Scholarly article on topic 'Emission Results of Amine Plant Operations from MEA Testing at the CO2 Technology Centre Mongstad'

Emission Results of Amine Plant Operations from MEA Testing at the CO2 Technology Centre Mongstad Academic research paper on "Earth and related environmental sciences"

CC BY-NC-ND
0
0
Share paper
Academic journal
Energy Procedia
Keywords
{MEA / monoethanolamine / "atmospheric emission" / nitrosamine / nitramine / "amine degradation products"}

Abstract of research paper on Earth and related environmental sciences, author of scientific article — Anne K. Morken, Bjarne Nenseter, Steinar Pedersen, Milan Chhaganlal, Jane K. Feste, et al.

Abstract Extensive atmospheric emission monitoring has been conducted at the CO2 Technology Centre Mongstad (TCM DA) during amine based post-combustion CO2 capture. The TCM DA amine plant was operated with an aqueous monoethanolamine (MEA) solvent system, treating flue gas from a combined heat and power (CHP) plant. Emission monitoring was conducted by a Fourier Transform Infrared (FTIR) Spectroscopy analyzer, a Proton Transfer Reaction Time-of-Flight Mass Spectrometry (PTR-TOF- MS) analyzer, and manual isokinetic sampling followed by off-line analysis in the laboratory. Atmospheric emissions of MEA were very low throughout the entire campaign, ranging from a few to a few hundred parts per billion (ppb, 1 ppb = 10-9 v/v). Atmospheric emissions of MEA amine based degradation products such as nitrosamines and nitramines were below detectable levels. Atmospheric emissions of ammonia (NH3) were in the low ppm range. Methylamine was emitted at low ppb range. Absorber wash water sections were found to effectively reduce atmospheric emissions from amine based solvent system.

Academic research paper on topic "Emission Results of Amine Plant Operations from MEA Testing at the CO2 Technology Centre Mongstad"

CrossMark

Available online at www.sciencedirect.com

ScienceDirect

Energy Procedia 63 (2014) 6023 - 6038

GHGT-12

Emission results of amine plant operations from MEA testing at the

CO2 Technology Centre Mongstad

ab b b ab

Anne K. Morken ' , Bjarne Nenseter , Steinar Pedersen , Milan Chhaganlal' , Jane K. Festeab, Rita B0e Tyborgnesab, 0yvind Ullestadab, Helge Ulvatnab, Liang Zhuc, Tomas Mikovinyc, Armin Wisthalerc, Toine Centsad, Otto M. Badee, Jacob Knudsene, Gelein de Koeijerb, Olav Falk-Pedersenaf, Espen S. Hamborgab*

aCO2 Technology Centre Mongstad (TCMDA), 5954 Mongstad, Norway bStatoilASA, P.O. Box 8500, 4035 Stavanger, Norway cUniversity of Oslo, Department of Chemistry, P.O. Box 1033 Blindern, 0315 Oslo, Norway dSasol Technology, PO Box 5486, Johannesburg 2000, South Africa eAker Solutions, P. O. Box 222, 1326 Lysaker, Norway fGassnova SF, Dokkvegen 10, 3920 Porsgrunn, Norway

Abstract

Extensive atmospheric emission monitoring has been conducted at the CO2 Technology Centre Mongstad (TCM DA) during amine based post-combustion CO2 capture. The TCM DA amine plant was operated with an aqueous monoethanolamine (MEA) solvent system, treating flue gas from a combined heat and power (CHP) plant. Emission monitoring was conducted by a Fourier Transform Infrared (FTIR) Spectroscopy analyzer, a Proton Transfer Reaction Time-of-Flight Mass Spectrometry (PTR-TOF-MS) analyzer, and manual isokinetic sampling followed by off-line analysis in the laboratory.

Atmospheric emissions of MEA were very low throughout the entire campaign, ranging from a few to a few hundred parts per billion (ppb, 1 ppb = 10-9 v/v). Atmospheric emissions of MEA amine based degradation products such as nitrosamines and nitramines were below detectable levels. Atmospheric emissions of ammonia (NH3) were in the low ppm range. Methylamine was emitted at low ppb range.

Absorber wash water sections were found to effectively reduce atmospheric emissions from amine based solvent system.

* Corresponding author. Tel.: +47 56 34 52 20. E-mail address: eham@tcmda.com

1876-6102 © 2014 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/3.0/).

Peer-review under responsibility of the Organizing Committee of GHGT-12

doi:10.1016/j.egypro.2014.11.636

© 2014TheAuthors. Publishedby ElsevierLtd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/).

