1. IAPG Natural Gas Congress Molecular Sieves:is your regeneration procedure optimized? CECA – VETEK Author Peter Meyer Presenter Bob Davenport

2. IAPG 2008  A non optimized regeneration procedure can harm the molecular sieves and reduce significantly their life time  New unit: if the regeneration gas is recycled the recycled water content has to be taken in account, if the regeneration gas flow rate is too short the unit will not work  Knowing how to optimize the procedure can help debottlenecking a unit Why optimize the regeneration procedure?

3. IAPG 2008  cap TSA  cap PSA Industrial units Regeneration: how does it work?

4. IAPG 2008 TSA Regeneration: how does it work? This presentation will focus on Natural Gas Drying regenerated by Thermal Swing Adsorption (TSA). Regeneration procedure: 1)Switch including possibly pressure change (depress.) 2)Heating (Purge? Two step heating? Heating ramp?) 3)Cooling (dry/wet gas?) 4)Switch including possibly pressure change (repress.)

5. IAPG 2008 Regeneration: parameters Heating step: how much heat? -Heat up the molecular sieves -Heat up the vessel (internal/external insulation) -Remove water -Push out desorbed water

6. IAPG 2008 Regeneration: temperature influence The quantity of regeneration gas depends on the inlet temperature during the heating. Below a minimum temperature the water dew point spec could not be reached (too high residual water content), the maximum temperature depends on the type of molecular sieve. Regeneration temperature Quantity of regeneration gas Minimum!!

7. IAPG 2008 Regeneration: uncomplete regeneration A sudden decrease of adsorption time is the sign for a bad regeneration (accumulation of water)  happens very often shortly after start up, possibility to recover sieves Adsorption cycles Adsorption time

8. IAPG 2008 Regeneration: uncomplete regeneration Make sure to have a plateau at the outlet during heating. Make sure to have a small temperature difference between inlet and outlet during heating.

9. IAPG 2008 Regeneration: Maximum temperature 3A : 230°C (446°F) for saturated gases, up to 260°C (500°F) for unsaturated gases 4A : normally 250°C (482°F), up to 290°C (554°F) with precautions For information 5A/13X : 300°C (572°F) in case of sweetening, but if there is NO water on the sieves

10. IAPG 2008 Regeneration: pressure influence More regeneration gas (quantity) is needed if the regeneration pressure is at a high pressure. Two cases for pressure range: Low pressure – heating limited The regeneration gas has to bring in the energy for heating and desorption (regeneration temperature above boiling temperature of water at regeneration pressure) High pressure – stripping limited The regeneration gas has additionally to strip off (push out) the desorbed water. The limit between both is around 30-35 bars. Example: a regeneration at 60 bar (870 psia) may require perhaps 25% more regeneration gas quantity.

11. IAPG 2008 CASE STUDY 3 adsorber system, 2 in adsorption, 1 in regeneration 500 MMSCFD, 60 bar (870 psia), 30°C (86°F), Saturated gas, 4A molecular sieve, Adsorption time 16 hrs, Regeneration recycled (air cooler sat. @ 55°C (131°F), 30 bar (435 psia))

12. IAPG 2008 CASE STUDY Case 1: Regeneration at 250°C, 482°F and 30 bar, 435 psia Case 2: Regeneration at 250°C, 482°F and 60 bar, 870 psia Case 3: Regeneration at 290°C, 555°F and 60 bar, 870 psia Case 4: Regeneration optimization to minimize hydrothermal damage and improve molecular sieves performance. (Case 5: Correction of case 2 supposing only case 1 flow rate available, internal heat insulation) Fixed: pressure drop during adsorption, no stand-by time.

13. IAPG 2008 CASE 1 Case 1: Regeneration at 250°C, 482°F and 30 bar, 435 psia Procedure Depress 15 min Heating Cooling Repress 15 min 7 hrs 30 min available 100% F P Case 1 Vessel diameter : 100% Adsorbent weight : 100% 100% F R1

14. IAPG 2008 CASE 2 Case 2: Regeneration at 250°C, 482°F and 60 bar, 870 psia Procedure Depress 0 min Heating Cooling Repress 0 min 8 hrs 00 min available 30 min more Case 2 Vessel diameter : 100% Adsorbent weight : 104% Regengasquantity : 123% Heater / Compressor - Capacity ?? 100% F P 115% F R1

15. IAPG 2008 CASE 3 Case 3 Vessel diameter : 100% Adsorbent weight : 102% Regengasquantity : 115% Case 3: Regeneration at 290°C, 482°F and 60 bar, 870 psia Procedure Depress 0 min Heating Cooling Repress 0 min 8 hrs 00 min available Steel: max. design temperature ?? 100% F P 108% F R1

