A Way of Qualifying Amine Based Capture Technologies with ... · the Amine Qualification Toolbox. 2. Basis and Approach The research program, named “Health and Environmental Technology

1876-6102 © 2016 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).Peer-review under responsibility of the Programme Chair of the 8th Trondheim Conference on CO

2 Capture, Transport and Storage

doi: 10.1016/j.egypro.2016.01.025

Energy Procedia 86 ( 2016 ) 239 – 251

ScienceDirect

The 8th Trondheim Conference on CO2 Capture, Transport and Storage

A way of qualifying Amine Based Capture Technologies with respect to Health

and Environmental Properties

Laila Iren Helgesena*, Erik Gjernesa aGassnova SF, Dokkvn. 10, Porsgrunn, Norway

Abstract

The Norwegian full scale CO2 capture project Mongstad (CCM) gained extensive knowledge and experience during seven years of preparation and planning including activities to qualify technology. Initially amine based capture technologies were the only viable choice of technology and it was very early stated the need for more knowledge to assist assessing the health and environmental (H&E) risk of these technologies. Emissions to air of possibly carcinogenic substances were flagged as a potential showstopper in an early phase of the project. A comprehensive research program was initiated to identify and close knowledge gaps with respect to formation potential and toxicity level of these compounds known as nitrosamines and nitramines. This has been scientifically founded work done by universities, research institutes and scientific bodies. The results consist of well documented procedures for sampling and chemical analysis, atmospheric chemistry reaction schemes and kinetics, several dispersion and deposition models and simulation results of environmental fate and impact on human health. The latter has been established through own toxicity testing and extensive literature search for toxicity data. Knowledge gained during technology qualification for CCM was to a large extent site specific for Mongstad and technology specific covered by non-disclosure agreements. However, a lot of the basic H&E information needed to carry out risk assessment is of general character and possible to transfer to capture technology suppliers and capture plants under planning and installation elsewhere. This has been done by Gassnova through making publicly available an impressive set of reports, 63 all together, on its web page. This paper will present how this comprehensive research program was executed and how the CCM project used the developed methods, models and procedures in the process of qualifying capture technology for the CCM plant. All in all this practice has proven to be both well fitted and useful in differentiating between the five capture technologies from Aker Clean Carbon, Mitsubishi Heavy Industries, Siemens, Huaneng Ceri Powerspan Joint Venture and Alstom Carbon Capture competing for the CCM contract. The CCM project was terminated before final choice of technology was conducted. © 2015 The Authors. Published by Elsevier Ltd. Peer-review under responsibility of the Programme Chair of The 8th Trondheim Conference on CO2 Capture, Transport and Storage.

* Corresponding author. Tel.: +47-97952698

E-mail address: [email protected]

Available online at www.sciencedirect.com

© 2016 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).Peer-review under responsibility of the Programme Chair of the 8th Trondheim Conference on CO

2 Capture, Transport and Storage

http://crossmark.crossref.org/dialog/?doi=10.1016/j.egypro.2016.01.025&domain=pdf



240 Laila Iren Helgesen and Erik Gjernes / Energy Procedia 86 ( 2016 ) 239 – 251

Keywords: Amine based CO2 Capture; Technology Qualification; Health and Environment Risk Assessment; Nitrosamines and Nitramines;

1. Introduction

The Norwegian full scale CO2 capture project Mongstad (CCM) gained extensive knowledge and experience during seven years of preparation and planning including activities to qualify CO2 capture technologies. It is generally accepted that even though the project was stopped in 2013 this knowledge and experience should be utilized in technology qualification by later projects [1]. It is Gassnova’s perception that the CCM project would have been able to select a CO2 post combustion capture technology for the combined heat and power plant at Mongstad with qualified technology and acceptable health and environmental risk and on the basis of technology vendor competition. It is the task of Gassnova to disseminate as much as possible of the experience and learning gained from the CCM project in an applicable way to all interested parties.

The greatest learning has beyond doubt been in the area of health and environmental (H&E) risk assessment of amine based capture technologies where emissions to air of possibly carcinogenic substances were flagged as a potential showstopper in an early phase of the project. A comprehensive program was thus started to identify and close knowledge gaps with respect to formation potential and toxicity level of these compounds known as nitrosamines and nitramines.

Early theoretic studies [2] concluded that nitrosamines and nitramines could form in the atmosphere from emitted amines when exposed to OH radicals and nitrogen oxides. Experimental studies also reported nitrosamine emissions when amines were exposed to nitrogen oxides under CO2 capture process conditions [3] and nitramine formation under atmospheric conditions [4]. Thus it became evident that more knowledge was needed to enable H&E risk assessments for CO2 capture processes using amine as the capture medium. Development of chemical analysis methods became necessary to identify and quantify the nitrosamines and nitramines potentially formed and emitted. The importance of various capture process conditions favoring the formation of these compounds became important as well as the atmospheric chemistry and kinetics to be applied for amines, nitrosamines and nitramines when emitted, formed and eventually decomposed. The latter had to be established and implemented in dispersion and deposition models. This would make it possible to quantify how much of these compounds are ending up in surrounding air and drinking water in the environment around the capture plant and finally these quantities could be compared with air and drinking water quality criteria normally set by the authorities.

