Refiners have many options to convert high-aromatic streams into ULSD

May 15, 2006
US refiners face both a challenge and an opportunity in converting high-aromatic streams, such as FCC light cycle oil or coker light gas oil , into low-sulfur products for use in the ultralow-sulfur diesel (ULSD) pool or other distillate products.

C.K. Lee, Steve J. McGovern - PetroTech Consultants, Mantua, NJ

John A. Zagorski - Middough Consulting Inc.,Philadelphia

US refiners face both a challenge and an opportunity in converting high-aromatic streams, such as FCC light cycle oil or coker light gas oil , into low-sulfur products for use in the ultralow-sulfur diesel (ULSD) pool or other distillate products.

Current and future aromatic or cetane specifications can limit the amount of these streams that can be blended into ULSD.

Refiners can use different processing options to convert these low-quality streams successfully into salable products:

  • Low-pressure hydrotreating (<500 psi hydrogen partial pressure). Coprocessing limited amounts of light cycle oil (LCO) or coker light gas oil (CLGO), with virgin kerosine or light gas oil (LGO), to produce ULSD (<15-ppm sulfur). The remainder of LCO and CLGO is used in higher-sulfur products.
  • High-pressure hydrotreating (>500 psi hydrogen partial pressure). Coprocessing more LCO or CLGO, with virgin kerosine or LGO, or processing LCO and CLGO alone in dedicated units to produce ULSD (<15-ppm sulfur).
  • Hydrocracking. Converting LCO or CLGO with other high-boiling distillates to produce naphtha, kerosine-jet fuel, and ULSD.
  • This article compares the influence of processing severity (low vs. high hydrogen partial pressure) and processing scheme (hydrotreating vs. hydrocracking) on product quality (density, cetane, aromatics) and hydrogen consumption. It also discusses the economics for selecting the optimal processing scheme for a generic refinery. The optimal strategy is site dependent.

    A US Environmental Protection Agency survey has indicated that most refiners plan to use hydroprocessing technologies, via a combination of revamping exiting units and installing new units, to convert the various mid-distillate streams into low-sulfur products. For high-complexity refineries with FCC or thermal conversion units, there is a greater challenge of optimizing feed selection to the various hydroprocessing units.

    ULSD

    Starting in June 2006, US refiners must supply large volumes of ULSD to the on-road diesel market. Initially, at least 80% of on-road diesel must contain less than 15-ppm sulfur when delivered to the end user.

    In 2007, the various off-road diesel products will also begin a staged reduction to 15-ppm sulfur from a current maximum of 0.5% sulfur. By 2010, all on-road diesel must contain less than 15-ppm sulfur and all other diesel must be less than that by 2012.

    Europe is also reducing the sulfur content of its distillate products. All diesel fuel must currently be less than 50-ppm sulfur; this limit drops to 10 ppm in 2009. Some countries have encouraged the early introduction of 10-ppm diesel through tax credits. Most other countries are also lowering the sulfur content of diesel to similar levels.

    Although diesel consumption accounts for more than half of the distillate produced by US refiners, current regulations do not require similar ultralow-sulfur levels in jet fuel, heating oil, and other industrial distillates. Maximum jet-fuel sulfur is currently limited to 3,000 ppm worldwide, although the actual average sulfur levels are much lower.

    In the US, individual states regulate the maximum sulfur in residential heating oil and vary from 2,000-5,000 ppm with some states considering reductions to 500 ppm or less. Europe has regulations in place to reduce heating oil sulfur to 1,000 ppm.

    The sulfur content of the small amounts of commercial heating oil and industrial distillates that are produced vary widely and can be more than 10,000 ppm.

    Table 1 shows a breakdown of the various uses of distillate products in the US in 2004, according to the US Energy Information Administration.

    Click here to enlarge image

    In 2004, low-sulfur diesel was 45% of the total distillate and kerosine market. The other high-sulfur diesels that are subject to the new off-road specifications will increase the amount of ULSD to almost 60% of the total distillate market. Kerosine accounts for about 30% and the other high-sulfur distillates are 10%.

    Any kerosine that is blended with ULSD must also meet the 15-ppm standard. A recently proposed federal regulation limits the sulfur content of fuels used in stationary turbines to 500 ppm, and discussions are under way to lower the sulfur content of fuels used in stationary diesel engines.

    No pending proposals, however, call for reducing the sulfur content of jet fuel below its current limit of 3,000 ppm.

