Feature articles

Raising hydrocracker performance in ULSD production

15/01/2008

Oil & Gas Journal

Customized catalyst systems, state-of-the-art reactor internals, and outstanding technical cooperation help North Atlantic Refining, Ltd. to maximize profitability

Introduction

Refiners everywhere face the challenge of producing fuels to meet increasingly stringent specifications and remaining competitive. Changing regulations, poor operational performance and sub-optimal yield structure based on feedstock or catalyst choices can negatively impact profitability. Hydrocracking is one of the major conversion processes in refineries and contributes a significant part of refinery profitability. It is an extremely robust and versatile conversion process.

Depending upon local market demand and refinery economics, hydrocrackers have been designed to produce primarily either naphtha or distillate products. The projected future increase in distillate demand and the more stringent quality specifications (e.g. ULSD, cetane, aromatics) increase the incentive for hydrocrackers to be operated in the most optimal and profitable ways.

Hydrocrackers that were designed as state-of-the-art 20 to 30 years ago in an environment of relatively low crude prices may not be as profitable as a unit designed recently. Common requests from the refinery to improve profitability and reliability of an existing hydrocracker include how to increase feed rate, how to process more difficult feeds (e.g. higher feed end point, FCC cycle oil, coker gas oil, DAO, synthetic feed), how to increase cycle life, how to produce the most profitable yield slates and how to improve product qualities. The wish list could be extremely long such that there is no unique solution. A close cooperation between the refinery, head office process engineering, technology provider and catalyst supplier is essential to identify profit enhancement opportunities and implement cost effective solutions for an existing hydrocracker.

North Atlantic Refining, Ltd. (North Atlantic) located in Newfoundland, Canada is equipped with a 37000 barrel per day hydrocracker. Over the years, the hydrocracker faced similar challenges as the rest of industry. The need to extend the cycle life and improve the product qualities became more obvious and urgent. By working closely with Criterion Catalysts & Technologies, Zeolyst International and Shell Global Solutions, North Atlantic has been able to justify and implement several key improvements to the hydrocracker including tailored catalyst systems, state of the art reactor internals, replacement of air-coolers and avoiding feedstock contamination.  The joint effort has resulted in a run length that is on target to be 1 year longer than the best cycle to date, more profitable product yield structure and ULSD quality distillates compared to previous cycles.

North Atlantic Refining, Ltd

The history of the North Atlantic Refinery includes ownership by a number of different companies since its construction. Shaheen Resources originally built this oil refinery between 1971-1973 with its first shipment of crude oil being refined in May 1973. Shaheen operated the refinery until 1976 at which time the company went bankrupt and the refinery was shut down. The oil refinery was refurbished and brought back online 10 years later, under new owners, Newfoundland Processing Ltd (NPL). In August 1994, North Atlantic Refining Ltd. purchased the refinery from NPL, and after a major overhaul, has been transforming sour crude into some of the cleanest fuels on the world market since that time. Harvest Energy Trust acquired North Atlantic Refining Ltd in October 2006. The North Atlantic refinery is now the second leading oil refinery (utilization) in Canada and gainfully employs over 700 Newfoundlanders.

The 115,000-barrel-per-day oil refinery is strategically located in the island portion of the province of Newfoundland and Labrador, in the country of Canada, on the far eastern side of North America. This position gives the refinery unique access to petroleum product markets in Europe and the U.S. eastern seaboard, and puts it in close proximity to sour crude supplies from Russia, Venezuela, and the Persian Gulf. At North Atlantic, the company has capitalized on innovations that were designed into the facility on the drawing board. When the refinery was built in the early 1970s, it was unusual to build a hydrocracker in this size of a facility. However, Shaheen invested in a hydrocracking unit. Additionally, investment was made at this facility to build one of North America’s largest refinery docks. Time has revealed the beneficial consequence of those decisions. 