In 2011 Norwegian environmental authorities presented temporary guidelines for acceptable levels of nitrosamines and nitramines in air and drinking water [5]. This was done to accomplish the emission permit for the CO2 Test Centre at Mongstad (TCM), Norway. At that time limited knowledge was available on emission composition as well as quantities from a capture plant and a conservative approach was chosen based on setting the toxicity of the sum of all nitrosamines and nitramines in the emission equal to one of the most toxic nitrosamines - nitrosodimethylamine (NDMA).

Gassnova has made publicly available on its web-page [6,7] all in all 63 reports documenting the achievements gained under the CCM technology qualification program. The present paper is a summary report explaining how this knowledge was established and used by the CCM project to enable qualifying amine capture technologies intended to be used at Mongstad. The outcome is an extensive set of methods, models and procedures which together constitute the Amine Qualification Toolbox.

2. Basis and Approach

The research program, named “Health and Environmental Technology Qualification Program for Amines” and abbreviated H&E TQP Amine, was established under the framework of technology qualification, a method frequently used when introducing new technologies in industrial applications [8]. Initially a technology assessment was carried out for the relevant capture technologies to be used by CCM according to Table 1 below.

Laila Iren Helgesen and Erik Gjernes / Energy Procedia 86 ( 2016 ) 239 – 251 241

Table 1: Technology assessment criteria and scores Technology maturity Proven Limited field history New or unproven Well known within project 1 2 3 New to project 2 3 4

Where 1=no new technical uncertainties, 2=new technical uncertainties, 3=new technical challenges, 4=demanding new technical challenges

This assessment revealed solvents’ H&E properties, sampling and analysis of process liquids and emissions particularly of the nitrosamine and nitramine compounds and finally washing and demisting of emissions to air as the most demanding challenges for CO2 capture with amine based solvents. They were new to the project and new/unproven, i.e a score of 4 in Table 1. Thus, it was decided to address this identified lack of knowledge through a comprehensive research program. The program was split into three phases:

Carry out a literature survey to establish a basis of knowledge in the area of nitrosamines and nitramines including

a gap analysis proposing what is needed to meet and close the gaps Establish methods and procedures on sampling and analysis, formation and degradation chemistry as well as

dispersion models and knowledge on toxicity and test them on non-proprietary solvents to prove its relevance in amine based CO2 capture

Use the methods and procedures to qualify the different amine and amino acid based capture technologies

Several universities and scientific institutes and bodies were invited to contribute in the development work. However, limited time was available for a peer review process of all the findings and results which is the normal way of establishing new scientific knowledge. The procurement process of each study program was set up with three parallel studies whenever feasible to ensure a broadest possible scientific approach. In this way a comprehensive set of reports were produced, of which all are publicly available. Universities and scientific bodies involved in the work have as a direct spin-off published at least 10-12 peer review articles in relevant journals and at various conferences all over the world.

3. Method Development and Choices

Cutting edge work has been executed by all universities and institutes involved and valuable method development has been achieved in all phases following the compounds in question from their identification and quantification in capture process and emission samples as well as chemical and physical properties both in the atmosphere and in the environment. Resulting methodology in all areas of importance to map the behavior of nitrosamines and nitramines are described in the following sub-chapters.

3.1. Sampling and Analysis Procedures

When the CCM project was initiated there were no international standard describing specifically the sampling and analysis of amines and degradation products of amines in flue gas. Three different institutes/organizations took part in the work of mapping the applicable standards and proposing development work needed to achieve a set of new standards. Standardization work is time consuming thus focus was put on establishing well documented sampling and analytical procedures applicable for amines and degradation products. All procedures as developed and applied in the CCM-project are documented in the “Sampling and analytical procedures” document issued by Statoil [9]. Apart from describing each analytical procedure, comprising specific non-alcohol based and alcohol based nitrosamines, total nitrosamines, nitramines, solvent amines, alkyl amines, aldehydes and ammonia also advice on how to carry out sampling, sampling preservation as well as transport conditions is given.

It was very early concluded that iso-kinetic sampling was necessary in gas phase to ensure a representative sample was taken [10,11,12]. Amines used in CO2 capture and the corresponding nitrosamines and nitramines are highly water soluble substances and any aerosol formation or entrained liquid would bring these compounds to the environment in the aqueous phase. The sampling train for collecting amines for analysis consisted of a condensate


trap followed by impingers filled with sulphuric acid. The same set up was used for collecting nitrosamines and nitramines only by exchanging the sulphuric acid with a sulfamic acid solution.

