The refining industry, especially

Transcription

1 Maximising distillate while minimising bottoms An FCC catalyst with reduced resistance to diffusion raises diesel output without the traditional bottoms penalty ALLEN HANSEN, ADRIAN HUMPHRIES, STEPHEN MCGOVERN and BARRY SPERONELLO Rive Technology, Inc The refining industry, especially in the US, has undergone significant changes during the last 5-10 years. Crude oil prices have gone from about $30/ bbl to over $140/bbl, back down to $35/bbl, and have recently hovered around $100/bbl. The primary transportation fuels, gasoline, diesel and jet fuel, currently account for over 80% of US refinery output. Achieving these high yields requires converting heavy fuel oil into lighter products. For the last 50 years, the refining industry has relied on the fluid catalytic cracking (FCC) process as the primary conversion unit in the refinery. Over 90% of refineries in the US include an FCC unit. The FCC unit converts heavy fuel oils into more desirable lighter products such as distillate and gasoline blend stocks, alkylation feed, LPG and the less desirable light gases and coke. Until recently, most FCC units were operated to maximise gasoline production; however, that is changing because of shifting demand for transportation fuels. Gasoline consumption peaked in 2007 and has continued to decline since then (see Figure 1). Continuing improvements in vehicle fuel efficiency and the imposition of greenhouse gas (GHG) emission limits from light duty vehicles will continue to reduce gasoline demand. Average fuel economy requirements for US light duty vehicles are expected to increase from 30.1 miles per gallon in 2012 to an equivalent of 35.5 mpg in 2016 and 54.5 mpg in 2025, based on required reductions in vehicle GHG emissions. 1 A recent U.S. product consumption, 1000 bpd Jan 1983 Finished motor gasoline Distillate fuel oil Jul 1988 Dec 1993 National Research Council report 2 concluded that it is feasible to reduce light duty vehicle fuel consumption to 20% of 2005 levels by Distillate consumption dipped with the 2008 recession, but has been slightly increasing since then. Jun 1999 Dec 2004 Jun 2010 Nov 2015 Figure 1 US product consumption data Source: Energy Information Administration, Distillate price gasoline price / Jan 1983 Jul 1988 Dec 1993 Jun 1999 Dec 2004 Jun 2010 Nov 2015 Figure 2 Relative gasoline and distillate wholesale prices (Energy Information Administration) In addition to the changing demand structure, the relative prices of gasoline and diesel fuel in the US have also changed in recent years (see Figure 2). Until August 2004, monthly average wholesale prices for refiner gasoline had been higher than diesel prices 95% of the Catalysis

2 Butenes Butanes Propenes Propanes Figure 3 Typical FCC reaction network Gasoline (3) k 34 LPG (4) time. Since then, monthly average wholesale prices for refiner diesel have been higher than gasoline almost 75% of the time. Diesel prices have been higher than gasoline in all but two of the last 36 months. This is a dramatic reversal in the relative historical prices of gasoline and diesel. k 13 k35 k 36 Gasoil (1) Dry gas (5) k 12 k 14 k 15 k 16 k 46 k 23 k 25 LCO (2) k 24 k 26 Coke (6) In addition to the changing demand structure, the relative prices of gasoline and diesel fuel have changed in recent years Refiners reaction Refiners have been gradually shifting their operations to meet this changing demand and price structure, increasing distillate production at the expense of gasoline. The FCC unit is a major gasoline producer in the refinery. Many refiners have shifted their FCC operations to distillate mode, at least during part of the year, to help meet this changing demand structure. There are several operational changes that refiners can make to their FCC operations in order to increase distillate production at the expense of gasoline. These include: Minimising diesel fractions in the FCC feed Changing the cut point between gasoline and light cycle oil (LCO) products Reducing cracking severity by increasing feed temperature, lowering riser outlet temperature, or lowering catalyst activity Changing the FCC catalyst to a more distillate selective catalyst (lower zeolite/matrix ratio). The first three items are operational changes that any refiner can implement. Removing diesel fractions from the feed eliminates the cracking of good quality distillate to lighter products. Changing the cut point between FCC product gasoline and LCO directly shifts gasoline to distillate without changing any other product yields or the overall FCC volume swell. Reducing cracking severity increases LCO yield by reducing upgrading of