ROSE® Process Offers Energy Savings for Solvent Extraction

ROSE® PROCESS OFFERS EIIIERGY SAVINGS FOR SOLVEft En'RACTIOR

J. A. Gearhart and S. R. Nelson Kerr-McGee Refining Corporation

Oklahoma City, Oklahoma

ABSTRACT

Kerr-McGee has developed and commercialized an energy-efficient solvent extraction process known as ROSE® (Residuum Oil Supercritical Extraction) in which the extractio~ solvent is recovered as a supercr i tical flu id. The energy requirement for supercritical solvent recovery is substantially lower than for conventional solvent processes using evaporation, compression and condensation. The ROSE technology is applicable to a wide variety of both organic and inorganic solvents. Energy savings of forty to fifty percent have been demonstrated commercially with ROSE compared to traditional solvent recovery. The dollar value of the savings is related to the solvent-to-feed ratio employed in the process and the cost of steam, fuel and electr ici ty. Kerr-McGee has a bench scale pilot plant available for feedstock evaluation. Four commercial ROSE units and a large pilot plant have been placed into operation during the past five years. Two additional units are under construction and others are in various stages of design.

This paper descr ibes the ROSE process and illustrates its flexibility with respect to the types of feedstocks and variety of solvents which are utilized.

IN'!'RODUCTION

Conventional solvent extraction has been utilized for more than half a century to separate the more valuable constituents from heavy oils and residuals while rejecting the less des irable mater ials. For example, propane deasphalting rejects high molecular weight viscous asphaltic and resinous materials from residuals while producing premium quality lube oil stocks and FCC and hydrocracker feedstocks. The ROSE process produces identical products more efficiently through the use of super­critical solvent recovery.

Solvent recovery in conventional solvent extraction processes consists of evaporating all of the solvent in a series of progressively lower pressure flashes, followed by steam stripping to produce a virtually solvent-free product. After the solvent has been vaporized it must be condensed and pumped from the condensing pressure back to the pressure required for extraction. It is well known that the quality of the oil extracted from petroleum residues can be improved by increasing the solvent-to-charge ratio. This requires an increased energy demand in the solvent recovery

system proportional to the increase in the sol­vent-to-oil ratio.

The excessive energy requirements of conventional solvent recovery techniques had little punitive impact until the cost of crude oil, and con­sequently the cost of energy, increased drama­tically du ring the past decade. The increased cost of both raw rna ter ial and energy has encou~­

aged the refining and chemical process industri~s to explore every conceivable technique for redue­ing raw material requirements and conservi~g energy.

The oil embargo of 1974 and subsequent rapid increase in the cost of crude, and hence energy, were not anticipated twenty years ago when Kerr­McGee's Research and Development group initiat~d the experimental work which led to the develop­ment of ROSE (Residuum oil Supercritical Extral: ­tion). The initial objectives were not ~o conserve energy, but rather to explore tre potential of recovering more valuable products from the heavy non-distillable portion of a barrel of crude. From this early work Kerr-Mc~e

developed a work ing knowledge of the distinct~e

products that could be produced through solvent extraction.

The three basic products are referred to as Asphaltenes, Resins, and Deasphalted Oil (DAO).

Asphaltenes are the heaviest, highest mOlecular weight, most viscous and most hydrogen-deficient type of hydrocarbon found in crude petroleJm. Asphaltenes are precipitated from petrol~um residues when the peptizing agents in ~he

residues are dissolved in solvents such as nor~al

pentane. Other solvents can be used, though the asphaltene yield and properties will vah. Asphaltenes may be described as dark brownito black friable solids that have Ring and B 11 softening points ranging from 700 to 200: C. Elemental analyses of asphaltenes show relativ~ly high carbon to hydrogen ratios ranging from ~.o to 11.6 by weight. Because of the complex ~nd

compact lattice-type molecular structure :of asphaltenes, it is extremely difficult to tdd hydrogen, or to remove sulfur, nitrogen or met~ls

from asphaltenes. Rydrotreating and cracking catalysts are rapidly coked, de-activated and poisoned by the contaminants present in asphaltenes.

Resins may be classified as an intermediate fraction, lighter than asphaltenes but heavier than deasphal ted oi 1. In the absence of these intermediate components, asphaltenes can be

823

ESL-IE-83-04-128

Proceedings from the Fifth Industrial Energy Technology Conference Volume II, Houston, TX, April 17-20, 1983

insoluble in the oil fraction. Resins, there­fore, can play an impor tan t role in the manu­facture of asphalt because of their peptizing action which permits the asphaltenes to be solubilized in the oil medium to produce a homo­geneous colloidal petroleum asphalt sOlution. Resins affect the penetration, viscosity, ductility and stability of asphalts. Resins can be hydrotreated, hydrocracked, thermally cracked or even coked to produce more valuable products. Resins are less viscous, lower in metals, sulfur and carbon as well as less hydrogen-deficient than asphaltenes.

