Environmental Impacts of Post-Combustion Capture - American

Chapter 10

Downloaded by UNIV OF NORTH CAROLINA on July 22, 2013 | http://pubs.acs.org Publication Date (Web): May 3, 2012 | doi: 10.1021/bk-2012-1097.ch010

Environmental Impacts of Post-Combustion Capture: New Insight M. I. Attalla,* P. Jackson, and K. Robinson Energy Transformed Flagship, CSIRO Energy Technology, 10 Murray-Dwyer CCT, Mayfield West NSW 2300, Australia *e-mail: [email protected]

This chapter reports on the propensity of amine-based post combustion CO2 capture (PCC) solvents to form deleterious nitrosamine compounds in the presence of NOx. The formation and emission of these materials from the PCC process has not been investigated in detail. The entrainment of these degradation products in the scrubbed flue gas and subsequent release to the environment is an issue that requires attention, particularly because of their hazardous nature. A possible mitigation method that can reduce emissions is the UV treatment of the solvent liquor, scrubbed gas and/or wash water to destroy any RR′N-NO1,2 materials that can form during the PCC process.

Introduction Despite amine-based post-combustion CO2 capture (PCC) being an established technology it has been used for decades to remove CO2 from small volume/flow industrial gas streams many challenges must be overcome before large scale CO2 capture from power station flue gas can be realized. One challenge involves addressing the degradation of the solvent (by both oxidative and thermal processes) that occurs under typical flue gas conditions i.e. in the presence of O2, NOx and SOx. As well, there is the potential release of volatile solvents and solvent degradation products to the atmosphere via (i) entrainment in droplets

© 2012 American Chemical Society In Recent Advances in Post-Combustion CO2 Capture Chemistry; Attalla, M.; ACS Symposium Series; American Chemical Society: Washington, DC, 2012.


or (ii) volatilization during the flue-gas scrubbing process. The associated environmental impacts need to be given serious consideration. If solvent-based PCC technology is deployed at an industrial scale, potentially millions of tonnes of solvent will be used per annum. Hence, the small-scale emission of solvents and solvent degradation products to the environment appears inevitable. Recent reports by Braten et al. (1) and Stangeland and Shao (2) clearly indicate that nitrosamine and nitramine generation is feasible when organic amines are utilized in a PCC process. Nitrosamines, nitramines and other N-centred species are potentially hazardous to both mammals (potential carcinogens) and aquatic life-forms (3). The formation and emission of these materials from the PCC process has not yet been fully explored. Given the potential magnitude of (fugitive) amine release if solvent-based PCC technology is adopted globally, research efforts in areas such as: i.

understanding the photochemistry of slipped nitrosamines and other degradation by-products, their atmospheric half-lives and ii. understanding the impact of slipped amines and degradation byproducts on the biosphere e.g. organism toxicity, are warranted and necessary. This case study demonstrates that secondary amines used in PCC formulations, such as piperazine (PZ) and diethanolamine (DEA), react slowly with NOx in the presence of O2 to form nitrosamine derivatives under controlled laboratory conditions. The experimental conditions under which these N-nitrosamine derivatives form are not dissimilar to the conditions found inside a typical CO2 flue-gas absorption column e.g. P ~ 1 atm; gas composition 81.4 % N2, 12.8 % CO2, 0.8 % NO and 5.0 % O2; temperature of 60 °C. The principal differences between the laboratory-based experiments and those of a typical capture plant are: i.

the absence of sorbent cycling, which is performed at an industrial scale to regenerate the amine from the carbamate derivative and liberate the captured CO2, and ii. elevated nitric oxide levels (used as a proxy for NOx) which are normally present at 300-700 ppm in flue streams derived from black coal (4). Elevated levels of nitric oxide and nitrogen dioxide will simply accelerate the rate of amine degradation and the formation of N-nitrosoamines. The structure of the amines investigated in this study are presented in Figure 1. This chapter focuses mainly on the results for piperazine, which has been studied more exhaustively than the other secondary amines.

208 In Recent Advances in Post-Combustion CO2 Capture Chemistry; Attalla, M.; ACS Symposium Series; American Chemical Society: Washington, DC, 2012.


