Conversion Extraction Desulfurization (CED) Phase III .../67531/metadc786886/m2/1/high_re… · improvements to the process, which was renamed “CED 2002”. Petro Star and Degussa

Conversion Extraction Desulfurization (CED) Phase III FINAL TECHNICAL REPORT REPORTING PERIOD August 2001 – December 2002 PRINCIPAL AUTHOR: James Boltz, Petro Star Inc. DATE ISSUED: March 2005 DOE AWARD NUMBER: DE-FC26-01BC15281 SUBMITTING PARTY:

Petro Star Inc 3900 C Street, Suite 401 Anchorage, AK 99503

PRINCIPAL SUBCONTRACTORS:

Degussa Corporation 379 Interpace Parkway Parsippany, NJ 07054

Koch Modular Process Systems, LLC 45 Eisenhower Drive Paramus, NJ 07652 Hetagon Inc. 25652 Santo Drive Mission Viejo, CA 92691

DISCLAIMER: This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency thereof, nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any commercial product, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof.

ABSTRACT This project was undertaken to refine the Conversion Extraction Desulfurization (CED)

technology to efficiently and economically remove sulfur from diesel fuel to levels below 15-

ppm. CED is considered a generic term covering all desulfurization processes that involve

oxidation and extraction. The CED process first extracts a fraction of the sulfur from the diesel,

then selectively oxidizes the remaining sulfur compounds, and finally extracts these oxidized

materials.

The Department of Energy (DOE) awarded Petro Star Inc. a contract to fund Phase III of the

CED process development. Phase III consisted of testing a continuous-flow process,

optimization of the process steps, design of a pilot plant, and completion of a market study for

licensing the process.

Petro Star and the Degussa Corporation in coordination with Koch Modular Process Systems

(KMPS) tested six key process steps in a 7.6-centimeter (cm) (3.0-inch) inside diameter (ID)

column at gas oil feed rates of 7.8 to 93.3 liters per hour (l/h) (2.1 to 24.6 gallons per hour). The

team verified the technical feasibility with respect to hydraulics for each unit operation tested

and successfully demonstrated pre-extraction and solvent recovery distillation. Test operations

conducted at KMPS demonstrated that the oxidation reaction converted a maximum of 97% of

the thiophenes.

The CED Process Development Team demonstrated that CED technology is capable of reducing

the sulfur content of light atmospheric gas oil from 5,000-ppm to less than 15-ppm within the

laboratory scale. In continuous flow trials, the CED process consistently produced fuel with

approximately 20-ppm of sulfur. The process economics study calculated an estimated process

cost of $5.70 per product barrel.

The Kline Company performed a marketing study to evaluate the possibility of licensing the

CED technology. Kline concluded that only 13 refineries harbored opportunity for the CED

process. The Kline study and the research team’s discussions with prospective refineries led to

the conclusion that there were not likely prospects for the licensing of the CED process.

TABLE OF CONTENTS

1.0 EXECUTIVE SUMMARY ................................................................................................ 1 2.0 INTRODUCTION .............................................................................................................. 3 3.0 THE ORIGINALLY PROPOSED WORK ........................................................................ 5

3.1 Task 1 - Laboratory-Scale Pilot Plant Validation ............................................................... 5 3.2 Task 2 - Bench-Scale Process Development Unit .............................................................. 6 3.3 Task 3 - Diesel Engine Testing of Processed Fuel ............................................................. 7 3.4 Task 4 - Design of a 50-BPSD Pilot Plant.......................................................................... 8 3.5 Task 5 - Conceptual Design and Cost Estimate of a 5,000-BPSD Plant ............................ 9 3.6 Task 6 - Further CED Research ........................................................................................ 10 3.7 Task 7 - Business Plan ...................................................................................................... 12

4.0 PROJECT PLAN REVISIONS ........................................................................................ 13 5.0 EXPERIMENTAL............................................................................................................ 18

5.1 Task-1, Pilot-Scale Process Validation Tests ................................................................... 19 5.2 Task-6, Develop and Optimize CED Process ................................................................... 21

6.0 RESULTS AND DISCUSSION....................................................................................... 22

6.1 Extractions ........................................................................................................................ 22 6.2 Task 6, Subtask 2: Improved Separations......................................................................... 24 6.3 Task 6, Subtask 3: IMPROVED OXIDATIONS ............................................................. 25 6.4 Summary of 2001 Work.................................................................................................... 26 6.5 2002 Work ........................................................................................................................ 27

7.0 CONCLUSIONS............................................................................................................... 39 8.0 REFERENCES ................................................................................................................. 40 9.0 LIST OF ACRONYMS AND ABBREVIATIONS ......................................................... 40 Figures Figure 1: Original Project Gantt Chart..........................................................................................16 Figure 2: Revised Project Gantt Chart ..........................................................................................17 Tables Table 1: KMPS CED 2000 Process Step Testing........................................................................20

TABLE OF CONTENTS (Cont.) Appendices Appendix A: Petro Star, Inc. LAGO Specifications ......................................................................43 Appendix B: Koch Pilot-Scale Test #1..........................................................................................45 Appendix C: Kline & Company Market Study............................................................................116 Appendix D: CED 2002 Summary Report for 5,100 BPSD Process Unit ..................................195 Appendix E: Koch Pilot-Scale Testing Final Report...................................................................217

Petro Star Inc Page 1 Conversion Extraction Desulfurization Phase III 3/11/2005 1.0 EXECUTIVE SUMMARY On September 20, 2001, the US Department of Energy (DOE) contracted Petro Star Inc. (Petro Star) to develop a method to remove sulfur from diesel fuel. Petro Star calls this process Conversion Extraction Desulfurization (CED). CED is considered a generic term covering all desulfurization processes that involve oxidation and extraction. Petro Star originally named the process “CED 2000”. Petro Star and Degussa Corporation (Degussa) analyzed the equipment performance and developed refinements to the CED 2000 process. This led to important improvements to the process, which was renamed “CED 2002”. Petro Star and Degussa in coordination with Koch Modular Process Systems (KMPS) tested six key process steps in a 7.6-centimeter (cm) (3.0-inch) inside diameter (ID) column at gas oil feed rates of 7.8 to 93.3 liters per hour (l/h) (2.1 to 24.6 gallons per hour). During the testing, the researchers varied several key process operating conditions to determine the effect on desulfurization efficiency and equipment capacity. The team verified the technical feasibility with respect to hydraulics for each unit operation tested and successfully demonstrated pre-extraction and solvent recovery distillation. The CED process must have a sulfur removal efficiency of 99.8% to attain 15 parts per million (ppm) of sulfur in diesel fuel. To attain this goal, the CED oxidation reactor must oxidize at least 99.5% of the sulfur compounds in 5,000-ppm sulfur diesel fuel. Test operations conducted at KMPS demonstrated that the oxidation reaction converted a maximum of 97% of the thiophenes. However, subsequent bench-scale results gave Petro Star reason to be optimistic. Further laboratory work performed by Petro Star and Degussa suggested modifying the oxidation process to attain the desired efficiency. Degussa Corporation worked extensively on developing process engineering models of the CED process that were based on realistic feed compositions and thermodynamics. Degussa used the Aspen PlusTM software to develop a fully converged, steady-state flowsheet model. Degussa based the model on rigorous thermodynamic methods that they developed using experimental and literature data. In addition, Degussa developed a systematic feed characterization method for generating simulated feeds for the CED process models. The CED Process Development Team demonstrated that CED technology is capable of reducing the sulfur content of light atmospheric gas oil from 5,000-ppm to less than 10-ppm within the laboratory scale. In continuous flow trials using rented equipment, the sulfur reaction capability of the process fell short of the 15-ppm goal by producing fuel with approximately 20-ppm of sulfur. As a result, the technical development for the project was scaled back considerably while a marketing study was completed. The Kline Company performed a marketing study to evaluate the possibility of licensing the CED technology. Kline found that there were a total of 143 refineries in the US of which 93 have crude oil capacity less than 150,000 barrels per day, and therefore qualify for various regulatory exceptions to the mid 2006 compliance for producing ultra-low sulfur diesel fuel. Kline interviewed top management in 43 of the 93 refineries. The conclusion from these interviews was that only 13 refineries harbored opportunity for the CED process.

Petro Star Inc Page 2 Conversion Extraction Desulfurization Phase III 3/11/2005 Degussa personnel visited 4 of the 13 refineries in December 2002 over a four-day period. All four refineries were in the Petroleum Administration for Defense Districts (PADD) 3 (Gulf Coast States) & PADD 4 (Mountain States) regions of the US. All of these refineries use moderate pressure hydrodesulfurization to produce highway diesel fuel with sulfur content less than 500-ppm. The management of these refineries had an open mind toward alternate desulfurization technologies, especially if the technology could produce ultra-low sulfur fuel at a lower operating cost. Unfortunately, only one of the three refiners was interested in delaying compliance past mid 2006. Based on the information in the Kline Marketing Study and on the four refinery visits, the research team decided that further investment in the technical development of the CED Process was not warranted.

Petro Star Inc Page 3 Conversion Extraction Desulfurization Phase III 3/11/2005 2.0 INTRODUCTION

Petro Star, Inc. is a refiner of diesel fuel in Alaska and supplies diesel to the general public,

military, and remote villages. Petro Star realized that impeding federal regulations to limit sulfur

concentrations in diesel would affect marketability of their refined petroleum products.

Hydrodesulfurization (HDS) is the standard method to remove sulfur. However, HDS processes

consume large amounts of hydrogen, require exotic catalysts that are easily neutralized and

operate under severe temperature and pressure conditions. These conditions result in expensive

capital and operating costs along with disposal problems associated with spent catalysts and the

by-product formation of elemental sulfur. In addition, HDS cannot remove sulfur from the more

complex thiophenic compounds without severe and costly treatment. These increased costs may

force small and medium size refineries out of the low-sulfur fuel market.

Therefore, Petro Star began a research program to develop another cost-effective method to

remove sulfur compounds from refined petroleum products. Their Phase I efforts developed the

Conversion Extraction Desulfurization (CED) process to chemically remove sulfur from diesel

fuel.

The CED process does not require costly hydrogen processing, high pressures, or high

temperatures. The process mildly oxidizes sulfur compounds and removes the compounds by

solvent extraction at near-ambient conditions. In addition to removing sulfur, the CED process

removes nitrogen-containing compounds and aromatics that adversely affect diesel fuel quality

and engine emissions.

Petro Star Inc Page 4 Conversion Extraction Desulfurization Phase III 3/11/2005 Petro Star, with the support of the Alaska Science and Technology Foundation, initiated Phase II

research to develop a laboratory-based CED process. Three patents on the CED process were

filed during Phase II. These were patents #6,160,193: Method of Desulfurization of

Hydrocarbons; #6,274,785: Method of Desulfurization of Hydrocarbons; and #6,596,914:

Method of Desulfurization and Dearomatization of Petroleum Liquids by Oxidation and Solvent

Extraction. Petro Star completed the Phase II work and submitted a final report (Petro Star,

2000). Cost evaluations resulting from the Phase II work indicated that the CED process would

be cost-effective for small and medium-size refineries manufacturing low-sulfur diesel.

Therefore, Petro Star sought assistance from DOE to develop Phase III.

Since Petro Star is a small refiner, they do not have an extensive research staff or laboratory.

Most of the work for the CED project was subcontracted to other organizations with Petro Star

providing oversight. Complete project reports for work done by each of the other organizations

are in the appendices.

Degussa’s work included the process chemistry, design, and economics for a 5,100 barrel per

standard day (BPSD) CED process unit (Appendix D). Much of their process design work was

based on the pilot-scale experiments performed by Koch (Appendix B and Appendix E). A

market evaluation by Kline (Appendix C) was used to help Petro Star evaluate the merits of

continuing research on the CED process.

Petro Star Inc Page 5 Conversion Extraction Desulfurization Phase III 3/11/2005 3.0 THE ORIGINALLY PROPOSED WORK

Petro Star originally proposed a work plan that was divided into seven tasks with various

subtasks. The original work plan is summarized below.

3.1 Task 1 - Laboratory-Scale Pilot Plant Validation

The goal of this work was to collect process information on the oxidation and extraction steps at

a laboratory-scale level. Petro Star would use this information as the basis for scale-up

decisions.

Subtask 1 - Provide Information

The Petro Star Team would assemble information required for designing experiments to

collect data on the pre-extraction, oxidation, and sulfone extraction steps. Petro Star

would provide this information to personnel at Koch Extraction Technologies (KET) who

would perform the laboratory measurements.

Subtask 2 - Laboratory-Scale Testing

KET would perform tests and collect data on the CED process at the KET Houston

facility. Degussa would provide hydrogen peroxide for the oxidation tests and technical

consulting on the safe operation of handling and using the peroxide. Geoffery E. Dolbear

and Associates (GEDA), Inc. would observe the KET testing to monitor progress and

provide quality control of the data collected. KET would model the chemical reactions

and engineering parameters. Results reported would include tabulated data and

appropriate interpretation. Petro Star would use this information as the basis for scale-up

decisions.

Petro Star Inc Page 6 Conversion Extraction Desulfurization Phase III 3/11/2005

Subtask 3 - Review of Data and Modeling

GEDA engineers would review the results of the KET study and incorporate the

information into the chemical and engineering process models from prior development.

GEDA would use these models to predict process behavior, product compositions, heat

and mass balances, and process economics.

3.2 Task 2 - Bench-Scale Process Development Unit

Petro Star engineers and chemists would design, assemble, and operate a bench-scale, continuous

test process development unit (PDU) in their Anchorage, Alaska laboratory. GEDA would

support Petro Star to build the PDU. The PDU would have a feed rate of about five barrels per

day. Petro Star would use the PDU for measurement of extraction data in test runs lasting a few

hours. Unlike the KET laboratory-scale unit, Petro Star will operate the PDU continuously for

long periods of time.

Information collected during testing of the continuous-run PDU would be useful in the design of

a future pilot plant. Development and operation of the PDU unit was considered critical to the

long-term development and commercialization of the technology.

Subtask 1 - Design and Implementation of PDU upgrades

Petro Star would work with GEDA to design operations and control improvements for

the existing PDU. These system improvements would allow safe, reliable and routine


operation of the PDU. Petro Star would purchase the material, instrumentation, and

equipment.

Subtask 2 - Safe Operation Studies

Petro Star in cooperation with Degussa would perform continuous oxidation tests to

establish safe operating parameters. Phase I and Phase II batch laboratory studies showed

that the compositions and temperatures used in routine oxidations of petroleum gas oils

are well outside the range that can lead to uncontrolled reactions and runaway conditions.

Work under this task would further define these ranges under continuous operation.

Subtask 3 - Operation and Data Evaluation

Petro Star chemists and engineers would operate the PDU with the objective of

demonstrating safe routine operations for pre-extraction, oxidation, and sulfone

extraction. Petro Star would analyze the feeds and products to determine chemical and

physical properties. Information obtained in this task would help develop process

procedures and demonstrate operation conditions that reliably predict process behavior in

a future commercial plant.

3.3 Task 3 - Diesel Engine Testing of Processed Fuel

Low-sulfur diesel produced in the laboratory-scale pilot plant would be tested in a commercial

engine by FEV Engine Technology at their Auburn Hills, Michigan facility. The goal of this

task was to verify the capability of the fuel to reduce emissions and to ensure that the

Petro Star Inc Page 8 Conversion Extraction Desulfurization Phase III 3/11/2005 desulfurized fuel has the ability to be used with sulfur-sensitive after-treatment devices that may

be used to meet future emissions regulations.

Subtask 1 - Preparation of Test Samples

Petro Star would send a sample of low-sulfur diesel, estimated at 250 gallons from the

laboratory-scale pilot plant at the KET Houston facility to FEV along with a sample of

the untreated feed LAGO.

Subtask 2 - Engine Testing

FEV would operate a test engine on low-sulfur CED diesel product and the untreated

LAGO. FEV would use standard calibrations to maintain constant test boundaries. The

engine would be equipped with pressure sensors, thermocouples and other

instrumentation. GEDA would maintain regular contact with FEV to monitor progress.

Subtask 3 - Data Review

GEDA would review FEV’s test results and prepare a summary report detailing any

unexpected results and recommendations.

3.4 Task 4 - Design of a 50-BPSD Pilot Plant

The next logical step in the commercial development of the CED process would be construction

of an integrated pilot plant. Petro Star and GEDA would prepare process flow diagrams and

equipment specifications for a pilot plant facility. The goal of this task would identify costs of

scale-up and operation of a pilot plant.