heavy products into transportation fuels. The lower cracking severity impacts all other yields in addition to LCO and gasoline. Lower value heavy fuel oil yield increases while alkylation feed decreases. Decreasing riser outlet temperature can also decrease gasoline octane, which may impact other refinery operations to maintain gasoline pool octane. Many refiners have implemented these operational changes; however, catalyst changes provide yet another powerful lever to improve LCO yields. Catalytic opportunities Catalyst changes can increase LCO yield while minimising the negative aspects of distillate mode FCC operation. ZSM-5 additives can be used to increase gasoline octane and light olefin production at reduced severity to maintain alkylation feed and gasoline octane. However, the net result is still a reduction in total transportation fuel yield. Increasing the matrix activity and reducing the zeolite content of the catalyst will also improve overall distillate selectivity, but generally is accompanied by an increase in coke selectivity. Beyond the catalytic opportunities described above, overcoming the diffusion limitations of today s catalysts can open the door to enhanced LCO selectivity while minimising any trade-offs. The conversion of large VGO and resid molecules to LCO, gasoline and gases is usually modelled as a series of parallel and series reactions, such as the network shown in Figure 3, as proposed by numerous researchers such as Heydari et al. 3 All light products can be formed directly from the larger, heavy molecules through the parallel reactions, but the series pathways are also significant. The cracking reactions are usually considered as first order in the reactant. Series, first order reactions usually result in a maximum of the yield of the reaction intermediates as feed conversion increases. This is true of the FCC process. As conversion is increased, LCO, then gasoline, then LPG, all reach a maximum yield and then decline as conversion is increased. Although the reactions are modelled as simple first order reactions, the rate constants are actually lumped parameters that include other physical processes such as the diffusion of the feed and product molecules into and out of the catalyst. In rapid conversion processes like FCC, these inter- and intra-particle diffusion transport rates can be the controlling steps in determining the overall rate of reaction and overall product selectivity in a series reaction network. Catalyst technologies that improve interand intra-particle diffusion transport rates can therefore be a key component of an LCO maximisation strategy. Advantage of Molecular Highway technology For the last 40 years, essentially all modern FCC catalysts have used faujasite (Y zeolite) as the primary, active cracking component in the 38 Catalysis

3 catalyst. The catalyst particles are a physic-chemical mixture of the zeolite crystals, an active or inert binder and other components that can assist in the cracking reactions or provide another function such as metals trapping. The faujasite particles are introduced into the catalyst formulation as individual particles that are 1-5 microns in diameter, while the final catalyst particles have an average particle size of about microns. The zeolite is a molecular sieve with most of its active sites inside the small pores of the individual crystallites. The size of these pores is similar to (or often smaller than) that of the molecules that are being converted, so molecular access to all of the interior active, catalytic sites is limited by diffusion. These diffusional resistances can impact selectivity as well as activity. If the primary cracked fragments are delayed in leaving the zeolite crystal, then they have a higher probability of re-cracking to lighter products. This re-cracking will reduce the yield of intermediate products such as LCO and gasoline and increase the yield of less valuable, lighter gases, together with coke. Rive Technology has developed methods to reduce these diffusional resistances and increase the yield of the more valuable products by introducing a uniform series of larger diameter holes into the zeolite crystals themselves, prior to or after incorporation into the finished catalyst. This creates a network of intermediate-sized mesopores or molecular highways within zeolite crystals, including type Y zeolite used in FCC catalyst (see Figure 4). These 40Å diameter mesopores are larger than the 7.5 Å diameter micropores which exist naturally in Y zeolite, so they provide both improved access for large feed molecules to the interior of the zeolite crystals (where the active sites are located), and improved escape of distillate molecules from the zeolite before they re-crack to light gases and coke. The Molecular Highway technology process begins with standard Y zeolite and then further processes it Figure 4 Modified zeolite pore structure Figure 5 Photomicrographs of modified zeolite Catalysis