    Based on the results of US Department of Energy studies, residential heating oil trade groups have espoused the use of low-sulfur diesel (LSD, 500-ppm sulfur) as a premium heating oil. Its benefits are higher efficiency and lower maintenance costs due to reduced deposits on the heat exchange surface. Some states, however, effectively preclude this use by taxing distillate fuels based on sulfur content, not end use.

    Even before the sulfur specs were tightened, demand for higher-sulfur distillates has been declining while the consumption of LSD has grown (Fig. 1). Although the annual average residential heating oil consumption is about 400,000 b/d, it varies from a peak of more than 700,000 b/d during the heating season to essentially zero in the summer.

    Click here to enlarge image

    The number of homes heated with fuel oil is declining at a rate of more than 100,000 homes/year. Other industrial users are also switching from high-sulfur distillate fuels to lower-sulfur fuels or other energy sources.

    In addition to the governmental sulfur regulations for distillate products, the aromatics content, density, and cetane are also controlled.

    Table 2 summarizes these specifications. Other specifications such as distillation, flash point, etc., are not shown.

    Click here to enlarge image

    Diesel engine manufacturers historically set the density and cetane specifications to enhance engine performance, although they are now part of the ULSD specifications for engine emission controls. The aromatic or polynuclear aromatic (PNA) content has been controlled to limit diesel engine emissions.

    Some areas, such as California and Sweden, have aromatic specifications of less than 10%. California, however, allows more aromatics if the measured emissions are no more than from a 10% aromatics reference fuel.

    Because of the historic development of the cetane and density specifications, there are significant differences between them in the US and Europe. Europe primarily refines crudes from the Middle East, North Africa, the North Sea, and Russia. Virgin distillates from these crudes generally have cetane numbers greater than 50 after desulfurization.

    The US processes more sour, heavy crudes and syncrudes with virgin distillate cetane numbers of less than 45. The distillate-to-gasoline ratio is higher and the amount of cracked distillates is lower in Europe than in the US; therefore the distillate pool is affected less by poor-quality distillates.

    Recent studies in the US and Europe have shown that fuel composition, especially aromatics and PNAs, have little effect on emissions for the newer diesel engines that use post-treatment to reduce NOx and particulate emissions. Neither the US nor Europe is currently proposing reductions in aromatic or PNA specifications.

    The higher severities needed to produce ULSD will naturally reduce the average aromatic and PNA content of the distillate pool.

    There is a beneficial effect of increasing cetane on emissions and engine performance, especially for older diesel engines. There is no measurable difference, however, between natural cetane based on fuel composition and increased cetane due to the use of cetane-improver additives.

    Different grades of premium diesel with cetane above the minimum specification are sold in the US. The locomotive and marine diesels that are used in the large, slower-speed locomotive and marine engines can have cetane levels below 40.

    Impact on refining

    The refining industry has invested in equipment to produce these ultralow-sulfur products. EPA surveys have shown that about 80% of the on-road ULSD in the US will be produced from revamped units, not new units. This was possible due to improvements in catalyst technology, hardware modifications, and allocating some of the more difficult-to-desulfurize distillate streams to the higher-sulfur products.

    The shrinking market for high-sulfur distillates now also requires hydrotreating some of these difficult streams that were moved to higher-sulfur products.

    Table 3 shows typical properties of the various distillate streams in the refining industry. FCC LCO and coker distillate are generally high in sulfur, especially the difficult-to-remove hindered dibenzothiophenes. They are also high in aromatics and nitrogen. These molecules compete with sulfur molecules for active catalytic sites and hydrogen in the reactor. This competition increases the severity required to remove sulfur from these streams.

    Click here to enlarge image

    Aromatic saturation and nitrogen-removal reactions also increase hydrogen consumption of these streams. These highly aromatic streams can consume 3-5 times as much hydrogen as virgin distillates when producing ULSD.

    In addition to being more difficult to desulfurize, these highly aromatic streams can result in a ULSD product that does not meet the cetane or density specifications, especially when a refiner processes high-aromatic crudes. Including more kerosine in the ULSD blend will lower its density but has very little effect on the cetane. Reducing the aromatic content of the ULSD allows it to meet both the density and cetane specification.

    There are several options for treating these high-aromatic streams to produce ULSD, usually considered in this order:

    • Desulfurize limited quantities of high density, high-aromatic streams in existing hydrodesulfurization (HDS) units along with the virgin components to produce ULSD. The properties of the virgin distillates and high-aromatic streams as well as the severity of the existing distillate HDS unit determine the amounts that can be blended into ULSD. This is the current operation in most refineries.
    • Desulfurize and increase the cetane of more highly aromatic streams in new HDS units, designed for higher severity-higher pressure and lower space velocity using nickel-molybdenum (NiMo) catalyst. Even the virgin distillate from some crudes does not always meet the density, cetane, and aromatic specifications and require higher severities.
    • Use two-stage aromatic saturation units, designed at lower pressure, but with interstage H2S and NH3 removal. This allows the use of higher-activity noble metal catalysts in the second stage for aromatic saturation.
    • Hydrocrack the high-aromatic streams at either partial or full conversion to “roll the aromatics down” into the gasoline range where they have more value.
    • Use cetane-improver additives to increase the cetane.