Now, the clean fuel technology hydrocracker gives the company a growing advantage in producing low-sulphur, clean fuels from lower-cost Russian, Venezuelan, and Middle Eastern sour crudes. This type of processing gets the job done in the new millennium as crude quality decreases while the demand for clean petroleum products increases. This refinery has a sour crude configuration and that is advantageous considering much of the proven oil reserves around the world are light to medium sour. On top of all that, North Atlantic can take advantage of lower freight costs as larger ships can tie up at the ice-free dock all year around. While the achievements are impressive, the company continues to work towards the future and the more challenging goals that have been set for tomorrow.

Challenges and Opportunities for North Atlantic Hydrocracker Operation

The hydrocracker at North Atlantic is a series flow, single stage hydrocracker with two parallel reactors and each reactor contains 3 beds (Figure 1). For the last decade, the first bed in each reactor typically has contained a combination of demetallization and pretreat catalyst to remove feed contaminants and reduce nitrogen slip to protect the cracking catalyst in the second and third bed of each reactor.

Figure 1: Simplified unit flow diagram for the North Atlantic hydrocracker.

The refinery has faced numerous issues concerning the performance of this hydrocracker. In the early 90’s a good cycle length was 9 months. As catalysts improved, and with additional focus on operational issues, the cycle length was extended to approximately 3 years. However, there were still issues with high deactivation rates due to processing visbreaker gasoils, with high pressure drop incidents due to upstream unit upsets, with inability to control the temperature in the High Pressure Separator during high ambient temperature periods, and with high radial temperature profiles in each of the catalyst beds. Therefore, it was decided to address all the above issues during the current cycle. The specific areas of technical improvement to the hydrocracker were:

• Optimizing feedstock selection and opportunity crude processing
• Applying an advanced tailored catalyst system provided by Criterion & Zeolyst
• Process improvements and installation of Shell Global Solutions reactor internals
  

Feedstock Improvements (technical cooperation)

Normally, the processing of visbreaker gasoils is a profitable activity. However, due to operational and separation issues, the Visbreaker Gasoils were more contaminated than desired for processing in the hydrocracker leading to accelerated deactivation of the catalyst. Criterion and Zeolyst, as part of their continuous technical support, provided North Atlantic with the yield and activity impact for removing the Visbreaker Gasoil from the feed diet, confirming North Atlantic’s decision to remove this component from the feed diet until the quality issues could be addressed.

Another critical factor that influences hydrocracker profitability is unplanned contamination of the feedstock. During the review of the previous cycle, it was determined that the line used to transfer the imported gasoil to tankage prior to feeding the hydrocracker was also used for transfer of #6 Oil to the wharf. #6 Oil contains higher metals, sulfur, nitrogen, and carbon residue than the typical hydrocracker feed. When the procedure was reviewed, additional steps were added to confirm that the line was properly flushed so that the hydrocracker feed would not be contaminated with #6 Oil. These steps were deemed important because the transfer of imported gasoil for the hydrocracker only occurs on a monthly or longer frequency.

Catalyst Improvements

The close cooperation between North Atlantic and Criterion/Zeolyst started over a decade ago when a first cracking catalyst (Z-673) was custom developed by Zeolyst for North Atlantic’s hydrocracker design, feedstock and operating constraints. Subsequent improvements in demetallization catalyst and pretreat catalyst have allowed North Atlantic to further optimize the catalyst load by including a more distillate selective catalyst and to further reduce the light gas production that constrains the North Atlantic hydrocracker at the end of run. Most recently, after extensive evaluation, a stacked bed of Zeolyst’s high distillate selective cracking catalysts, Z-673 and Z-623, was chosen for this cycle. Another important catalyst parameter applied to this cycle is the new TX shaped version of Z-673 and Z-623 for pressure drop reduction and other benefits.