The pilot plant design would incorporate all the necessary components to process petroleum gas

oils to low-sulfur diesel fuel. The plant would recover and recycle solvent streams to test the

potential buildup of problem compounds. The design would also include the necessary

equipment to prepare marketable chemical or fuel products from the sulfur-rich extract isolated

in the process. The design would allow testing of any steps identified as critical or poorly

defined in the laboratory-scale and PDU studies. The design and supporting information would

be summarized in a written report.

3.5 Task 5 - Conceptual Design and Cost Estimate of a 5,000-BPSD Plant

GTC Technology Corporation (GTC) of Houston, Texas would generate a conceptual design and

estimate capital and operating costs.

Subtask 1 - Provide Information

Petro Star, GEDA, and Degussa would provide to GTC, information to serve as starting

point for the conceptual design. This information would include the data and modeling

results from Koch and process engineering data from prior development. Degussa would

provide information on the safe handling of hydrogen peroxide and peracetic acid

solutions.

Petro Star Inc Page 10 Conversion Extraction Desulfurization Phase III 3/11/2005 Subtask 2 - Conceptual Design

GTC would prepare a conceptual process design which includes capital and operating

cost estimates. Personnel from GEDA would maintain routine contact with GTC to

monitor progress and provide necessary information.

Subtask 3 - Design Review

GEDA would review the completed conceptual design and cost estimate. Cost

information would be used to revise and update the process economics and business

model.

3.6 Task 6 - Further CED Research

Petro Star identified opportunities for extending and improving CED technologies. Additional

opportunities and problems would emerge from the laboratory-scale and PDU testing. The goal

of this task would address these opportunities and problems.

Subtask 1 - Characterizations of Other Crude Oils

Petro Star, supported by GEDA, would test atmospheric gas oils from typical crude oils

other than from Alaska North Slope (ANS) such as East Texas and California crude.

These tests would be performed using the continuous PDU.

Subtask 2 - Improved Separations

It is likely that separations can be improved using variations on the aqueous acetic acid

extraction solvent system previously developed. Such variations may involve the addition

of lesser amounts of various polar solvents; examples of such solvents are methanol and


isobutyric acid. There is also the potential of substituting continuous on-stream liquid

chromatography for one or both of the solvent extraction steps. Such possibilities would

be explored as needed, using PDU measurements, to improve economics or to solve

problems identified in the work by KET.

Subtask 3 - Improved Oxidations

Hydrogen peroxide consumption is the most important cost element in the CED Process.

Previous studies have shown the potential for dramatically reducing peroxide

consumption through control of peroxide oxidation selectivity to make sulfoxides.

Working with Degussa and GEDA, Petro Star would extend these studies to minimize

peroxide costs.

Subtask 4 - Process Engineering

GEDA would apply the methods of chemical process engineering to explore

opportunities and solve problems that arose in the course of laboratory-scale and PDU

testing, engine testing, and plant design.

Subtask 5 - Extract Products

Experiments with solid adsorbents showed the potential for isolating the sulfur-rich

molecules from the aromatics using relatively simple physical processes. Such a

separation has the potential for generating a valuable aromatic stream for fuel, solvent, or

petrochemical applications. Degussa and GEDA would perform studies to demonstrate

and enhance this separation and characterize the aromatic stream.


Studies of the composition and behavior of the sulfur-rich component revealed that the

molecules can be cracked thermally and with high-temperature caustic to yield a variety

of products, depending on conditions. A survey of potential applications for the untreated

and cracked products has revealed a variety of opportunities based mostly on their

potential as surface active agents. An example is as anti-wear agents for lubricating oils.

Degussa and GEDA would extend this chemistry in cooperation with potential user

companies as a way to make valuable use of these commercially unusual materials.

3.7 Task 7 - Business Plan

In prior development Travis/Peterson Environmental Consulting, Inc. (TPECI) of Anchorage,

AK, performed some preliminary market research. The goal of this task would be to revise and

update this research with information arising from the test and design program.

Subtask 1 - Update Market Information

GEDA would revise and improve the market assessment previously developed. This

assessment would include compiling information detailing sizes and capabilities of

refineries around the world.

Subtask 2 - Extend Contacts with Potential Users and Partners

A low-key and effective prior effort led to contacts with a number of refiners and process

companies with potential interest in the CED process. The effort emphasized personal

contacts and delivery of appropriate technical papers at professional meetings. This effort


would be continued with the ultimate goal of connecting with refiners who would license

and use CED in their efforts to provide low-sulfur diesel fuels.

Subtask 3 - Business Plan Update

TPECI would collect information from the various engineering and marketing subtasks of

this program for use in updating and extending the earlier business plan for carrying the

CED process to commercialization. The plan would include market size and type, process

economics, and would provide recommendations to the Petro Star management on its

future course of action.

Figure 1 is the original Gantt chart, outlining the originally proposed work schedule.

4.0 PROJECT PLAN REVISIONS

Between the time that the DOE proposal was submitted and the project got underway, Petro Star

initiated several changes to the project plan. The primary reason for the changes was that early

pilot-scale experiments showed that the original CED process needed to be modified

significantly because the pilot-scale process did not constantly remove sulfur compounds to

project goals. This would require new process engineering and additional pilot-scale testing. At

the same time Petro Star decided to modify the CED team and partner with the Degussa

Corporation. Degussa brought considerable technical expertise to the project and could perform

most of the tasks originally assigned to subcontractors and consultants. Degussa analyzed the

weaknesses of the CED 2000 process that was causing the problems in the pilot-scale

Petro Star Inc Page 14 Conversion Extraction Desulfurization Phase III 3/11/2005 experiments. Through considerable new process engineering, Degussa developed the improved

CED 2002 process.

Figure 2 contains a revised Gantt chart that displayed the new schedule. Specifically, the

revisions to the project plan included the following:

Task 1: Pilot Scale Process Validation Tests. These tests would be completed as originally

proposed, but Degussa and Petro Star performed most of the work originally assigned to GEDA

and other contractors. This work would include providing information to Koch Modular Process

Systems KMPS (KET changed their name to KMPS), supervising tests, reviewing data and

incorporating it into models.

Task 2: Bench Scale PDU. This task was replaced by a new Task 2 called Continued Pilot Scale

Process Validation Tests. This change of plans became necessary because the original CED

process had to be substantially modified and improved to make it work effectively in the Pilot

Scale.

Task 3: FEV Engine Testing. Petro Star deleted this work from this scope of work in the DOE

grant. Petro Star reallocated these funds to Task 1, Pilot Scale Process Validation Tests.

Task 4: Pilot Plant Conceptual Design for 50 BPSD. Degussa would complete the conceptual

engineering design and cost estimate for the 50-bpsd pilot plant.


Task 5: 5,000 BPSD plant design and cost estimate. Degussa would complete the design and

cost estimate for a 5,000 BPSD for a plant that continuously manufactures low-sulfur fuel.

Task 6: Develop and optimize CED process:

• Subtask 1: Test process on other US and Canadian crude. Degussa would complete this

work using their systematic feed characterization method for generating simulated feeds

for the CED 2000 process models.

• Subtask 2: Improved Separations. Degussa would investigate techniques to improve

separations.

• Subtask 3: Improved oxidations. Both Petro Star and Degussa would complete this work.

• Subtask 4: Process Engineering. Degussa would analyze the system engineering to

improve process efficiency.

• Subtask 5: Extract Products. Degussa would study the extract products that result from

the desulfurization process and investigate techniques to minimize fuel loss, increase

solvent recycling, and develop useful extract by-products.

Task 7: Business Plan. The Kline Company would conduct market research on the licensing of

the process. Degussa and Petro Star would complete the business plan.

Figure 1

C E D P h a s e III P ro g ra m P la nM o d ifie d 6 A p ril 0 1

M in d ic a te s a p ro je c t m ile s to n eQ 1 Q 2 Q 3 Q 4 Q 5 Q 6

J u n e 0 1 S e p t 0 1 D e c 0 1 J a n 0 2 M a r 0 2 J u n 0 2 S e p t 0 2T a s k , S u b ta s k , a n d T a s k D e s c rip tio n 1 2 3 4 5 6 7 8 9 1 0 1 1 1 2 1 3 1 4 1 5 1 6 1 7 1 83 .2 .1 P ilo t-s c a le p ro c e s s va lid a tio n te s ts

.1 P ro v id e in fo rm a tio n

.2 S u p e rv is e te s ts

.3 P ilo t S c a le T e s tin g a t K o c h E x tra c tio n T e c h n o lo g ie s M

.4 R e v ie w d a ta a n d in c o rp o ra te in to m o d e ls

3 .2 .2 B e n c h -s c a le P D U.1 D e s ig n a n d im p le m e n t u p g ra d e s o f c u rre n t u n it.2 E q u ip m e n t fo r u p g ra d e s.3 C h e m ic a l a n d p h ys ic a l te s tin g.4 S a fe o p e ra tio n s tu d ie s.5 O p e ra tio n a n d d a ta e va lu a tio n

3 .2 .3 D ie s e l e n g in e te s tin g o f p ro c e s s e d fu e l.1 P re p a re te s t s a m p le s.2 S u p e rv is e te s ts.3 E n g in e te s tin g a t F E V M.4 R e v ie w d a ta a n d s o lve p ro b le m s th a t a r is e fro m u n e x p e c te d re s u lts

3 .2 .4 P ilo t p la n t d e s ig n fo r 5 0 B P S D.1 P re p a re d e s ig n s a n d e s tim a te c o s ts M

3 .2 .5 5 0 0 0 B P S D p la n t d e s ig n a n d c o s t e s tim a te.1 P ro c e s s e n g in e e rin g d e s ig n.2 C o n c e p tu a l e n g in e e rin g d e s ig n.3 D e s ig n R e v ie w.4 P re p a re 5 0 0 0 B P S D p la n t d e s ig n a n d c o s t e s tim a te s M.5 R e p o rtin g a n d e va lu a tio n M M

3 .2 .6 D e ve lo p a n d o p tim ize C E D P ro c e s s.1 O th e r U S a n d C a n a d ia n c ru d e.2 Im p ro ve d s e p a ra tio n s.3 Im p ro ve d o x id a tio n s.4 P ro c e s s e n g in e e rin g.5 E x tra c t p ro d u c ts M

3 .2 .7 B u s in e s s P la n.1 U p d a te m a r k e t in f o r m a t io n M.2 E x te n d c o n ta c ts w ith p o te n t ia l u s e r s a n d p a r tn e r s.3 U p d a te B u s in e s s P la n M

R e p o r ts M

G E D A 4 /6 /2 0 0 1

Figure 2

C E D P h a s e I I I P r o g r a m P l a nR e v i s e d M a r c h 2 0 0 2

M in d ic a t e s a p r o je c t m i le s t o n eQ 1 Q 2 Q 3 Q 4 Q 5 Q 6

J u n e 0 1 S e p t 0 1 D e c 0 1 J a n 0 2 M a r 0 2 J u n 0 2 S e p t 0 2 D e c 0 2 M a r 0 3T a s k , S u b t a s k , a n d T a s k D e s c r i p t i o n 1 2 3 4 5 6 7 8 9 1 0 1 1 1 2 1 3 1 4 1 5 1 6 1 7 1 8 1 9 2 0 2 1 2 2 2 3 2 43 . 2 . 1 P i lo t - s c a le p r o c e s s v a l id a t io n t e s t s

. 1 P r o v id e in f o r m a t io n

. 2 S u p e r v is e t e s t s

. 3P i lo t S c a le T e s t in g a t K o c h E x t r a c t io n T e c h n o lo g ie s M

. 4 R e v ie w d a t a a n d in c o r p o r a t e in t o m o d e ls

3 . 2 . 2 C o n t in u e d P i lo t S c a le P r o c e s s V a l id a t io n T e s t s

. 1T e s t s 1 & 2 - D e v e lo p R F P / R e c e iv e B id f r o m K M P S

. 2 T e s t s 1 & 2 - P i lo t S c a le T e s t in g a t K M P S

. 3T e s t s 1 & 2 - R e v ie w d a t a a n d I n c o r p o r a t e in t o M o d e ls M

. 4 T e s t s 3 & 4 - D e v e lo p R F P / R e c e iv e B id s

. 5 T e s t s 3 & 4 - P i lo t S c a le T e s t in g

. 6T e s t s 3 & 4 - R e v ie w D a t a a n d I n c o r p o r a t e in t o M o d e ls M

3 . 2 . 3 D ie s e l e n g in e t e s t in g o f p r o c e s s e d f u e l ( D e f e r r e d ). 1 P r e p a r e t e s t s a m p le s. 2 S u p e r v is e t e s t s. 3 E n g in e t e s t in g a t F E V. 4 R e v ie w d a t a a n d s o lv e p r o b le m s t h a t a r is e

f r o m u n e x p e c t e d r e s u l t s

3 . 2 . 4 P i lo t p la n t C o n c e p t u a l D e s ig n f o r 5 0 B P S D

. 1P r e p a r e C o n c e p t u a l D e s ig n a n d E s t im a t e C o s t s M

3 . 2 . 5 5 0 0 0 B P S D p la n t d e s ig n a n d c o s t e s t im a t e. 1 P r o c e s s e n g in e e r in g d e s ig n. 2 C o n c e p t u a l e n g in e e r in g d e s ig n. 3 D e s ig n R e v ie w

. 4P r e p a r e 5 0 0 0 B P S D p la n t d e s ig n a n d c o s t e s t im a t e s M M

. 5 R e p o r t in g a n d e v a lu a t io n M M M M3 .2 .6 D e v e lo p a n d o p t im iz e C E D P r o c e s s

.1 O th e r U S a n d C a n a d ia n c r u d e

.2 Im p r o v e d s e p a r a t io n s

.3 Im p r o v e d o x id a t io n s

.4 P r o c e s s e n g in e e r in g

.5 E x t r a c t p r o d u c ts M

3 .2 .7 B u s in e s s P la n.1 U p d a te m a r k e t in f o r m a t io n M

. 2E x te n d c o n ta c ts w ith p o te n t ia l u s e r s a n d p a r tn e r s

.3 U p d a te B u s in e s s P la n M M

R e p o r ts M

Petro Star Inc Page 18 Conversion Extraction Desulphurization Phase III 3/11/2005 Some of these tasks were not completed because the project team reached a conclusion that the

CED process was not technically and economically feasible. The tasks that were not completed

were:

• Task 3 – Diesel Engineer Testing of Process Fuel.

• Task 4 – Pilot Plant Conceptual Design for 50-BPSD.

• Task 6.1 – Test CED Process on other US and Canadian Crude Sources.

• Task 6.2 – Improve Separations.

• Task 6. 5- Extract Products Development.

• Task 7.3 – Update Business Plan.

These tasks were put on hold until the three critical-path tasks were completed. The critical-path

tasks were:

1. Achieving 15ppm or less total sulfur content in the continuous-flow, pilot-scale

apparatus.

2. Economic analysis of the CED process at full commercial scale (5,000-BPSD).

3. Market analysis of licensing the CED process to other refiners.

5.0 EXPERIMENTAL

Petro Star conducted the process development work in accordance with the work plan for Task 1,

Pilot-Scale Process Validation Tests, and Task-6, Develop and Optimize CED Process.

Petro Star provided a straight-run Light Atmospheric Gas Oil (LAGO) for the process feedstock.

Petro Star distilled the LAGO from ANS crude oil. The LAGO exhibited a nominal boiling

Petro Star Inc Page 19 Conversion Extraction Desulphurization Phase III 3/11/2005 range of 175° C – 345° C (347° – 653° F) and contained about 0.4-w% sulfur (Please see

Appendix A for the detailed specifications). This LAGO is a typical fuel that Petro Star

produces. Degussa supplied 70-w% hydrogen peroxide for all experimental activities. Degussa

also provided expertise for all hydrogen peroxide safety issues, provided onsite training of

personnel on safe handling procedures, and reviewed the pilot plant set up for safe operations.

5.1 Task-1, Pilot-Scale Process Validation Tests

The Task 1 objectives were to test the individual process steps with continuous-flow equipment,

evaluate the proposed preliminary process design, and identify any process design problems and

operating deficiencies. A secondary objective consisted of collecting additional data to assist in

optimizing the process and refining the process models.

Koch Modular Process Systems, LLC (KMPS) completed the pilot-scale continuous-flow

process work at their Houston, Texas laboratory. Appendix B contains the pilot-scale plant

configuration and the detailed KMPS report on the testing that was done.