The recovered deasphalted oil (DAO), on the other hand, is a relatively clean product con­taining substantially lower metals and carbon residue than either asphaltenes or resins. The DAO can be directed to such downstream processes as hydrotreating, hydrocracking or catalytic cracking without having to suffer the severe penalties imposed by the high metals and carbon residue contained in the asphaltenes and resins.

Kerr-McGee's research efforts and experimental work during the late 1950's were surprisingly successful and led to a significant breakthrough in solvent extraction technology. Not only were the asphaltenes, resins and oils separated by the newly developed techniques, but the revolu­tionary discovery was made that one could utilize the supercritical properties of the extr~c~i~n solvent to improve the efficiency and flexlblllty of extraction processes. The ROSE process recovers 85 to 93% of· the extraction solvent as a supercri tical flu id, thereby con­suming a fraction of the thermal energy required for. evapora tive recovery. OVerall energy savlngs of 40 to 50% compared to conventional solvent recovery techniques have been demon­strated commercially. The economic magnitude of the potential savings is directly related to the solvent-to-feed ratio employed by the process and the cost of steam, fuel and electricity.

In solvent extraction it is recognized that increasing the extraction temperature yields progressively greater amounts of insoluble material. In the ROSE process the yields from the asphaltene and resin separators in a 3-stage unit or the combined asphaltenelresin (AIR) separator yield in a 2-stage unit are controlled by adjusting the separators' operating tempera­tures. The oil separator, operating at super­cri tical tempera ture and pressure, is str ictly a solvent recovery device. The DAO yield is set by the Resin (3-stage) or AIR (2-stage) separators temperature. The beauty of the ROSE process is that 85 to 93% of the extraction solvent is recovered in the oil separator with­out evaporation. Further, the recovered solvent is recycled through exchangers which recover a major portion of the energy necessary to achieve supercritical conditions. The energy required for supercritical solvent recovery is substan­tially lOwer than that required by any other method including conventional evaporation.

Solvent losses also tend to be minimized with this solvent recovery technique.

PROCESS DBSCRIP'l'IOII

The simplicity of operation of the ROSE process is illustra ted schematically in Figure 1. The flow sequence follows: The feedstock, wh ich may be either long or short residuum, or a blend thereof, is charged thru a mixer M-l, where it is contacted with several volumes of light hydrocarbon solvent at elevated temperatures and pressures. The mixture passes to the asphal tene separa tor V-I, where the heavy asphal tene frac­tion is withdrawn as a liquid on level control from the bottom and passes thru a heater H-l to a flash tower T-l where the contained solvent is flashed and stripped from the asphaltene product.

The solvent-res in-oil phase flows from the top of V-I thru heat exchanger E-l, where it is heated by the recycle solvent before entering the resin separator V-2. As a result of the increased temperature, with a corresponding decrease in solubility, a second phase separation takes place yielding an intermediate "resin" fraction. The resin fraction is withdrawn on level control from the bottom of the second separator vessel V-2 and stripped of its solvent in tower T-2.

The remaining solvent-oil solution proceeds overhead from V-2 thru heat exchanger E-4 where it is heated by the recycle solvent before flow­ing thru heater H-2, in which the temperature is elevated above the critical temperature of the solvent.

At this increased tempera ture, there is a decrease in the solubility of the oil in the solvent and a resultant phase separation occurs in the oil separator V-3. The oil phase is withdrawn from the bottom of V-3 and stripped of its solvent in tower T-3. The substantially oil­free supercri tical solvent phase flows overhead from V-3 thru exchangers E-4 and E-l, giving up heat to the solvent-oil solution from V-2 and the solvent-res in-oil solution from V-I, before being further cooled in exchanger E-2 for recycling back thru the process via the recycle solvent pump P-l. In many applications E-2' s duty is negligible.

The small amount of solvent associated with the heavy phases wi thdrawn from vessels V-I, v-2 and V-3 (between 7 and 15% of the total extraction solvent) is recovered by conventional stripping in T-l, T-2 and T-3 and condensed in E-3. The recovered solvent is then collected in the solvent surge drum S-1. This relatively small volume of solvent is pumped by the make-up solvent pump P-2 into the large volume of high pressure solvent upstream of the recycle solvent pump, P-l. The recycle solvent pump P-l develops only the differential pressure necessary to overcome the separator system pressure drop.