A report validating the analytical procedures was issued [13] by the chemical analysis contractor carrying out all the analyses work during the technology qualification program. High resolution LCMS was used for analyzing nitrosamines and nitramines. Worth mentioning, is the total nitrosamine method [14] which is a useful tool to determine the level of total nitrosamines including both identified and unidentified nitrosamines. The method is based on direct injection of a preserved sample into a reaction chamber of tri-iodide coupled to a chemilluminescense detector. This method also fitted very well with the quality criteria set by the Norwegian authorities where the sum of all nitrosamines was to be used in risk assessment [5]. Two recognized peer review articles have been published under this initiative [15,16].

3.2. Process formation

Solvent-based post-combustion plants will have minor release of amine and amine degradation products to the atmosphere along with the cleaned flue gas. Operational conditions as well as the specific solvent used will influence quantity and composition of emission to air [17]. Through several external studies [18,19,20] a solvent test protocol and a rig design for investigating solvent stability were developed. The primary purpose of the rig and protocol was to identify formation of possible harmful components and verify the analysis methods developed by CCM on actual solvent and condensate matrices. The rig allows stress testing of the solvent at different operating conditions such as high temperature, high O2 and NOX content in the feed gas. A rig prototype was developed and fabricated by Sintef [21], and the test protocol was further refined [22]. See Figure 1 below for set up and process flow diagram of the Solvent Degradation Rig (SDR).

Figure 1: Set-up and process flow diagram of the Solvent Degradation Rig (SDR) (Image courtesy of Sintef)

Test conditions used when executing the test protocol on monoethanolamine (MEA) are listed in Table 2 below. This simulates real capture conditions both in the absorber and stripper through a closed loop absorber/desorber cycle.


Absorber temperature was 40oC and the lean and rich loading around 0.2 and 0.45 mole CO2/mole amine respectively. Table 2 shows a wide range of O2, temperature and NOX conditions. Exact testing conditions were adapted to each of the solvents actually tested by the CCM project.

Table 2: Process conditions for a 14 weeks’ SDR test protocol using 30% MEA Condition Tstripper [oC] O2 [mol%] NOX [ppmv] Duration Standard 120 12 5 5 weeks High O2 120 18 5 3 weeks High stripper T 140 12 5 3 weeks High NOX 120 12 50 3 weeks

Experiences with the usage of the SDR proved very useful in demonstrating both formation and destruction of nitrosamines under extreme capture process conditions as well as learning and experience for the development of analytical tools already mentioned. A peer review article has been published under this initiative [21].

3.3. Atmospheric Chemistry

Theoretical studies had shown that nitrosamines and nitramines could form in the atmosphere from amines when the latter is emitted from CO2 capture plants [23]. This was confirmed through experiments in the Euphore reactor in Valencia, Spain under the Atmospheric Degradation of Amines (ADA) program [4]. The first approach was a conservative worst case scenario where a certain percentage of the emitted amine was instantaneously converted to nitrosamines and nitramines at stack exit, being 2 and 7 % respectively [24] followed by a dispersion simulation. For a more precise quantification atmospheric chemistry and kinetics were established after thorough investigation of which conditions did contribute and which did not.

As illustrated in Table 3 below, it became clear from the work done that formation of nitrosamines and nitramines are completely dominated by gas phase reactions. Of these the reaction of amines with OH radicals during daytime [4] and NO3 radical during nighttime [25] both forming an amino radical are the reactions necessary to consider. The role of the chlorine and bromine radicals have shown no importance under conditions relevant for Mongstad and have not been included [26,27,28]. Loss processes for nitrosamines and nitramines are of minor importance apart from the photolysis of nitrosamines during sunlight exposure [29,30]. Nitrosamine and nitramine formation from amines reacting in aqueous phase are insignificant and can be neglected [31,32]. Aqueous phase is only considered as a loss process for amines in the atmosphere due to their solubility in water. Partitioning to the aqueous phase based on Henry Law constants are thus included in the overall atmospheric model.

Table 3: Importance of atmospheric chemistry conditions for amines, nitrosamines and nitramines Nitrosamine/Nitramine Chemistry Gas Phase Aqueous Phase Formation, Daytime Amine + OH None important Formation, Nighttime Amine + NO3 None important Destruction, Daytime Nitrosamine + hν None important Destruction, Nighttime None important None important

The atmospheric chemistry model which was established by the CCM-project had a generic approach and meant

to be applicable for all amines and amino acids although the investigations have been based on the following amines only: ethanolamine, diethanolamine, triethanolamine, pyrrolidine, methylamine, dimethylamine, trimethylamine, ethylamine, diethylamine, triethylamine and piperazine.

Based on the findings in Table 3 a generic gas phase reaction scheme has been put up for an amine when a hydrogen atom is subtracted from the amino group, leaving an amino radical with three different routes of reaction as described by Figure 2 below.