4 CSO yield, wt% Figure 6 Bottoms yield for Rive catalysts versus coke to chemically burrow a network of surface-accessible mesopores throughout the crystal without materially affecting its crystal structure. The high resolution field emission scanning electron photomicrographs in Figure 5 illustrate the transformation. 4 The top photo is of a high quality Y zeolite representative of the type used in modern FCC catalysts. It has angular prismatic crystals and flat, sharp crystal faces characteristic of highly crystalline material. Each crystal face also contains millions of 7.5 Å diameter micropore windows that are too small to see at even the X X magnification of these images. The lower photo shows a sample of the same type zeolite after it has been processed to create the additional mesopore network. The particles retain the prismatic, sharp angular faces characteristic of crys- LCO /CSO, wt/wt tallinity. They also contain millions of micropore openings that are too small to resolve at this magnification. In addition, though, the crystals now also show an extensive network of mesopore openings that were created during the processing. These unique molecular highways are hydrothermally stable in the FCC unit and result in a materially improved product yield structure and increased middle distillate yield in catalytic cracking. Catalyst performance: impact of enhanced diffusion With conventional catalysts, the yield of low value bottoms increases when FCC severity is lowered to increase LCO yield. Rive catalysts mitigate this problem by increasing the yield of transportation fuels, while reducing bottoms yield, allowing refiners to better optimise their distillate operations Figure 7 LCO/CSO Selectivity for Rive catalysts versus coke Improved pore structure and reduced diffusional resistances allow larger molecules to access the zeolite and cracked products to exit before over-cracking to produce coke. This improved balance between cracking and coke can be seen in Figure 6. Bottoms (CSO) yield is plotted versus coke yield in this figure for two refineries where the incumbent catalyst was tested against a commercial catalyst with Molecular Highway technology, produced and jointly developed by W.R. Grace. These results were generated in an ACE unit, but these yield differences have been confirmed with controlled trials at each refinery. 5,6 In commercial operations, the refiner can effectively control the unit severity and coke yield by adjusting operating conditions, including riser temperature, feed temperature and possibly catalyst cooler duty. Accordingly, the modified catalyst has been shown to produce less low value bottoms at a given coke yield. This provides an opportunity to maintain or reduce bottoms yield while lowering unit severity, which translates into higher yields of more valuable products. Additionally, many refiners use the ratio of LCO to bottoms yields (LCO/CSO) as a measure of desirable catalyst selectivity since it is an indicator of the catalyst s ability to produce good yields in a reduced severity environment. Figure 7 compares the ratio of LCO/CSO yields for commercial Rive catalysts versus the incumbents at the two refineries. In each case, LCO selectivity is improved over CSO at all levels of severity (coke yield). Similarly, gasoline and LPG yields increase with Rive catalyst at a given severity (see Figure 8). The yield of gasoline plus LPG consistently increases relative to the yield of low value bottoms. In these examples, we have combined gasoline and LPG since some refiners prefer to maximise LPG and olefins, while others need to minimise LPG. Both gasoline and LPG olefins are valuable, and the split between them can be controlled through fine tuning of the catalyst formulation (for instance rare earth content and 40 Catalysis