H24

ESL-IE-83-04-128

Proceedings from the Fifth Industrial Energy Technology Conference Volume II, Houston, TX, April 17-20, 1983

An al terna te and somewhat simpler configura tion of a ROSE unit, a 2-stage instead of a 3-stage unit, may be all that is required in some applications. By simply leaving out V-2, .E-4 and T-2, as shown wi thin the dashed boundanes, one may combine the asphal tenes and res ins into a single product, which is withdrawn from the bottom of asphal tene/res in separator V-l. The deasphalted oil fraction is then recovered from the oil separator V-3. This 2-stage configura­tion can also be operated to produce asphaltenes in V-l and reject all or any portion of the resins overhead from V-l to combine them with the deasphaltened oil product from V-3.

m'ILITY REQUlREMEIftS

The utility requirements of the ROSE process are relatively insensitive to variations in either solvent-to-charge ratio or solvent selection (rang ing from propane thru pentane). For example, increasing the solvent-to-charge ratio from 5 to 1 up to 10 to 1 results in an insig­inificant increase of about 10% in the elec­trical power requirement with no significant change in the other utili ties. By way of com­parison with conventional extraction in which all of the solvent is vaporized, doubling the solvent-to-charge rat io results in s ignif icantly higher utility requirements, both in electrical load and heater duty.

Figures 3 and 4 illustrate why the ROSE process has these inherent advantages:

Figure 3: In conventional solvent recovery, all of the solvent used to extract the DAO leaves the extractor as a liquid at a relatively low temperature and high pressure (Point A). This material (solvent plus MO) is heated and flashed, to some point, labeled E, in the vapor region of the solvent's enthalpy diagram. At Point E virtually all of the extraction solvent is vaporized, leaving a small amount of solvent to be stripped. This remaining solvent is stripped, with steam at a higher temperature and a much lower pressure (Point F).

Figure 4: In the ROSE process, the extraction solvent and DAO start out at the same tempera­ture and pressure as was used in the conven­tional case (Point A). To make the comparison easily understood a 2-stage ROSE unit is used. The DAO plus Solvent flow from the extractor (V­I) (see Figure 1) through exchanger E-l, gaining heat from the recycle solvent. The DAO plus solvent leave E-l at Point B. Heater H-2 provides the energy necessary to heat the DAO plus solvent from Point B to Point C. At Point C, 85 to 93% of the solvent is recovered as a supercri tical fluid, ready to recycle back through exchanger E-l, heating the DAO plus solvent from Point A to Point B. As the recycle solvent heats the DAO plus solvent, it is in turn cooled from Point C to Point D. The recycle solvent is cooled from Point D to Point A by a combination of mixing with the cold recovered solvent from the product stripping

section (T-l and T-3) and trim cooling which is provided by E-2.

The residual solvent in the DAO from V-3 is recovered by flashing and stripping, in a like manner as that presented for the conventional case. In the ROSE case, only 7 - 15% of the extraction solvent is heated to Point E, compared with heating 100% of the solvent in the conventional case.

Knowing that horizontal distance in Figures 3 and 4 is proportional to change in the solvent's enthalpy and that only 0.5% of the extraction solvent must be steam stripped in both cases (Point E to Point F); further, assuming the worst case, tha t only 85% of the solvent is recovered in the DAO separator, it can be seen that supercr it ical solvent recovery requires only 34% of the heat energy required for evaporative recovery.

A comparison of overall utility requirements is shown in Table 1 for a ROSE unit versus a con­ventional propane deasphalter (PDA) . Total utility requirements for the ROSE process are substantially less (approximately one-third) than those required for a conventional PDA unit at normal solvent-to-charge ratios of 10 or 12 to 1. The unique solvent recovery system, which was discussed above, makes the utility reduction possible. Plant acceptance test runs, conducted on a licensee's commercial solvent extraction unit, before and after conversion to the ROSE process, have demonstra ted these savings in utilities.

To demonstrate that this newly discovered tech­nology could be practiced on a commercial scale, a 120 m3/SD plant was constructed by Kerr-McGee in 1954. This plant operated successfully over a six-year period processing feedstocks ranging from 85-100 penetration asphalt to a topped crude with only the naphtha fraction removed. Feedstock gravities ranged from 90 to 180 API, and DAO yields ranged from 56 to 93 volume per­cent of charge. Typical results from this operation were reported earlier (Gearhart and Garwin, 1976).