Figure 2: A generic atmospheric gas phase chemistry scheme for amines (Image courtesy of Tel-Tek)

A hydrogen atom can also be subtracted from the alkyl or hydroxyl part of the amine both leading to non-toxic products indicated in Figure 2 above by a branching ratio (α) of 10-60% to amino radical with further possible reaction to nitrosamines and nitramines. It has been proved [25,27] that the ratios between different reaction coefficients of the amino radical are the following:

k2/k3 = 0.6 (1) k4/k3 = 0.22 (2) k5/k3 = 3 x 10-7 (3)

A fixed value was set on the k3 = 3.2 x 10-13 taken from the literature [33]. This leads to a complete setup of

equations for the atmospheric formation of nitrosamines and nitramines:

(4)

(5) Where BNitrosamine is 0 for primary amines and else:

(6)

(7)

The reactions given above are valid for the reaction of amines with the OH radical. A similar set of reaction

coefficients and equations for the reaction with the NO3 radical can be derived from the Wayne correlation [25]. The only loss reaction to be implemented in the dispersion model is the photolysis of the nitrosamines and this is

done by the following expression:


(8) This equation will apply to directly emitted nitrosamines as well as nitrosamines formed from amines after emission

[29]. JNO2 is the photolysis rate of nitrogen oxide, which is a frequently used reference number in photolysis chemistry. The approach when implementing the atmospheric reaction into the dispersion models has been of conservative

nature, i.e. not to underestimate formation reactions and not to overestimate loss processes, and can be summarized as follows:

Take the sum of all non-volatile amines in the emission Take the sum of all volatile amines in the emission Take the sum of all nitrosamines in the emission Take the sum of all nitramines in the emission Check if the reaction scheme with accompanying rate data and branching ratios (Figure 2) can be used directly or

if adjustment is needed (e.g. no imine formation) Include formation scheme and photolysis rate expression in the dispersion model Allow water partitioning by the equilibrium Henry law constant (select typical values for the sums) Let the dispersion model run and calculate the relevant parameters and the output results

This will result in an annual average ground level concentration of nitrosamines and nitramines in the surrounding

air and in a water source as is explained further in the next chapter. This approach illustrated the fact that chemistry takes time and that the concentration in air and drinking water was reduced by a factor of 10 compared to the worst case approach of instant reaction at the stack exit.

Several peer review articles and conference papers have been published as a spinoff of this comprehensive work, e.g. [34].

3.4. Dispersion, Deposition and Environmental Fate

The small amounts of emitted and formed nitrosamines and nitramines are likely not detectable in the environment. Impacts can thus only be simulated by dispersion modeling and it is normally a recommended approach to do this by using more than one model.

Dispersion models selected for use by CCM were all equipped with possibilities to include chemical reactions in the setup. Additionally they were all capable of running annual average ground level concentrations and dry and wet deposition rates within acceptable run time and computer cost.

A generic figure describing what a dispersion model normally is set up to handle is seen in Figure 3 below.

Figure 3: A generic illustration of dispersion model properties (Image courtesy of CERC)


Several models were evaluated and ADMS, an in-house Gaussian plume model developed by Cambridge Environmental Research Consultants (CERC), and Calpuff an open source Gaussian puff model supported by United States Environmental Protection Agency (USEPA) and in the CCM project operated by Det Norske Veritas (DNV), was selected for implementation of amine chemistry. In the last round also the Norwegian Institute for Air Research (NILU) was running simulation cases with their WRF-EMEP model, which combines the Weather Research and Forecasting (WRF) numerical weather prediction model (NWP) with the Unified European Monitoring and Evaluation Program model (EMEP). Both models are open source, well documented and widely used. For more thorough description of all models it is recommended to read the following reports [35,36,37].

Experience from the atmospheric modeling gave good and accurate risk pictures of the different capture technologies. Even though the results differ somewhat between the different dispersion models the combined usage of more than one model gave a broader understanding of the estimated exposure. An overall experience is that by incorporating atmospheric chemistry and aqueous partitioning the model results are still conservative but much more realistic than a worst case approach [38].

Calculations of annual average ground level concentrations of sum nitrosamines and nitramines demonstrated the characteristics of the weather conditions at Mongstad with rather strong and prevailing wind from north and southwest and peak concentrations very close to or even within the refinery. At this short distance very little amine would have had time to reacted to nitrosamine/nitramine making the emitted nitrosamines an important constituent of the annual average ground level concentration [39,40].

At and near Mongstad the precipitation is one of the highest in Norway with up to 3000 mm rainfall annually. This in combination with highly soluble amines used for capture and corresponding nitrosamines and nitramines formed it was expected that wet deposition was the most important source of pollution ending up in lakes and rivers. Calculations did however show that dry deposition, i.e. compounds in air landing on ground due to its reactivity and interaction with the roughness of the landscape did also contribute a considerable part of the total deposition [36].

