EVALUATION OF PROGRESSIVE DISTILLATION Dan Dobesh Jesse Sandlin

EVALUATION OF PROGRESSIVE DISTILLATION Dan Dobesh – Jesse Sandlin Dr. Miguel Bagajewicz 04. 29. 2008

This presentation is not about this Insurance Company

Not about this one either…

$Our Mission “Analyze progressive crude fractionation, a technology patented in 1987 that claims to$

Our Mission “Analyze progressive crude fractionation, a technology patented in 1987 that claims to be more energy efficient than conventional fractionaltion. ”

Punchline “Progressive Distillation can reduce the heat duty requirement of the distillation process by 17% for a heavy crude, and use 16% less furnace heat utility while producing more valuable products for a light crude. ”

Overview 1) Background: – Distillation Specifications – Conventional Crude Distillation – Progressive Crude Distillation 2) Methodology 3) Results 4) Accuracy & Limitations

Petroleum Value Chain Petroleum Refining Petroleum Products Petroleum Production http: //en. wikipedia. org/wiki/Oil_refinery www. freddiesasphaltoval. com/ http: //www. lakewoodconfer ences. com/direct/dbimage/ 50241031/Plastic_Toy. jpg http: //en. wikipedia. org/wiki/Image: Oil_well. jpg www. ehow. com/how_2041 839_siphon-gas-car. html Fuels Solvents Lubricants Plastics Detergents Nylon Polyesters

Oil Refinery Schematic Over 2% of the energy content in a crude stream is used in distillation. * Distillation accounts for about 40% of energy use in a refinery. ** Diagram Source: http: //en. wikipedia. org/wiki/Oil_refinery * Bagajewicz, Miguel and Ji, Shuncheng. “Rigorous Procedure for the Design of Conventional Atmospheric Crude Fractionation Units. Part I: Targeting. ” Ind. Eng. Chem. Res. 2001, 40, 617 -626 **Haynes, V. O. “Energy Use in Petroleum Refineries. ” ORNL/TM-5433, Oak Ridge National. Laboratory, Tennessee, September (1976).

Overview 1) Background: – Distillation Specifications – Conventional Crude Distillation – Progressive Crude Distillation 2) Methodology 3) Results 4) Accuracy & Limitations

Light Crude Feed 10000 9000 8000 Barrels/day 7000 6000 5000 4000 3000 2000 1000 -164 -89 -42 13. 2 50 59 73 86 100 114 128 142 155 169 183 197 211 225 239 253 266 280 294 308 322 336 350 364 378 391 405 419 439 467 495 523 549 573 597 622 663 720 781 827 0 Normal Boiling Point (NBP) of Component (°C) Petroleum crude component boiling points range from -161 C (CH 3) to over 827 C (C 40 H 82+)

Heavy Crude Feed Conventional Distillation Products Heavy Crude 10000 9000 8000 Barrels/Day 7000 6000 5000 4000 3000 2000 1000 -164 -89 -42 13. 2 50 59 73 86 100 114 128 142 155 169 183 197 211 225 239 253 266 280 294 308 322 336 350 364 378 391 405 419 439 467 495 523 549 573 597 622 663 720 781 827 0 NBP of Component (°C) Petroleum crude component boiling points range from -161 C (CH 3) to over 827 C (C 40 H 82+)

ASTM D 86 -07 b, “D 86 Point” • American Society for Testing and Materials (ASTM): international organization that is a source for technical standards • Rigorously developed method for quantitatively testing the boiling range of a petroleum product (1) Oil sample heated in glass flask using electric heater (2) Vapor is condensed and collected (3) Temperature versus amount collected is recorded • Not applicable to products containing large amounts of residual

Product Specifications Generated from Pro/II Computer Model 6000 This graph compares the boiling point range of the five products 5000 Barrels/day 4000 3000 2000 Naphtha Kerosene 1000 Diesel Gasoil 0 -164 -42 13. 2 50 73 100 128 155 183 211 239 266 294 322 350 378 405 439 495 549 597 663 781 NBP of Component (°C) Resid Crude Feed