Since then the ROSE process has been utilized by four domestic refiners to produce a wide spectrum of products including lube oils, a variety of asphalt products, incremental feedstocks for downstream processing in catalytic crackers, together wi th seve ral specialty products. Two additional commercial ROSE units are in the engineering design and construction stage. An 8 m3/SD ROSE pilot plant has been in operation in Japan since early 1982.

A propane deresining un i t was converted to the ROSE process in the latter part of 1979 for a licensee whose plan t produces lube oil and resin from Pennsylvania crude residues. The conversion resulted in a significant capacity

825

ESL-IE-83-04-128

Proceedings from the Fifth Industrial Energy Technology Conference Volume II, Houston, TX, April 17-20, 1983

increase and a substantial reduction in utility requirements. Feedstock and product analyses from a test run in the converted plant are summarized in Table 2.

A grass roots 795 m3/SD 2-stage pentane ROSE unit was constructed at one of Kerr-McGee's refineries in 1979 to produce asphalt and cat cracker feedstock from either atmospheric or vacuum residue. This unit has been modified recently to produce asphaltenes from the first stage and a resin-oil fraction from the second stage. Kerr-McGee is conducting a series of exploratory tests to develop commercial uses for the asphaltenes and asphaltene-resin blends. Typical results for this ROSE unit are shown in Table 3.

A third commercial ROSE unit was completed and placed into operation in January 1981. This 3­stage 510 m3SD grass roots ROSE unit is capable of utilizing isobutane, normal butane or pentane solvent. The feedstock is a waxy vacuum residue from a variety of crudes. The products are asphalt blend stocks and deasphalted oil (DAO). The DAO is charged directly to a fluid catalytic cracker without further treatment. This plant has opera ted well above design charge rates for extended periods. Operating results are shown in Table 4.

The installation and commissioning of a fourth ROSE unit was completed in December 1981. This plant, a 3-stage 715 m3/SD grass roots facility, was initially designed to use normal pentane solvent to produce 1630 c Ring and Ball softening point asphaltenes, resins and DAO. The inherent flexibility of the ROSE process made it possible to start up this plant on normal butane solvent instead of the pentane for which it was designed. This unit currently produces asphalt blending stocks and DAO for catalytic cracking feedstock. Operating results are listed in Table 5.

construction is in progress on the fifth com­mercial ROSE unit, a 1908 m3/SD 3-stage pentane unit with completion and start-up due shor tly. This plant will be capable of producing high softening point asphaltenes, which will be flaked and marketed as supplemental fuel for coal-fired boilers. The resins will either be blended to fuel oil or combined with the DAO for catalytic cracking feedstock.

The process des ign has been comple ted and detailed engineering is in progress on a sixth ROSE unit. This will be a 795 m3/SD 2-stage unit designed to use normal butane solvent initially, but will have the flexibility to use either isobutane or normal pentane. The asphal­tene-resin product will be blended into fuel oil, and the recovered DAO will be charged to a catalytic cracking unit.

FEKDSTOClt EVALUATION PACILI'!'IES

A 0.16 m3/SD bench-scale ROSE pilot plant facili ty is available at Kerr-McGee I s Technical Center in Oklahoma City, Okla., to evaluate feed­stocks for potential licensees as well as to set the design basis for commercial ROSE units. This facility is also used for continued research and development to find further improvements and utili za tions of ROSE technology. The choice of solvent is dictated by the nature of the feedstock and the desired specifications for the materials to be produced.

In addi tion to residual feedstocks der ived from crude petroleum, oil from tar sands and oil shales, pyrolysis tar residues and tall oil have been processed in the ROSE pilot plan t. Experi­mental work has been planned for upgrading very heavy crudes. Deasphal tening in the field would result in a lower-viscosity crude with improved transportation and processing Characteristics. This upgrading technique would be applicable to many domestic as well as foreign crudes.

ROSE PROCESS VERSUS DELAYED COt!NG

Tradi tionally, cokers have been destined to the role of "garbage can" for refiners who process heavy crudes and have an excess of asphal t or heavy residual fuel production. When no satis­factory or profitable market existed for these materials, a coker was considered to- be the obvious answer because it served a dual purpose: First, a coker affords a means of increasing the value of the products made from a barrel of crude. Second, a coker removes the limitation on crude runs imposed by the excess residuals such as asphalt or fuel oil. Delayed cokers also play an important role in the production of specialty products, such as electrode coke for the metallurgical industry.

