate removal units. Based on the company’s longstanding experience in adsorption technology and its application in PP technology^e when designing the purification sections of the PP plant, the co-authors’ technology company is ideally positioned to support process design and implementation for improved C₃ purification. 

In industry, hybrid adsorbents are sometimes considered for such applications. Hybrid adsorbents combine the adsorption properties of a zeolite molecular sieve with a carrier material, moderating the adsorption capacity. Competitors offering hybrid adsorbents typically highlight that no preloading step is needed. However, due to the dilution of zeolite, adsorption capacity hybrid products require larger unit designs. Therefore, while the preloading step can be avoided, the energy demand and regeneration requirements will be larger.

PART 3: COPROCESSING BIOFEEDSTOCK IN THE COMMERCIAL UNIT

Lab testing to minimize risks

The Cepsa Research Center is committed to pathways to decarbonization set by the 2030 Positive Motion strategy. Cepsa promotes research and open innovation as key drivers to increase the technology readiness levels of novel processes. This project was an example of close work involving research and development (R&D), energy parks and open innovation schemes, as well as cooperation with other innovative companies.

The conversion area at the research center specializes in FCC and deep-conversion processes, providing technical support to the energy parks’ industrial FCCUs. Furthermore, it develops research projects in this theme. In this case, the research center carried out pilot plant tests for coprocessing vegetable oil. The primary aim was to derisk operations by providing preliminary data before proceeding with an industrial trial. Real VGO, Ecat and vegetable oil from the San Roque Energy Park were used in these tests. The experiments were performed in a Davison circulating riser (DCR) pilot plant, which was adjusted to mimic the industrial San Roque FCCU conditions. The DCR is a fluidized-bed pilot plant that functions like a real FCCU. Several pilot plant runs were performed, covering a coprocessing rate of 5% and 10% vegetable oil in the total feed.

The experimental results were expressed as the gap in yield between a Base Case and the coprocessing cases (TABLE 3) and were minor (< 1%) in all cases. For typical FCC products, negative effects were observed, except in coke yield. This loss in yield could be explained due to the formation of water, CO and CO₂ during the cracking of the carboxylic moiety of triglycerides. The production of gases will increase if the coprocessing ratio increases. In any case, the vegetable oil did not have an impact on the yield and quality of FCC products. In addition, operational issues were not observed during the tests. Therefore, the main conclusion of the study was that an industrial trial for coprocessing 5% vegetable oil in the FCCU could be carried out without immediate risks to safe and stable operation.

Commercial unit trial

As part of its decarbonization Positive Motion program, the Cepsa Gibraltar refinery proceeded with a biofeedstock coprocessing trial in its FCCU. The vegetable oil used had a density of 0.915 g/cm³, similar to the VGO quality typically processed at the unit, while the oxygen content was 11.5 wt%. A commercial trial was conducted to increase non-fossil feed from 0 wt% to 5 wt% in a stepwise fashion. The trial was conducted at constant operating conditions, which reflected typical operations. Key trial parameters are listed in TABLE 4.

 

No major operating issues were observed during the trial. Increased sour water production was observed and handled accordingly. The yield deltas—detailed in TABLE 5—were minor and were within typical variations often encountered with feed quality or type changes. It was concluded that this vegetable oil was processable in the FCCU with minimal impact and that it did not constrain the unit. Given this positive experience, the refinery team is considering further expanding the feedstock window with second-generation biogenic oils to further minimize the carbon intensity of the FCC products.

Takeaway

The refining industry must adapt quickly to new ways of thinking, and to a new way of imagining processes to successfully navigate the ongoing energy transition. Close collaboration between trusted partners is a key enabler to accelerate adaptation to the changes in the business environment.

Cepsa’s Positive Motion program is an ambitious package of strategic initiatives to develop leadership in the energy transition going forward. These initiatives include expanding the company’s energy parks’ feedstock qualities and adapting processes to minimize product carbon intensity.

The co-authors’ technology company has joined Cepsa’s San Roque Energy Park (and R&D Center) as a trusted partner for this journey. This article has highlighted the importance of innovative companies partnering to deploy the right catalyst technologies and technical services to defossilize the FCCU. HP

NOTES

^a W. R. Grace

^b The unit’s technology is licensed by Honeywell UOP

^c ACE™ pilot plant

^d W. R. Grace’s ZAVANTI™ additive

^e W. R. Grace’s Unipol® PP technology

LITERATURE CITED

¹ Gonzalez, R. and B. Riley, “Identifying value from CO₂ reduction in the FCC unit,” Catalagram, December 2022.

² Gonzalez, R., M. Bescansa, A. Fernandez, A. Mena and C. Rivas, “Defossilizing the FCCU via coprocessing of biogenic feedstocks: From laboratory to commercial scale,” Hydrocarbon Processing, July 2023.

³ Lee, G., S. Brandt and D. Holder, “Maximizing renewable feed coprocessing at an FCC,” PTQ, July 2023.

⁴ Cipriano, B., C. Cooper and S. Brandt, “Paving the way to low-carbon propylene from the FCC unit,” Decarbonization Technology, November 2023.

⁵ Gonzalez, R., J. Llano, B. Aramburu, R. Larraz and C. Chau, “FCC catalyst for maximum propylene,” PTQ, 4Q 2016.

⁶ Gonzalez, R., et al., “A successful case of resid-to-propylene maximization using premium catalyst technology,” Hydrocarbon Processing, October 2022.

⁷ Serban, S., C. Ekeocha, B. Cipriano and U. Singh, “Maximizing yields and profits from the FCC unit,” PTQ, 3Q 2021.

⁸ Seiser, R., J. L. Olstad, K. A. Magrini, R. D. Jackson, B. H. Peterson, E. D. Christensen and M. S. Talmadge, “Coprocessing catalytic fast pyrolysis oil in an FCC reactor,” Biomass and Bioenergy, 2022.

⁹ Harding, R. H., X. Zhao, K. Qian, K. Rajagopalan and W.-C. Cheng, “Fluid catalytic cracking selectivities of gasoil boiling point and hydrocarbon fractions,” Industrial and Chemical Engineering Research, 1996.

The Authors

Related Articles

From the Archive

Comments