Deposition is giving the value of how much of a component is falling to the surface per time and area. From that value there is a need to calculate how much is ending up in a certain drinking water source and the corresponding concentration reached. This is done by establishing a run off model, where the catchment of a lake is modeled based on the knowledge of that lake and the transport of water to and from it. Centre for Ecology and Hydrology (CEH), based in United Kingdom and Norwegian Institute for Water Research (NIVA) based in Norway were engaged to do this modeling work using a so-called fugacity 3-model describing the lake catchment with different compartments and the transport between them as indicated in Figure 4 below. For further reading about this model see [41].

Figure 4: A generic illustration of a fugacity 3 model describing a lake catchment (Image courtesy of DNV)

The actual drinking water recipient chosen was the lake Rotevatn approximately 10 km south of the Mongstad site.

Specific data for this lake can be found in [42]. During the time of transport to and retention in the lake all organic


substances are subject to biodegradation at a certain rate. Limited data was available on biodegradation of different nitrosamines and nitramines. It was though generally stated that nitramines were more stable in the environment than nitrosamines mainly due to the fact that nitrosamines are prone to photolysis while nitramines are not. EPISuite data has been generated for model compounds of both nitrosamines (74 g/mole) and nitramines (90 g/mole). Both primary degradation, measured in days, to the first metabolite (fast) and ultimate, complete degradation to CO2 and water (slow) have been estimated in all the compartments illustrated in Figure 4 above.

Table 4: Estimated degradation time (days) of nitrosamines and nitramines in different lake compartments Compound Aquatic degradation Soil degradation Atmosph.

degradation Sediment degradation

Primary Ultimate Primary Ultimate Primary Primary Ultimate Nitramine 2.3 15 4.6 30 0.72 21 135 Nitrosamine 0.7 23 5 38 4.2 2.7 207

Using the data in Table 4 together with all other required input to the model an annual average concentration of

sum nitrosamines and nitramines in the drinking water source Rotevatn could be calculated.

3.5. Health Impact and Toxicity

Efforts have been made to collect toxicity data for nitrosamines and nitramines through literature searches and generation of new data. As already mentioned nitrosamines and nitramines are potential carcinogens and testing on rats has led to classifying two nitrosamines, i.e. N-nitrosodimethylamine and N-nitrosodiethylamine, as probably carcinogenic to humans while a group of seven nitrosamines, as possibly carcinogenic according to International Agency for Research on Cancer [43]. These are the most potent nitrosamines and based on a total of 23 experiments on rats an average value for T25† and HT25‡ were calculated for these most potent carcinogens [44].

T25=0.075+0.045 mg/kg bw/d, range 0.0088 – 0.19 mg/kg bw/d HT25=0.021+0.013 mg/kg bw/d, range 0.0023 – 0.054 mg/kg bw/d Other nitrosamines like for instance N-nitrosodietanolamine is 25 times less potent than the above calculated

average value, N-nitrosopiperazine 100 times less potent and N-nitrosodiphenylamine more than 1000 times less potent. The two only nitramines where test data is available are N-nitromethylamine and N-nitrodimethylamine being at least 10 times less potent according to the above calculations.

The CCM project did also carry out own toxicity testing on a set of five nitramines. Apart from acute oral toxicity, cytotoxicity, eco-toxicity, sensitization, skin irritation and skin and eye corrosion also three individual genotoxicity tests were applied. All results can be found in [45]. Efforts were made to establish Quantitative Structural Activity Relationship (QSAR) as a tool to estimate carcinogenicity of non-tested nitrosamines and nitramines [46]. However, data on experimentally derived carcinogenicity of nitrosamines was too sparse to support such an application.

4. Discussion

4.1. Applicability and Constraints

It can be argued that the results obtained by the CCM project are site specific and only valid for the Mongstad region and the west coast of Norway. To a certain extent this is correct. Any new capture project will have to consider changes in composition of flue gas to be treated and choice of technology including type of solvent and operating

† T25=Chronic dose rate usually expressed in units of mg/kg bodyweight/day which will give tumors at a specific tissue site in 25% of the animals after correction for spontaneous incidents and within the standard lifetime of the species ‡ HT25=T25/(bwhuman/wanimal)0,25 ,an adjustment formula to interpolate between animals and humans


conditions influencing the emission profile. Local conditions such as meteorology and topography as well as ambient air quality, especially the atmospheric NOX content, will in deed decide the dispersion pattern and exposure levels of harmful compounds. However, the methodology on how to carry out technology qualification and assess the health and environmental risk as described in this document is of generic character and could be applied for other amine and amino acid based CO2 capture projects elsewhere in the world.

4.2. Simplifications and still undone work

Such an enormous amount of data and results called for a need to prioritize and simplify while still retaining a realistic approach however on the conservative side. Particularly the dispersion modeling can be a time and resource consuming exercise when a whole year is to be simulated with simultaneous chemistry and dispersion running. It is thus important to keep the run time of the models as low as possible by making the number of compounds and chemical reactions low without compromising the validity of the results obtained. Careful choice of model compounds and reactions has been a must and has led to a manageable set of data within a reasonable timeframe.