5 activity), and through operational changes (such as riser temperature and cat-to-oil ratio). Therefore, the total is most important and transportation fuels overall are increased relative to bottoms with Rive catalysts, at any level of severity. Economics The impact of modified pore structure and increased diffusion has been shown to produce improved yields. In addition to the two sets of catalyst comparisons described above, which have been vetted through controlled refinery trials, numerous additional studies have produced the same trends on a variety of feed types comparing incumbent catalyst formulations against catalysts with Molecular Highway technology. This provides substantial flexibility to refiners to re-optimise their operations within their specific economics and constraints, with uplifts generally estimated to be worth between $1.00/bbl and $2.50/bbl of FCC feed. Summary In contrast to the traditional approach of lower operating severity and lowering the zeolite activity of conventional FCC catalyst, Rive Technology has introduced Molecular Highway technology to maximise distillate production. While the traditional approach drops LPG into gasoline, gasoline into LCO and LCO into bottoms, Molecular Highway technology enables refiners to crack more bottoms into valuable transportation fuel at lower riser temperatures. FCC catalysts formulated with this novel technology have been commercially proven to achieve uplifts much greater than were previously achieved in FCC units. Benefits of catalysts with Molecular Highway technology include: Less bottoms yield at fixed operating conditions Optimised LCO yield at lower severity, with no increase in bottoms yield Alleviation of constraints such as air rate, wet gas rate and regenerator temperature at lower severity, CSO, wt/wt / Gasoline+LPG Figure 8 Gasoline + LPG selectivity for Rive catalysts versus coke with no increase in bottoms yield, leading to: Better optimised yield structure The ability to run more feed The opportunity to process heavier, cheaper feedstocks Higher recycle rates in order to minimise bottoms yield. Molecular Highway is a mark of Rive Technology. References 1 EPA and NHTSA Set Standards to Reduce Greenhouse Gases and Improve Fuel Economy for Model Years Cars and Light Trucks, Office of Transportation & Air Quality, EPA-420-F , Aug Transition to Alternative Vehicles and Fuels, National Research Council of the National Academies, Heydari M, AleEbrahim H, Dabir B, Study of seven-lump kinetic model in the fluid catalytic cracking unit, American Journal of Applied Sciences, 7 (1), 71-76, Li K, Valla J, Garcia-Martinez J, Realizing the commercial potential of hierarchical zeolites: new opportunities in catalytic cracking, ChemCatChem, 6 (1), 46-66, Jan Krishnaiah G, Speronello B, Duncan H, Molecular Highway TM technology for FCC catalysts in a commercial refinery, AFPM Annual Meeting, AM-12-25, Dight L, Krishnaiah G, Speronello B, Hansen A, Crosby J, Rive Molecular Highway TM catalyst delivers over $2/50/bbl uplift at Alon s Big Spring, Texas Refinery, AFPM Annual Meeting, AM-13-03, Allen Hansen is a Consultant to Rive Technology and founder and Vice President of ClinChain Inc., providing modelling and advanced optimisation consulting services to the process industries. He has 24 years of industrial experience including extensive work in FCC spanning hardware design, modelling, technical service, and hands-on operations. Previously, he worked for Aspen Technology, and Mobil Research and Engineering. He holds a bachelor s degree in chemical engineering from the University of Delaware, and a master s and PhD in the same field from the University of Illinois Urbana-Champaign. Adrian Humphries is Director of Technical Services at Rive Technology. He has over 30 years of experience in the catalyst industry, serving in a variety of senior leadership roles at Albemarle Corporation, Akzo Nobel, Filtrol, Hermes Catalysts, and Quanta Technologies in various product development and marketing roles, many of which focused on fluid catalytic cracking and additives. He holds a PhD in chemistry and catalysis from the University of Bristol, England. Stephen McGovern is the FCC Technology Advisor to Rive Technology. He has been active in the refining and petrochemical industries for over 40 years, and is currently a principal of Petrotech Consultants, an independent consulting firm specialising in refining technologies and economics. Previously, he was with Mobil Technology Company in process development and refinery technical support in the areas of catalytic cracking and hydroprocessing. He holds BS and MS degrees in chemical engineering from Drexel University and a PhD in chemical engineering from Princeton University. Barry Speronello is a Research Fellow at Rive Technology. He has over 30 years of service at Engelhard Corporation and then BASF Catalysts LLC after its acquisition of Engelhard. He retired from BASF in 2009 as a Research Fellow and joined Rive Technology where he has worked since. He is an inventor on 53 US Patents, primarily in the areas of catalytic cracking, emission control catalysis, and oxidation chemistry, and holds a PhD in ceramic engineering from Rutgers University. Catalysis