The research team selected the continuous-flow equipment for each individual process. KMPS

had several Scheibel-, Karr- and packed-columns of different sizes. A variety of packing types

were available at their laboratory for the packed-column. In some cases, KMPS used a smaller

Scheibel column to establish process viability before scaling up to larger packed columns.

KMPS collected product samples for each test run and analyzed them or submitted them to Petro

Star and Degussa for analysis.

Petro Star Inc Page 20 Conversion Extraction Desulphurization Phase III 3/11/2005 KMPS tested the following process steps with the indicated equipment capacity and sizes:

Table 1 – KMPS CED 2000 Process Step Testing

CED 2000-Process Step Test Equipment Type & Size Capacity Description Size, ID cm, L, cm Feed (l/h)

1 - Pre-extraction of thiophenes Packed Column, 7.6 x 610 27 – 50

2 - Oxidation of thiophenes Karr Column, 7.6 x 610 54 – 81

3 – Sulfone extraction

Scheibel Column, 7.6 x (36 stages) Packed Column, 7.6 x 610

8 – 13 30 – 60

4 - Raffinate washing Scheibel Column, 7.6 x (36 stages) Packed Column, 7.6 x 610

22 – 23 86 – 93

5 – Adsorption. Not tested. Development and demonstration to be conducted on smaller bench-scale equipment.

6 – Solvent recovery distillation Packed Column, 7.6 cm x 488 cm 39

7 – Extract washing Karr Column, 2.5 cm x 366 cm

plate stack

7 - 12

The project team anticipated several potential operating problems. Hydraulic problems included

the formation of emulsions, column flooding, impediments to phase separation, insufficient

density gradients, and dispersed phase droplet size limitations. Thermodynamic problems

included the formation of homogeneous azeotropes between hydrocarbon and solvent, and

heterogeneous azeotropes between hydrocarbon and water. None of these concerns developed

into serious process impediments, although some relatively minor process adjustments were

required to avoid areas of process instability.

Petro Star Inc Page 21 Conversion Extraction Desulphurization Phase III 3/11/2005 Based on the initial hydraulic stability of the individual process steps in the continuous-flow

equipment, Petro Star and Degussa decided to gather operating data for process optimization.

KMPS therefore expanded each test to include variations in operating parameters such as

temperature, flow rate, solvent/feed ratios, solvent composition, and counter-flow dispersed and

continuous phases. Steady-state operation was obtained in all cases. KMPS averaged three to

five different tests at steady-state within a single day for each process step.

5.2 Task-6, Develop and Optimize CED Process

The research team designed the oxidation reaction process as a co-current, two-phased, plug-

flow reactor with the gas oil feed, oxidant, and catalyst mixed together at the inlet to the reactor.

KMPS used a Karr Column (7.6 cm ID x 610 cm length) as the reactor and preheated the feed to

the desired reaction temperature utilizing an external heater.

During the continuous-flow tests of the oxidation reaction (Process Step 2), KMPS observed

93% to 97% conversion of thiophenic compounds. Previous work suggested the oxidation

process must convert at least 99.5% of the thiophenes for the solvent to extract the sulfur

compounds from the diesel fuel to a sulfur level of 15-ppm. Petro Star and Degussa therefore

implemented laboratory work under Task-6, Subtask-3, “Improved Oxidations” to determine the

cause for the deficiency. This work was conducted within lab-scale batch equipment and

covered the investigation of several important operating variables. Appendix B contains a

summary of all variables investigated.

Petro Star Inc Page 22 Conversion Extraction Desulphurization Phase III 3/11/2005 6.0 RESULTS AND DISCUSSION

The results of the continuous-flow testing completed during this period will be discussed in two

parts. The first includes the suitability of the equipment used, the operability of the process,

concerns for column hydraulics and phase separations, and related process concerns. The second

part includes an evaluation of the product purity, processing efficiencies, and material balances

based on test sample analytical results.

6.1 Extractions

KMPS tested four extraction process steps during countercurrent continuous-flow conditions in

Scheibel, Karr, and packed columns. KMPS used a Scheibel column (7.6 cm ID x 36 stages) and

a Karr column (2.5 cm x 366 cm plate stack) to assess the initial process efficiency. Phase-

mixing in these columns was more controllable than with packed columns.

The 610-cm (240-in) long packed column contained sections of 13-cm (5.1-in) long, 316-SS

Flexipac-2Y-1 structured packing. KMPS used this column to demonstrate process performance

within low-cost industrial-type equipment. Tests demonstrated the counter-current packed

column configuration was hydraulically adequate for the pre-extraction and sulfox extraction

columns. It was clear, however, the raffinate wash column and extract wash column will require

a configuration other than the counter-current packed column arrangement. Mechanically

agitated columns proved to be hydraulically adequate for the wash columns.

Petro Star Inc Page 23 Conversion Extraction Desulphurization Phase III 3/11/2005 KMPS first observed whether the extractions created any physical operating problems such as

column flooding, formation of emulsions, difficult phase separations, insufficient density

gradients, and dispersed phase droplet size limitations.

Testing the four extraction process steps within various column types generated important data

with respect to hydraulic behavior and column operation. Phase separations occurred without the

formation of interface-layers or emulsions. The droplet velocities of the dispersed phases in the

continuous-flow processes were satisfactory for the intended industrial operation. The surface

tension and viscosity of the phases allowed visually acceptable dispersion in the pre-extraction

column and sulfox extraction column. The raffinate washing operation, however, exhibited

unacceptably large droplet sizes when operated in the packed column and was abandoned for a

mechanically stirred column. By alternating which phase was continuous, KMPS discovered

that this choice has only a small effect on the hydraulic operability.

KMPS found the column loadings at steady-state were in the range of 2.2 to 3.5 l/h-cm2 (540 –

859 gph/sq ft). These results allowed the calculation of distillation column diameters for a

5,000-BPSD commercial plant. Appendix D describes the column dimensions.

Petro Star Inc Page 24 Conversion Extraction Desulphurization Phase III 3/11/2005 The KMPS extraction tests (See Appendix B for the complete results) accomplished the

following:

1) Operating data was collected for determining the minimum solvent/feed ratios.

2) Data was collected to validate and fine-tune the process design models.

3) The three gas oil extractions were operated in accordance with the preliminary

design operating conditions.

4) The sulfone extraction appeared insufficient with respect to sulfur removal

efficiency. It was unclear whether this was due to partial oxidation or mass

transfer limitations in the extraction column.

5) The thiophene pre-extraction was tested in both solvent and recycle-solvent mode

with good results in both cases.

6.2 Task 6, Subtask 2: Improved Separations

The improved process engineering models developed by Degussa provide reasonable

representations of each distillation step in the CED 2000 process. The extract distillation to

recover solvent for recycle required continuous-flow testing because of the presence of

azeotropic compounds, the non-ideality of the solvent system, and the need to validate the

process engineering model due to the complexity of the feed.

Petro Star Inc Page 25 Conversion Extraction Desulphurization Phase III 3/11/2005 KMPS conducted continuous-flow testing of the extract distillation, which proceeded without

significant problems. These tests accomplished the following:

1) Although the original process engineering model did not consider the formation

of azeotropes, the azeotropic behavior as predicted by Degussa was well within

the acceptable limits for obtaining a viable recycled solvent in accordance with

the preliminary process design. However, the presence of azeotropes does have a

significant impact on the viability of the solvent purification distillation.

2) The recovery of solvent in this process step can be increased without penalty from

the assumed limit of 15-w% by reducing the residual solvent in the bottoms

product to less than 10-w%.

3) The column experienced hydraulic problems only when operated at high solvent

removals (approximately 10w% solvent remaining in bottoms product). At this

point, significant foaming began in the sump of the pilot-scale column. It is

unclear whether this was due to the chemical composition of the system or due to

the method of heat input to the column.

6.3 Task 6, Subtask 3: IMPROVED OXIDATIONS

The maximum conversion of thiophenics to oxidized species was 97%. Since a conversion of at

least 99.5% is required, the technical feasibility of this step could not be confirmed.

Petro Star Inc Page 26 Conversion Extraction Desulphurization Phase III 3/11/2005 Petro Star and Degussa investigated the causes of the thiophene conversion deficiency and

determined that two possible classes of side-reactions depleted the oxidant before the desired

thiophenic-compound conversion was achieved. These reactions included the oxidant reacting

with the non-sulfur containing compounds in the hydrocarbon feed and oxidant decomposition.

This oxidant depletion is, as yet, an unsubstantiated contributor to low thiophene conversion. It

is highly likely the slight excess of oxidant originally specified in CED 2000 would be

insufficient even if side reactions and decomposition were not present.

6.4 Summary of 2001 Work

KMPS performed four extraction process steps and one critical distillation without hydraulic

problems and with only minor adjustments in temperatures and solvent ratios. Preliminary

estimates by KMPS indicated a packed column height of 16.5 m (54 ft) for the sulfox extraction

packed column and 12.2 m (40 ft) for the raffinate wash Scheibel column. In addition, the

thiophene pre-extraction column and the solvent recovery distillation column were successfully

demonstrated. Consequently, the research team could now optimize these process steps when

the sample results were finalized.

A process step that presented difficulty was the oxidation reaction, which was short of the

desired 99.5% conversion of thiophenic compounds. Petro Star and Degussa determined side-

reactions depleted the reagent before the thiophene reaction was finished. Depletion of oxidant

and catalyst due to decomposition and side reactions may not be the only cause of insufficient

conversion. The small excess of oxidant would probably be insufficient with the specified

residence time even if no side reactions or decomposition were present.

Petro Star Inc Page 27 Conversion Extraction Desulphurization Phase III 3/11/2005 The two critical findings of the 2001 work were:

1. the need to develop an improved oxidation process that can produce diesel in continuous

flow at sulfur levels less than 15 ppm, and

2. the economic analysis showed that the extract value and quantity has a much larger

impact on commercial feasibility than the consumption of oxidant. The investigation of

alternate process schemes began. These investigations focused on minimizing the

quantity of extract.

6.5 2002 Work

The combined work of the Petro Star and Degussa labs produced a process in which the

oxidation reaction utilizes excess oxidant. The most promising results were obtained when a dual

oxidant was utilized and the 2nd stage oxidation was conducted at lower temperatures. This

improves the selectivity of the oxidant and therefore allows recycle of unreacted oxidant to the

1st stage oxidizer. From this experimentation, a new CED process scheme was developed that is

based on maximizing fuel yield rather than minimizing oxidant consumption. It utilizes two-

stage oxidation with excess oxidant. This new process scheme is called CED 2002. In order to

obtain an accurate estimate of the conversion cost for the CED 2002 process, the team completed

conceptual process engineering for the 5,000-BPSD plant (See Appendix D). The conceptual

process engineering included:

• Process flow diagrams;

• Material & heat balances;

• Equipment sizes;

• Layouts;

Petro Star Inc Page 28 Conversion Extraction Desulphurization Phase III 3/11/2005

• Factored capital cost estimates;

• Process economics with a sensitivity analysis;

• Open issues list; and

• Study list for future optimization work.

It was decided that the conceptual phase of the project would continue until the new process

could be demonstrated in continuous flow pilot-scale trials. The team prepared a bid request for

continuous flow testing of the CED 2002 process, and issued it to KMPS as the basis for a

quotation.

In February 2002, Petro Star and Degussa formed a Limited Liability Corporation (LLC) called

Arcticlear LLC. The purpose of the LLC was to develop and market the CED process. They

developed a program for consistency testing and optimization of the process. Three laboratories,

one in Allendale, NJ, one in Anchorage, AK, and one in Wolfgang, Germany began conducting

experiments using the same detailed testing protocol. All three laboratories were obtaining end

point sulfur measurements in the same range, between 15 ppm and 20 ppm. This was

considerably better than the scattered results obtained before the detailed protocol was

implemented.

By adjusting key reaction parameters, the laboratories were able to generate product fuel with

sulfur levels lower than the required 15 ppm. However, to generate these sulfur levels, 40

minutes of residence time was needed in the 2nd stage reactor. Unfortunately, this extended

residence time results in a very expensive 2nd stage reactor. By analyzing the experimental data,

Petro Star Inc Page 29 Conversion Extraction Desulphurization Phase III 3/11/2005 it was determined that the 2nd stage reaction is not mass transfer controlled, but rather kinetically

controlled. Based on this realization, a new concept for the 2nd stage reactor was developed. This

new concept allows the 2nd stage oxidizer to operate under homogenous conditions. The new

concept was tested in the laboratory and the results were as predicted. As a result, the reactor

shape is no longer determined by minimal velocity criteria. Thus, extended residence times can

be implemented without significant increases in capital cost.

To evaluate the competitiveness of the CED technology against conventional and non-

conventional desulfurization technologies, the Degussa and Petro Star team began evaluating the

strengths and weaknesses of competitors. In an effort to obtain information regarding two

prominent competitors, UniPure and Phillips S-Zorb, a representative from Degussa attended the

non-conventional desulfurization technologies session of the American Institute of Chemical

Engineers 2002 Spring National Meeting in New Orleans, Louisiana. A great deal of

information on the competitors, their technologies, and their target markets was obtained. In

addition, a recently published patent of UniPure was reviewed. This patent in conjunction with

the conference information greatly increased the understanding Degussa has regarding the

primary competitor in the oxidative desulfurization arena.

The consistency testing in the laboratory showed that reactions on the laboratory scale

consistently yielded LAGO with a sulfur content less than 15 ppm. The optimization of reactor

conditions was ongoing. Since the primary objectives for the oxidation system were achieved,

secondary goals were being investigated.

Petro Star Inc Page 30 Conversion Extraction Desulphurization Phase III 3/11/2005 Continuous Flow Testing of the Oxidation System and Sulfox Extraction System

Clarification of the quotation from KMPS on the continuous flow testing of the Two-Stage

Oxidation System and the Sulfox Extraction System was completed. These clarifications

included the incorporation of the new reactor concept that allows for extended residence times in

the 2nd Stage Oxidizer without the need for excessive capital investment. The purchase orders

necessary to support this testing program were issued.

As planned, KMPS setup the continuous flow equipment for testing the Oxidation System.

Hydrotesting of the process side of all equipment items was completed on the last day of the first

week. However, KMPS failed to hydro-test the jacket side of the 2nd Stage Oxidizer.

While hydrotesting the jackets, it was discovered that two of the segments were defective and

had to be replaced. Several other piping problems were discovered and correction had to be

made. In addition, several cracks in the glass shell of 2nd Stage Oxidizer were noted and

additional repairs were necessary. As a result of these system problems, the oxidation tests could

not commence as scheduled.

Process fluids were introduced into the oxidation equipment. Due to operational difficulties, the

system was shutdown immediately. As a result of this incident, the second day of testing yielded

no results.

Process fluids were again introduced into the oxidation equipment. Shortly after fluid feeding

started, one of the many pumps used in the process failed. The system was stopped and the

Petro Star Inc Page 31 Conversion Extraction Desulphurization Phase III 3/11/2005 pump was replaced. Feed was started again by mid afternoon. Sulfur results at the end of the 2nd

Stage Oxidizer proved to be poor. A brainstorming session concluded that manual level control

of the liquid-liquid interfaces in the two decanters was causing drastic swings in system

flowrates. This high variability in mass flow caused unsteady state conditions to dominate the

system.

The changes in the control philosophy suggested by the CED Process Development Team

resulted in smooth operation of the system in the recycle mode. Unfortunately, the level of

unoxidized sulfur in the product was substantially above the desired concentration of 15 ppm.

The product contained approximately 20 to 27 ppm of unoxidized sulfur. Based on the results of

a simultaneous laboratory experiment on the same fuel, it was concluded that the old fuel being

used for the test might be the reason for the unsuccessful test.

To ensure stable operation and to generate sufficient material for the testing of the Sulfox

Extraction System, the team decided to operate the system in open loop mode. Namely, the

recycle oxidant was discontinued and the fresh oxidant was feed to both the 1st Stage Oxidizer

and the 2nd Stage Oxidizer simultaneously. In addition, the old fuel was replaced by fresh

material recently obtained from the Petro Star refinery in Valdez. The final level of unoxidized

sulfur in the system effluent had dropped to approximately 17 to 18 ppm.

Based on these positive results and the desire to generate enough material for the testing of the

Sulfox Extraction System, it was decided to extend the testing of the Oxidation System to the

first two days of the following week. The system was operated in open loop mode. Higher than

Petro Star Inc Page 32 Conversion Extraction Desulphurization Phase III 3/11/2005 expected oxidant flows were necessary to reduced the level of unoxidized sulfur to the 12 to 15

ppm range. However, sufficient material was generated to satisfy the requirements for the

testing of the Sulfox Extraction System.