Product Gaps Explanation D 86 5% point heavy component - D 86 95% point light component 390⁰ C - 360⁰ C = 30⁰ C 0. 2 0. 18 0. 16 0. 12 0. 1 D 86 95% point light component D 86 5% point heavy component 0. 08 0. 06 0. 04 Positive gaps indicate more distinct separation. 0. 02 0 -263. 2 -128. 2 -43. 6 31. 1 121 138 163 187 212 237 262 287 312 337 362 387 412 437 462 487 512 537 562 587 612 637 662 687 713 737 761 786 823 873 924 974 1021 1064 1107 1152 1225 1329 1438 1521 Composition 0. 14 NBP of Component

Overview 1) Background: – Distillation Specifications – Conventional Crude Distillation – Progressive Crude Distillation 2) Methodology 3) Results 4) Accuracy & Limitations

Conventional Distillation

0. 16 -0. 04 -164 -42 50 86 128 169 211 253 294 336 378 419 495 573 663 827 Resid NBP of Component Composition 0. 035 0. 03 NBP of Component 0. 015 0. 02 0. 005 0 0. 2 0 0 NBP of Component 0. 5 0 -164 -42 50 86 128 169 211 253 294 336 378 419 495 573 663 827 0 NBP of Component -164 -42 50 86 128 169 211 253 294 336 378 419 495 573 663 827 Composition 0. 2 NBP of Component -164 -42 50 86 128 169 211 253 294 336 378 419 495 573 663 827 0. 01 Composition Crude Composition NBP of Component -164 -42 50 86 128 169 211 253 294 336 378 419 495 573 663 827 -164 -42 13. 2 50 73 100 128 155 183 211 239 266 294 322 350 378 405 439 495 549 597 663 781 Composition Conventional Distillation Simulation Naphtha 0. 025 Kerosene Diesel Gasoil

Gaps – Conventional Distillation Products 6. 00 E+03 Light Crude D 86 95% point anchors products on the right side, gaps change the left side 5. 00 E+03 3. 00 E+03 Naphtha Kerosene 2. 00 E+03 Diesel Gasoil 1. 00 E+03 Resid Crude Feed 73 10 0 12 8 15 5 18 3 21 1 23 9 26 6 29 4 32 2 35 0 37 8 40 5 43 9 49 5 54 9 59 7 66 3 78 1 0. 00 E+00 -1 64 -4 2 13. 2 50 Barrels/Day 4. 00 E+03 NBP of Component (°C)

Conventional = Indirect Takes the heaviest component as the bottom product in each column. Lighter components are sent to the next column. Source: Smith, Robin, Chemical Process Design

Conventional = Indirect Stacking these columns on top of each other is essentially conventional distillation. Bagajewicz, Miguel and Ji, Shuncheng. “Rigorous Procedure for the Design of Conventional Atmospheric Crude Fractionation Units. Part I: Targeting. ” Ind. Eng. Chem. Res. 2001, 40, 617 -626

Conventional = Indirect Stacked columns Stacking these columns from the on indirect top of each other is sequence. essentially conventional distillation. Bagajewicz, Miguel and Ji, Shuncheng. “Rigorous Procedure for the Design of Conventional Atmospheric Crude Fractionation Units. Part I: Targeting. ” Ind. Eng. Chem. Res. 2001, 40, 617 -626

Overview 1) Background: – Distillation Specifications – Conventional Crude Distillation – Progressive Crude Distillation 2) Methodology 3) Results 4) Accuracy & Limitations

Patent: Process for Distillation of Petroleum by Progressive Separations • This is an expired patent for crude fractionation that is now being commercialized by Technip. • Main idea is to heat components only as much as necessary. • Several companies are excited by this concept that promises large energy savings. • A new refinery is being built in central Germany using this concept.