Peer-review under responsibility of the Organizing Committee of GHGT-12

Keywords: MEA, monoethanolamine, atmospheric emission, nitrosamine, nitramine, amine degradation products

1. Introduction

The CO2 Technology Centre Mongstad (TCM DA) has in collaboration with partners undertaken several months test using the non-proprietary aqueous monoethanolamine (2-aminoethanol, MEA) solvent system at 30 wt% and 40 wt% in an attempt to characterize the performance and atmospheric emissions from such operations [1,2,3]. The operations were carried out at a considerably large scale of about 50.000 Sm3/h of flue gas supply flow rates from a combined heat and power (CHP) plant, as described elsewhere [1,2,3]. In the CHP plant, the natural gas is combusted in a gas turbine and the flue gas content and characteristics are similar to those of a combined cycle gas turbine (CCGT) power plant. TCM DA has made significant investment in equipment and instrumentation for monitoring of stack emissions. Continuous efforts are being done to improve sampling methods, sampling lines and the instrumental analysis.

Quantitative emission data from a representative CO2 capture plant is one remaining knowledge gap in the assessment of health and environmental risks posed by the amine-based post combustion capture (PCC) technology [4]. A health risk analysis for the emissions to air from the amine plant TCM DA was recently published [5]. The emission permit granted to TCM DA by the Norwegian Environmental Agency (Milj0direktoratet) in November 2011 regulates the emission levels for solvent amines, alkylamines, aldehydes and ammonia [6]. It also sets requirements for online monitoring and how to calculate the nitrosamine and nitramine environmental concentrations by a dispersion calculation method. The air and drinking water concentrations of 0.3 ng/m3 and 4 ng/L respectively were associated with negligible excess risk level for cancer (10-6) after lifelong exposure to nitrosodimethylamine (NDMA). Since all amines that are emitted to air from the absorber stack may undergo photo-oxidation in the atmosphere and be converted to nitramines or nitrosamines they will contribute to the environmental concentrations as calculated by the dispersion simulation method. In the granted permit the total sum of nitrosamines and nitramines must be below the given limits. Therefore both amine emissions and direct emissions of nitrosamines and nitramines will contribute to the total environmental budget of the harmful compounds.

Although sampling and analysis of flue gases in general are well known, the wet flue gas containing solvent amine, amine degradation products and other trace components give many sampling and analytical challenges. Very limited standard methods are established for such a task. Several studies were undertaken by international experts for the, now terminated, Carbon Capture Plant Mongstad (CCM) project, and much of the work is available for the public [7]. The CCM project developed a toolbox for qualifying amine based solvent technologies, consisting of the steps liquid sampling, isokinetic gas sampling, sample preservation and sample logistics, sample work-up and analytical procedures, atmospheric chemistry including dry and wet deposition, dispersion modelling including local Mongstad weather conditions, toxicology assessment of major degradation products as nitramines and nitrosamines, solvent degradation rig and test protocol for solvent stress testing as well as process emission reducing technologies. The analytical measurement chain was essential in the toolbox and it is also the basis for the current work.

This work is part of a continuous effort of gaining better understanding of the performance potential of the nonproprietary aqueous MEA solvent system, conducted by TCM DA and its affiliates and owners, in order to test, verify, and demonstrate CO2 capture technologies [1, 2, 3]. The purpose of the current work is to provide results which quantify the amounts and the compositions of atmospheric emissions sampled and analyzed during amine plant operations treating CHP flue gases. A thorough overview and discussion of available equipment and instrumentation for monitoring of stack emissions will be given. The results are believed to provide realistic emission figures for emission monitoring and control for any future large scale carbon capture and storage (CCS) project due to the considerable size of the TCM DA amine plant.

2. Instrument and analysis

A description of the TCM DA amine plant is given elsewhere [1,2,3].

2.1 Overall system description and instrument position

Removing CO2 from flue gas by using post-combustion amine based CO2 capture reduces the emission of greenhouse gases to the atmosphere, but inevitably causes some emissions of amines and amine related degradation products to the atmosphere. Thus, qualitative and quantitative analysis of the emitted components is very important, but this task is far from trivial. For practical purposes, analytical instruments are preferably placed at ground level, but in that case a long sampling line (often >50 m) is required to direct the treated flue gas from the top of the absorber into the apparatus at ground level. This sample line has to be heated to well above the dew point of the gas to avoid condensation and possibly unwanted adsorption and/or reaction of emitted components.