The extraction equipment was setup and the Sulfox Extraction System was tested. Results were

better than expected. A total of nine steady state operating points were obtained. Three distinct

solvent-to-feed ratios and three solvent compositions were tested. At the design solvent-to-feed

ratio and solvent composition, the total sulfur in the raffinate was reduced to 23 ppm. This

represented a drastic improvement over the results obtained during the testing conducted in the

summer of 2001. It also confirmed the team’s suspicion that the testing of the oxidation system

conducted in the summer of 2001 only resulted in unoxidized sulfur levels of approximately 150

ppm.

Degussa prepared a summary outlining raw material consumption and waste generation during

all previous testing at KMPS. Travis/Peterson Environmental Consultants, Inc. utilized this data

to prepare the necessary hazardous waste plan required by the Department of Energy contract

held by Petro Star Inc.

The technical reports issued by KMPS on the continuous flow testing of the Oxidation System

and Sulfox Extraction System were reviewed. Minor errors were noted. However, to avoid

additional charges from KMPS, it was decided that Degussa personnel would complete the final

reports.

Petro Star Inc Page 33 Conversion Extraction Desulphurization Phase III 3/11/2005 Degussa personnel worked with KMPS to coordinate the cleanup phase of the pilot plant work

that was conducted in May. This included coordination of sample shipments to both Degussa and

Petro Star research laboratories. In addition, arrangements were made to have hazardous waste

transported to a disposal facility and disposed of by incineration. Both the transportation and

disposal were conducted by Onyx Environmental Services.

The Operating Committee approved the continuation of the continuous flow testing of the

Oxidation System. The CED Process Development Team prepared for the 2nd phase of testing

by issuing the necessary technical and commercial documents. In addition, a schedule was

issued and the shipment of equipment, supplies, and chemicals was coordinated.

The Aspen steady state flowsheet simulation of the new oxidation scheme was developed. This

simulation model was used to study the effect of the first stage decanter heavy phase purge on

the size of the Peracetic Acid Reactor.

Continuation of the Continuous Flow Testing of the Oxidation System

The CED Process Development Team conducted testing of the two-stage oxidation system at the

pilot plant facilities of KMPS in Houston, Texas. This testing was a continuation of oxidation

system testing. The goal of the testing was to demonstrate the technical feasibility of the process

(product containing less than 15-ppm sulfur on a gas oil basis) while simultaneously satisfying

economic feasibility constraints (conversion cost of less than $2.50 dollars per barrel). The six-

day test consisted of three 16-hour operating days and three 9-hour analytical days.

Petro Star Inc Page 34 Conversion Extraction Desulphurization Phase III 3/11/2005 The team decided that operation in a closed loop (i.e., recycle mode) was critical to satisfying

both the technical and economic requirements. Operation in open loop mode would not be

satisfactory for demonstrating that a recycle process was technically feasible. In addition,

utilization of fresh oxidant in both the 2nd Stage Contactor and 1st Stage Oxidizer leads to very

high oxidant requirements and unfavorable economics.

Then several operational problems occurred. These problems included pump failures and

leaking rotameters. The process was fed gas oil containing 3,850 ppm of sulfur. Approximately

4.5 molar equivalents of oxidant were fed to the 2nd Stage Contactor. The oxidant to the 1st Stage

Oxidizer was supplied by recycling heavy phase effluent from the decanter downstream of the

2nd Stage Contactor. The recycle oxidant quantity in terms of molar equivalents was very steady

for the last eight hours of operation. On average, the system achieved unoxidized sulfur levels of

20 ppm. The lowest unoxidized sulfur level achieved was 17 ppm.

Due to the unacceptably high sulfur content in the product, the Team decided to raise the oxidant

level to 5.0 molar equivalents. The lowest unoxidized sulfur value achieved was 15 ppm.

However, results ranging from 20 to 24 ppm were more typical.

Throughout the testing, the losses of active oxygen in the 2nd Stage Contactor were unacceptably

high. At times, these losses accounted for nearly 40 percent of the oxidant consumption. The

team conducted several tests in the 2nd Stage Contactor to determine the effects of decomposition

without gas oil present and the effect of mass transfer on the active oxygen losses. The team

found that peracetic acid held in the Scheibel column for sixty minutes loses only a small amount

Petro Star Inc Page 35 Conversion Extraction Desulphurization Phase III 3/11/2005 of active oxygen. Since the steady-state residence time is one to two minutes, this is an unlikely

source of losses. Next, the team added previously oxidized gas oil and peracetic acid solution to

the column. The agitator rotational speed was lowered as far as possible while maintaining

intimate mixing. It was shown that the rate of active oxygen decomposition was significantly

lower than that observed in steady-state operation. Based on this finding, the team determined

that the higher interfacial areas present during steady-state operation might account for the

accelerated active oxygen losses.

The team encountered several problems while operating the continuous flow system. These

problems included failure of the oxidant pump, failure of the chiller system, and electrical

problems with the Scheibel column motors. The unit was started in open loop mode with 2.5

molar equivalents of oxidant fed to both the contactor and the first-stage reactor. Once steady

state was achieved, large samples were taken from the intermediate location of the pipe reactors

(residence time of approximately 35 minutes). This material was then contacted with an

additional 2.5 molar equivalents of oxidant in the lab. The heavy phase was removed and the

light phase was held for an additional 35 minutes. The unoxidized sulfur content was found to

be 12 ppm after this third reaction stage. It is possible that a multistage reactor (greater than two

stages) may be required to make the CED process technically feasible. After the samples had

been collected for the third stage test, the system was returned to closed loop mode with 5.0

molar equivalents of oxidant fed to the contactor. The lowest unoxidized sulfur content in the

effluent was again 15 ppm.

Petro Star Inc Page 36 Conversion Extraction Desulphurization Phase III 3/11/2005 Arcticlear Operating Committee Meeting

An Arcticlear Operating Committee Meeting was held during the week of July 22, 2002 in

Anchorage, Alaska. The primary purpose of the meeting was to discuss the latest results from

the continuous flow testing of the CED 2002 process and decide on a suitable path forward.

The overall conclusion of the Arcticlear Operating Committee was that the CED Process

Development Project was in a “no go” state based on the failure of the continuous flow testing

program to achieve its predefined objectives. The Kline marketing study would be conducted as

soon as possible. Laboratory R&D activities in Anchorage and Allendale would continue based

on the current task list. All other engineering activities were placed on hold, except that

necessary for completion of report writing and supporting the laboratory efforts.

Work on completing the technical reports issued by KMPS on the continuous flow testing of the

Oxidation System and Sulfox Extraction System began. These were the tests conducted in May

2002. Degussa supplemented the KMPS report by adding the analytical data and an analysis of

the process performance.

The Arcticlear Operating Committee approved the scope of work and proposal from Kline and

Company in August. Degussa prepared a draft questionnaire and presentation to be used for the

survey.

Waste Removal from KMPS

A considerable amount of waste material was generated during the continuous flow testing of the

oxidation system and extraction system in the KMPS pilot plant in Houston, Texas. The

Petro Star Inc Page 37 Conversion Extraction Desulphurization Phase III 3/11/2005 Arcticlear Operating Committee decided to send Steve Bonde, the Petro Star senior research

chemist, to Houston to oversee the waste removal. A total of 48 full or partially full containers,

holding approximately 11,000 pounds of material, were removed. The team adhered to the

project Hazardous Substance Plan. Waste types, amounts, and final disposal is documented in

the Hazardous Waste Report.

Kline Company began the marketing study. The Arcticlear Board of Managers and Operating

Committee approved a draft questionnaire for the survey. This plan was to contact 50 of the 143

US refineries in operation in the US today. The primary focus would be on the refineries in the

Petroleum Administration for Defense Districts (PADD) 4 (upper mid-west and mountain states).

The study would also include face-to-face meetings with a select subgroup of these refineries.

Degussa personnel would visit at least 3 refineries to obtain a first hand impression.

All waste generated during the pilot plant trials at KMPS in Houston was successfully removed.

Steve Bonde, the Petro Star senior research chemist, supervised waste removal.

A meeting was held in New Jersey to review progress on the Kline Marketing Study. Of the 143

possible refineries in the US, approximately 48 have been isolated for contact. These 48

refineries do not currently have hydro-treating capabilities and are owned by companies that

process less than 150,000 BPD of crude. Of the refineries contacted thus far, at least four small

refineries in PADD 3 and PADD 4 were interested in a face to face meeting with representatives

from Degussa. In late November, Degussa personnel contacted the interested refineries to

arrange meetings.

Petro Star Inc Page 38 Conversion Extraction Desulphurization Phase III 3/11/2005 During 2002, the CED Process Development Team showed that CED technology is capable of

reducing the sulfur content of light atmospheric gas oil from 5,000-ppm to less than 10-ppm on

the laboratory scale. In continuous flow trials using rented equipment, the sulfur reaction

capability of the process fell short of the 15-ppm goal by producing fuel with approximately 20-

ppm of sulfur. As a result, the technical development for the project was scaled back

considerably, while a marketing study was completed in December 2002.

In December , the Kline Company presented their findings in a meeting held in Parsippany, New

Jersey. Kline found that there are a total of 143 refineries in the US of which 93 have crude oil

capacity less than 150,000 barrels per day, and therefore qualify for various regulatory

exceptions to the mid 2006 compliance for producing ultra low sulfur diesel fuel. Kline

interviewed top management in 43 of the 93 refineries. The conclusion from these interviews

was that only 13 refineries harbored opportunity for the CED process.

Degussa personnel visited 4 of the 13 refineries in December 2002 over a four-day period. All

four refineries are in the PADD 3 (Gulf Coast States) & PADD 4 (Mountain States) regions of

the US. The four refineries are Montana Refining in Great Falls, Montana, Sinclair Oil

Corporation in Salt Lake City, Utah, , LaGloria Refining in Tyler, Texas and Navajo Refining

Company in Artesian, New Mexico. In all of these refineries, moderate pressure

hydrodesulfurization (HDS) is currently used to produce on-road diesel fuel with a sulfur content

less than 500-ppm. The management of all these refineries had an open mind toward alternate

desulfurization technologies, especially if the technology could produce ultra low sulfur fuel at a

lower operating cost. A considerable amount of valuable marketing information was obtained

Petro Star Inc Page 39 Conversion Extraction Desulphurization Phase III 3/11/2005 from each visit. Unfortunately, only one of the three refiners was interested in delaying

compliance past mid 2006. Therefore, based on the Kline Marketing Study results and on the

four refinery visits, the Arcticlear Operating Committee in conjunction with the Arcticlear Board

of Managers decided that further investment in the technical development of the CED Process

was not warranted.

7.0 CONCLUSIONS

The early pilot-scale work showed that the original the CED 2000 process would not achieve the

desired sulfur removal to 15 ppm. The project team therefore had to re-engineer the CED 2000

process. This resulted in the new and improved CED 2002 process. The CED 2002 process

needed to be tested in Koch’s continuous-flow, pilot-scale configuration. Because of this

additional work, the CED team decided to take a critical path approach to complete this project.

Therefore, the project team focused on three “critical path” activities to determine the merits of

the CED process. They were:

1. The pilot-scale CED process must reduce the diesel fuel sulfur content to 15 ppm or

lower. Test results indicated that the CED process could not consistently achieve sulfur

levels below 20 ppm.

2. Petro Star determined that the CED process cost could not exceed $2.50 per product

barrel if the process was to be economically feasible for Petro Star to employ the CED

process at its Valdez, Alaska refinery. The process economics study in the CED 2002

Summary Report, 5,100 BPSD Process Unit (see Appendix D) calculated an estimated

process cost of $5.70 per product barrel, which was too expensive for the Valdez

refinery.

Petro Star Inc Page 40 Conversion Extraction Desulphurization Phase III 3/11/2005

3. Petro Star/ Degussa could not assume the costs of commercializing the CED process

unless there were prospects for licensing the process to other refineries. The Kline study

(Appendix C) and the research team’s discussions with prospective refineries led to the

conclusion that there were not likely prospects for the licensing of the CED process.

Once these three conclusions were reached, the project team decided not to pursue the CED

project any further. Therefore, the CED Process Development Project was closed as of

December 31st, 2002. Thus, the following non-critical path tasks were not completed:

• Task 3 – Diesel Engineer Testing of Process Fuel.

• Task 4 – Pilot Plant Conceptual Design for 50-BPSD.

• Task 6.1 – Test CED Process on other US and Canadian Crude Sources.

• Task 6.2 – Improve Separations.

• Task 6. 5- Extract Products Development.

• Task 7.3 – Update Business Plan.

8.0 REFERENCES

Petro Star, 2000. Chemical and Biological Upgrading and Desulfurization of Petroleum Feedstocks. Final Report to the Alaska Science Technology Foundation. Report #ASTF 98-2-083. February, 2000. 5pp.

9.0 LIST OF ACRONYMS AND ABBREVIATIONS

ANS Alaska North Slope

BPSD Barrels per standard day

DOE US Department of Energy

CED Conversion Extraction Desulfurization

Petro Star Inc Page 41 Conversion Extraction Desulphurization Phase III 3/11/2005 CED 2000 Conversion Extraction Desulfurization process based upon original assumptions

at the beginning of this DOE project.

CED 2002 Modified and improved Conversion Extraction Desulfurization process after the

initial pilot-scale, continuous-flow experiments.

CPE Conceptual Process Engineering

cm Centimeters

FEV FEV Engine Technology

ft Feet

GEDA Geoffery E. Dolbear and Associates

gpd Gallons per day

GTC GTC Technology Corporation

HDS Hydrodesulfurization – the standard industry method for removing sulfur that uses

extremely high temperatures and pressures.

ID Inside Diameter

in Inches

KET Koch Extraction Technologies – original name for KMPS.

KMPS Koch Modular Process Systems, LLC, subcontractor to Petro Star Inc.

LAGO Light Atmospheric Gas Oil produced by Petro Star Valdez Refinery

l/h Liters per hour

l/h-cm2 Liters per hour per square centimeter

m Meters

O.D. Outside diameter

PADD Petroleum Administration for Defense Districts

Petro Star Inc Page 42 Conversion Extraction Desulphurization Phase III 3/11/2005 PDU Process Development Unit

ppm Parts-Per-Million

SI SI is an abbreviation for “Le Systeme International d’Unites.”

TPECI Travis/Peterson Environmental Consulting, Inc.

ULSD Ultra Low Sulfur Diesel

W% Percent weight

APPENDIX A

PETRO STAR, INC. LAGO SPECIFICATIONS

Page 44

APPENDIX B

KOCH PILOT-SCALE TEST #1

Page 46

Page 47

Page 48

Page 49

Page 50

Page 51

Page 52

Page 53

Page 54

Page 55

Page 56

Page 57

Page 58

Page 59

Page 60

Page 61

Page 62

Page 63

Page 64

Page 65

Page 66

Page 67

Page 68

Page 69

Page 70

Page 71

Page 72

Page 73

Page 74

Page 75

Page 76

Page 77

Page 78

Page 79

Page 80

Page 81

Page 82

Page 83

Page 84

Page 85

Page 86

Page 87

Page 88

Page 89

Page 90

Page 91

Page 92

Page 93

Page 94

Page 95

Page 96

Page 97

Page 98

Page 99

Page 100

Page 101

Page 102

Page 103

Page 104

Page 105

Page 106

Page 107

Page 108

Page 109

Page 110

Page 111

Page 112

Page 113

Page 114

Page 115

APPENDIX C

KLINE & COMPANY MARKET STUDY

Page 117

Page 118

Page 119

Page 120

Page 121

Page 122

Page 123

Page 124

Page 125

Page 126

Page 127

Page 128

MARKET OPPORTUNITY ASSESSMENT FOR CED PROCESS WITH SMALL REFINERS

A final report to

December 18, 2002

Page 129

1

PRESENTATION OVERVIEW

Refiner interview process

Universe of potential refiners for CED process

Review of refiner interviews by Petroleum Administration for Defense District (PADD)

Need for CED process to treat to 7 ppm

Findings, opportunities, obstacles

Page 130

2

OBJECTIVE

The primary objective of this assignment is to provide Degussa Corporation with a survey of small refiners to assess the marketopportunities for the CED process in diesel fuel desulfurization with small refiners

This engagement will also be used to generate potential leads for the Degussa/Petro Star team to pursue

Page 131

3

PETROLEUM ADMINISTRATION FOR DEFENSE DISTRICTS (PADD)

PADD District 4PADD District 2

PADD District 1-N

PADD District 1-SPADD District 3

PADD District 5-a

a- Includes Alaska and Hawaii.