16. IAPG 2008 CASE STUDY Case 1: Regeneration at 250°C, 482°F and 30 bar, 435 psia Case 2: Regeneration at 250°C, 482°F and 60 bar, 870 psia Case 3: Regeneration at 290°C, 555°F and 60 bar, 870 psia (possibility to reach lower dew point) Case 4: ??? MS QuantityRegengas quantity Regengas Flow rate CASE 1100% 100% F R1 CASE 2104%123%115% F R1 CASE 3102%115%108% F R1 HYDROTHERMAL DAMAGING

17. IAPG 2008 End of Adsorption Hydrothermal damaging

18. IAPG 2008 Start of Heating Cold saturated section Heated section Hydrothermal damaging

19. IAPG 2008 Heating proceeds... Steam Fog Formation Cold Saturated Section Hot Section Hydrothermal damaging

20. IAPG 2008 Water Retro-condensation during Heating Vaporization Zone Condensation Zone REFLUX Water Droplets Crust and Lumps formation Hydrothermal damaging

21. IAPG 2008 CASE 4 Case 4: Regeneration at 290°C, 482°F and 60 bar, 870 psia Procedure Depress 0 min 30 min intermediate heating Heating Cooling Repress 0 min 7 hrs 30 min available Case 4 Vessel diameter : 100% Adsorbent weight : 102% Regengasquantity : 123% 100% F P 115% F R1

22. IAPG 2008 CASE STUDY Case 1: Regeneration at 250°C, 482°F and 30 bar, 435 psia Case 2: Regeneration at 250°C, 482°F and 60 bar, 870 psia Case 3: Regeneration at 290°C, 555°F and 60 bar, 870 psia Case 4: Regeneration at 290°C, 555°F and 60 bar, 870 psia, interm. heating Advantage of case 4: lower water dew point at outlet. MS QuantityRegengas Quantity Regen flow rate CASE 1100% 100% F R1 CASE 2104%123%115% F R1 CASE 3102%115%108% F R1 CASE 4102%123%115% F R1

23. IAPG 2008 CASE STUDY Case 1: Regeneration at 250°C, 482°F and 30 bar, 435 psia Case 2: Regeneration at 250°C, 482°F and 60 bar, 870 psia Case 3: Regeneration at 290°C, 555°F and 60 bar, 870 psia Case 4: Regeneration at 290°C, 555°F and 60 bar, 870 psia, interm. heating Case 5: keep regeneration gas flow rate and MS height of case 2, internal heat insulation, diameter 3.5 m, increase feed flow rate, decrease adsorption time MS QuantityRegengas quantity Regen flow rate CASE 1100% 100% F R1 CASE 2104%123%115% F R1 CASE 3102%115%108% F R1 CASE 4102%123%115% F R1

24. IAPG 2008 CASE 5 (versus case 2) Case 2: Regeneration at 250°C, 482°F and 60 bar, 870 psia Procedure Depress 0 min Heating Cooling Repress 0 min 4 hrs 30 min available Adsorption 9 hrs Case 2 Vessel diameter : 94.6% Adsorbent weight : 88% Regengasquantity : 61% Pressure drop 0.7 bar (233%) 140% F P 100% F R2

25. IAPG 2008 CASE STUDY Internal insulation, 3.7m  3.5 m internal diameter Higher feed flow rate : 140% Higher pressure drop during adsorption : 233% Adsorption time down from 16hrs to 9hrs (shorter life time) MS QuantityRegengas quantity Regen flow rate CASE 2100% 100% F R2 CASE 588%61%100% F R2

26. IAPG 2008 How to prevent hydrothermal damaging? Hydrothermal damaging happens when liquid water is present on the molecular sieves at high temperature: One should try to change the regeneration procedure in order to prevent desorption of water when the molsieve bed is not yet heated up almost homogenously thus limiting water condensation at top layers  intermediate heating step + higher regeneration gas flow rate

27. IAPG 2008 How to prevent hydrothermal damaging? Original procedure: - int. temperature too high, too much water desorbed - plateau of outlet temperature showing water re-vaporization

28. IAPG 2008 How to prevent hydrothermal damaging? New procedure: - int. temperature lower - smoother increase of outlet temperature

29. IAPG 2008 How to prevent hydrothermal damaging?

30. IAPG 2008 Conclusion When looking at your molecular sieve unit and its regeneration procedure: -Don’t underestimate the pressure influence on the regeneration gas quantity -Think about hydrothermal damaging -Optimization does not cost a lot but lengthens the life time of the molecular sieves Don’t hesitate to ask the nice and knowledgeable guys from CECA to help you. THANK YOU!