Progressive Crude Distillation Patent

Technip’s Progressive Brochure

Naphtha Progressive Crude Distillation Kerosene Gaps 0. 12 0. 18 0. 08 0. 16 0. 14 0. 06 Composition 0. 1 0. 04 0. 02 0. 1 NBP of Component (°C) 0. 05 0 Gasoil 0. 12 0. 1 0. 08 Diesel 0. 06 Kerosene NBP of Component Gasoil 0. 04 Resid Diesel 0. 02 0. 16 0 0. 14 Component 164 -42 3. 2 50 73 100 128 155 183 211 239 266 294 322 350 378 405 439 495 549 597 663 781 NBP of Component (°C) NBP of (°C) -164 -42 13. 2 50 73 100 128 155 183 211 239 266 294 322 350 378 Crude Composition 500 0 1800 1600 1400 1200 1000 800 600 400 200 0 1. 60 E+03 1. 40 E+03 1. 20 E+03 Naphtha 1. 00 E+03 Kerosene 8. 00 E+02 6. 00 E+02 4. 00 E+02 2. 00 E+02 0. 00 E+00 0. 12 Composition 1000 0. 2 0. 15 Top 3 Top of 1. 80 E+03 Column 2 of Column NBP of Component 2000 1500 0. 25 -164 Composition -42 13. 2 50 73 100 128 155 183 211 239 266 294 322 350 378 405 439 495 2500 0. 02 NBP of Component 0 Barrels/Day 3000 0. 06 0. 04 -164 -42 50 86 128 169 211 253 294 336 Barrels/Day 378 419 495 573 663 827 Barrels/Day 3500 0. 08 -164 13. 2 73 128 183 239 294 350 405 -164 495 597 -42 781 50 86 128 169 211 253 294 336 378 419 495 573 663 827 4000 Diesel 0. 1 -164 -42 13. 2 50 73 100 128 155 183 211 239 266 294 -164 322 -42 350 13. 2 378 50 405 73 439 100 Composition 495 128 549 155 597 183 663 211 781 239 266 294 322 350 378 405 439 495 549 597 663 781 Top 0 of Column 1 0. 12 0. 1 0. 08 0. 06 0. 04 0. 02 NBP of Component

Gaps – Progressive Distillation 6000 Light Crude 5000 Barrels/Day 4000 Naphtha Kerosene 3000 Diesel Gasoil Resid 2000 Crude Feed 1000 0 -164 -42 13. 2 50 73 100 128 155 183 211 239 266 294 322 350 378 405 439 495 549 597 663 781 NBP of Component (°C)

Progressive = Direct Takes the lightest component as the top product in each column. Heavier components are sent to the next column. Source: Smith, Robin, Chemical Process Design

Conventional vs. Progressive Summary One main column Many columns Indirect Direct Recover heavy components first Recover light components first

Overview 1) Background: – Distillation Specifications – Conventional Crude Distillation – Progressive Crude Distillation 2) Methodology 3) Results 4) Accuracy & Limitations

Simulation Development Method 1) 2) 3) 4) 5) 6) Build PRO/II progressive crude simulation Obtain correct D 86 95% points Synchronize product gaps Mimimize heat duty Compare to conventional heat duty Determine areas for improvement

Simulation Assumptions • SRK is a valid thermodynamic model for hydrocarbon systems • Pseudocomponents represent crude composition • PRO/II provides a close representation of reality

Basis of Comparison PRO/II Conventional Simulation, 260 ⁰C steam

PRO/II Computer Model(s) Progressive Model – 4 column direct Furnace heat duty = 89 MW This is higher than 58. 7 MW for conventional distillation Previous work suggested that this setup provided no furnace heat utility benefit over conventional distillation. Our results verify this.

Initial Complex Simulation Too many products for conventional comparison • Unnecessarily complicated

PRO/II Computer Model Patent Vacuum distillation for residual product is not important for comparison

Second Type Simulation Each column has a reboiler • Too much furnace heat utility: 200+ MW

Third Type Simulation All seven columns have steam input • Furnace utility is lower, but steam utility his very high

F*Cp MW Heating Supply-Demand Temperature ⁰C Demand Curve – dark line showing heat needed by system Supply boxes – heat utility able to be recovered from system • Heat can be transferred down and left by second law • Heat can only move right across pinch via a pumparound

Final Type Simulation Replaced steam with reboilers in the first series of columns