Figure 1. Structures of the amines examined for nitrosamine formation. These amines are commonly used in PCC solvent formulations.

Establishing the Formation of N-Nitrosamine Derivatives For the investigation of nitrosamine formation from amine-based PCC absorbents, a typical experiment involved bubbling a simulated flue gas stream (of the composition outlined previously) at a flow rate of 1.34 L/min, through a 200 g mass of: 1. 2.

15 wt % aqueous solution of piperazine 30 wt % solutions of monoethanolamine (MEA), 2-amino-2-methyl1-propanol (AMP), diethanolamine (DEA), methyldiethanolamine (MDEA) and 2-piperidinemethanol (2-PM).

All experiments were carried out at a reaction temperature of 60°C, with a total run time of 15 hours. Reaction mixtures were sampled at 3 hour intervals for analysis by positive ion electrospray ionization-mass spectrometry (ESI-MS). Broad-scan (m/z 50 – m/z 250) ESI-mass spectra were used to screen for the species of interest and optimize mass spectrometer parameters. Measures were adopted to exclude light from the reaction mixture, including the use of aluminium foil and opaque (brown glass) reaction and sample vessels.


Nitrosamine Derivatives of Secondary Amines N-Nitrospiperazine


The broad scan positive ion ESI-mass spectrum collected for a 15 wt % aqueous piperazine reaction mixture, following exposure to the simulated flue gas stream for 3.5, 7.0 and 15.0 hrs respectively, are shown in Figure 2.

Figure 2. Broad-scan spectra of the aqueous piperazine reaction mixture after exposure to simulated flue gas for (1) t = 3.5 hours, (2) t = 7 hours, and (3) t = 15 hours. m/z 116, N-nitrosopiperazine, is labelled with an asterisk. N-nitrosamine can be seen to increase with time relative to the piperazine (m/z 87) peak. 210 In Recent Advances in Post-Combustion CO2 Capture Chemistry; Attalla, M.; ACS Symposium Series; American Chemical Society: Washington, DC, 2012.


The broad-scan spectra of the reaction mixtures were quite complex and contained many species. For example, protonated and sodiated peaks identified in the piperazine reaction mixture include (H2O)(CH3CN)H+, (CH3CN)Na+ or H2NO3+, (CH3CN)2H+, PZH+, (CH3CN)2Na+, PZ-OH+ (the N-oxide of piperazine), PZNOH+ (N-nitrosopiperazine), PZ-CO2H2+ (piperazine carbamic acid derivative), PZ-NO2H+ (piperazine nitramine) and ON-PZ-NOH+ (1,4-dinitrosopiperazine). The abundance of the peak at m/z 116 (N-nitrosopiperazine) increases with time in the presence of oxygen, and a peak at m/z 145 also evolves after 15 hrs, which corresponds to protonated N,N-dinitrosopiperazine. There was found to be no appreciable formation of N-nitrosopiperazine in samples exposed to a simulated flue gas without oxygen (after t = 3.5 hours and t = 14 hours, synthetic flue gas composition 81.4 % N2, 17.8 % CO2, 0.8 % NO). The low energy collision-induced dissociation (CID) spectrum of protonated N-nitrosopiperazine (m/z 116) is presented in Figure 3. Peaks corresponding to loss of 30 (NO) and 31 (HNO) mass units dominate the spectrum. Deuterium labelling studies confirmed the identity of these losses, and that m/z 116 is indeed protonated N-nitrosopiperazine (5).

Figure 3. Low energy CID spectrum of putative N-nitrosopiperazine. Two major products are observed corresponding to loss of NO (m/z 86) and HNO (m/z 85).



The mass losses can be rationalised according to Scheme 1, with loss of 31 mass units resulting from protonation of the NO group, whereas protonation of the ring nitrogen at the 4-position leads to loss of NO•. Evidence that the nitrogen in the four position is protonated is the peak corresponding to loss of ammonia (17 mass units, m/z 99). The peaks at m/z 56, 57 originate from losses of methanimine from m/z 85, 86 respectively (refer to Scheme 1). The peak at m/z 30 (NO+) is likely to arise from charge-trapping upon dissociation of the parent ion, to produce neutral piperazine after rearrangement.