Many residuals contain excessive contaminants (sulfur and metals) , thereby producing an inferior quality coke which is unsuitable for many applications. The coker gas oil, coker naphtha and sour gas are difficult products to utilize in most refineries, and less desirable for downstream processing than uncracked or virgin material. Furthermore, a significant portion of the gas oil contained in the residual feedstock to a coker is cracked, producing gas and coke.

The ROSE process offers numerous advantages over coking: (1) ROSE yields a higher liquid recovery of virgin gas oil, which is a better feedstock than coker gas oil for downstream processing in hydrotreaters, cat crackers or hydrocrackers; (2) no sour naphtha or sour gas is produced; and (3) the capital investment and operating costs are significantly less for ROSE than for a coker. Table 6 illustrates the comparative utility costs for a delayed coker versus ROSE.

H26

ESL-IE-83-04-128

Proceedings from the Fifth Industrial Energy Technology Conference Volume II, Houston, TX, April 17-20, 1983

Based on vacuum residuum from Arabian Light crude as a feedstock, product yield and quality estimates have been developed for a delayed coker and a ROSE unit. ROSE unit yields and produc t quali ties are shown for both n-pentane and n-butane solvents. The delayed coker data is summarized in Table 7A and the ROSE data is shown in Table 7B. By utilizing n-pentane sol­vent the asphaltene yield from a ROSE unit only 20 wt% while a coker would produce 32.8 wU coke. If cleaner OAO is required, progres­sively lighter solvents can be utilized as is demonstra ted in Table 7B for n-butane. OAO can be processed in a fluid catalytic cracker or hydrocracker, yielding products which do not require the extensive upgrading that coker products necessitate.

POTRN'rIAL USES FOR ROSE PRODUCTS

Asphaltenes can be burned either as a liquid or as a solid fuel like coke or coal. Other potential uses for asphaltenes are: as a sub­stitu te for coke in metallic ore reduction and in cement manufacture 1 as a blending component in asphal t manufacture 1 as either a solid or liquid fuel for fluidized bed combustionl as feedstock for partial oxidation to produce synthetic fuel gas which, in turn, can be con­verted to hydrogen, ammonia or methanoll and as an additive in drilling mud. Asphaltenes typically have Ring and Ball softening points ranging from 700 to 2000 C and a heating value of 39,500 kilojoule/kilogram. Their ash content is nil. Asphaltenes are very friable, having a Hardgrove Gr indabili ty Index of 120+, compared to about 50 to 60 for coal and 40 to 60 for delayed coke. A typical asphaltene contains 60­70 wt% volatile and 30-40 wt% fixed carbon.

Resins can be used as coker feedstock to produce premium quality anode grade coke for the aluminum industry, provided the sulfur and vanadium content is not excessive. This utili­zation has been demonstrated on a pilot-plant scale. Resins are a valuable ingredient in asphalt, providing ductility and viscosity as well as improved homogeneity. Resins and asphaltenes can be blended in a variety of com­binations to produce binders, protective coat­ings, asphalt cements, etc.

OAO can be produced in a wide range of specifi­cations by the ROSE process to manufacture lubricating oils as well as incremental feed­stock for either hydrocracking or cat cracking. The ROSE process can also be used to separate OAO into multiple fractions.

CSD PROCESS - BRIEF DESCRIP'l'I~ AND APPLICA'l'I~

As an ou tgrowth of ROSE technology, Ker r-McGee has developed another critical solvent extrac­tion process for separating solids (mineral matter and unreacted coal) from coal liquids as well as for removing solids from shale oil and tar sands. This process operates near the critical point of the extraction solvent and is

called the Critical Solvent Oeashing process, or CSO.

Figure 2 is a simplified CSo flowsheet. Vacuum bottoms feed and the extraction solvent are mixed and fed to a first-stage settler. The heavy phase, a slurry containing 10-20 wt% of the coal liquids plus deashing solvent, along with essentially all of the coal solids, is "let-down" to about atmospher ic pressure. The deashing solvent is flashed and recovered, and the remaining solids leave the system as a dry flowable powder called Ash Concentrate. The light phase, which contains the remaining 80-90 wt% of the coal liquids, is heated and passed to a second-stage settler. The heavy phase is let-down to flash off deashing solvent and produce SRC (solvent refined coal). The pressure of the light phase is reduced, and it passes to a third-stage settler. The heavy phase, after flashing off the deashing solvent, becomes Light SRC, which either becomes a product for sale or is recycled back to the coal liquefaction step to become a part of the coal-dissolving solvent.