TCM DA applies different measurement techniques to monitor and quantify the amounts and concentrations of emitted compounds. Some of the analyzer techniques currently applied on a permanent basis are;

Fourier Transform Infrared (FTIR) Spectroscopy

Proton Transfer Reaction Time of Flight Mass Spectrometry (PTR-TOF-MS)

Manual isokinetic sampling technique with impingers and subsequent off-line laboratory analysis (carried out by TCM DA, Statoil CP Laboratory, SINTEF and Ramboll)

Online gas phase concentration measurements are also performed at ground level (via a sample line) using a Fourier Transform Infra-Red (FTIR) Spectroscopy instrument and a Proton Transfer Reaction - Time of Flight -Mass Spectrometry (PTR-TOF-MS) device. This online equipment is placed in an analyzer house at ground level. At the absorber top, isokinetic sampling is performed on a regular basis. There is an analyzer house and a shelter on the top of the amine absorber where all the equipment is located, as seen in Figure 1.

Figure 1: Emission sampling set-up on the top of the amine absorber. Stack configuration (left) and sampling control from analyzer house

Extracted gas is sampled from the stack through an impinger train containing absorption liquids. By onsite measurement of the gas flow and laboratory analysis of the impinger liquids, the gas phase concentration of different components can be determined. The measurement system is shown schematically in Figure 2 and the techniques are further explained in the sections below.

Flue Gas outlet

Manual Gas Sampling

'-Stack ID

Isokinetic Gas Sampling

Flow Gas velocity (m/sj

A Pilot tube (S-type) B Thermocouple C Access port / Flange (3 - 4") D Split ot gas stream E isokinetic master flow

1. Nozzle

2. Probe

3. Gas cooler (optional)

4. Condesate flask with cooling

5. Impingers f absorption flasks with cooling

6. Drying tower

7. Pump

8. Flowmeter

9. Gas meter with thermometer

PTR-TOF

Figure 2: Schematic drawing of the emission monitoring set-up at the TCM DA amine plant

2.2 Sampling lines

The sampling line bundle installed at TCM DA is 101 meters long. It consists of 3 separate lines made from the following materials, respectively:

• PFA Teflon®

• Electro-polished stainless steel

• Sulfinert®-treated passivated stainless steel

All lines can be heated to 140 °C. Sample transfer via a heated sampling line has several benefits over placing the equipment at the top of the absorber:

• Easy access to the analyzer for maintenance and calibration and to utilities such as power, gas supplies, etc.

• Increased physical space for the analyzer

• Safer operations

Some negative aspects are however:

• Delayed analyzer response

• Potential degradation reactions and adsorption effects in the sampling line

Potential sample line effects are rarely reported in open literature. It is generally accepted is that the sample path should be kept as short as possible, and that the line temperature should be well above the dew point. However, increasing the temperature too much may lead to unwanted decomposition, to potential formation of nitrosamines, and to other sampling artefacts. Switch between different sample lines should be avoided due to memory effects. The effects of different sample lines were investigated by Cents et al [8].

2.3 FTIR analyzer

The FTIR model Anafin 2000 is employed at TCM DA to measure standard gas phase components (CO2, water, NOx, SOx) as well as amines, aldehydes and ammonia. The analyzer operates at wave numbers between 500 and 7000 cm-1, with a resolution of 2 cm-1. The path length is 7 meters. The detection limit for amines, aldehydes and ammonia is on the order of 1 ppmv. According to the discharge permit from the Norwegian Environmental Agency (Milj0direktoratet), TCM DA is allowed to emit 6 ppmv of total amines as a daily average [6]. For this purpose, the detection limit of the FTIR instrument is satisfactory.

The FTIR is connected via heated sampling lines to sampling probes at the absorber inlet (downstream DCC), absorber outlet and desorber overhead condenser outlet. An automatic stream selector makes it possible to program the plant's control system to alternate between the different measuring locations as desired. The FTIR is calibrated for a list of standard flue gas pollutants, including CO2, SO2, NH3, etc., as well as solvent amines and some volatile degradation products e.g. aldehydes. The instrument is not set up for measuring alkyl amines, nitrosamines and nitramines.

The FTIR technique has the advantage that the sample is measured without any preconditioning, hence reducing the risk for analytical artefacts. To avoid water condensation, the FTIR gas cell is heated to 85°C and the sampling lines are heated to 120°C. Target compounds contained in mist or droplets are likely to be evaporated at these temperatures. The FTIR monitor thus measures the total content of analytes in the flue gas. A draw-back of the FTIR technique is the interference from water vapor which results in a relatively high detection limit. The experience is that NH3 and amines can be detected down to 1 ppm levels. This is also in accordance with earlier measurements of gaseous emissions in post combustion carbon capture [9, 10].