Model substances of nitrosamines and nitramines have also been applied in the run off model. This has of course been feasible due to the fact that guideline air and drinking water quality criteria are at present in Norway based on the sum of all nitrosamines and nitramines. For a specific solvent chosen for a specific project in the future both toxicity and degradation data for nitrosamines and nitramines formed should be established and used for an even better precision than was obtained by the CCM project.

4.3. Quality Assurance

A group of experts on analytical chemistry, atmospheric modeling, human health and toxicology has carried out an independent evaluation of the Amine Qualification Toolbox. The conclusions from this evaluation states that the tool box is adequate and fit for the purpose of comparing and choosing between amine based capture technologies to be considered for use at the full scale Mongstad plant [44]. A few points of concern and recommendations for improvements have been put forward. The main finding was that the used tolerable cancer risk level of 10-6 is found not to be in accordance with the Norwegian drinking water regulations which use the World Health Organisation (WHO)’s guideline of 10-5 and also not according to the risk level accepted for many chemicals both under the United States and European Union legislation framework.

4.4. Health and Environmental Risk Assessment

The CCM project based its risk assessment on the same assumption as recommended by Norwegian Institute of Public Health (NIPH) [5]:

Risk estimate calculated for Nitrosodimethylamine (NDMA) shall be used for the total concentration of

nitrosamines and nitramines in air and drinking water A maximum limit of 0.3 ng/m3 of sum nitrosamines and nitramines shall be used for air quality limitations Maximum 4 ng/l of sum nitrosamines and nitramines shall be used for drinking water quality limitations

In addition, Rotevatn was selected by CCM as the drinking water recipient in the Mongstad region. The selected acceptance criteria are on the conservative side. However, this choice added value to the project

through a less complex evaluation, less need for animal testing and a simplified communication to third parties. By using one value for sum of nitrosamines and nitramines the risk evaluation was also less complex with respect to solvent information and confidentiality and this approach could with success be used in combination with the total nitrosamine analysis method as described above.

The Norwegian environmental authorities granted the emission permit for the Technology Centre Mongstad (TCM) in 2011 [47] based on the same acceptance criteria as listed above as well as rather limited understanding of emission profiles and the process and atmospheric chemistry of amines, nitrosamines and nitramines. As of today after having


developed a broader and deeper understanding of the properties of nitrosamines and nitramines both in general and in amine based capture processes as such it is clear to see that a risk assessment based on individual substances and according to the WHO guidelines would be the recommended procedure.

5. Conclusions

CO2 capture with amine based post combustion technologies can create potentially toxic and carcinogenic degradation products within the process as well as within the atmosphere when NOX and amines can react with each other. These substances may present a risk of cancer depending on the level and duration of the exposure. For the CCM project to be able to evaluate the proper H&E risk from amine based CO2 post combustion capture plants a set of thorough methods and procedures named the Amine Qualification Toolbox [48] were developed. The toolbox consists today of methods, procedures and models within the following scientific areas: Solvent degradation rig and test protocol Liquid and iso-kinetic gas sampling Sample preservation, logistics and workup Analytical procedures Atmospheric emission chemistry, both wet and dry Dispersion and lake catchment modelling Toxicity assessment of nitrosamines and nitramines

This toolbox was used as the scientific framework of qualifying post combustion capture technologies tendering

for the CCM project and proved to be a very powerful and useful tool in the assessment of each technology. The CCM project was terminated before a specific technology was chosen for execution. Even though the chosen criteria for qualification were on the conservative side, the CCM project would if continued been able to select a capture technology with both acceptable H&E risk and technology vendor competition. It is highly recommended that this practice is implemented in any future post combustion project based on amine or amino acid solvents.

Acknowledgements

Gassnova is thankful to all contributors to the development of the Amine Qualification Toolbox being either universities or institutes in the forefront on this area of research or the companies involved in technology qualification and testing of capture technology for the CCM project. It is an impressive and also important work which has been carried out under this research and development program.

The authors greatly acknowledge Gassnova for allowing time and resources to prepare this publication which hopefully will contribute to a general acceptance of the Amine Qualification Toolbox as the state of the art way of qualifying amine based CO2 capture technologies, and assessing their health and environmental risk.

References

[1] Kaasa S, Presentation at Tekna Conference: “CO2 konferansen 2014”, Trondheim, 09.01.2014 [2] Knudsen S, Karl M, Randall S, Summary Report: Amine Emissions to Air during Carbon Capture, NILU report no OR8/2009, ISBN 978-82-

425-2076-0, March 2009, Kjeller, Norway [3] Fostås B, Gangstad A, Nenseter B, Pedersen S, Sjøvoll M, Sørensen AI, Effects of NOx in the flue gas degradation of MEA, Energy Procedia,

Volume 4, p. 1566-1573, 10th International Conference on Greenhouse Gas Control Technologies (GHGT 10) [4] Nielsen, C. et.al., Atmospheric Degradation of Amines (ADA), ISBN 978-82-992954-7-5, University of Oslo, 2012 [5] Låg M, Lindeman B, Instanes C, Brunborg G, Schwarze P, Health effects of amins and derivatives associated with CO2 capture, ISBN 978-82-