Scheme 1. Fragmentations of N-nitrosopiperazine

Although derived using different ionisation methods, there is excellent agreement between the electron ionisation mass spectrum of N-nitrosopiperazine (6) and the ESI-mass spectrum presented in Figure 3.


Detection of the Nitrosoamine Derivatives of Other Secondary Amines: N-Nitrosodiethanolamine and N-Nitroso-2-piperidinemethanol


Analysis of the reaction mixtures (t= 9.0 hrs) for 2-PM and DEA using broad-scan MS revealed the presence of peaks at M/z 145 (putative N-nitroso-2-piperidinemethanol) and M/z 135 (putative N-nitrosodiethanolamine) respectively. The ions were mass-selected and analysed using MS/MS. The low-energy MS/MS spectra are presented in Figure 4.

Figure 4. MS/MS spectrum of: (top) m/z 145 (N-nitroso-2-PM), and (bottom) N-nitroso-DEA.



Characteristic losses of 30 Da (NO) and 31 Da (HNO) are evident in both spectra, which identifies these ions as protonated N-nitroso species (albeit these peaks are significantly weaker in the MS/MS spectrum of N-nitroso-2-PM). 2-PM is a “sterically hindered” heterocyclic alkanolamine, and the observation of its N-nitroso derivative suggests that while this aspect of alkanolamine molecular structure may have a small effect on the kinetics, its chemistry with O2/NOx is otherwise typical of unhindered alkanolamines i.e. steric hindrance does not preclude the formation of certain derivatives, but the derivatives might be more prone to hydrolysis or other chemistry. This could render the detection of such species difficult.

Nitrosamine Derivative of Primary, Tertiary and ‘Hindered’ Amines To determine whether primary, tertiary and ‘hindered’ amines could form nitrosamines, 30 wt % aqueous solutions of MEA, MDEA and AMP (respectively) were also exposed to the synthetic gas mixture for up to 15 hours. The resulting reaction mixtures were analysed in the same fashion, described vide infra. No evidence of nitrosamines were found, although analysis of the MDEA mixture revealed trace amounts of an ion at M/z 106, which was determined to be protonated DEA. The origin of the DEA could be either (i) degradation of MDEA induced by the presence of NOx in the synthetic gas stream, or (ii) an impurity in the original material. MS analysis of the MDEA purchased from the manufacturer precludes the presence of DEA as an impurity in the stock. A peak at M/z 147 was also detected in negative ion mode for the MDEA reaction mixture (possibly [MDEA+NO-H]−). The MS/MS spectrum of M/z 147 was largely devoid of fragments aside from an intense peak at m/z 62 (spectrum not shown). It is concluded this ion is not a nitroso derivative but a nitrate salt, since M/z 62 corresponds to NO3−.

Mitigating Methods for the Reduction of Emissions There are a number of mitigation methods available that can reduce emissions from an amine-based PCC plant, including the removal of NOx from the flue gas stream prior to the absorption process and minimisation of gaseous emissions by a water wash. However, these methods would have a negative impact on the capital cost of the PCC process and may be viewed as impractical. One approach that is currently being investigated is the UV treatment of the solvent liquor, gaseous emissions and/or wash water to destroy any materials such as nitramines and nitrosamines that can form during the PCC process. This section reports on the thermal- and photo- stability of N-nitrosopiperazine and UV treatment as a possible mitigation strategy in the reduction of these materials from an operational amine-based PCC plant. 214 In Recent Advances in Post-Combustion CO2 Capture Chemistry; Attalla, M.; ACS Symposium Series; American Chemical Society: Washington, DC, 2012.