    Several factors affect the economics of the various options. These factors are site specific, which makes the best option also site specific.

    Conventional, two-stage HDS

    These high-density streams can consume 3-5 times as much hydrogen during desulfurization to ULSD vs. less-aromatic virgin streams. If more aromatic saturation is required to meet a density or cetane specification, then the hydrogen consumption will be even higher. Hydrogen consumed in producing ULSD from these highly aromatic feeds dominates the overall cost to desulfurize these feeds.

    Table 4 shows the overall operating costs for typical low (<800 psig) and high (>1,000 psig) pressure HDS units.
    Click here to enlarge image

    Some hydrogen is always consumed in distillate desulfurization. Desulfurizing to 10-15 ppm consumes more hydrogen than producing 500 or 5,000-ppm sulfur products. Hydrogen consumption is a strong function of feed quality and operating pressure (Fig. 2).

    Click here to enlarge image

    Off-road diesel specifications have pushed many more refineries out of hydrogen balance. The naphtha reformer no longer produces sufficient hydrogen to satisfy the distillate desulfurization requirements.

    Refiners satisfy additional hydrogen requirements by either purchasing hydrogen or building a hydrogen plant. Recent increases in natural gas prices, however, have increased the cost of manufacturing hydrogen in the US.

    Click here to enlarge image

    Fig. 3 shows the effect of natural gas price increases. Because highly aromatic feeds consume more than 400 scf of hydrogen/bbl and natural gas prices are more than $6/Mcf, the cost of hydrogen exceeds all other HDS operating costs.

    High hydrogen consumption arises from denitrogenation and aromatic-saturation reactions that occur in parallel with desulfurization reactions. At the conditions required to produce 5,000 or 500-ppm sulfur product, little nitrogen removal or aromatic saturation occurs. Some PNAs are converted to monoaromatics, but this reaction is equilibrium constrained at typical LSD conditions.

    The extent of aromatic saturation is much higher at the higher pressure and lower space velocity of ULSD units designed for aromatic feeds. The combination of more severe conditions and higher feed aromatic content leads to much higher hydrogen consumption.

    If the processing objective is to reduce the product’s aromatic content for either density or cetane control, the hydrogen consumption will be essentially the same for both a high-pressure unit with NiMo catalysts or a lower pressure, two-stage aromatic saturation unit with noble metal catalysts in the second stage. The main differences between these two approaches to aromatic saturation are in the investment cost for the two units, operating costs, and catalyst costs.

    For many larger units, the two-stage process will have lower investment and other operating costs, although the hydrogen costs will be the same. Using noble metal catalysts, however, can add more uncertainty to the catalyst cost. Noble metal prices are typically more volatile than molybdenum prices due to their use as investments. Large oil companies often have inventories of noble metals they can use for this service, but smaller companies must purchase or lease their metals on the open market.

    Hydrocracking

    The high aromatic content of LCO causes the low cetane and high density of FCC LCO. Due to the nature of the reactions that occur on the Y-zeolite catalysts that are used in FCC units, the primary hydrocarbon types in LCO are:

    • Methyl PNAs (2-ring naphthalenes and 3-ring phenanthrenes with 1-4 methyl groups attached to the rings).
    • Methyl monoaromatics (mostly partially hydrogenated 2 and 3-ring aromatics).
    • Methyl benzothiophenes and dibenzothiophenes.
    • Normal paraffins.

    There are also minor amounts of 2 and 3-ring naphthenes, olefins, nitrogen compounds, and other sulfur molecules. There are essentially no highly branched paraffins or long alkyl side chains on any of the rings because these compounds are cracked quickly in the FCC unit.

    LCO is therefore a blend of high-cetane paraffins and low-cetane aromatics and PNAs. Measurements on pure components have shown that normal paraffins in the diesel boiling range have cetane numbers of about 100. (The compound cetane, which boils at about 550° F. has a cetane number of 100 by definition.)

    Monomethyl paraffins have cetane numbers around 70, and the cetane number decreases rapidly with additional branching. Highly branched paraffins have cetane numbers less than 50.