Page 132

4

THE SPECIFIC ISSUES TO BE TARGETED WITH REFINERS WILL INCLUDE SUCH ITEMS AS:

How are they currently complying with the 500ppm standardAre credits from large refiners an integral piece of their compliance strategy?What might these credits cost?What process options are available to them?Do they want to be in the diesel market once the 15ppm standard is fully implemented?Are they willing to take a wait and see attitude and hold off investments until 2008?Will some of them be looking to enter the 15ppm diesel market early (to possibly take advantage of high prices during short supply)?Given an attractive process, would some of the refiners currently not producing on-highway diesel enter the market?With the small refiners without hydrotreating, what is the largest, smallest, and average capacity of on-highway diesel production?

Page 133

5

THE SPECIFIC ISSUES TO BE TARGETED WITH REFINERS WILL INCLUDE SUCH ITEMS AS:

What is the per barrel conversion cost for an HDS unit (including the supporting H2

plant, amine unit and sulfur plant) scaled down to the minimum, maximum and average capacity?What is the total conversion cost per barrel range necessary to compete in this market with a non-conventional technologySmall refiners’ internal analysis of their costs to produce 15ppm diesel (to the extent possible)Would a small refiner favor a process with lower variable or fixed costs even if the total treatment costs would be the same?What are the future plans for other distillate streams including off-highway diesel, etc.?What are the transportation (rail, truck and pipeline) issues associated with shipping high sulfur streams to large refiners along with a buy-back arrangement for ULSD?

Page 134

6

REFINER INTERVIEW PROCESS

Refiners

On highwaydiesel production

Commitment toproducing ULSD

Distillate Hydrotreating or

diesel desulfurization

•Upgrade/Additional hydrotreating capacity

•Other alternatives

• Add hydrotreater• Non-conventional

diesel desulfurization

Yes

No Yes

No

Yes No

Identify those refineries producing on-highway diesel and committed to ULSD

Introduce the CED process as an alternative to hydrotreating for the production of ULSD

Obtain agreement to receive and review the CED Powerpoint presentation

Willingness to enter into discussions with Degussa

Page 135

7

REFINER INTERVIEW PROCESS

PADD 1 PADD 2 PADD 3 PADD 4 PADD 5

Refineries Refineries

Interviews Interviews

Findings / Issues / Observations / Opportunities

Refineries

Interviews

4

3

17

8

Refineries

Interviews

30

14

Refineries

Interviews

15

11

27

7

TOTAL

Refineries93

Companies43

Total Universe is 143 refineries93 represent refineries with crude capacity >150KBD

Page 136

8

REFINER UNIVERSE

PADDTotal

RefineriesExcluded

>150000BPSD Diesel Distillate No DHT1 12 6 0 2 42 27 10 2 5 103 56 23 7 4 224 15 0 7 2 65 33 6 6 5 16

143 45 58 Page 137

9

PADD 1

Page 138

10

PADD 1 UNIVERSE

Refinery Location St.Diesel Desulf.

Distillate HT

No Distillate HT

1 United Refining Co. Warren PA 14,000

2 Coastal Eagle Point Oil Co. Westville NJ

3 Young Refining Corp. Douglasville GA4 Ergon-West Virginia Inc. Newell WV

PADD 1 INTERVIEWS

PAD Name City St. Contact Title email Yes No Yes No Yes No Yes No1 United Refining Co. Warren PA David Wortman Planning Group [email protected] x x x x2 Coastal Eagle Point Oil Westville NJ Mark Anderson Process Technology Mgr [email protected] x x x x3 Young Refining Corp. Douglasville GA David Kilgore Refinery Mgr [email protected] x x x x4 Ergon-West Virginia Inc. Newell WV Neil Stanton Refinery Mgr x x ? ? x

Comp Int Rcvd CED.ppt Interest Ref to Deg

Page 139

11

COASTAL EAGLE POINT OIL COMPANYWestville, New Jersey

Coastal- Eagle Point, NJ hydrotreats 40KBD on-road dieselOptions to treat to 15ppm include:

– Revamp one of two hydrotreaters with current capability of treating to 200ppm

– Conversion of a catalytic dewaxer

Mark Anderson, Process Technology Manager for El Paso Energy reviewed our CED PowerPoint proposal and replied back with the following comments:

“We looked at this technology when it was being touted by Petro Star about a year ago. At the time the economics looked uninteresting for two reasons:

1. The sulfones were extracted from the diesel into sulfolane or DMSO, which also extracted some 15% of the feed diesel as aromatics. Petro Star looked on the bright side, claiming their process improved cetane number. That claim was valid, but pointless; the downgrade of the aromatics to the value of residual fuel containing 6 times the sulfur of the diesel offsets any imaginable value of cetane improvement. The cost of downgrade would be on the order of $0.75/bbl of feed diesel, or $1.00 per barrel or product diesel.

Page 140

12

COASTAL EAGLE POINT OIL COMPANYWestville, New Jersey

2. The chemical cost is a killer, and worse so for producers of high sulfur diesel. With the price of 50% peroxide at $0.50/lb, the cost of peroxide amounted to $4,500 per ton of sulfur removed. That compares to $1,000 to $2,000 of H2 consumption per ton of S removed by HDS. The advantage in other operating costs would have had to be $0.50/bbl to offset that alone.

Since then, it appears that acetic acid has been adopted for the solvent, which would probably reduce the extraction of aromatics. I notice that there are no longer any claims about improvements in cetane number, so perhaps that issue has quietly faded into appropriate oblivion.We figure that the technology might make sense at small scale for handling product contaminated with S, such as pipeline interfaces, where the operating cost would be secondary to capital.Now peroxide is under priced, but with 50% now at $0.20/lb., the operating cost is much less of a drawback. Of course, Degussa’s interest is to expand consumption of H2O2 so the price will go up again. Have the process economics changed enough to make the process economically competitive?”

Page 141

13

YOUNG REFININGDouglassville, GeorgiaDavid Kilgore, Assistant Refinery Manager

Off-road diesel production onlyEssentially an asphalt refiner producing roofing grade asphalt for Owens CorningInterested in entering the on-road diesel business using alternative process technologiesProduction is 1KBD #2 oil of which 80% for other applications and 20% off-road dieselNo estimates of conversion costs

Page 142

14

PADD 2

Page 143

15

PADD 2 UNIVERSE

Refinery Location St.Diesel Desulf. Distillate HT

No Distillate HT

1 Somerset Refinery Inc. Somerset KY

2 Countrymark Cooperative Inc. Mount Vernon IN3 Murphy Oil USA Inc. Superior WI 5,5004 Gary-Williams Energy Corp. Wynnewood OK5 Sinclair Oil Corp. Tulsa OK 15,5006 Tesoro West Coast Mandan ND7 Premcor Refining Group Hartford IL 7,6008 Marathon Ashland Petroleum LLC St. Paul Park MN9 Marathon Ashland Petroleum LLC Canton OH

10 Marathon Ashland Petroleum LLC Detroit MI11 National Cooperative Refining Assoc. McPherson KS 25,700 9,60012 Ultramar Diamond Shamrock Corp. Ardmore OK 33,17613 Sunoco Inc. Tulsa OK

14 Farmland Refining LLC Coffeyville KS 21,00015 Frontier Oil Corp. El Dorado KS 31,860

16 Sunoco Inc. Toledo OH17 BP Toledo OH

Page 144

16

PADD 2 INTERVIEWS

PAD Name City St. Contact Title email Yes No Yes No Yes No Yes No1 2 Somerset Refinery Inc. Somerset KY Jan Acrea Vice President [email protected] x x x x

2 2 Countrymark Cooperative Inc. Mount Vernon IN John DeatonRefinery Mgr & Chief Technology Officer [email protected] x x x x

3 2 Murphy Oil USA Inc. Superior WI x x ? ? x4 2 Gary-Williams Energy Corp. Wynnewood OK Steven George Asst Refinery Mgr x x x x5 2 Sinclair Oil Corp. Tulsa OK x x ? ? x6 2 Tesoro West Coast Mandan ND x x ? ? x7 2 Premcor Refining Group Hartford IL Mike Dennis Operations Mgr x x ? ? x8 2 Marathon Ashland Petroleum LLC St. Paul Park MN Chester Shamlin Engineering Mgr x x ? ? x9 2 Marathon Ashland Petroleum LLC Canton OH Todd Sandofer Engineering Mgr x x ? ? x

10 2 Marathon Ashland Petroleum LLC Detroit MI x x ? ? x11 2 National Cooperative Refining Assoc. McPherson KS Rick Johnson Engineering x x x x12 2 Ultramar Diamond Shamrock Corp. Ardmore OK x x ? ? x13 2 Sunoco Inc. Tulsa OK Dirk Morris Clean Fuels [email protected] x x x x14 2 Farmland Refining LLC Coffeyville KS K. Osborne Refinery Mgr [email protected] x x x x15 2 Frontier Oil Corp. El Dorado KS Jim Stump Refinery Mgr x x x x

16 2 Sunoco Inc. Toledo OHChris McCormick Mike Bukowski

Engineering Engineering x x ? ? x

17 2 BP Toledo OH Doug Rundell Refining Tech. Group [email protected] x x x x

Comp Int

Rcvd CED.ppt Interest

Ref to Deg

Page 145

17

SOMERSET REFINING INC.Somerset, Kentucky

Somerset Refining is the 4th smallest U.S. refinery with total crude capacity at 5.5KBDSomerset operates 2 hydrotreaters for naphtha and keroReformer unable to make sufficient hydrogen required to treat to 15ppm

– Current process technology unable to treat to 15ppm– Investment in additional hydrotreating is required but not the direction the

company is willing to takeSomerset has met and is in discussions with UniPure about its process technology

– Somerset has a confidentiality agreement in place with UniPure– Next meeting mid January,2003

Somerset is not convinced UniPure fully interested in Somerset and indicated that UniPure has something in place with a larger Gulf refiner

Page 146

18

SOMERSET REFINING INC.Somerset, Kentucky

Jan Acrea, Vice President of Refining for Somerset Refining inquired if Petro Star has a pilot plant on line or plans for one going forward

He re-iterated his concerns with UniPure and offered the use of Somerset’s infrastructure to Petro Star/Degussa to construct and operate a pilot plant

– A prompt reply to Mr. Acrea is necessary before end December to avoid any possible conflicts with the confidentiality agreements in place between Somerset and UniPure

Page 147

19

SELECT FINDINGS FROM PADD 2 INTERVIEWS

Countrymark Cooperative Inc.– Hydrotreating is the option going forward and we are 2 weeks shy from receiving

a “Schedule A” report on the process design plan (8-9KBD diesel production)– Aware of Petro Star and spoke with a representative years ago

Questioned disposal of sulfone extract

Gary Williams Energy Corp.– Views extraction as a match for existing hydrotreating as a polishing step– Plant never had a diesel hydrotreater (8KBD diesel production)– Signed deal with UOP to revamp hydrocracker

Will add straight run high sulfur diesel, unconverted to the hydrocracker and treat to below 10 ppm

– Estimated conversion costs for the hydrocracker revamp are $2,500/bbl– No amine or sulfur plant-estimated cost to construct at <$10.0M Page 148

20


National Cooperative Refining Association & Farmland Refining Inc.– Decided on conventional hydrotreating (50KBD diesel production)– Design and implementation plans in place

Sunoco, Tulsa, Oklahoma– Off-road diesel production only (15-17KBD)– Wait and see approach to EPA ruling on direction of off-road diesel

Hopeful EPA will exempt railroad diesel market

Sunoco, Toledo, Ohio– On- road production through hydrocracker and most likely revamp unit to produce ULSD– 30KBD diesel capacity

BP– 5 U.S fuels refineries hydrotreating to 500ppm using distillate desulfurizers installed in early

’90s– BP will evaluate all conventional and non-conventional process technologies and have

decision made about process technology adopted by end, 2003Secure bids-2004Construction-2005Operational-2006

Page 149

21

PADD 3

Page 150

22

PADD 3 UNIVERSERefinery Location St.

Diesel Desulf. Distillate HT

No Distillate HT

1 Alon USA Big Spring TX 22,000

2 Coastal Refining & Marketing Inc. Corpus Christi TX 14,000

3 Chevron Products Co. El Paso TX 11,300

4 Hunt Refining Co. Tuscaloosa AL 10,800

5 Crown Central Petroleum Corp. Pasadena TX 10,000

6 Lion Oil Co. El Dorado AR 8,500

7 Giant Refining Co. Bloomfield NM 3,000

8 Placid Refining Co. Port Allen LA 12,000

9 Phillips Petroleum Co. Borger TX 9,200

10 Ultramar Diamond Shamrock Corp. Three Rivers TX 6,600

11 Giant Refining Co. Gallup NM 4,000

12 Cross Oil & Refining Co. Inc. Smackover AR

13 Berry Petroleum Co. Stephens AR

14 AGE Refining & Manufacturing San Antonio TX

15 Calcasieu Refining Co. Lake Charles LA

16 Coastal Mobile Refining Co. Mobile Bay AL

17 Ergon Refining Inc. Vicksburg MS

18 American International Refining Inc. Lake Charles LA

19 Canal Refining Co. Church Point LA

20 Trifinery Petroleum Services Corpus Christi TX

21 Shell Chemical Co. Saint Rose LA

22 LaGloria Oil & Gas Co. Tyler TX

23 Navajo Refining Co. Artesia NM

24 Marathon Ashland Petroleum LLC Texas City TX

25 Valero Energy Corp. Krotz Springs LA

26 Valero Energy Corp. Houston TX

27 Shell Chemical Co. Saraland AL

28 Valero Energy Corp. Corpus Christi TX

29 Murphy Oil USA Inc.Meraux El Dorado

LA AK

30 CITGO Petroleum Lake Charles LA

Page 151

23

PADD 3 INTERVIEWS

PAD Name City St. Contact Title email Yes No Yes No Yes No Yes No1 3 Alon USA Big Spring TX Gordon Leaman Refinery Mgr [email protected] x x x x2 3 Coastal Refining & Marketing Inc. Corpus Christi TX D.T. Warner Refinery Mgrs Office x x ? ? x3 3 Chevron Products Co. El Paso TX Nicole Sparoli? Refinery Mgrs Office x x ? ? x4 3 Hunt Refining Co. Tuscaloosa AL Steve Jackson Refinery Mgr x x ? ? x5 3 Crown Central Petroleum Corp. Pasadena TX Gary Munson Technology Mgr [email protected] x x x x

6 3 Lion Oil Co. El Dorado ARSteve Cousins Paul Fisher

VP & Refinery Mgr Process Technology Mgr [email protected] x x x x

7 3 Giant Refining Co. Bloomfield NM Chad King Refinery Mgr x x ? ? x8 3 Placid Refining Co. Port Allen LA Joey Hagmann Operations [email protected] x x x x9 3 Phillips Petroleum Co. Borger TX Thurman Nance x x x x

10 3 Ultramar Diamond Shamrock Corp. Three Rivers TX Cathy? ? x x x11 3 Giant Refining Co. Gallup NM Ned Davis Refinery Mgr x x x x12 3 Cross Oil & Refining Co. Inc. Smackover AR Micky Steel Refinery Mgr x x ? ? x13 3 Berry Petroleum Co. Stephens AR x x x x14 3 AGE Refining & Manufacturing San Antonio TX Sam Ramahi Engineering [email protected] x x x x

15 3 Calcasieu Refining Co. Lake Charles LARodney Nelson Jody Beret

Refining Mgr Asst Refinery Mgr [email protected] x x x x

16 3 Coastal Mobile Refining Co. Mobile Bay AL Ken Ready Refining Mgr x x ? ? x17 3 Ergon Refining Inc. Vicksburg MS Ken Dillard Refinery Mgr [email protected] x x x x18 3 American International Refining Inc. Lake Charles LA Bill Dean Refinery Mgr x x ? ? x19 3 Canal Refining Co. Church Point LA x x x x20 3 Trifinery Petroleum Services Corpus Christi TX x x x x21 3 Shell Chemical Co. Saint Rose LA x x x x22 3 LaGloria Oil & Gas Co. Tyler TX Chuck Morgan Operations Mgr [email protected] x x x x23 3 Navajo Refining Co. Artesia NM Jim Resinger Refinery Mgr [email protected] x x x x24 3 Marathon Ashland Petroleum LLC Texas City TX Mike Armbruster Refinery Mgr x x ? ? x25 3 Valero Energy Corp. Krotz Springs LA x x ? ? x26 3 Valero Energy Corp. Houston TX x x ? ? x27 3 Shell Chemical Co. Saraland AL Rick Mykitta Operations Mgr x x x x28 3 Valero Energy Corp. Corpus Christi TX x x ? ? x

29 3 Murphy Oil USA Inc.Meraux El Dorado

LA AK

Jim Koontz Fred Green

Refinery Mgr Corp. Engineering Group x x ? ? x

30 3 CITGO Petroleum Lake Charles LA Lee Joyce [email protected] x x x x

Comp Int


Ref to Deg

Page 152

24

GIANT REFINING INC.Gallup, New MexicoNed Davis, Refinery Manager

Presently making ULSD using a “proprietary” desulfurization process it intends to market and license

DIESEL FUEL NEWS, August 5, 2002 reports that:– Giant Refining will start-up the world’s first “diesel isotherming”

hydroprocessing technology on a 4KBD unit running 60% straight-run distillate and 40% light-cycle oil. The new technology, developed by Arkansas-based Process Dynamics and Oklahoma based Linde Process Plants, features a 4’ diameter, 12’ long reactor that will pre-treat feeds to an existing hydrotreater and ultimately yield a 7-8ppm ULSD from conventional 500ppm on-highway feedstocks. The reactor is claimed to cost about half as much (< $500 per daily barrel capacity) and be about half the size of a new, conventional hydrotreating reactor, use only about 60% of the catalyst volume, consume about 100 standard cubic feet of H2/bbl and have an opex of about $0.41/bbl, offset by a 1% product volume swell due to H uptake.