As a result of the successful demonstration of the CSO process in the SRC pilot plant at Wilsonville, Ala., and in subsequent bench-scale development work, International Coal Refining Company has licensed the CSo process as the solids-removal method for the proposed 5,500 metric ton/day SRC-I demonstration plant.

In addition to the pilot plant, Kerr-McGee has bench-scale facilities for evaluating samples for potential CSO licensees to test not only coal liquids slurry samples but also shale oil or tar sands oil samples.

caiCLOSIONS

The ROSE process has emerged as a competitive, well established and commercially proven solvent extraction process. No other solvent extraction process offers greater flexibility, either in choice of solvent, in ability to process as many different types of feedstocks, or in potential recovery of higher-quality products with more attractive market values. ROSE has the addi­tional advantages of lower capital investment and substantially lower operating costs when compared with conventional solvent extraction processes. Evaporative type solvent recovery systems can be readily converted to utilize ROSE technology with a resultant reduction in energy requirements.

The energy-efficient ROSE process offers an economical and effective method for rejecting metal containing carbonaceous and highly viscous components from heavy residual or low-gravity crude oil feedstocks that are destined for downstream hydrotreating, hydrocracking or catalytic cracking. Each of these downstream processes is very catalyst-sensitive to such contaminants because of their adverse effect on catalyst life and activity.

827

ESL-IE-83-04-128

Proceedings from the Fifth Industrial Energy Technology Conference Volume II, Houston, TX, April 17-20, 1983

The flexibility of the ROSE process can be utilized by refiners to improve their processing capabilities and profitability. These appli ­cations include: debottlenecking or replacing cokers; relieving overloaded vacuum units; producing premium qual i ty lubricating oils; manufacturing paving, roofing and specialty asphalts; substantially reducing or eliminating heavy fuel oil produc tion; and recover ing incremental feedstocks for hydrocracking or cat cracking.

In summary, the poten tial advan tages offered by the ROSE process are:

• Lower capital investment• Lower utility costs• Lower maintenance costs• Wide range of solvent selection• Maximum yield and quali ty of recovered

oil• Ability to process a wide range of

feedstocks• Easy conversion of conventional

solvent extraction units

The advantages offered by the ROSE process and their economic impact can be readily evaluated for your particular refinery configuration. With the continuing decline in average crude gravity, coupled with higher sulfur content of the crude, it is imperative that more effective processing of the residual portion of crude petroleum be employed to remain competitive.

Gearhart, J.A., and Garwin, Leo, "ROSE process improves resid feed", Hydrocarbon Processing Vol. 55, No. 5 125-8 (1976)

Gearhart, J.A., "Solvent treat resids", Hydro­carbon Processing, Vol. 59, No. 5 150-1 (1980) •

TABLE 1

COMPARISON OF UTILITY COSTS

Conventional Propane Units/m3

Deasphalting of Charge Rate Cost/m3

Electrical Power, kWh 17.6 $0.050/kWh $ 0.88 Steam, kg 610 $0.018/kg $10.98 Fuel Fired, MJ 910 $0.006/MJ $ 5.46 Cooling Water, m3 37 $0.013/m3

~

Total $17.80

ROSE Process

Electrical Power, klfu 17.6 $0.050/kWh $ 0.88 Steam, kg 34 $0.018/kg $ 0.61 Fuel Fired, MJ 640 $0.006/MJ $ 3.84

Total $ 5.33

Difference in utility costs/m3 residuum charge

S2H

ESL-IE-83-04-128

Proceedings from the Fifth Industrial Energy Technology Conference Volume II, Houston, TX, April 17-20, 1983

TABLE 2

PENNZOIL ROSE UNIT CHARGE AND PRODUCTS ANALYSES

Vacuum Residue Resin Oil

Yields, Wt% on Crude 16.9 3.2 13.7 LV% on Crude 15.0 2.7 12.3 Wt% on Vac Resid LOO 18.9 81.1 LV% on Vac Resid 100 17.9 82.1

Specific Gravity, 15°C/4°C 0.906 0.959 0.895

Nitrogen, Wt% 0.03 0.09 0.02 Sulfur, Wt% 0.08 0.14 0.07 Conradson Carbon, Wt% 2.7 11.2 0.67 Nickel, wppm 0.2 0.8 0.01

Viscosities, mPa'S (cst) 42 @ 99°C 109 @ 135°C 11.3 @ 135°C 208 @ 60°C 490 @ 99°C 27.9 @99°C