F*Cp MW Heating Supply-Demand Temperature ⁰C

Specifications

Variables

Controller-Variable Systems 1) 2) 3) 4) Naphtha-kerosene gap varies with steam flowrate in Column 1 Kerosene-diesel gap varies with steam flowrate in Column 2 Diesel-gas oil gap varies withsteam flowrate in Column 3 D 86 95% points are obtained by varying the condenser duty Column 2 Column 1 Column 3

days… weeks MONTHS After hours of red simulations and Red Bulls… Happy hour

Final Simulations • Conventional: four simulations – 260 ⁰C steam, 135 ⁰C steam – Heavy feed, light feed • Progressive: eight simulations – Reboilers, steam – 260 ⁰C steam, 135 ⁰C steam – Heavy feed, light feed – High heat exchanger temperatures, low heat exchanger temperatures

Overview 1) Background: – Distillation Specifications – Conventional Crude Distillation – Progressive Crude Distillation 2) Methodology 3) Results 4) Accuracy & Limitations

Conventional vs. Progressive Light Crude 9% Decrease 15% Decrease

Conventional Distillation Products 6000 Light Crude 5000 Barrels/day 4000 Naphtha Kerosene 3000 Diesel Gasoil 2000 Resid Crude Feed 1000 0 -164 -42 13. 2 50 73 100 128 155 183 211 239 266 294 322 350 378 405 439 495 549 597 663 781 NBP of Component (°C)

Progressive Distillation Products Light Crude 6000 5000 Barrels/Day 4000 Naphtha Kerosene 3000 Diesel Gasoil Resid 2000 Crude Feed 1000 0 -164 -42 13. 2 50 73 100 128 155 183 211 239 266 294 322 350 378 405 439 495 549 597 663 781 NBP of Component (°C)

Progressive Heat usage Light crude heat utility diagram Progressive Distillation Hot Utility Light Crude 400 Cold Utility 350 300 T 250 200 150 100 50 0 -100 0 100 200 300 400 H The intersection that is unaccounted for is the cold and hot utility

Progressive Heat usage F*Cp MW Light Crude Temperature ⁰C

Conventional vs. Progressive Heavy Crude 9% Decrease 14% Decrease

Conventional Distillation Products Heavy Crude 16000 14000 Barrels/Day 12000 Naphtha 10000 Kerosene 8000 Diesel Gasoil 6000 Resid Crude Feed 4000 2000 0 -164 -42 13. 2 50 73 100 128 155 183 211 239 266 294 322 350 378 405 439 495 549 597 663 781 NBP of Component (°C)

Progressive Distillation Products Heavy Crude 16000 14000 Barrels/Day 12000 Naphtha 10000 Kerosene 8000 Diesel Gasoil 6000 Resid Crude Feed 4000 2000 0 -164 -42 13. 2 50 73 100 128 155 183 211 239 266 294 322 350 378 405 439 495 549 597 663 781 NBP of Component (°C)

Progressive Heat usage Heavy crude heat utility diagram Progressive Distillation Heavy Crude 400 350 300 T 250 200 150 100 50 0 -100 0 100 200 H 300 400

Progressive Heat usage F*Cp MW Heavy Crude Temperature ⁰C

Our Conclusion “Progressive Distillation can reduce the heat duty requirement of the distillation process by at least 17% for a light crude, and at least 16% for a heavy crude, while producing similar amounts of products. ”

Economic Analysis • 120, 000 BPD plant • Gross profit = Product sales – Utility costs • Progressive provides gross profit increase of $10. 2 million each year using light crude feed and $27. 3 million each year using a heavy crude feed

Vacuum Economic Analysis • Gas oil and residue profits are recovered in equal amounts in both cases • Progressive provides gross profit increase of $25. 7 million each year using light crude feed and $57. 2 million each year using a heavy crude feed

Overview 1) Background: – Distillation Specifications – Conventional Crude Distillation – Progressive Crude Distillation 2) Methodology 3) Results 4) Accuracy & Limitations

Limitations • Different column sequences and setups may offer lower heat utility • Optimum setup is based on composition of crude feed • Simulations are a simplification of reality • Heat exchanger network in the simulation is not optimized

Accuracy • D 86 95% point comparisons between conventional and progressive are within 0. 1 degrees Celcius • Product gap comparisons between conventional and progressive are within 1. 0 degrees Celcius • Flowrate comparisons between conventional and progressive are within 10 cubic meters per hour

Questions