Resid hydrocracking better than delayed coking in case studies

Feb. 16, 1998
Case Studies [114,894 bytes] To highlight the advantages of two popular vacuum residue upgrading routes, coking and hydroprocessing, IFP North America Inc. completed a detailed economic comparison. The study compares delayed coking plus hydrotreating vs. the H-Oil process (residue hydrocracking) at two conversion levels, moderate (65 vol %) and high (85 vol %), and on two different feedstocks, Arab Light/Heavy vacuum residue and Isthmus/Maya vacuum residue. The six cases studied are shown in

UPGRADING BOTTOMS-2

Lawrence Wisdom, Eric Peer, Pierre Bonnifay
IFP North America Inc.
Princeton, N.J.
To highlight the advantages of two popular vacuum residue upgrading routes, coking and hydroprocessing, IFP North America Inc. completed a detailed economic comparison.

The study compares delayed coking plus hydrotreating vs. the H-Oil process (residue hydrocracking) at two conversion levels, moderate (65 vol %) and high (85 vol %), and on two different feedstocks, Arab Light/Heavy vacuum residue and Isthmus/Maya vacuum residue. The six cases studied are shown in the box.

The cases enable refiners to examine the effects of heavy and very heavy crude on process performance, product quality, and overall economics. Hydrotreating was added to the delayed coking cases to meet the product quality specifications mandated by many refiners. Each case processes 38,000 b/sd of a nominal 565° C+ vacuum residue.

In the first three cases, the feed is derived from a 50/50 blend of Arabian Light and Heavy crudes which have been selected by many companies as a benchmark for upgrading studies. Key properties for the Arabian feedstock are shown in Table 1 [16,192 bytes].

H-Oil and coking processes

A simplified flow diagram of the H-Oil process integrated within a typical refinery is shown in Fig. 1 [43,692 bytes].

The H-Oil plant consists of a single train with two reactors in series. All products are routed directly to the product pools or to downstream treating units. The light naphtha is normally routed to the gasoline pool or to an isomerization unit. The heavy naphtha is blended with virgin naphtha and hydrotreated prior to reforming.

The H-Oil diesel is sent directly to tankage, while the VGO is mixed with virgin VGO prior to being routed to a fluid catalytic cracking (FCC) unit. Unconverted H-Oil bottoms are blended with FCC light cycle oil (LCO) and routed to the low-sulfur fuel-oil pool.

Moderate-conversion H-Oil bottoms are blended to 1.0 wt % sulfur-fuel oil, while high-conversion bottoms are blended to 1.5 wt % sulfur fuel oil. The use of high-activity, second-generation catalysts allows production of high quality products and stable fuel oil.

A simplified flow diagram of the delayed coker plus hydrotreating unit (HTU) case is shown in Fig. 2 [44,566 bytes]. In the delayed coking case, all liquid products must be hydrotreated prior to routing to final products or to downstream processing units.

Coker products contain high levels of sulfur and nitrogen, as well as diolefins, olefins, and aromatics, which must be saturated in order to meet stability and product quality specifications.

Product quality

A comparison of product quality between the H-Oil and delayed coker (before hydrotreating) products is presented in Table 2 [29,618 bytes]. The H-Oil product quality from either moderate or high-conversion operation is better than the delayed coker but comparable after downstream hydrotreating. Normally, the naphtha, full-range gas oil (diesel), and vacuum-gas oils are hydrotreated in three separate units to efficiently and economically meet the processing objectives.

A comparison of C4+ liquid product yields is presented in Fig. 3 [47,605 bytes]. The two H-Oil cases produce about the same volume of total products; however, the high-conversion case produces proportionally more C4-565° C. products and less 565° C.+ product than the moderate-conversion case. The average C4+ liquid product rate for the two H-Oil cases is 40,500 b/sd, or nearly 107 vol % of fresh feed.

As shown in Fig. 3, the delayed coker case produces much less liquid product. The total C4+ liquid product in the delayed coker case is only 24,200 b/sd, equal to about 64 vol % of fresh feed. This is more than 40% less liquid product than is produced by the H-Oil unit. However, the distillate-product yield is about the same as the moderate-conversion H-Oil case but is significantly lower (32%) in comparison to the high-conversion H-Oil case.

Cost comparison

Investment costs were calculated for each case. The estimates are total installed costs, including catalyst, fractionation, light ends recovery, and offsites. Ancillary units, such as hydrogen production and sulfur recovery, are omitted from this analysis.

The investment costs for the first three cases are presented in Fig. 4 [51,332 bytes]. As shown, the delayed coker alone is less expensive than either H-Oil unit, which explains why it was so popular from 1960 to 1980 before strict product quality requirements came into effect. When the required hydrotreating is added, however, the cost of the coker case exceeds the H-Oil unit investment.

The delayed coker costs about $230 million, and the hydrotreaters cost about $90 million, for a total of $320 million. The moderate-conversion H-Oil plant is estimated to cost $254 million, and the high-conversion H-Oil plant $275 million.

