ENZYMES DESULFURIZING DIESEL FUEL IN PILOT PLANT TESTS

Anne K. Rhodes
Refining/Petrochemical Editor

Energy BioSystems Corp., The Woodlands, Tex., is collecting data from a new 5 b/d, continuous-operation, biocatalytic desulfurization (BDS) pilot plant. The plant, located at codeveloper Petrolite Corp.'s St. Louis facility, has been operating since early March.

GOALS

EBC vice-president of research and development, Daniel Monticello, says refiners worldwide will need to add 10-15 million b/d of desulfurization capacity in the next 10 years. This equates to capital expenditures of about $30 billion and $6 billion/year in operating expenses.

EBC does not intend to replace any existing refinery desulfurization capacity with its new technology. Instead, the company sees the process as a supplement to existing refinery technologies.

To produce a successful commercial-scale BDS process, Energy BioSystems must improve:

• Catalyst specific activity (kg sulfur removed/hr/kg catalyst)

• Catalyst stability (longevity, or catalyst "half life")

• Manufacturing costs ($/kg cost of catalyst fermentation).

In 1990, specific activity was 0.1 desulfurization units (DSU), as defined by EBC. This figure increased to 5.0 DSU in 1992 and is now 30 DSU, as a result of research and development efforts.

Monticello says the company hopes, through conventional fermentation optimization techniques, to increase catalyst activity 3 to 5-fold in 1995 and 10 to 20-fold over the following year. This optimization involves molecular genetics, or DNA manipulation, of the Rhodococcus erythropolis bacteria.

Monticello believes the 10 to 20-fold increase in activity is likely, and says precedents have been set for scale-up o this magnitude. The two examples he cited were:

• Detergent proteases - These enzymes were developed to remove protein stains from clothing and are no found in most laundry detergents

• High-fructose corn syrup - A common sweetener that used to be produced by a chemical process but is no produced biologically as a result of what Monticello calls "metabolic debottlenecking."

EBC's catalyst initially exhibited a 6-10 hr stability. The desired stability is 20 hr, and recent laboratory data indicate a 24 hr stability for the catalyst in its present form. These results indicate that EBC is not far from the target stability, says Monticello.

The fermentation yield of the catalyst originally was 1 g/1. EBC has boosted this to 60 g/l., which is better than most of today's industrial fermentations, according to Monticello.

EBC's original plan for step-wise increase in the size of the fermentation reactor was: Shake flask (~250 ml) - 1 l. - 14 l. - 300 l. - production. The company recently became capable of producing 350 1. batches of catalyst at its headquarters in The Woodlands.

"In 18 months, we have grown from zero fermentation capacity to 30 liters to 350 liters," said EBC president and CEO John Webb.

Process issues to be addressed by the pilot plant include reactor technology, separations, product quality, and scale-up. The sequence for process scale-up is:

• Shake flask (~250 ml)

• Stirred reactor (1 l.)

• Bench-scale continuous reactor (1 gpd)

• Pilot plant (0.5-5 b/d)

• Production (10,000 b/d).

This means, if the project continues as planned, EBC is now just one step away from the commercial scale.

PARTNERSHIPS

EBC has lined up five alliances for the development and commercialization of the BDS process:

• Petrolite

Petrolite paid for the $1.5 million St. Louis pilot plant, supplied $5.4 million in research and development funding, and will share in the gross profits from the process. And because Petrolite has a large refinery service business with an established infrastructure in operating refineries, the company was chosen to provide technical service for commercial BDS units.

• M. W. Kellogg Co.

Kellogg built the pilot plant, and will be the exclusive supplier of basic engineering and design packages for commercial BDS units.

• Total

Total supplies the diesel feedstock for the pilot plant. The company also plans to build and operate a BDS pilot plant in France at its own expense and, if results are successful, will build a commercial BDS unit.

• Koch Refining Co.

Koch will provide refinery operations support and product quality testing for development of a gasoline BDS process. Koch also will provide EBC with a gasoline stream for biocatalytic desulfurization.

• Texaco Inc., Exploration & Production Technology Division

Texaco E&P will provide additional petroleum engineering and analytical sources for the development of a BDS process for desulfurizing mildly sour crude oil. Texaco will choose the target crude.

PILOT PLAN

The BDS process uses enzymes to remove organically bound sulfur from petroleum streams at mild temperatures and atmospheric pressure (Fig. 1(81590 bytes).

The pilot plant was designed by EBC, Petrolite, and Kellogg. Kellogg constructed the unit in Houston and shipped the two stackable sections to Petrolite's headquarters in St. Louis.

The catalyst slurry and oil are mixed in a 40-gal bioreactor that is jacketed, with either steam or coolant, for temperature control. The reaction mixture is sent to a three-phase centrifuge, where the aqueous phase is separated from the desulfurized oil.

An electrostatic precipitator removes additional water from the oil product. A series of three filters removes any remaining catalyst.

Sodium hydroxide is injected to neutralize the aqueous phase. This neutralization produces a sodium salt from the by-product sulfate. The water and biocatalyst are recycled to the reactor.

Process by-products are spent biocatalyst and sulfate. In the pilot plant, the sulfate is produced in the form of sodium sulfate bee se sodium hydroxide is use for neutralization. The sodium sulfate produced is harmless and can be processed in the refinery sewer system.

If ammonia is used for neutralization, the process byproduct is ammonium sulfate, which can be used as fertilizer.

The waste catalyst stream can be processed via several options. The sterilized catalyst can be disposed of as a listed landfill waste. Alternatively, depending on site-specific conditions, the stream can be regenerated and recycled, or incinerated.

Monticello says a 10,000 b/d BDS unit is expected to produce no more than 0.5 ton/day of spent biomass.

The pilot plant is operated by a distributed control system. This unit is linked electronically to EBC's offices in The Woodlands, where the data can be downloaded and analyzed.

Feedstock for the pilot plant is a middle distillate stream shipped from France by Total. The feed is hydrotreated, and is equivalent to a stream Total wishes eventually to biocatalytically desulfurize on a commercial scale.

This feed is particularly suited to biocatalytic desulfurization because BDS removes the remaining sulfur species, which are resistant to hydrotreating. Total supplied 16 metric tons of the feed, which is enough to operate the Plant for more than 230 days.

RESULTS

Objectives of the pilot plant studies include:

• Validating and refining the computer simulations used to control the process

• Establishing the process design basis.

So far, the results from pilot plant operations have met expectations, says Monticello.

The projected 45% desulfurization rate has been achieved, within a few percent. Monticello notes that this rate was simply the target for the initial evaluation experiments, and that the process is capable of desulfurizing almost to extinction.

COMMERCIAL APPLICATION

To commercialize the process, EBC must continue to improve the productivity of the biocatalyst to a competitive economic level and develop a reactor system that allows control of key process parameters for optimum desulfurization.

The company expects capital costs of a commercial BDS unit to be about 50% of the cost of a traditional hydrodesulfurization (HDS) unit. And operating costs are projected to be on the order of 10-15% less than for HDS.

The diesel BDS process can be used as either a pretreatment or polishing step. Thus, hydrotreated diesel is just one possible feedstock for the technology.

Because the biocatalyst is most specific for removal of dibenzothiophene and related molecules, the desulfurization rate is not affected by whether the process lies before or after hydrotreating in the refinery flow scheme.

EBC also plans to develop BDS processes for gasoline and crude oil desulfurization. In the more distant future, possible applications for "biorefining" include: nitrogen reduction, metals removal, viscosity reduction, and other advanced biotransformation processes, possibly including alkylation and cracking.

Copyright 1995 Oil & Gas Journal. All Rights Reserved.