Page 153

25

VALERO ENERGY CORP.San Antonio, TexasJack Olesky, Technology Group

Valero operates 12 refineries of which 3-4 will require alternative desulfurization process due to either a lack of hydrogen or insufficient hydrotreating capacity

Valero likely to install UniPure process using formic acid as opposed to acetic acid

Valero believes that UniPure is further along the development path than other alternative technologies and will get to market faster

Estimated grass roots HDS installation costs are $1,600 to $2,000/bbl– UniPure is estimated to be ½ the cost of HDS

Krotz Springs, Louisiana reportedly the first Valero refinery to install the UniPure process technology-2003

Page 154

26

LaGLORIA OIL & GASTyler, TexasChuck Morgan, Operations Manager

The Tyler, TX LaGloria Oil & Gas refinery is committed to remaining in the on-highway diesel business going forward

Mr. Morgan indicated that the conventional method of hydrotreating with proven equipment and technology is the most likely approach the refinery will take to produce ULSD

He agreed to receive a copy of the CED Powerpoint presentation

10/30/02, after reviewing the presentation, Mr. Morgan replied back with the following comments:

– “It raises more questions than it answers. Needs a lot more discussion on vessels and safety, metallurgy needed to contain process and size. I would like to talk but as I told you our timeline is very short and I'm not sure this process is a sure thing. Remember for the refinery this cannot be an experiment since the whole business fails if the process fails.”

– 10/31/02 Forwarded request to John Tarabocchia to contact Mr. Morgan

Page 155

2721

The Artesia, NM Navajo refinery is committed to remaining in the on-highway diesel business going forward

Mr. Resinger indicated that hydrotreating is the most likely approach to produce ULSD however capital expense and an anticipated hydrogen balance crunch were identified as barriers to entry

He agreed to receive a copy of the CED Powerpoint presentation

10/31/02, after reviewing the presentation, Mr. Resinger replied back with the following comment:

– “Please arrange a discussion with the Degussa representative. My biggest concern continues to be the "disposal" of the sulfone rich product”

– 10/31/02 Forwarded request to John Tarabocchia to contact Mr. Resinger

NAVAJO REFINING COArtesia, New MexicoJim Resinger, Refinery Manager

Page 156

28

ERGON REFININGVicksburg, MississippiJim Dillard, Refinery Manager

The Vicksburg, MS Ergon Refinery produces off-highway diesel only

The refinery has a hydrotreater but the diesel production is not hydrotreated

Refinery is taking a wait and see approach until firm regulations on off-highway diesel are announced

Described the CED process and Mr. Dillard requested a copy of the CED Powerpoint presentation

Page 157

29


Alon USALion Oil Company (25KBD diesel production)

– Expanding conventional hydrotreating process technology to make ULSD– Target treat level is 8ppm– “With new non-traditional technology, no one wants to be serial #001”

Crown Central Petroleum– 30-40KBD primarily off-road diesel– Using SZorb process technology for gasoline production– Likely adopt SZorb technology for diesel if decision is made to make ULSD

Placid Refining– “Can CED treat LCO?”– Existing hydrotreaters-believes CED can be used as a finishing step– “How much are actual operating costs?”– Placid is concerned about the cost of hydrogen peroxide

(13KBD diesel, 3.5KBD LCO) Page 158

30


Calcasieu Refining Company– Options considering to make ULSD by 2008: (6-8KBD diesel)

Additional hydrotreating capacityBuyback arrangement with larger refinerNon-conventional technology

– “Is a pilot plant in operation?”– “If the process works, we’d be interested in adopting the technology”

CITGO– ULSD refinery gate target is 7ppm to achieve 15ppm at retail

+- 4ppm testing variability– “What happens to the olefins in CED?”– Side reactions, yield loss (how much)-considers 10% too high– “What is the nitrogen compound reaction?”– “Where is the peracetic acid formation with hydrogen done? Separate vessel?”– “Activation energy at 200F w/o a catalyst?”– “CED process not suitable for big units with high throughput, specifically because the

technology has yet to be commercially proven?

Page 159

31

PADD 4

Page 160

32

PADD 4 UNIVERSE

Refinery LocationDiesel Desulf.

Distillate HT

No Distillate HT

1 Frontier Refining Inc. Cheyenne, WY 16,5002 Cenex Harvest States Laurel, MT 14,5003 ExxonMobil Corp. Billings, MT 10,500 13,5004 Chevron Products Co. Salt Lake City, UT 10,200 6,500

5 Little America Refining Co. Casper, WY 8,5006 Flying J Inc. Salt Lake City, UT 7,0007 Wyoming Refining Co. Newcastle, WY 4,0008 Conoco Inc. Billings, MT 8,1009 Phillips Petroleum Co. Woods Cross, UT 1,700

10 Montana Refining Co. Great Falls, MT11 Silver Eagle Refining Inc. Woods Cross, UT12 Ultramar Diamond Shamrock Corp. Denver, CO13 Tesoro West Coast Co. Salt Lake City, UT14 Sinclair Oil Corp. Sinclair, WY15 Conoco Inc. Commerce City, CO

Page 161

33

PADD 4 INTERVIEWS

PAD Name City St. Contact Title email Yes No Yes No Yes No Yes No1 4 Frontier Refining Inc. Cheyenne WY Thor Forseth Planning Mgr x x x x2 4 Cenex Harvest States Laurel MT Pat Kimmet Refinery Mgr [email protected] x x x x3 4 ExxonMobil Corp. Billings MT Brad Fuller Facilities Planner x x x x4 4 Chevron Products Co. Salt Lake City UT Jim Newton x x ? ? x

5 4 Little America Refining Co. Casper WYMike Palumbo Paul Moote

Operations [email protected] x x x x

6 4 Flying J Inc. Salt Lake City UT Jeff Utley Plant Mgr [email protected] x x x x7 4 Wyoming Refining Co. Newcastle WY Ernie Hamlet x x ? ? x8 4 Conoco Inc. Billings MT Steve Geiger Asst Refinery Mgr [email protected] x x ? ? x9 4 Phillips Petroleum Co. Woods Cross UT Ron Marelli Engineering Mgr [email protected] x x x x

10 4 Montana Refining Co. Great Falls MT Leland Griffin Refinery Mgr [email protected] x x x x11 4 Silver Eagle Refining Inc. Woods Cross UT K.B. Carroll Plant Supervisor x x ? ? x12 4 Ultramar Diamond Shamrock Corp. Denver CO Jim Clary Refinery Mgr x x ? ? x13 4 Tesoro West Coast Co. Salt Lake City UT Matthew Baebler Engineering Mgr x x x x14 4 Sinclair Oil Corp. Sinclair WY Paul Moote [email protected] x x x x15 4 Conoco Inc. Commerce City CO Brian Lever Facility Mgr [email protected] x x x x

Comp Int


Ref to Deg

Page 162

34

LITTLE AMERICA REFINING/SINCLAIR OIL CORP.Salt Lake City, UtahPaul Moote, Planning

Sinclair Corporation is committed to remaining in the on-highway diesel market going forward

After describing the CED process, Mr. Moote requested a copy of the CED Powerpoint presentation

10/25/02- Mr. Moote replied back and requested to be put in contact with a representative from Degussa

– Forwarded request to John Tarabocchia

Page 163

35

PHILLIPS PETROLEUMWoods Cross, UtahRon Marrelli, Engineering Manager

The Woods Cross, UT Phillips Petroleum refinery is committed to remaining in the on-highway diesel business going forward

After describing the CED process, Mr. Marrelli had two immediate concerns– The extent of the real time research committed to the CED process– Alternative uses for CED waste stream

Attended a NPRA Clean Fuels meeting in Philadelphia the week of 10/14/02

Requested a copy of the CED Powerpoint presentation

10/28/02- Mr. Marrelli replied back and requested to be put in contact with a representative from Degussa

– Forwarded request to John Tarabocchia Page 164

36

MONTANA REFININGGreat Falls, MontanaLeland Griffin, Refinery Manager

The Great Falls, MT Montana Refining refinery is committed to remaining in the on-highway diesel business going forward

– Current diesel production is 1.5 KBD

Refinery is configured using a high pressure hydrotreater running 4 days/wk for FCC gas oil and 3 days/wk for diesel

Mr. Griffin has some familiarity with the CED process and requested a copy of the CED Powerpoint presentation

Questioned alternative applications to the sulfone CED process extract (pharmaceuticals & plastics) rather than blend stock

10/29/02-Mr. Griffin requested to be contacted by a representative from Degussa– Forwarded request to John Tarabocchia

Page 165

37

FRONTIER REFINING INC.Cheyenne, WyomingThor Forseth, Planning Manager

The Cheyenne, WY Frontier refinery is committed to remaining in the on-highway diesel market going forward

Frontier installed a high-pressure hydrotreater about 10 years ago and is convinced current technology can produce 15ppm ULSD

Not interested in CED process and declined CED Powerpoint presentation

Page 166

38

CENEX HARVEST STATESSalt Lake City, UtahPat Kimmet, Refinery Manager

The Salt Lake City, UT Cenex Harvest States refinery is committed to remaining in the on-highway diesel market going forward

Familiar with sulfone technology but concerned about the by product from CED

Summarized Cenex's position as leaning heavily on construction of a new hydrotreater and currently working funding issues

Wants to see a fully functioning pilot plant before committing to non-conventional technology

Reviewing CED Powerpoint presentation but indicated that the likely process will be additional hydrotreating Page 167

39

FLYING J INC.Salt Lake City, UtahJeff Utley, Refinery Manager

The Salt Lake City, UT Flying J refinery is committed to remaining in the on-highway diesel business going forward

Refinery currently has a diesel hydrotreater in operation and will utilize catalyst technology to produce ULSD

Mr. Utley reported that current hydrotreating process can produce 16ppm diesel

Mr. Utley remarked that refiners will have to 1st quarter 2003 to have decided on a technology to meet 2006 deadline

– Equipment availability and skilled craftsmen to install will be a limiting factor


Subsequent discussion with Mr. Utley indicates no further interest in CED– Likely scenario will be hydrotreating

Page 168

40

CONOCO INC.Billings, MontanaSteve Geiger, Assistant Refinery Manager

The Billings, MT Conoco refinery is committed to remaining in the on-highway diesel business going forward

Currently evaluating two desulfurization options:– Installation of a new hydrotreater– Phillips S Zorb process


Suggested contacting Mr. Gary Poffenbarger, Director of Emerging Strategies for Conoco in Houston

Page 169

41

CONOCO INC.Commerce City, ColoradoBrian Lever, Facility Manager

The Commerce City, CO Conoco refinery is committed to remaining in the on-highway diesel business going forward

Options available to the refinery to meet ULSD include the Phillips S Zorb process and additional hydrotreating capacity

As a result of the Conoco Phillips merger, this refinery has been targeted by the FTC to be sold

– The S Zorb process will be considered but not a definite given the sale of the refinery

After describing the CED process, Mr. Lever requested a copy of the CED Powerpoint presentation

Subsequent discussion with Mr. Lever: – Has forwarded presentation internally for review

Extract downgraded to residual fuel oil value?Yield low at 90%?

Page 170

42

TESORO WEST COASTSalt Lake City, UtahMatthew Baebler, Engineering Manager

The Salt Lake City, UT Tesoro West Coast refinery is committed to remaining in the on-highway diesel business going forward

The refinery has excess hydrogen and the process technology currently being evaluated is the installation of a 15 KBD hydrotreater to produce ULSD

The Phillips S Zorb process has been considered but indications are that it does not appear economical for refineries producing <30 KBD diesel

Contact was not interested in the CED process and declined to receive a copy of the CED Powerpoint presentation

Tesoro's “Mirastar” retail locations offer fast refueling service at a rapidly growing number of Wal-Mart stores in the western United States

– Mirastar retail outlets are constructed exclusively at Wal-Mart stores and offer discounts on gasoline when customers use the Wal-Mart Shopping Card

– The Mirastar network of stations is expanding rapidly and is scheduled to include more than 80 locations by early 2002

Page 171

43

EXXONMOBIL CORP.Billings, MontanaBrad Fuller, Facilities Planning

ExxonMobil’s corporate approach to ULSD will not include any grass roots HDS installation

Expansion or revamp of existing process technology only

ExxonMobil’s opinion that the market will not reward a grass roots investment

Estimates total grass roots conversion costs to 15ppm for a small refinery (55KBD) at about $10/bbl

Page 172

44

PADD 5

Page 173

45

PADD 5 UNIVERSERefinery Location St.

Diesel Desulf.

Distillate HT

No Distillate HT

1 U.S. Oil & Refining Co. Tacoma WA 5,5002 Tesoro Alaska Co. Kenai AK3 Tosco Refining Co. Ferndale WA 24,0004 Tesoro Hawaii Corp. Kapolei HA5 Tesoro West Coast Co. Anacortes WA6 Tenby, Inc. Oxnard CA7 Anchor Refining Co. McKittrick CA8 Greka Energy Santa Maria CA9 Tricor Refining Bakersfield CA

10 San Joaquin Refining Co. Inc. Bakersfield CA11 Kern Oil & Refining Co. Bakersfield CA 6,50012 Paramount Petroleum Corp. Paramount CA 8,000 8,00013 Ultramar Diamond Shamrock Corp. Wilmington CA 30,000 30,000

14 Valero Energy Corp. Benicia CA 11,50015 World Oil Co. South Gate16 BP Kuparuk AK17 BP Prudhoe Bay AK18 Petro Star Inc. North Pole AK19 Petro Star Inc. Valdez AK

20 Equilon Enterprises LLC Anacortes WA 15,80021 Equilon Enterprises LLC Bakersfield CA

22 Phillips 66 Refining Co.

Los Angeles (Carson & Wilmington) CA 50,000 35,000

23 ExxonMobil Corp. Torrance CA 17,50024 Equilon Enterprises LLC Martinez CA 70,800 8,600

25 Chevron Products Co. Barber's Point HA26 Equilon Enterprises LLC Wilmington CA

27 Phillips 66 Refining Co.Rodeo & Santa Maria CA 22,100

Page 174

46

PADD 5 INTERVIEWS

PAD Name City St. Contact Title email Yes No Yes No Yes No Yes No1 5 U.S. Oil & Refining Co. Tacoma WA Harvey Van Refinery Mgr [email protected] x x x x2 5 Tesoro Alaska Co. Kenai AK Lauren Wigfield [email protected] x x x x3 5 Tosco Refining Co. Ferndale WA Mr. James Commercial Mgr x x ? ? x4 5 Tesoro Hawaii Corp. Kapolei HA Frank Clouse Technical Director x x ? ? x5 5 Tesoro West Coast Co. Anacortes WA Russ Crawford Refinery Mgr [email protected] x x ? ? x6 5 Tenby, Inc. Oxnard CA x x ? ? x7 5 Anchor Refining Co. McKittrick CA x x ? ? x8 5 Greka Energy Santa Maria CA x x ? ? x9 5 Tricor Refining Bakersfield CA Merlman Giddy Refinery Mgr x x ? ? x

10 5 San Joaquin Refining Co. Inc. Bakersfield CA Pat Oveson Refinery Mgr x x ? ? x11 5 Kern Oil & Refining Co. Bakersfield CA Tony Boutlik Operations Mgr x x x x12 5 Paramount Petroleum Corp. Paramount CA Ryan ? Refinery Mgr x x ? ? x13 5 Ultramar Diamond Shamrock Corp. Wilmington CA Refinery Mgr x x ? ? x14 5 Valero Energy Corp. Benicia CA Al Middleton [email protected] x x x x15 5 World Oil Co. South Gate Refinery Mgr x x x x16 5 BP Kuparuk AK x x x x17 5 BP Prudhoe Bay AK x x ? ? x18 5 Petro Star Inc. North Pole AK x19 5 Petro Star Inc. Valdez AK x20 5 Equilon Enterprises LLC Anacortes WA Homer Dawson x x ? ? x21 5 Equilon Enterprises LLC Bakersfield CA x x ? ? x

22 5 Phillips 66 Refining Co.