Effect of Heat, UV Light and NOx Concentration on the Formation of Nitrosamines


In order to establish the effect of heat and UV light on the formation of nitrosamines, a method was developed for the analysis of N-nitrosopiperazine using HPLC-MS. This method has been published in the journal Rapid Communications In Mass Spectrometry (5). The results are presented in Figure 5. Figure 5 demonstrates that N- nitrosopiperazine is thermally stable at 150 °C (typical amine regeneration conditions in a CO2 stripping column), however it does degrade slowly when exposed to UV irradiation between 400-310 nm (maximum irradiance 2.6 μW/cm2 at 350 nm).

Figure 5. Plot of nitrosopiperazine sample degradation when exposed to UV light (◊) and heat (□).

Effect of Reducing NOx Concentration on the formation of N-Nitrosopiperazine The detection and formation of the N-nitrosoamine derivatives of secondary amines, discussed at the beginning of this chapter, were investigated in the presence of a high NOx concentration (8000 ppm), much higher than that experienced in a typical flue gas stream form a coal-fired station. For this reason, NOx concentration was adjusted to resemble Australian flue gas compositions (300-700 ppm (4)) and to examine the propensity of N-nitrosopiperazine to form over longer periods of exposure. An aqueous 15 % wt piperazine solution was exposed to a synthetic flue gas with this NOx concentration (0.07 %), and the N2 concentration in the stream was adjusted to maintain the gas flow at 1.34 L/min. The experiment was run at 60 °C over 9 days (1215 L NOx, ~ 50 moles at 298 K). The reaction mixture was then sampled and analysed using multiple reaction monitoring-MS. Both transitions (M/z 86 ← 116, M/z 85 ← 116) recommended above were used in this study. The traces for each transition are shown in Figure 215 In Recent Advances in Post-Combustion CO2 Capture Chemistry; Attalla, M.; ACS Symposium Series; American Chemical Society: Washington, DC, 2012.


6 for: (i) an N-nitrosopiperazine standard, (ii) a blank injection between the standard and the unknown, and (iii) the reaction mixture. The results demonstrate that N-nitrosopiperazine forms at the lower NOx concentrations.

Figure 6. MRM traces for M/z 116 ← 86 (left) and M/z 116 ← 85 (right) for: (top row) N-nitrosopiperazine standard solution; (centre row) blank injection; (bottom row) PZ reaction mixture exposed to synthetic gas stream containing 700 ppm NOx for 9 days.

Conclusion N-nitrosamine formation is facile for secondary amines when exposed to gas streams containing NOx, however the nitrosamines formed could be destroyed using UV irradiation. While N-nitrosopiperazine appears to be stable at temperatures close to those typically used to regenerate this amine, other nitrosamines formed from alkanolamines may be more thermally labile, and mitigation strategies may not be needed. Work in the area of N-nitrosamine degradation is ongoing. It should be noted that the conditions under which this study was performed did not include solvent cycling; the effect of this process on the N-nitroso species is currently not known, although it is likely to have little effect on N-nitrosopiperazine.


References 1.

2.

3.


4.

5. 6.

Bråten, H. B.; Bunkan, A. J.; Bache-Andreassem, L.; Solimannejad, M.; Nielsen, C. J. Final report on a theoretical study on the atmospheric degradation of selected amines; Report: NILU: OR 77/2008. Stangeland, A.; Shao, R. Amines used in CO2 capture – health and environmental impacts; Report: The Bellona Foundation, Oslo, Norway, 2009. www.bellona.org. Wang, P. G.; Xian, M.; Tang, X.; Wu, X.; Wen, Z.; Cai, T.; Janczuk, A. J. Chem. Rev. 2002, 102, 1091–1134. Cottrel, A. J.; McGregor, J. M.; Jansen, J.; Artanto, Y.; Dave, N.; Morgan, S.; Pearson, P.; Attalla, M. I.; Wardhaugh, L.; Yu, H.; Allport, A.; Feron, P. H. M. Energy Procedia 2009, 1, 1003–1010. Jackson, P.; Attalla, M. I. Rapid Commun. Mass Spectrom. 2010, 24, 3567–3577. Linstrom, P.J.; Mallard, W.G. NIST Chemistry webbook, NIST standard reference database number 69. http://webbook.nist.gov/chemistry/, 2009.


Environmental Impacts of Post-Combustion Capture - American

Recommend Documents