2.4 PTR-TOF-MS analyzer

The PTR-TOF-MS (model PTR-TOF 8000) used is manufactured by Ionicon Analytik (Innsbruck, Austria). The PTR technique has been widely used for environmental volatile compound measurements for over a decade. Its measurement principle is based on soft ionization, via proton transfer, followed by high mass resolution mass spectrometric analysis. At TCM DA, the PTR-TOF-MS instrument subsamples from the main sample line through a heated (100-130°C) Siltek inlet line. The sample flow is diluted by a factor of 10 to 20 with bottled synthetic (zero) air, to avoid ion signal titration caused by high ammonia levels. The PTR-TOF-MS is able to measure amines, ammonia, aldehydes, ketones, carboxylic acids, nitramines and nitrosamines which are all important target compounds in amine based CO2 capture. The analytical setup at Mongstad is described in recent publications by Zhu et al. [11, 12].

2.5 Manual gas emission sampling

The analytical value chain applied for manual gas emission sampling and analysis is schematically shown in Figure 3. This value chain governs the measurement and is described in details below.

Figure 3: Schematic drawing of the emission measurement value chain

Sampling. The TCM DA stack is designed to achieve flow conditions suitable for isokinetic flue gas sampling as specified in the standard EN 15259:2007. The stack is insulated to minimize condensation. Sampling nozzles are located at a level 2 meters below the stack exit. A sampling system from Paul Goethe GmbH in Germany is used for allocation of a gas emission sample. The equipment is operated from an associated control unit (iTES). The special sampling equipment configuration is assembled for amine emissions based on experiences from the CCM project [7]. Isokinetic gas sampling principles are used to secure representative sampling from a ducted gas stream where two-phase conditions (particles or droplets with diameter > 1 ^m) are present or may occur. From an amine absorber the presence of droplets in the flue gas has to be considered, hence isokinetic gas sampling is an assurance for representative samples.

Capture of analyte. The double tube sampling probe was cooled with pressurized air in order to start condensation of the extracted gas sample stream. Typical amine emission analytes are captured by two principles, condensation and liquid absorption. It is experienced that the main sampling step is condensation. The condensate flask is kept cool in an ice bath and has a size and design to maximize the condensation capacity. In this way the gas is dried and further downstream split to subsequent impinger trains or solid adsorbents. It is further experienced that only for the most volatile components like NH3, small alkyl amines and aldehydes the second trap based on stepwise liquid absorption or solid phase adsorption is significant. In case of mist formation in the absorber, submicron aerosols will enter the sampling train. It is known that aerosols potentially can have limited retention through liquid

sampling systems. In order to improve the capture of aerosols, a high capacity condensation step is followed by jet-impinger flasks to force agglomeration. However mist is normally not associated with the CHP flue gas.

The condensates were preserved with sulfamic acid at site directly after sampling to avoid potential nitrosation of secondary amines [7]. Ammonia and the different amines were absorbed in 0.05 M sulphuric acid, the aldehydes and ketones were adsorbed on DNPH cartridges (Sep-Pak DNPH-Silica Long Body Cartridges, Waters). For nitrosamines and nitramines 10 g/L sulfamic acid solutions were used as second sampling step.

Sample work-up. Samples were brought to TCM laboratory and immediately cooled or frozen for storage until analysis. As a principle the condensate sample was prepared for analysis first and subsequent absorbent solutions were prepared and analyzed secondly, with various experimental techniques (Table 1). This often includes extensive laboratory work.

Analysis. Expected degradation and emission products from a MEA based solvent system, were assessed from a recent study using the solvent degradation rig for stress testing of MEA [13]. The target analytes for the current work is given in Table 1. These compounds cover the requirements set forth by the Milj0direktoratet in the emission permit.