Strieter Analysis

Asphaltenes, \~t% 0.1 0.5 0 Resins, Wt% 12.1 27.4 8.5 Oils, Wt% 87.8 72.1 91.5

TABLE 3

KERR-McGEE ROSE UNIT CHARGE AND PRODUCTS ANALYSES

Vacuum Residue Pitch Oil

Yields, Wt% on Crude 18.4 5.9 12.5 LV% on Crude 16.0 4.8 11.2 Wt% on Vac Resid 100 32.0 68.0 LV% on Vac Resid 100 30.1 69.9

Specific Gravity, 15°C/4°C 0.978 1.041 0.951

Nitrogen, Wt% 0.6 1.0 0.4 Sulfur, Wt% 1.2 1.6 1.0 Conradson Carbon, Wt% 13 27 6.7 Nickel, wppm 60 160 20 Vanadium, wppm 120 320 30 R & B Softening Point, °c 24 82

Viscosities, mPa's (cst) 360 @ 99°C 1600 @ 149°C 260 @82°C iDO @ 127°C 190 @ 191°C 220 @ 99°C

Strieter Analysis, Wt%

Asphaltenes 9.2 27.8 0.5 Res ins 23.4 30.6 20.0 Oils 67.4 41.6 79.5

H29

ESL-IE-83-04-128

Proceedings from the Fifth Industrial Energy Technology Conference Volume II, Houston, TX, April 17-20, 1983

Yields, Wt% on Crude LV% on Crude Wt% on Vac Resid LV% on Vac Resid

Specific Gravity, 15°C/4°C

Nit ragen, Wt% Sulfur, Wt% Conradson Carbon, Wt% Nickel, wppm Vanadium, wppm R & B Softening Point, °c Penetration @ 25°C

Viscosities, mPa's (cst)

Strieter Analysis, Wt%

Asphal tenes Resins Oils

ROCK

Yields, Wt% on Crude LV% on Crude Wt% on Vac Resid LV% on Vac Resid

Specific Gravity, 15°C/4°C

Nitrogen, Wt% Sulfur, Wt% Conradson Carbon, Wt% Nickel, wppm Vanadium, wppm R & B Softening Point, °c

Viscosities, mPa.s (cs t)

Strieter Analysis, Wt%

Asphaltenes Resins Oils

PESTER ROSE UNIT

Vacuum Residue

15.1 13.0 100 100

0.985

0.5 1.5 14 34 86

420 @ 93°C 110 @ 121°C

12.5 25.0 62.5

ISLAND ROSE UNIT

Vacuum Residue

7.2 6.1 100 100

1.001

0.6 1.5 18 30 75 38

406 @ 118°C 106 @ 149°C

15 17 68

TABLE 4

CHARGE AND PRODUCTS

Asphaltene

4.1 3.3 27.2 25.0

1.073

0.9 2.4 35 100 270 102

420 @ 199°C 335 @ 204°C

40.8 26.8 32.4

TABLE 5

CHARGE AND PRODUCTS

Asphaltene

2.2 1.7 31 28.1

1.104

0.8 2.3 41 90 220 127

1360 @ 210°C 390 @ 238°c

49 23 28

ANALYSES

Resin Oil

1.6 9.4 1.3 8.4 10.3 62.5 10 .0 65.0

1.014 0.946

0.6 0.3 2.0 1.1 19 4.0 42 4.0 89 5.5

23.5

300 @ 138°C 330 @ 66°C 126 @ 157°C 79 @ 93°C

6.5 1.1 54.6 19.4 38.9 79.5

ANALYSES

Resin Oil

0.5 4.5 0.4 4.0 7.0 62 7.0 64.9

1.003 0.956

0.6 0.5 1.4 1.1 17 7.0 18 4 33 6

395 @ 118°C 440 @ 79°C 117 @ 135°C 159 @ 99°C

1.2 a 27 13 72 87

H30

ESL-IE-83-04-128

Proceedings from the Fifth Industrial Energy Technology Conference Volume II, Houston, TX, April 17-20, 1983

TABLE 6

COMPARISON OF UTILITY COSTS

DELAYED COKER VS

Delayed Coker Units/m3 of Charge

Electrical Power, Steam, kg Fuel Fired, MJ

Total

kWh 5.4 7.4

3085

ROSE Process

Electrical Power, Steam, kg Fuel Fired, MJ

Total

kWh 17 .6 34

640

Difference in utility costs/m3 residuum charge

ROSE

Rate

$O.OSO/kWh $0.018/kg $0.006/MJ

$O.OSO/kWh $0.018/kg $0.006/MJ

Cost/m3

$ 0.27 $ 0.13 $18.51

$18.91

$ 0.88 $ 0.61 $ 3.84

U:ll

83]