As discussed in the next section, the replacement of the older generation internals with new internals from Shell Global Solutions, allows the catalyst volume in the reactors to be increased by 11%. Additional catalyst volume with the same loading method and same catalyst size would lead to increased pressure drop. However, the unit had previously faced premature end of run due to high pressure drop across the catalyst beds. There were two pressure drop issues. The first issue was excessive pressure drop across the first bed due to particulates and crust formation. The second issue was the total pressure drop across the reactor loop from compressor discharge to suction. The additional catalyst volume allowed all of the partners to consider installing additional demetallization catalyst and grading material to deal with crust formation, but the total pressure drop issues still needed to be addressed. Criterion and Zeolyst developed the patented TX Trilobe shape to assist with the mitigation of the increased pressure drop due to the increased catalyst load. This catalyst shape was presented in a previous NPRA paper (1) and a detailed description can be found in another NPRA paper (2).

Figure 2: Cross section of the TL trilobe shape compared to the new TX trilobe shape. The TX trilobe shape further reduces the diffusion path and simultaneously decreases pressure drop through a higher void fraction.

In addition to the new catalyst shape for the cracking catalysts, this cycle has also taken advantage of Criterion’s improvements in demetallization and pretreat catalysts with the installation of RM-5030 and DN-3300 in place of the RN-412 and DN-3100 that was loaded during the previous cycle. The decision was made based on the commercial experience of up to 15°F improvement in HDN activity for DN-3300 over DN-3100.

Process Improvements (Reactor Internals)

As mentioned above, an additional issue was the inadequate control of the High Pressure Separator temperature during summer operation. While cleaning the air coolers has shown benefit in the past, the decision was made to replace all 16 banks of the air coolers to take advantage of the improvements in air cooler design developed during the last 30 years. While the cost of maintenance and cleaning was enough to justify the replacement of the air coolers, further justification was provided by Criterion and Zeolyst based on the run length benefits of improving the purity of the recycle gas in the hydrocracker.

Another important operating issue addressed was the poor radial temperature distribution. Having a non-uniform temperature profile across the catalyst bed leads to non-profitable operation due to some catalyst being over-worked relative to the rest of the catalyst in the bed. In severe cases, high radial temperature deviations can lead to run limitations due to non-selective yields or temperature limitations. Evaluation of the technologies available in the market place led North Atlantic to select Shell Global Solutions internals specifically their Ultra-Flat Quench (UFQ) interbed internals, High Dispersion (HD) distribution trays and top bed Filter Tray (3,4). 

The location installed a filter tray at the top of each reactor. In addition, Ultra Flat Quench and HD Trays were installed between each bed. As shown in the following figure, the bed outlet radial temperature profiles decreased from 50°F to an average of 10°F. The uniform bed outlet temperatures post-revamp show that exotherms are equal on the various bed locations as proof of a uniform distribution of the feed. In addition, the improved product selectivity (shown later) is a strong indicator for a “plug-flow” type of reactor and thus a good uniform feed distribution in each reactor bed.

Figure 3: Bed radial temperature profile. This figure shows the original bed radials, the radials following the change-out of the top bed (previous cycle), and finally the radials after the replacement of the internals and the catalyst (current cycle) showing minimal increase as the cycle continues.

An additional benefit of the new internals was increased catalyst bed volume. The new internals allowed for approximately 11% additional catalyst volume. This additional volume was used to increase the active catalyst loaded into the reactors as well as additional grading material and demetallization catalyst. 

When compared on a constant bed depth basis, the Start-of-Run pressure drop decreased substantially due to the use of the TX catalyst shape for the cracking catalyst (2 of the 3 beds in the reactor) as shown in the following figure. Also, compared to the past, the dP increase in the current cycle has been mitigated through the use of the Shell top bed filter tray and top bed grading system, both of which are capable of removing and storing fines without immediately plugging the catalyst system. The first two curves in the figure display the reactor pressure drop for the previous cracking catalyst cycle that included a top bed replacement. The third curve is for the current cycle showing the actual pressure drop. The lower rate of increase of the pressure drop confirms that the current cycle will not be pressure drop limited.