8082-462-2, Norwegian Institute of Public Health, April 2011 [6] Link to Gassnova web page: http://www.gassnova.no/no/prosjekter52/fullskala-mongstad/resultater-fra-de-innledende-studiene-av-helse-og-

miljøaspekter-ved-bruk-av-aminer-i-co2-fangst [7] Link to Gassnova web page: http://www.gassnova.no/no/prosjekter52/fullskala-mongstad/resultater-fra-andre-runde-med-studier-av-helse-og-

miljøaspekter-ved-bruk-av-aminer-i-co2-fangst [8] Technology Qualification, Recommended Practice, DNV-RP- A203, Oslo, July, 2013


[9] Gangstad A, Nenseter B, Sampling and analytical procedures for potentially harmful components related to amine based CO2-capture, Statoil doc, 01.09.2014, Trondheim, Norway

[10] Halliburton B, Day S, Lavrencic S, Riley K, Azzi M, Establishing sampling and analytical procedures for potentially harmful components from post-combustion amine based CO2 capture, Task1: Design of Sampling Points for Treated Flue Gas, CSIRO report no EP 104693, 08.09.2010, Newcastle, Australia

[11] Wittgens B, Einbu A, Zahlsen K, Hunnes KE, Establish sampling and analytical procedures for potentially harmful components post combustion amine based CO2 capture, subtask 1: Design of sampling points in treated flue gas, Sintef report no F16137, 15.12.2010, Trondheim, Norway

[12] Järvinen E, Hoffren H, Vilhunen S, Design of sampling points in treated flue gas (manual and online sampling), Ramböll report no 89105265-01, 20.08.2010, Espoo, Finland

[13] Keränen J, Teräsahde P, Tykkä T, Jarova E, Järvinen E, Luste S, Chemical analysis of samples from CO2-capture test facilities – Validation report, Ramböll report no 82141163-01, 04.01.2013, Espoo, Finland

[14] Mitch W, Optimization and Application of a Total N-nitrosamine Assay to Solvents and Washwaters Associated with an Array of Amine Solvent Structures, Final report, 13.12.2012, Yale University, New Haven, USA

[15] Dai N, Shah A, Hu L, McKague B, Mitch W, Measurement of Nitrosamine and Nitramine Formation from NOX Reactions with Amines during Amine-Based Carbon Dioxide Capture for Postcombustion Carbon Sequestration, Environmental Science & Technology, 2012, 46, 9793 - 9801

[16] Dai, N., W. Mitch, Effects of Flue Gas Composition on Nitrosamine and Nitramine Formation in Postcombustion CO2 capture Systems, Environmental Science & Technology, 2014, 48, 7519 - 7526

[17] Mitch W, Predicted products of solvent decay and experimental plan, Task 1 report, CCM Project, 16.01.2012, Porsgrunn, Norway [18] DaSilva E, Booth A, Loeppky R, Ma’Mun S, Karl M, Heimstad E, Johnson M, Feilberg K, Protocol for Evaluation of Solvents – Process and

Atmospheric Chemistry, Sintef report no 801572, 19.06.2010 [19] Eimer, D., S. Fredriksen, H. Herrmann, D. Hoffmann, J. Hovland, K. Jens, C. Nielsen, Test Protocol for the evaluation of amine solvents for

Carbon Capture, Tel-Tek report no 2210040-1, 25.10.2010, Porsgrunn, Norway [20] Jackson, P., D. Angove, M. Azzi, M. Attalla, CO2 Capture Mongstad – Theoretical evaluation of the potential to form and emit harmful

compounds, Task 3: Test Protocol, CSIRO report, Rev. 03, 18.10.2010, Newcastle, Australia [21] A, DaSilva E, Haugen G, Grimstveit A, Lauritsen KG, Zahlsen K, Vassbotn T, A new test rig for studies of degradation of CO2 absorption

solvents at process conditions; comparison of test results and pilot plant data for degradation of MEA, Energy Procedia, Volume 37, Pages 717-726, GHGT-11

[22] Haugen G, Einbu A, Chikukwa A, DaSilva E, Grimstveit A, Process Protocol, SINTEF report F22608, Trondheim, Norway, 27.04.2012 [23] Bråten H, Bunkan A, Andreassen LB, Solimannejad M, Nielsen CJ, Final report on a theoretical study on the atmospheric degradation of

selected amines, NILU report no OR77/2008, 16.03.2009, Kjeller, Norway [24] Karl M, Brooks S, Wright R, Knudsen S, Worst Case Studies on Amine Emissions from CO2 Capture Plants (Task 6), NILU report no

