1. Corrosion evaluation of MEA solutions by SEM-EDS, ICP-MS and XRD Georgios Fytianos Seniz Ucar Andreas Grimstvedt Hallvard F. Svendsen Hanna Knuutila 8 th Trondheim Conference on CO 2 Capture, Transport and Storage. TCCS-8 16 - 18 June 2015

2. 2 Outline Introduction Motivation Methodology Results Conclusions

3. 3 Amine treating Units Although the continuous improvement in the capture efficiency there are various operational problems such as corrosion of process equipment and solvent degradation

4. 4 Degradation and Corrosion Amine degradation: Decreases the efficiency of CO 2 capture Solvent loss Unwanted compounds and emissions Corrosion: Severe operational problems Increases the maintenance budget Corrosion and Degradation are closely tied

5. 5 Research Motivation Process parameters that determine the extent of corrosion: 1.Temperature 2.CO 2 loading 3.Amine type and concentration 4.Degradation products Little information on corrosivity available Little information on corrosivity available

6. Degradation Products We tested different acids: Some of them increase corrosion Other Degradation Compounds: We tested 12 compounds Degradation ProductsCAS OZD497-25-6 HEEDA111-41-1 HEIA3699-54-5 HEI1615-14-1 HEF693-06-1 HEA142-26-7 BHEOX1871-89-2 HEPO23936-04-1 HEGly5835-28-9 DEA111-42-2 Bicine150-25-4

7. 7 Degradation products We chose 3 techniques to evaluate corrosion ICP-MS SEM-EDS XRD ICP-MS SEM-EDS XRD After a first screening, we chose two degradation products:

8. Corrosion Evaluation: Overview Weight loss technique with metal coupons is one of the most used for the calculation of the corrosion rate. A number of electrochemical methods for corrosion measuring exist potentiodynamic polarization techniques are among the most popular When studying corrosion it is of great importance to examine Corrosion rate Kinetics Corrosion products Corrosion type Links between degradation products and corrosion

9. Corrosion Evaluation: Overview ICP-MS : Inductively Coupled Plasma Mass Total Metal Concentration in the liquid gives information about the relative corrosivity SEM- EDS : Scanning Electron Microscopy-Energy Dispersive X-ray Spectroscopy Surface morphology- Elemental Mapping (homogeneous corrosion or not) XRD : X-ray Powder Diffraction Identification of corrosion products

10. 10 Methodology

11. Experimental Setup Stainless steel cylinders (316 stainless steel tubes with an outer diameter of ½ inch and equipped with Swagelok® end caps) 9 g of loaded solution injected into the cylinder(30wt% MEA+1wt% degradation product) Cell put in forced convection oven at 135 o C Experiments run for 5 weeks (two replicates) Samples analysed for Fe, Cr and Ni by ICP-MS as an indication of corrosivity LC-MS analysis for MEA and degradation products SEM-EDS for surface characterization XRD for corrosion products identification

12. 12 Results

13. Degradation MEA concentration after 5 weeks for the different solutions HeGly (1wt%)Bicine (1wt%) Week 20.48%0.77% Week 50.06%0.59% Thermal Stability

14. Corrosion Results

15. Metal Concentration

16. 16 Surface morphology (A)before the experiment, (B)End of exp. MEA solution (C)End of exp. MEA+bicine, (D)End of exp. MEA+HeGly Scale bars denote 100 µm

17. Elemental mapping wt % MEAMEA+BicineMEA+ HeGly Fe64.663.864.4 Cr18.318.018.2 Ni12.513.112.8 Mo2.42.62.5 Mn1.6 Si0.40.50.4

18. Corrosion Products XRD data showed formation of highly crystalline siderite, FeCO 3, on the cylinder surfaces for all the solutions. Additional low intensity peaks at 43.5° associated with elemental iron, Fe, were observed in the presence of 1 wt% degradation product+MEA solutions. MEA+HeGly MEA+Bicine MEA

19. LIQUID SAMPLE Liquid sample analysis with ICP-MS can work as a first screening of relative corrosivity among solutions ICP-MS alone is not enough to study corrosion in post- combustion CO 2 capture STAINLESS STEEL SEM itself can be used for surface morphology but does not give any details about the composition of the surface. EDS can be used for elemental composition

20. Conclusions The combination of ICP-MS, SEM-EDS and XRD was used for corrosion evaluation of MEA solutions. ICP-MS: Additon of Bicine or HeGly increases the corrosion SEM-EDS: Surface morphology of 316 SS changed with addition of degradation product Overall the results from SEM support the findings from the liquid analyses since ICP-MS showed higher relative corrosivity with the addition of HeGly and bicine.

21. Acknowledgements The work is done under the SOLVit SP4 project, performed under the strategic Norwegian research program CLIMIT. The authors acknowledge the partners in SOLVit, Aker Solutions, Gassnova, EnBW and the Research Council of Norway for their support.

22. 22 Thank you for your attention

23. 23 APPENDIX Degradation Products Structure Materials and Methods

24. 24 Degradation Products Eirik F. da Silva et. al. : Understanding MEA degradation in post-combustion CO2 capture

25. 25 Materials and Methods A high resolution Thermo Fischer Element 2 (Bremen, Germany) ICP-MS was used for the analysis of metals in the liquid samples. The solutions were analyzed for Fe, Cr, and Ni by ICP-MS as an indication of corrosivity. For the ICP-MS analysis each sample was mixed at room temperature and 100 μL was pipetted into a sample tube. 100 μL of concentrated HNO 3 (ultra pure) are added and everything was diluted to 10 mL with water. Finally this solution was further diluted resulting in a total dilution of 10000 (1 +9999). Both SEM and EDS characterization were carried out by using a Hitachi S-3400N scanning electron microscope. For this purpose small pieces were cut from the cylinders and their surfaces were cleaned with ethanol to remove any deposited corrosion product prior to scanning. An acceleration voltage of 20.0 kV and a working distance of 10.0 μm were used, and samples were placed on stubs and scanned without coating. Aztec Energy software was used to process the EDS data. Qualitative characterization of deposited corrosion products was conducted via powder X-Ray Diffraction (XRD) (D8 Advance DaVinci, Bruker AXS GmBH). The pattern was collected in the 2θ range of 20-80° using a Cu X-ray tube, with a step size of 0.013° and a step time of 0.78 s. Precipitates formed on cylinder walls were collected gently after air-drying, and crushed with a mortar and pestle before XRD analyses. The PDF-4+ database (from the International Centre for Diffraction Data) was used for species identification.