Figure 4: Improved reactor pressure drop performance compared to the previous cycle. The TX trilobe shape, the installation of a filter tray, and the bed grading in the additional catalyst volume has allowed for the run length to be determined by catalyst activity instead of pressure drop.

Commercial Performance and Economic Benefit

As a result of all of these changes, the unit is showing a more profitable operation. The WABT has been lowered at higher throughput and higher conversion. The previous cycle was approximately 32 months with an intermediate change-out of pretreat catalyst. The current cycle was proposed to be 48 months, but it is on track for a cycle that will approach 5 years with an intermediate change-out of pretreat catalyst. This extension of the pretreat catalyst cycle length and the cracking catalyst cycle length translates into 1 less mini-turnaround and 1 less full turnaround during a 10 year period.

The overall reactor WABT for this cycle is seen in Figure 5 in comparison with the WABT from the previous cycle. This figure displays the WABT on a Days On Stream basis (DOS). The current WABT is lower even though the feed rate and the conversion have been higher and more consistent for this cycle than for the previous cycle.

Figure 5: WABT for current cycle compared to the previous cycle WABT. The unit is currently running 10–15oF below the comparable point in the previous cycle. This activity advantage would translate to a cycle length extension of at least one year.

While the overall WABT is lower than the WABT for the previous cycle, an additional benefit can be seen when the reactor bed WABT’s are examined individually. The lower radial temperature profile for each of the beds was one indication of better flow distribution in the reactor. Another indication of better catalyst utilization can be seen from the bed WABT’s shown in Figure 6. There are two improvements in the Bed 1 WABT. The first improvement is lower WABT due to the improved activity of the pretreat catalyst installed. The second improvement is the lower deactivation rate due to the improved flow distribution. This lower deactivation improvement is also present for the catalyst in Beds 2 & 3.

Figure 6: Bed WABTs for current cycle compared to the previous cycle bed WABTs. Each of the beds is currently operating at a WABT 10–15oF below the comparable point in the previous cycle, indicating an extension of the cycle by at least one year.

All of these WABT improvements are even more important when one looks at the
unit overall conversion change between this cycle and the previous cycle. In
addition to the WABT improvements, the unit has been running over 2.5% higher
overall conversion this cycle compared to the previous cycle as shown in the
following figure.

Figure 7: Cycle average conversion for current cycle compared to the previous cycle. The conversion level for this cycle has been averaging over 2% on feed higher than the previous cycle.

One of the critical end-of-run limits in the past has been excessive production of LPG due to poor selectivity. The combination of new internals with the better catalyst utilization, and the TX catalyst shape that further reduces non-selective overcracking have significantly lowered the LPG make from the hydrocracker as shown in Figure 8.

An additional interest for the location was increasing the total distillate make from the unit at constant overall conversion. Figure 9 displays the total distillate yield for this cycle and the previous cycle. There are two points of interest in this figure that demonstrate the impact of good flow and temperature distribution in the catalyst bed. The first point is the increased yield of distillate caused by the uniform usage of all of the catalyst in the bed (no overcracking in part of the bed leading to high radials). The second point is the increased yield stability due to the more uniform usage and deactivation of the catalyst.

Figure 8: LPG make for the current cycle compared to previous cycle. At this point in the cycle, the LPG make is approximately 300 bpd lower than the previous cycle. The reduction in LPG make will remove this constraint on the cycle length.

Figure 9: Total distillate production for current cycle compared to the previous cycle. The new internals and the TX catalyst shape lead to increased selectivity and increased yield stability.

With the significant shift of products from LPG and light naphtha to heavy naphtha and distillate, the C5+ liquid volume gain remained approximately constant with a slight decrease in hydrogen consumption. These additional improvements in the overall profitability of the unit can be seen in Figures 10 through 12. Figure 10 displays the API gain of the bottoms relative to the feedstock, Figure 11 displays the C5+ total volume gain across the unit and Figure 12 displays the hydrogen consumption.