OR78/2008, 12.03.2009, Kjeller, Norway [25] Nielsen CJ, Atmospheric Chemistry - Dark Chemistry, Tel-Tek report no 2211030-DC06, 01.11.2011, Porsgrunn, Norway [26] Feilberg K, University of Copenhagen, Atmospheric Chemistry – Chlorine Chemistry, Sintef report no F21201, 18.11.2011, Trondheim,

Norway [27] Nielsen CJ, Atmospheric Chemistry - Chlorine Chemistry, Tel-Tek report no 2211030-CC07, 08.11.2011, Porsgrunn, Norway [28] Nielsen CJ, Bunkan A, Tang Y, Atmospheric Chemistry – Chlorine Chemistry. Comparison of amine+Cl and amine+Br chemistry, Tel-Tek

report no 2211030-CC08, Porsgrunn, Norway [29] Nielsen CJ, Atmospheric Chemistry - Nitrosamine Photolysis, Tel-Tek report no 2211030-NP07, 28.10.2011, Porsgrunn, Norway [30] Larsen RW, University of Copenhagen, Atmospheric Chemistry – Nitrosamine Photolysis, Sintef report no 201202, 18.11.2011, Trondheim,

Norway [31] Herrmann H, Weller C, Atmospheric Chemistry - Aqueous phase chemistry, Tel-Tek report no 2211030-AQ05 v2, 23.11.2011, Porsgrunn,

Norway [32] Karl M, Herckes P, Mitch W, da Silva EF, Atmospheric Chemistry - Aqueous phase chemistry, Sintef report no F23082, 25.05.12, Trondheim,

Norway [33] Lazarou YG, Kambanis KG, Papagiannakopoulos P, Journal of Physical Chemistry, 1994, 98, 2110-2115 [34] Nielsen CJ, Herrmann H, Weller C, Atmospheric chemistry and environmental impact of the use of amines in carbon capture and storage

(CCS), Chem. Soc. Rev. 2012, 41, 6684-6704 [35] Yiannoukas S, Morale G, Vernon G, Modelling atmospheric dispersion for components from Post-combustion Amine-based CO2 Capture,

DNV report no EP024442, 05.10.2010, London, UK [36] Price C, Stocker J, Carruthers D, Strickland S, McHugh C, Modelling atmospheric dispersion of components from Post combustion Amine

based CO2 Capture, CERC report no FM867/R2/10, 11.10.2010, Cambridge, UK [37] Knudsen S, Cassiano M, Karl M, Slørdal L, Tarrasón L, Modelling atmospheric dispersion for components from Post combustion Amine

based CO2 Capture, NILU report no OR../2009, Kjeller, Norway [38] Price C, Atmospheric Chemistry Modelling - Variation Order 2 Dispersion sensitivity analysis, CERC report no FM901/VO2/R2/11,

18.11.2011, Cambridge, UK [39] Fowler T, Vernon G, Atmospheric Chemistry Modelling of Compounds from Post-combustion Amine-based CO2 Capture - Exec summary,

DNV report no PP011013, 14.06.2012, London, UK [40] Price C, Atmospheric Chemistry Modelling -Executive summary, CERC report no FM901/ES2/12, 11.05.2012, Cambridge, UK


[41] Yiannoukas S, Morale G, Williams R, Johnson A, Deposition and Soil Transport Modelling of Compounds from Post-combustion Amine-based CO2 Capture, DNV report no PP011015, 04.11.2011, London, UK

[42] Link to the national lake data base: http://www.nve.no/en/Water/NVEs-geographic-databases/The-National-Lake-Database-Innsjodatabasen/ [43] IARC Monographs on the Evaluation of Carcinogenic Risk to Humans Volume 17 (1978) Some N-Nitroso Compounds, Lyon, France [44] Baklanov A, Kromhout H, Sanner T, Greibrokk T, Report from Amine Reference Group after review of the TQP Amine phase 2 studies,

27.09.2012, Gassnova, Porsgrunn, Norway [45] Dye C, Fjellsbø L, Dusinska M, Nitramine Analysis Procedure Development and Screening Toxicity Study, NILU report no OR15/2011,

13.05.2011, Kjeller, Norway [46] Harju, M, Ravnum S, Pran E, Grossberndt S, Fjellsbø L, Dusinska M, Heimstad E, Alternative Approaches to Standard Toxicity Testing,

NILU report no OR71/2011, 16.12.2011, Kjeller, Norway [47] Climate and Pollution Agency, Permit for activities pursuant to the Pollution Control Act for CO2 Technology Centre Mongstad DA,

16.11.2011, Oslo, Norway, http://www.tcmda.com/en/Press-center/News/2011/Discharge-permit-granted/ [48] Gjernes E, Helgesen LI, Maree Y, Health and environmental impact of amine based post combustion CO2 capture, Energy Procedia, Volume

37, 2013, p. 735-742, 11th International Conference on Greenhouse Gas Control Technologies (GHGT 11)

Documents

A Way of Qualifying Amine Based Capture Technologies with ... · the Amine Qualification Toolbox. 2. Basis and Approach The research program, named “Health and Environmental Technology