Los Angeles (Carson & Wilmington) CA x x ? ? x

23 5 ExxonMobil Corp. Torrance CA x x ? ? x24 5 Equilon Enterprises LLC Martinez CA x x ? ? x25 5 Chevron Products Co. Barber's Point HA x x ? ? x26 5 Equilon Enterprises LLC Wilmington CA x x ? ? x27 5 Phillips 66 Refining Co. Santa Maria CA x x ? ? x

Comp Int


Ref to Deg

Page 175

47

U.S. OIL REFINING CO.Tacoma, WashingtonHarvey Van, Refinery Manager

The Tacoma, WA U.S. Oil refinery is committed to remaining in the on-highway diesel business going forward

Refinery has hydrotreating capacity currently and over the next 2+ months will test a new diesel desulfurization catalyst to determine if the refinery can produce ULSD

After explaining the CED process, Mr. Van stated he had heard of Petro Star but was not aware of Degussa

Mr. Van requested a copy of the CED Powerpoint presentation

10/30/02 follow up call with Mr. Van revealed no further interest in the CED process– Will make modifications to the hydrotreater to produce 30ppm diesel– One year later, additional modifications to produce 15ppm diesel Page 176

48

TENBY, INC. OXNARD, CA WORLD OIL CO. SOUTH GATE, CA ANCHOR REFINING CO. McKITTRICK, CA GREKA ENERGY SANTA MARIA, CA

Asphalt production only-CED not applicable

Page 177

49

KERN OIL & REFINING CO.Bakersfield, CaliforniaTony Boutlik, Refinery Manager

The Bakersfield, CA Kern Oil refinery is committed to remaining in the on-highway diesel business going forward

Current diesel hydrotreater reportedly is capable of producing 15ppm – No interest in alternative or non-conventional diesel desulfurization technologies to

current process

Gave brief description of the CED process however, Mr. Boutlik expressed no interest in receiving a copy of the CED Powerpoint presentation

Page 178

50

ULTRAMAR DIAMOND SHAMROCK/VALERO ENERGY CORPWilmington, CaliforniaRefinery Manager

The Wilmington, CA UDS refinery is committed to remaining in the on-highway diesel business going forward

Contact stated that alternative, supplemental, or non-traditional diesel desulfurization technologies to meet ULSD are not needed at this refinery

– Current hydrotreating capability suitable for 15ppm ULSD

Declined a copy of the CED Powerpoint presentation

Page 179

51

TESORO ALASKA CO.Kenai, AlaskaRefinery Manager

Indicated that decision on desulfurization technology is corporate, not refinery level

Referred to James Taylor, Head of Manufacturing in San Antonio, TX

Page 180

52

COMPETING TECHNOLOGIES

Overview of Non-Conventional Alternative Process TechnologiesUniPure’s ASR-2 Process Technology

– Converts diesel and gasoline containing up to 1,500ppm S to ULS products having <5ppm

– No hydrogen required

– Desulfurization achieved by removing oxidized sulfones by extraction

– For a 500ppm feed, hydrocarbons extracted with the sulfones is about 0.3% of the feed

– UniPure has produced 0.5ppm diesel

– Estimated capital cost for a 25KBD ASR-2 plant processing 500ppm diesel to 5ppm is about $25.0MM, or $1,000/bbl of installed capacity

– Estimated incremental conversion costs in the range of $0.05 to $0.08/gal

UniPure is in negotiations with Valero Energy, and Somerset Refining

Page 181

53

COMPETING TECHNOLOGIES

Overview of Non-Conventional Alternative Process Technologies

Linde Process– First commercial project

4KBDFeed: 60% Straight Run, 40% LCOStart up: September, 2002

Linde Process operational at Giant Refining’s Gallup, NM plantNext steps-license process technology

Page 182

54

TYPICAL HYDROTREATER COSTS

Type Capacity, BPD

Straight run feed, %

Capital, $/BPD - a

New 50,000 100 1,000

New 10,000 68 1,800

Revamp 50,000 100 600

Revamp 10,000 68 1250

a – excludes hydrogen, amine and sulfur units

Page 183

55

ISSUES IN ULSD IMPLEMENTATION

Technology that is capable of producing a 15 ppm ULSD at the refinery is not GOOD enough– As per the EPA analysis, refiners are required to produce ULSD

below 15 ppm, in fact, it should be around 7 ppm levels– EPA also believes that current technologies are capable of reaching

to 7 ppm levelsReasons

15 ppm ULSD is at the pumps

Inaccuracies and reproducibility problems in sulfur testing

Contamination in tanks and pipelines

Contamination during transportation

Page 184

SULFUR TESTING - ULSD

15 ppm Designated Test MethodASTM D 6428-99

Allowable Alternative Test MethodsASTM D 5453-99 and ASTM D 3120-96

• Pipelines set their point of origin specifications based on the compliance standard and reproducibility (R)

ULSD Compliance = 15 ± 16 ppm (using ASTM D 6428)

• Based on these results, refiners will need to blend to impossibly low levels of sulfur on each batch at large cost

Page 185

57

PIPELINE SHIPMENTS OF DISTILLATE FUELS BETWEEN PADDs

Page 186

58

PIPELINE SIZES, OPERATING RATES & INTERFACE VOLUMES

Diameter Distance Maximum Rate Interface Volume Arrival Time(Inches) (Miles) (BPH) (Barrels) (Minutes)

40 1,047 50,000 15,000 1836 1,047 37,000 12,500 20

16 14 6,000 200 216 106 3,900 600 916 206 7,000 890 8

12 24 3,350 130 212 254 4,200 530 8

10 7 4,000 50 18 24 2,650 60 1

Page 187

59

TYPICAL PRODUCT SEQUENCE IN A REFINED PRODUCTS PIPELINE

• The ULSD Rule prohibits any party downstream of the refiner fromdowngrading more than 20% of the annual volume of 15 ppm highwaydiesel

• EPA estimated that the interface volume will be 4.4% of the total volume of highway diesel transported by pipelines

• However, pipeline operators estimate that the interface volume will be in the range of 4 to 10%

• Pipelines are unlikely to build dedicated tankage or carry two grades of diesel

Page 188

60

PRELIMINARY PIPELINE TEST RESULTS

D5453 D6428Origin Lifting 10 7Atlanta 10 7Greensboro 10 7Greensboro Tankage 18 13Greensboro Local 18 13Greensboro North 22 17Selma 24 19

Page 189

61

CONTAMINATION DURING TRANSPORTATION

7,500 Gallons 15ppm ULSD contaminated with:

<500 ppm Diesel Fuel

<3000 ppm Jet/Diesel Fuel

<5000 ppm Heating Oil

7 Gallons or .1% + .5 ppm + 3 ppm + 5 ppm

37 Gallons or .5% + 2.5 ppm + 15 ppm + 25 ppm

75 Gallons or 1% + 5 ppm + 30 ppm + 50 ppm

Page 190

62

ISSUES RELATED TO CED PROCESS

Alternate uses/disposal of sulfone extract?

Value of sulfone extract?

Barrel/yield loss of CED process?

Pilot plant/commercialization timing?

Is CED capable of producing 5-8ppm ULSD?

Cost of hydrogen peroxide?

Actual capital and operating cost of CED?

Can CED treat all process streams? If so, provide results Page 191

63

Based on discussions with 43 U.S. refining companies, Kline & Company believes that opportunities currently exist for CED with at least 13 refiners…

Somerset Refining

Young Refining

Sunoco, Tulsa

LaGloria Oil and Gas

Navajo Refining

Ergon, Vicksburg

Calcasieu Refining

Little America Refining/ Sinclair Oil

Montana Refining

AGE Refining

Placid Refining

Tesoro, Anacortes

Phillips, Woods Cross

Page 192

64

SUMMARY

CED process must be capable of producing 7-8ppm ULSD and able to handle various refinery streams– It is important to provide pilot scale results to gain confidence with

potential customers– Cost items need to be more delineated – Need to address technical issues raised

Lack of hydrogen seems to be the major driver for refiners opting for non-conventional technologies

Small refiners are more concerned with the “technology risk” rather than the capital and operating cost at the current stage Page 193

65

SUMMARY

Articlear needs to demonstrate CED process on a pilot scale unit by mid-2003 as most of the small refiners will be evaluating and selecting the ULSD process by 2003– With the exception of PADD 4 small refiners, most of the small

refiners in other PADDs may not claim exemption

All non-conventional ULSD process technologies including S Zorb, UniPure, and Linde have at least one commercial license or an operating unit, unlike Articlear– It will be increasingly difficult to market the CED technology without a

commercial license

Articlear must take prompt advantage of the opportunity with Somerset Refining to use its facilities for commercial demonstration

Page 194

APPENDIX D

CED 2002 SUMMARY REPORT FOR 5,100 BPSD PROCESS UNIT

Page 196

Page 197

Page 198

Page 199

Page 200

Page 201

Page 202

Page 203

Page 204

Page 205

Page 206

Page 207

Page 208

Page 209

Page 210

Page 211

Page 212

Page 213

Page 214

Page 215

Page 216

APPENDIX E

KOCH PILOT-SCALE TESTING FINAL REPORT

CED 2002 Pilot Plant Testing

Test 2 – Final Repor t

Test 2 – Sulfox Extraction System TestingMay 23, 2002

ByCED Process Development Team

Robert D’AlessandroAndy Jones

John TarabocchiaSteve Bonde

Joe LallyAlan Henderson

August 9, 2002

� This report is a revised and extended version of a report initially issued

by Donald J. Glatz, Elizabeth Czerniawski, and Deva Hupaylo of KochModular Process Systems (KMPS). The report was entitled “SulfoxExtraction” and was issued on June 7, 2002.

Page 218

SULFOX EXTRACTION

OVERVIEW AND OBJECTIVES

Oxidized sulfur compounds (sulfones) in the gas oil product from the oxidation reactionstep are more polar than the unoxidized sulfur compounds (thiophenes) in the raw gas oil.The increased polarity makes these compounds extractable by a polar solvent such asacetic acid. During this extraction step, acetic acid solvent and oxidized gas oil arecontacted countercurrently in a packed extraction column. A majority of the sulfurcompounds are transferred into the extract phase (acetic acid).

The primary objectives for this test program were:

1. Demonstrate the technical feasibility of the sulfox extraction by measuring the totalconcentration of sulfur in the column raffinate.

2. Examine the removal efficiency of unoxidized and oxidized sulfur compounds bymeasuring the concentration of each in the column feed and raffinate.

3. Evaluate the effect of solvent to feed ratio (S/F) and amount of water in the aceticacid on sulfur extraction efficiency and overall hydrocarbon yield. The S/F targetswere 0.8, 1.25, and 2.0 on an “as is” weight-to-weight basis. The target waterconcentrations in acetic acid were 1.0, 2.5 and 5.0 percent.

4. Produce sufficient materials for solvent recovery testing and hydrocarbon recoverytesting.

The operating temperature and specific throughput (capacity) were not variables for thistest program. These were set at 45-49oC and 615 GPH/ft2 (combined feed rate of 1900cc/min), respectively.

EQUIPMENT SET-UP

The equipment set-up is shown in Dwg. No. D202382-03-140 for the 3” diameter x 24’packed column. The packed column had a 4” diameter top disengaging chamber and 6”diameter bottom chamber. It was packed with Flexipac-2Y-1 structured packing(316SS). Elements were 5-1/2” long and a 3” diameter Karr Column plate was placedbetween every other element to simulate SMVP packing.

The oxidized gas oil feed was charged from a 275-gallon Tote using one or two FMIpumps (0-1000 cc/min each) to set and control the volumetric feed rate. The acetic acidwas spiked to the proper water content and charged from 55-gallon drums. One or twoFMI pumps were utilized to set and control the volumetric rate from the drums to thecolumn. Heat exchangers were used to preheat both streams to the desired temperature.The interface was maintained in the column manually by controlling the bottom take-offrate using an FMI pump and needle valve by-pass combination.

Page 219

FEED MATERIAL PREPARATION

Prior to operating the extraction column, the feed and solvent were prepared as follows:

Oxidized Gas Oil Feed: The feed for the extraction step was a combination of thefollowing drums of product discharged from the Destruct Reactor (DR1) during thetesting program for the Two Stage Oxidation System:

Partial drum generated between 14:00 and 19:00 on Friday, May 17Partial drum generated between 11:30 and 14:00 on Monday, May 20Partial drum generated between 14:42 and 18:40 on Monday, May 20Full drum generated between 11:15 and 15:10 on Tuesday, May 21

The last 2 drums also contained discharge from the pipe reactor, which was held in thereactor at 62-65 oC for 68 minutes after shutting down the front end of the system. Thesematerials were blended and mixed by a circulating pump in a clean, new tote. A total of165 – 170 gallons of material was obtained. Upon sitting and cooling to roomtemperature, less than 1 gallon of heavy phase was decanted off the bottom of the totebefore starting the feed to the extraction column.

Acetic Acid Solvent - Three (3) full drums plus one partial drum of solvent wereprepared for the test program. These included one full drum each of 1% and 5% waterand approximately 1-1/3 drums of 2.5% water in acetic acid. The water was addeddirectly to the acetic acid drums and mixed with a drum mixer for several minutes.

COLUMN OPERATION

The column was initially filled with the continuous phase (acetic acid), then the dispersedphase (gas oil) was added and the interface set in the top disengaging chamber of thecolumn. Oxidized gas oil and acetic acid rates were set, the interface established andcontrolled in the top disengaging chamber, and heat inputs adjusted to establish therequired temperature profile in the column. Flow rates were monitored regularly (every30 minutes) and adjusted as required. For the first run, a total of four column turnovers(column volume divided by combined feed rates) were processed prior to sampling.After the initial run, and after conditions were adjusted (changed S/F and/or % water inacetic acid), three turnovers were processed prior to sampling.

Prior to collecting the samples for each run, the raffinate and extract rates were measured.First an attempt was made to stabilize the interface (bottoms take off rate). Thevolumetric rates were then measured by graduated flasks and stopwatch method.

Page 220

RESULTS AND DISCUSSION

The extraction tests were performed via a 3 x 3 variable design specified by Degussa.The outline for the runs was as follows:

RunNumber

Water Content ofAcetic Acid (wt%)

Solvent-to-FeedMass Ratio (S/F)

Oxidized FuelRate (mL/min)

Solvent Rate(mL/min)

1 2.5 0.80 1140 7502 2.5 1.25 950 9503 2.5 2.00 720 11804 5.0 2.00 720 11805 5.0 1.25 950 9506 5.0 0.80 1140 7507 1.0 0.80 1140 7508 1.0 1.25 950 9509 1.0 2.00 720 1180

After the first three runs with 2.5% water in the acetic acid, the system was shut down forapproximately 5 minutes while the acid feed was switched to 5% water and new extractand raffinate drums were installed. The same shutdown and switch occurred betweenruns 6 and 7 when changing to 1.0% water in the acetic acid.