Figure 10: Bottoms API gain for current cycle compared to the previous cycle.

With the improved yields and product properties, the diesel produced by this hydrocracker meets ULSD specifications during normal operation without issue. This improvement in operation has allowed for the reduction of sulfur analyses of the diesel to only those times when the unit is upset. Routine sulfur monitoring is now only conducted on the naphtha and bottoms streams since ULSD specifications will be met whenever the bottoms sulfur level is below 40 ppm.

Figure 11: C5+ total liquid yield for current cycle compared to the previous cycle.

Figure 12: Hydrogen consumption for current cycle compared to the previous cycle.

In addition to the improvement seen in the Fractionator Bottoms API, there is a
corresponding improvement in the sulfur and nitrogen levels remaining in the
fractionator bottoms as shown in Figures 13 and 14.

Figure 13: Fractionator bottoms sulfur content for current cycle compared to the previous cycle. This figure presents the previous cycle in two parts due to the top bed changeout.

Figure 14: Fractionator bottoms nitrogen content for the current cycle compared to the previous cycle. This figure presents the previous cycle in two parts due to the top bed changout.

All of these changes have allowed maximization of the catalyst activity to be utilized beyond the previous constraints of pressure drop, demetallization capacity, or non-selective yields.

Conclusions

Hydrocrackers continue to be an excellent source profit for a refiner due to the high quality diesel blending components as well as the naphtha that they produce. However, to maximize the profitability, improve reliability and meet more stringent product specifications (i.e. less than 10 ppmw sulfur specification for ULSD), refiners need to select the best catalyst systems, ensure good reactor flow distribution and closely monitor feed properties. 

By working closely with Criterion Catalysts & Technologies Company, Zeolyst International and Shell Global Solutions, North Atlantic has been able to justify and implement several key improvements to the hydrocracker including state of the art tailored catalyst systems, new reactor internals, replacement of air coolers and changes to avoid feedstock contamination.

The installation of Shell internals (filter tray, UFQ and HD trays) has resulted in improved flow distribution in the reactors, increasing the catalyst utilization as evidenced by the lower radial temperature profile in each of the catalyst beds, and making them more resistant to fouling. In addition, the same extent of conversion is being achieved at 70% of the previously required axial delta temperature.

The combination of new demetallization catalyst (RM-5030), improved pretreat catalyst (DN-3300) and tailored cracking catalyst system (TX trilobe shaped Z- 673/Z-623), has allowed North Atlantic to achieve yield improvements, pressure drop reductions, meeting ULSD specification for the entire cycle and record run length. The expected economic benefit is $3.5 Million per year. 

In summary, the ever-changing regulations with even more stringent fuel specifications will continue to pose challenges to the refiners. However, with proper application of technologies, significant profit uplift opportunities can be realized. North Atlantic appreciates that Criterion and Zeolyst will continue to work closely with the refinery to address future issues and create additional value for this location.

References

1. M. Hu, R. Anderson, R. Adarme, C. Ouwehand, J. Smegal, “The Era of ULSD – New Challenges and Opportunities for Hydrocracking Processes”, NPRA 2006 Annual Meeting, AM-06-46.
 
2. D.M. Altrichter, E.J. Creyghton, C. Ouwehand, J.A.R. van Veen and A.S. Hanna, “New Catalyst Technologies for Increased Hydrocracker Profitability and Product Quality”, NPRA 2004 Annual Meeting, AM-04-60.
 
3. ERTC, Nov-1998, Vienna, Austria, CED Ouwerkerk, E. S. Bratland, A. P. Hagan, B.L.J.P. Kikkert and M. C. Zonnevylle (Shell), “Performance optimization of fixed bed processes”.
 
4. ERTC Nov-2000, Rome, Italy, D. Pohl, F. Geerdes (SRS Salzbergen), J. Swain (Criterion), MC Zonnevylle (Shell), “Are you really getting the most from your Hydroprocessing reactors?”