    The PNAs have effective cetane numbers near zero; some are even negative. 1-Methyl naphthalene is 0 by definition, while 2,6-dimethyl naphthalene is -13. There are no reports of cetane measurements on 3-ring aromatics.

    Products from aromatic saturation also have low cetane numbers and do not always make a good diesel fuel. The only cetane data available on a full sequence of hydrogenated PNAs are for butyl naphthalene and its products of hydrogenation, butyl tetralin and butyl decalin.

    Table 5 shows the cetane values for these compounds and the hydrogen consumptions required to achieve them.

    Click here to enlarge image

    Fully hydrogenated PNAs still do not meet the current diesel cetane or density specifications. Diesel fuel must contain some paraffins to raise its cetane and lower its density. Distillates from syncrudes (for example Zuata in Table 3) have low cetane numbers even though they have been desulfurized to 350-ppm sulfur at relatively severe conditions. This is because they contain high levels of saturated PNAs and low paraffins relative to normal distillates.

    Some heavy crudes, such as those produced in California and offshore Brazil also contain low-cetane distillates. It is important to know the type and amount of saturates that are in the diesel to determine the true cetane number. Cetane index correlations do not accurately predict the cetane of these unusual distillates.

    It is difficult to saturate PNAs fully. The rate of reaction of PNAs to monoaromatics is fairly fast, even using NiMo catalysts; however, thermodynamic equilibrium often limits the extent of reaction, especially at end-of-run temperatures.

    The saturation of monoaromatics is slow over NiMo catalysts and requires either high pressure (>1,500 psig) or a two-stage unit with noble metal catalysts in the second stage. Increasing the cetane number of diesel via additional aromatic saturation (beyond that necessary to produce ULSD) can be expensive due to the high cost of hydrogen.

    Partial or full hydrocracking of high-aromatic distillate streams generally has better economics than high-severity hydrotreating to saturate aromatics for cetane uplift. The conditions necessary to produce ULSD from LCO or similar streams are essentially the same as those required for partial conversion hydrocracking; however, the hydrogen is used more efficiently in hydrocracking.

    In the past, hydrocracking units were expensive to build because a high pressure, low-space-velocity first stage was required to produce the low-nitrogen feed that was required for the second hydrocracking stage.

    Hydrodenitrogenation catalyst activity has increased ten-fold since the first hydrocracker was built and hydrocracking catalysts have become more active and more nitrogen tolerant. In fact, the conditions typically required to produce ULSD result in product nitrogen levels that are acceptable for hydrocracking catalysts.

    Replacing some of the hydrotreating catalyst with hydrocracking catalyst in a ULSD unit can actually improve desulfurization rates because the hydrocracking catalyst opens additional pathways (dealkylation and isomerization) for HDS of hindered dibenzothiphenes.

    Capital cost differences between a diesel HDS unit and a distillate hydrocracker have decreased significantly because the HDS severity requirements have increased and hydrocracking technology has improved.

    Relative hydrocracking rates are similar to the cracking rates in an FCC unit, i.e., the aromatics, naphthenes, and highly branched paraffins all crack before the normal and monomethyl paraffins. This leaves the high-cetane paraffins in the distillate product and rolls the lower-cetane aromatics and naphthenes down into the naphtha boiling range where they are more valuable.

    Like FCC, the cracking reactions give a significant volume swell and increase overall product value. Unlike FCC, the feed aromatic content does not limit the extent of reaction. The hydrogenation component of hydrotreating and hydrocracking catalysts partially saturates the aromatics, which allows them to crack.

    At current hydrogen and diesel values, the volume swell that results from PNA and aromatic saturation during ULSD production does not fully offset the cost of hydrogen that is required for aromatic saturation. The volume swell for hydrocracking, however, greatly exceeds that of aromatic saturation for the same hydrogen consumption.

    Fig. 4 shows the natural gas price that can be supported by the volume swell generated from aromatic saturation or hydrocracking. It also includes historic refinery-gate price data for natural gas and distillate from 1983 through October 2005.

    Click here to enlarge image

    All of the historic natural gas price data are above the breakeven cost that can be supported by the volume swell for aromatic saturation. Additional aromatic saturation, beyond that required for sulfur removal, adds additional operating costs to the refinery.

    Most of the price data, however, falls below breakeven values for hydrocracking, indicating revenue improvement from hydrocracking PNAs. Data that are above the breakeven cost are during periods of depressed distillate prices or high natural gas prices during the peak heating season.

    For Fig. 4, we used pure component data for various PNAs to develop the hydogen consumptions and volume swells.