Many times investment is shown as $/bbl of feed, or $/bbl of C4+ liquid product, as presented in Fig. 5 [54,295 bytes]. The lowest-cost option on either basis is the moderate-conversion H-Oil case. On a $/bbl of feed basis, this case is estimated to cost $6,700. The cost drops to $6,300/bbl of product, reflecting the C4+ product rate of 107 vol % of fresh feed discussed above.

Over the past 10 years, cost for H-Oil units has been declining as a result of major design improvements, which have enabled refiners to process more feed in a single-train plant. The coker plus hydrotreater case is estimated to cost $8,400/bbl of fresh feed, one-third more than the moderate-conversion H-Oil case. On a per-barrel-product basis, the coker case increases to $13,200/bbl, more than double the moderate-conversion H-Oil case.

Economic analysis

An economic analysis was performed on the three cases to determine which had the best return and shortest payout time. Operating costs, including purchased hydrogen, were calculated for each unit. The H-Oil cases have higher operating costs than the coker- plus-hydrotreater case due to higher hydrogen consumption and the daily addition of fresh catalyst.

Price of the vacuum residue feedstock was back calculated by removing FCC LCO to produce a high-sulfur fuel oil meeting a viscosity of 380 cSt at 50° C. Feedstock and product prices used in this analysis are shown in Table 3 [34,625 bytes]. The prices in this table correspond to differentials (diesel minus high sulfur-fuel oil) of $13.80/bbl. This price differential was sufficiently high to encourage many projects to move ahead during the periods of 1980-82, 1990-92, and was positively projected in the year 2000 and beyond.

The results of the analysis are shown in Fig. 6 [49,603 bytes]. The two H-Oil cases have similar rates of return and payout times. The internal rate of return (IRR) calculated for the moderate-conversion case is 27.5%, yielding a 3.2-year payout. For the high-conversion case, the IRR is slightly lower, 27.0%, but the payout time remains the same.

The combination of low liquid yields and higher investment cost give the delayed coker case the lowest rate of return and the longest payout time. The coker-plus-hydrotreater case has an IRR of 19.8%, nearly 30% lower than the moderate-conversion H-Oil case. The corresponding payout time is 4.5 years, nearly 16 months longer than that of either H-Oil plant.

Isthmus/Maya residue

A sensitivity study showing the effect of a heavier feedstock was also performed. The feedstock chosen was a blend of Isthmus and Maya vacuum residue derived from a 60/40 crude blend. Pertinent feedstock inspections are shown in Table 4 [14,629 bytes].

Compared with the Arabian vacuum residue shown in Table 1, this feedstock has a significantly higher specific gravity (1.064 vs. 1.039) and metals content (707 vs. 221 ppm (wt)). The higher Conradson carbon residue (CCR) content (27.8 vs. 24.6 wt %) will lead to lower liquid yield in the delayed coker. However, the higher metals and specific gravity will lead to higher operating costs for the H-Oil unit.

The same three scenarios were studied for this feedstock: moderate-conversion H-Oil with unconverted bottoms to 1.0 wt % sulfur-fuel oil, high-conversion H-Oil with unconverted bottoms to 1.5 wt % sulfur-fuel oil, and delayed coking plus hydrotreating.

Investment costs for the three new cases were calculated on the same basis as for the Arabian vacuum residue cases. The results are shown in Fig. 7 [59,630 bytes]. The H-Oil units are slightly more expensive for the Maya/Isthmus feed primarily as a result of higher hydrogen consumption with these heavier residues.

The delayed coker plus hydrotreater investment has also increased over the Arabian case as a result of greater coke production with the Maya/Isthmus feed. The net effect, the differential in investment cost between residue hydroprocessing and delayed coking, has been reduced when heavy high metal feedstocks are processed.

The economic analysis was repeated for the Maya/Isthmus feedstock to show the effects on rate of return and payout time for the heavier feedstock. Using the method described above, a price of $1.60/bbl was calculated for the Maya/Isthmus vacuum residue feedstock. The results are shown in Fig. 8 [53,365 bytes] along with the Arabian vacuum residue cases as a reference.

For both H-Oil cases, the IRR dropped from 27.5% to 22.4% in the moderate-conversion case and from 27.0% to 22.2% in the high-conversion case. The drops are a result of higher operating costs, primarily increased hydrogen consumption and catalyst addition rate associated with the heavier, high-metals feedstock. Due to increased coke make, the IRR of the delayed coker case dropped as well, from 19.8% to 18.9%.

The moderate-conversion H-Oil case remains the most economic, having an IRR nearly 20% higher than for delayed coking plus hydrotreating.

Even if price differentials between diesel and high-sulfur fuel oil drop to an extremely low level of $6.00/bbl, residue hydrocracking would result in a 13% IRR and would be slightly better than the delayed coking option because of the additional hydrotreating requirements.

Copyright 1997 Oil & Gas Journal. All Rights Reserved.