ESL-IE-83-04-128

Proceedings from the Fifth Industrial Energy Technology Conference Volume II, Houston, TX, April 17-20, 1983

----------------- --- - ---

TABLE 7A

COMPARISON OF DELAYED COKER AND ROSE YIELDS

FOR ARABIAN LIGHT VACUUM RESIDUUM

Coker Products Feedstock Lt. Arab.

l85-343

Vac. Resid H2S ~ C5/C6 Cr 185°C °c 343°C+ Coke

Wt% Yield 100 0.4 10.3 3.1 10.8 26.0 16.6 32.8 Vol% Yield 100 4.8 14.3 30.2 18.2 API Gravity 6.9 82 52 29.4 21 Sulfur, Wt% 4.0 0.8 0.9 2.4 3.8 6.9 Nitrogen, wppm 3100 50 1050 2100 7500 Con. Carbon, Wt% 20.8 K 11.4 11.3 11.1 Nickel, wppm 23 70 Vanadium, wppm 75 229

TABLE 7B

ESTIMATED ROSE YIELDS

FOR ARABIAN LIGHT VACUUM RESIDUill1

ROSE Products Feedstock Lt. Arab. n-Pentane n-Butane

Vac. Resid. Asphaltene DAO Asphaltene DAO

\olt% 100 20.0 80.0 32.8 67.2 Vol% 100 17 .8 82.2 30.2 69.8 API Gravity 6.9 - 7.4 10.7 - 7.4 12.3 K 11.4 11. 7 11. 7 Sulfur, lolt% 4.0 5.7 3.6 5.4 3.3 Nitrogen, wppm 3100 fi600 2200 5800 1800 Can. Carbon, Wt% 20.8 52 13 43 10 Nickel, wpprn 23 87 7 65 2 Vanadium, wpprn 75 290 22 214 8 R & B Softening Point, °c 177 135

832

ESL-IE-83-04-128

Proceedings from the Fifth Industrial Energy Technology Conference Volume II, Houston, TX, April 17-20, 1983

I I I I I I I I I I

FIGURE 1

;-------1

I,II I

1

II

I "'---"-1----.

START

1 ­ _

ASPHALTENES RESINS OILS

H33

ESL-IE-83-04-128

Proceedings from the Fifth Industrial Energy Technology Conference Volume II, Houston, TX, April 17-20, 1983

FIGURE 2

THREE STAGE CRITICAL SOLVENT DEASHING UNIT

VACUUM TOWER BOTTOMS

FEED

ALTERNATE

FEED SURGE TANK

MIXERFEED PUMP

HIGH PRESS

SOLVENT PUMP

1ST 2ND 3RD STAGE STAGE STAGE

SETTLER SETTLER SETTLER

DEASHING SOLVENT

PUMP

DEASHING DEASHING DEASHING SOLVENT SOLVENT SOLVENT

SEPARATOR SEPARATOR SEPARATOR NO.1 NO.2 NO.3

ASH SRC LIGHT CONCENTRATE PRODUCT SRC

PRODUCT

834

ESL-IE-83-04-128

Proceedings from the Fifth Industrial Energy Technology Conference Volume II, Houston, TX, April 17-20, 1983

Figure 3

CONVENTIONAL SOLVENT RECOVERY

SOLVENT ENTHALPY DIAGRAM

ISOTHERMS INCREASING TEMPERATURE

\ \ \

LIQUID

~-- A - EXTRACTOR V-1

E ­ FLASH

~--- F ­ STRIPPER

::> 1---------------;(/) VAPOR

w a:

(/) w a: Q..

~ Z (/)

« w a: U Z

INCREASING ENTHALPY

83'>

ESL-IE-83-04-128

Proceedings from the Fifth Industrial Energy Technology Conference Volume II, Houston, TX, April 17-20, 1983

Figure 4

SUPERCRITICAL SOLVENT RECOVERY

SOLVENT ENTHALPY DIAGRAM

ISOTHERMS INCREASING TEMPERATURE ...

VAPOR

\ \ \

Y--- A - EXTRACTOR V-'

B ­ H-2 INLET

1'----- C ­ DAO SEPARATOR -----10

D ­ RECYCLE SOLVENT EX E-' -- ­ ....

E-FLASH

1---- F ­ STRIPPER

LIQUID

w ex:: ::> I---------------j(J) (J) w ex:: a­t=) Z (J)

<X: w ex:: U Z

INCREASING ENTHALPY

836

ESL-IE-83-04-128

Proceedings from the Fifth Industrial Energy Technology Conference Volume II, Houston, TX, April 17-20, 1983