Operating log sheets for all Sulfox Extraction runs are attached. A summary of the keyoperating parameters and sulfur and acetic acid analyses are provided in Table 1. Theanalytical data was generated by Degussa and Petro Star personnel in the KMPSlaboratory. Table 2 shows the material balance with estimated fuel yield. Based uponthe data provided and observations by KMPS personnel during the test work, thefollowing points are made:

• As with previous gas oil / acetic acid testing, this system processed very well in thepacked column. Dispersed phase particle sizes appeared to vary somewhat,depending on the water concentration in the acetic acid. The smallest dispersed phaseparticles were about 1-2 mm diameter for 1% water in the acetic acid, ranging up toabout 2-4 mm diameter for 5% water. For all runs the hold up was highest in themiddle of the column (between elevations of 8’ and 14’ in the column), with less holdup above and below this region. The hold up appeared to increase slightly withdecreasing water concentration in the acetic acid.

• All runs were made with a total combined throughput of 1900 cc/min or a specificthroughput of 615 GPH/ft2. This rate appears to be conservative for this system, andhigher specific throughput should be attainable. Some future optimization of thecolumn capacity may be considered.

Page 221

%H2O S/F Capacity Packed# Date/Time Gas Oil Acid in Acid (wt.) GPH/sq.ft. Ht. (ft) #2 #5 #16 #21 Raff. as is Raff. Cor. Ext. as is Ext. Cor. Raff. Ext.

5/23/2002

FEED 1110.0 11.1

1 9:50 1140 750 2.5 0.8 615 24 45 37 40 48 30.0 35.4 980.3 8238 15.2 88.1

2 10:55 950 950 2.5 1.25 615 24 51 45 45 55 19.6 23.9 619.4 5530 18.0 88.8

3 11:50 720 1180 2.5 2.0 615 24 52 50 48 52 16.3 19.5 454.5 2877 16.5 84.2

4 13:10 720 1180 5.0 2.0 615 24 51 48 48 51 23.4 26.5 449.3 5349 11.7 91.6

5 14:10 950 950 5.0 1.25 615 24 53 44 47 55 27.4 31.5 748.0 8698 13.0 91.4

6 15:10 1140 750 5.0 0.8 615 24 53 46 45 56 36.3 42.7 1174.7 10777 14.9 89.1

7 16:15 1140 750 1.0 0.8 615 24 54 46 45 57 27.7 34.5 1074.0 7956 19.7 86.5

8 17:10 950 950 1.0 1.25 615 24 54 46 47 56 18.0 22.7 701.9 3734 20.8 81.2

9 18:15 720 1180 1.0 2.0 615 24 51 48 46 50 13.6 16.5 413.6 2298 17.6 82.0

Table 1

% Acetic AcidPacked Column Runs

Degussa - Petro Star, Sulfox Extraction, 202382

RUN Feed Rate, cc/min Temp. C Total Sulfur (ppm)

Revised: June 4, 2002

Page 222

Flow Raffinate Sulfur Fuel

# Date/Time Gas Oil A.Acid Raff. Extract Balance Fract. Acid Raff. Extract Feed Raff. Extract Balance Feed Raff. % Yield

5/23/2002

1 9:50 991 800 926.0 940.0 104.2% 0.152 30.0 980.3 1.100 0.028 0.921 86.3% 881.5 785.2 89.1%

2 10:55 826 1013 675.0 1132.0 98.3% 0.180 19.6 619.4 0.917 0.013 0.701 77.9% 734.7 553.5 75.3%

3 11:50 626 1259 640.0 1202.0 97.7% 0.165 16.3 454.5 0.695 0.010 0.546 80.1% 556.8 534.4 96.0%

4 13:10 626 1259 458.0 1464.0 102.0% 0.117 23.4 449.3 0.695 0.011 0.658 96.2% 556.8 404.4 72.6%

5 14:10 826 1013 811.0 1060.0 101.7% 0.130 27.4 748.0 0.917 0.022 0.793 88.9% 734.7 705.6 96.0%

6 15:10 991 800 1045.0 787.0 102.3% 0.149 36.3 1174.7 1.100 0.038 0.924 87.5% 881.5 889.3 100.9%

7 16:15 991 800 930.0 948.0 104.9% 0.197 27.7 1074.0 1.100 0.026 1.018 94.9% 881.5 746.8 84.7%

8 17:10 826 1013 726.0 1145.0 101.7% 0.208 18.0 701.9 0.917 0.013 0.804 89.1% 734.7 575.0 78.3%

9 18:15 626 1259 437.0 1020.0 77.3% 0.176 13.6 413.6 0.695 0.006 0.422 61.6% 556.8 360.1 64.7%

Total Sulfur (g/min) Fuel Rate (g/min)

Date: June 4, 2002

Table 2

Degussa - Petro Star, Sulfox Extraction, 202382

Material Balance and Yield

Run Feed Rate (g/min) Effluent Rate (g/min) Sulfur (ppm)

Page 223

• The raffinate analyses for total sulfur concentration (ppmw) are summarized below.The numbers are presented on a gas oil basis. In other words, the acetic acid is notincluded when calculating the overall sulfur content.

Water Content of Acetic Acid (wt%)Solvent-to-Feed Ratio 1.0 2.5 5.0

0.80 34.5 35.4 42.61.25 22.7 23.9 31.52.00 16.5 19.5 26.5

In addition, analyses were performed on the unoxidized sulfur content (ppmw) on agas oil basis. The results of these analyses are summarized below.

Water Content of Acetic Acid (wt%)Solvent-to-Feed Ratio 1.0 2.5 5.0

0.80 13.0 16.6 14.01.25 8.9 11.7 13.02.00 8.8 12.3 10.7

These data clearly show a trend of increasing raffinate sulfur concentration withincreasing water in the acetic acid (lower extraction efficiency) and decreasingraffinate sulfur concentration with increasing S/F (as expected).

• The combined oxidized gas oil feed was analyzed to contain 17 ppm of unoxidizedsulfur compounds (thiophenes).

• The overall mass balance around the extraction column was in the range of ±5% forall runs except #9. This level of accuracy is typically considered good for extractionruns. However, the sulfur balance values are generally greater than –10%. This is anindication that the interface may have been rising during the effluent rate checks,giving a higher than actual raffinate rate and lower than actual extract rate.

• The fuel yield was calculated by dividing the actual mass rate of fuel out by the massrate of fuel in (after subtracting out the acetic acid concentration). This value isdependent on accurate raffinate and extract flow rates. These flow rates may not havebeen sufficiently accurate due to variable liquid-liquid interface levels during ratemeasurement. It was expected that the fuel yield would show a correlation towardshigher yield for increasing water concentration in the acetic acid. However, the datais highly variable, and no general trend is evident.

Page 224

• Figure 1 shows the liquid-liquid equilibrium curve generated by KMPS for oxidizedsulfur in gas oil with 2.0 wt% water in the acetic acid. To generate this curve, KMPSused the data provided by Degussa, left out the first point (highest distributioncoefficient), and regressed a straight line through zero with the remaining data.Figures 1A, 1B, and 1C show the operating line and stepped off theoretical stages forRuns 1-3, where acetic acid with 2.5 wt% was used. These figures indicate that 2.2 –3.0 theoretical stages were achieved in the 24’ packed column. These figures wereprepared by KMPS and neglect the nonlinear nature of the equilibrium line and thehigh mutual solubilities of gas oil and acetic acid.

Page 225

run#1 graph Chart 1

Figure 1a. Sulfox Extraction, Run #1

0

100

200

300

400

500

600

700

800

900

1000

0 200 400 600 800 1000 1200

RAFFINATE, ppmw sulfur

EX

TR

AC

T, p

pm

w s

ulf

ur

LLE OP

Page 226

run#2 graph Chart 1

Figure 1b. Sulfox Extraction, Run #2

0.0

100.0

200.0

300.0

400.0

500.0

600.0

0.0 200.0 400.0 600.0 800.0 1000.0 1200.0


EX

TR

AC

T, p

pm

w s

ulf

ur

LLE OP

Page 227

run#3 graph Chart 1

Figure 1c. Sulfox Extraction, Run #3

0.0

50.0

100.0

150.0

200.0

250.0

300.0

350.0

400.0

450.0

500.0

0.0 200.0 400.0 600.0 800.0 1000.0 1200.0


EX

TR

AC

T, p

pm

w s

ulf

ur

LLE

OP

Page 228

CONCLUSIONS

Using the original objectives as a basis, the following conclusions can be made regardingthe results from this test program:

1. The technical feasibility of the sulfox extraction using acetic acid as the solvent wassuccessfully demonstrated. Excellent sulfur removal was achieved using the 3”diameter x 24’ SMVP packed column and sufficient data is on hand to providedesigns for production size extraction columns.

2. The percent sulfur extraction increases slightly as the acetic acid water content isreduced from 5% to 1%. Variations of gas oil yield based upon water content in theacetic acid cannot be identified from this test work.

3. At the completion of the extraction testing, approximately 3 x 50 gallons each ofraffinate and extract were generated for future testing requirements. A 1 x 50 gallondrum of both raffinate and extract were produced at each concentration of water in theacetic acid (1% - 2.5% - 5%).

RECOMMENDATIONS

KMPS recommends the design and use of an SMVP packed column for the 50 BPD plantproposed by Degussa and Petro Star. The operating conditions and results from this pilotplant test can be used as the basis for the scale up and design of that unit.

Regarding optimization, the results from this test work clearly demonstrate the effect ofS/F and water content in the acetic acid on the extraction of sulfur. However, the effectof water content in the acetic acid on the overall gas oil yield could not be determined.Any future work should make a strong effort to evaluate this variable. Future testingshould utilize digital scales to facilitate determination of a time-averaged mass flowratefor the raffinate and extract streams.

Page 229

PROCESS PERFORMANCE

The total sulfur content and unoxidized sulfur content was measured on a sample of theraffinate stream for each of the nine operating conditions. As illustrated in Figure 2, thetotal sulfur content decreased with increasing solvent-to-feed ratio for each of the solventwater contents. It is also evident that increasing water content results in less efficientextraction.

A similar result is obtained when the unoxidized sulfur content of the raffinate wasmeasured. Figure 3 illustrates this result. It appears that the measurement at a watercontent of 5.0 wt% and a solvent-to-feed ratio of 0.75 is erroneously low.

Prior to testing of the sulfox extraction system at KMPS, Petro Star and Degussagenerated liquid-liquid equilibrium (LLE) data for the system containing oxidized gas oil,acetic acid, and water. As the sulfur concentration in the gas oil decreases, the apparentdistribution coefficient for sulfur decreases. This results in a curved equilibrium line withan upward concavity. The reason for this decreasing distribution coefficient is twofold.First, the sulfur content of the oxidized gas oil is composed primarily of oxidizedorganosulfur species (i.e., sulfones). The gas oil also contains residual unoxidized sulfurspecies (i.e., thiophenes). Thiophenes have a much lower distribution coefficient(approximately 0.3) than the oxidized species. Second, the oxidized sulfur compoundsexhibit a range of distribution coefficients. This is due to aliphatic substituents attachedto the basic thiophenic structures. These alkyl substituents lead to a more aliphatic nature

Figure 2. Effect of Solvent-to-Feed Ratio on Raffinate Total Sulfur

0

5

10

15

20

25

30

35

40

45

0 0.5 1 1.5 2 2.5

Solvent-to-Feed Ratio (Mass Basis)

To

tal S

ulf

ur

Co

nte

nt,

Gas

Oil

Bas

is (

pp

m)

5.0 wt% H2O in Acetic Acid



Page 230

and a lower distribution coefficient. The oxidized sulfur species have distributioncoefficients ranging from 2.5 to 2.8.

Figure 4 shows the equilibrium curve with an acetic acid solvent containing 2.0 wt%water at 45oC.

Figure 4. Sulfox Extraction Equilibrium - 2 wt% Water in Acetic Acid

0

500

1000

1500

2000

2500

0.0 100.0 200.0 300.0 400.0 500.0 600.0 700.0 800.0

Gas Oil Sulfur Content (ppm) - Unadjusted Concentration

So

lven

t S

ulf

ur

Co

nte

nt

(pp

m)

- U

nad

just

ed C

on

cen

trat

ion

Equilibrium Line

Equilibrium Data

Figure 3. Dependence of Unoxidized Sulfur Content on Solvent-to-Feed Ratio

0

2

4

6

8

10

12

14

16

18

0 0.5 1 1.5 2 2.5

Solvent-to-Feed Ratio (Mass Basis)

Un

oxi

diz

ed S

ulf

ur

Co

nte

nt,

Gas

Oil

Bas

is (

pp

m S

)




Page 231

By forcing the linear regression through the origin, the effect of the low distributioncoefficients at low sulfur concentrations is compromised by the KMPS regressedequilibrium line. The KMPS equilibrium line results in a lower number of predictedtheoretical stages. Figure 5 demonstrates this effect. Although the equilibrium lines donot appear to be dramatically different, the sulfur removed in the raffinate stage variessignificantly. The equilibrium curve developed by Degussa predicts a sulfur reduction ofapproximately 60 ppm in the raffinate stage. The equilibrium curve of KMPS, however,predicts a sulfur reduction of approximately 85 ppm.

Even at low and moderate temperatures, acetic acid and light atmospheric gas oil(LAGO) exhibit a fairly large mutual solubility. At 45oC, the solubility of acetic acid inLAGO is approximately 16.5 wt%. At the same temperature, the solubility of gas oil inacetic acid is approximately 13.0 wt%. Figures 1a, 1b, and 1c prepared by KMPS treatthe feed (gas oil) and solvent (acetic acid) as immiscible, however. For immisciblesystems, the operating line passes through (Xf,Ye) and (Xr,Ys) where X and Y representthe gas oil and acetic acid sulfur concentrations, respectively, and the subscripts f, s, r,and e represent the feed, solvent, raffinate, and extract, respectively. For partiallymiscible systems, however, this is not the case. In partially miscible systems, aconsiderable amount of gas oil is transferred to the fresh acetic acid solvent in theraffinate stage and a considerable amount of acetic acid is transferred to the gas oil feedin the feed stage. Figures 6a, 6b, and 6c show the results from Runs 1, 2, and 3 when thenonlinear equilibrium line and partial miscibility are properly considered. It should benoted that the axes have been converted to gas oil concentrations on an acetic acid freebasis and acetic acid concentrations on a gas oil free basis.

Figure 5. Effect of Equilibrium Line on Theoretical Stage Determination

0

50

100

150

200

250

300

350

400

0 20 40 60 80 100 120 140 160

Gas Oil Sulfur Content (ppm)

So

lven

t S

ulf

ur

Co

nte

nt

(pp

m)

KMPS Regression of Equilibrium Data

Degussa Regression of Equilibrium Data

Operating Line

Page 232

Figure 6a. Run 12.5 wt% Water in Acetic Acid - Solvent-to-Feed Mass Ratio of 0.80

0.0

500.0

1000.0

1500.0

2000.0

2500.0

3000.0

0.0 200.0 400.0 600.0 800.0 1000.0 1200.0 1400.0

Gas Oil Sulfur Content (ppm) - Acetic Acid Free Basis

So

lven

t S

ulf

ur

Co

nte

nt

(pp

m)

- G

as O

il F

ree

Bas

is

Equilibrium LineInlet and Outlet ConcentrationsOperating Line

Page 233

Figure 6b. Run 22.5 wt% Water in Acetic Acid - Solvent-to-Feed Mass Ratio of 1.25

0.0

500.0

1000.0

1500.0

2000.0

2500.0

3000.0

0.0 200.0 400.0 600.0 800.0 1000.0 1200.0 1400.0


So

lven

t S

ulf

ur

Co

nte

nt

(pp

m)

- G

as O

il F

ree

Bas

is


Page 234

Figure 6c. Run 32.5 wt% Water in Acetic Acid - Solvent-to-Feed Mass Ratio of 2.00

0.0

500.0

1000.0

1500.0

2000.0

2500.0

3000.0

0.0 200.0 400.0 600.0 800.0 1000.0 1200.0 1400.0 1600.0


So

lven

t S

ulf

ur

Co

nte

nt

(pp

m)

- G

as O

il F

ree

Bas

is


Page 235

When the curvature of the equilibrium line and the partial miscibility are properlyutilized, the predicted number of theoretical stages is increased. The following tableshows the number of theoretical stages predicted by KMPS and Degussa.

Predicted Theoretical StagesRun Number KMPS Degussa

1 3.0 3.62 2.6 3.03 2.2 2.6

Page 236

Page 237

Conversion Extraction Desulfurization (CED) Phase III .../67531/metadc786886/m2/1/high_re… · improvements to the process, which was renamed “CED 2002”. Petro Star and Degussa

Documents