    The net hydrogen required for distillate hydrocracking is actually lower than it is for distillate aromatic saturation. This requires complete distillate PNA saturation, whereas hydrocracking only requires partial PNA saturation followed by cracking. The aromatics are then removed from the diesel by distillation.

    Although not shown in Fig. 4, cetane of the unconverted hydrocracker distillate is also higher. Hydrocracking could require an additional fractionator for the hydrocracked product from high-conversion units, but the additional investment is more than offset by the added product value.

    Distillate hydrocracking converts some potential distillate-range material to gasoline. This is not a problem in the US but can be a drawback in other parts of the world with greater distillate product demands.

    US refineries that have the highest-aromatic distillate pools are usually those that are running the heaviest crudes with the most FCC capacity. These refineries are often running higher-end-point naphthas to their naphtha reformers to maximize gasoline production. These refineries can maintain a proper gasoline:distillate balance by reducing virgin naphtha end point and replacing the lost reformer feed with hydrocracked naphtha.

    This swing in operation benefits gasoline and distillate production:

    • The hydrocracked naphtha is a better reformer feed than the heavy virgin naphtha.
    • The additional virgin kerosine produced by lowering naphtha end point improves overall distillate pool quality.

    European refineries are also including hydrocracking in their investment strategies; however, these hydrocrackers process vacuum gas oil and not atmospheric distillates. Hydrocracking vacuum gas oil produces higher distillate yields than FCC and is a better fit for the higher distillate demands in Europe.

    Cetane improvement additives

    The cetane of ULSD products that meet all specifications except cetane can be improved with cetane-improver additives. The most commonly used cetane improver is 2-ethyl hexyl nitrate. Many suppliers offer this under different product names.

    Cetane increases of 5-10 numbers are possible at treatment costs of about 0.1¢/gal-cetane number. The actual cetane uplift and treatment costs depend on the composition and base cetane of the diesel fuel, so actual costs are site specific. Its effectiveness decreases and cost increases as dosage level increases.

    Processing configuration

    ULSD specifications for the various off-road diesel products are more costly and difficult to meet than the on-road diesel specification, especially in the US. The various off-road products usually are more aromatic and higher in sulfur and nitrogen due to the higher concentrations of cracked distillates in these products.

    Processing heavier crudes also increases the aromatic content of the distillate pool. Desulfurizing these lower-quality streams requires more severe conditions and consumes more hydrogen.

    Partial hydrocracking of highly aromatic distillates can be more cost effective than hydrotreating to produce ULSD. Hydrogen consumption for partial hydrocracking for cetane uplift or aromatic reduction is lower than for aromatic saturation.

    In many locations that require additional manufactured hydrogen, the overall investment (including hydrogen plant) for a hydrocracking unit can be lower than that of a hydrotreating unit. Even if the investment is more, the incremental rate of return for the small incremental investment is high because of the larger volume swell in the hydrocracker.

    Cetane improver additives are also a more cost-effective way of increasing diesel cetane than aromatic saturation. The best off-road ULSD project can only be developed after careful evaluation of all of the various processing options. Because the “best” project is site specific, there is no general solution for all refineries. Each project must be evaluated using local economics, product supply and demand and existing refinery configuration.

    Based on a presentation to the National Petrochemical & Refiners Association Annual Meeting, Salt Lake City, Mar. 19-21, 2006.

    The authors

    Click here to enlarge image

    C.K. Lee ( [email protected]) is a principal of PetroTech Consultants, Mantua, NJ. His areas of expertise are hydroprocessing, catalytic reforming, and clean-fuel technologies. Previously, he worked for Mobil Technology Co. for more than 20 years. Lee holds a BS in chemical engineering from Cheng Kung University, Taiwan, and a PhD in chemical engineering from the University of Houston.

    Click here to enlarge image

    Steve McGovern ( sjmcgovern @hotmail.com) is a principal of PetroTech Consultants. His areas of expertise include hydroprocessing, catalytic cracking, clean fuel-technologies, and reactor design. Previously, he was a technology expert for Mobil Technology Co. where he worked for 27 years, primarily in process development and commercial operations support. McGovern holds a BS in chemical engineering from Drexel University, Philadelphia, and a PhD in chemical engineering from Princeton University, NJ. He is also a director of the Fuels and Petrochemicals Division of AIChE.

    Click here to enlarge image

    John A. Zagorski ( zagorsja@ middough.com) is a senior technical manager at Middough Consulting, Philadelphia. He has 28 years’ experience in petroleum refining in operations and as a consulting engineer. Zagorski holds a BS in chemical engineering from Lehigh University, Bethlehem, Pa.