Table 1. TCM DA sampling and analysis configuration and principle for different parameters / component groups

Parameters/ component groups Sampling Analysis

Amines (solvent) Condensate + 2x 0,05M sulphuric acid impingers + empty flask LC MS QQQ

Amines (alkyl) Condensate + 2x 0,05M sulphuric acid impingers + empty flask UPLC-MS/MS (Ramboll [7])

Ammonia Condensate + 2x 0,05M sulphuric acid impingers + empty flask Cation chromatography, IC-ECD

Aldehydes Condensate + 2x DNPH cartridges LS MS QQQ

Nitrosamines* (Specific, generic and TONO) Condensate + 2x 10 g sulfamic acid impingers + empty flask See * (Ramboll [7])

Nitramines Condensate + 2x 10 g sulfamic acid impingers + empty flask UPLC-MS/MS or GC-HRMS (Ramboll [7])

pH** - pH-paper [7]

Nitrite (NO3-)** - Anion chromatography, IC-ECD [7]

«Specific; CLLE extraction followed by UPLC-MS/MS or GC-HRMS . Generic; LLE followed by analysis on GC-HRMS. TONO; Quench of soluted nitrite followed by break of N-NO bond in a reaction chamber. Total NO released from the N-nitroso groups detected by chemiluminscence analyzer. **For sample preservation and work-up.

Amines, nitrosamines, and aldehydes were analyzed using an LC MS-MS QQQ (Agilent). The condensate from the first impinger was analyzed directly on the LC MS, the acidic impinger solutions were diluted before analysis. Ammonia was analyzed on an ion chromatograph (IC).

2.6 Additional analyzer techniques: Voice200 and PTR-QMS

TCM DA also tested a Voice200 analyzer from SYFT Technologies and a PTR-QMS 300 analyzer from Ionicon. These instruments operate on the same measurement principle as the PTR-TOF-MS but include cheaper and less specific quadrupole mass analyzers. Results from both analyzers compared well with the PTR-TOF-MS data. The results are not presented in this paper.

3. Results and Discussions

3.1 Analysis of Solvent samples

Solvent degradation processes were monitored during the course of the entire MEA campaign. The solvent amine, ammonia, and some degradation products were analyzed by TCM DA and Statoil CP laboratories. Alkyl amines, aldehydes, ketone, generic nitrosamines, solvent specific nitrosamines and nitramines were analyzed by Ramboll and SINTEF laboratories.

The concentration of the solvent amine was observed to remain stable over the extended period of the campaign indicating reasonable degradation rates of the solvent amine. The main degradation products of MEA were found to be amides, amino acids and other amines. Heat stable salts were also measured through the entire campaign, anions (OA, GA, FA, NO3-) by IC and total heat stable salts (HSS) by ion exchange and titration. Figure 4 displays the evolution of various degradation products and heat stable salts in the solvent. The components and amounts found were expected from an aqueous based MEA solvent system [9].

4000 2000

m ю oo ьл m тн тн m ьл

HEGly 4000

HEI 3000

HEPO 2000

t OA I N03-ÉGA I FA ■HSS

Hours of operation

Hours of operation

Figure 4: Results from some major degradation components (left figure) and heat stable salt formations (right figure) in the aqueous MEA solvent system during treatment of CHP flue gas

Two solvent specific nitrosamines, N-nitrosodiethanolamine (NDELA) and N-nitroso-2-hydroxyethyl-glycine (Nitroso-HeGly), were detected in the solvent as the degradation process progressed (Figure 5). The total concentrations of nitrosamines (TONO) were measured to be 797 ^mol/L.

Figure 5: Results from degradation of solvent amine MEA (04.02.2014) [13]

Since MEA is a primary amine it is not expected to form a stable nitrosamine. The identified compounds are thus formed from secondary amines occurring as impurities in the solvent or being formed during the degradation reactions. As is shown in Figure 5, there are still some unidentified nitrosamines in the used solvent sample. These nitrosamines are formed from high molecular weight amines and have low volatility. Only in the first water wash stage low quantities of nitrosamines were found (see below).

The solvent specific nitramine (MEA-NO2) was detected at a concentration of approximately 2 mg/L (Table 3).

3.2 Analysis of wash water samples

MEA was periodically measured in the wash water from both water wash sections. The wash water sections are specifically designed to physically absorb gaseous and entrained aqueous MEA before the depleted flue gas is emitted to atmosphere. Figure 6 shows that the liquid phase concentration in the first wash water section (Lower wash water - right y-axis) was about 100 times higher than the upper section (Upper Water Wash - left y-axis). The results from 16/12-2013 show higher results, the temperature in the flue gas was 47°C and this will give higher MEA concentrations. Going from 30 to 40 wt% MEA in the solvent, will also give higher MEA concentrations in the water wash sections and this is measured at 19/2-2014, where the solvent MEA concentration was 40 wt%. Methylamine and minor amounts of ethylamine were also found in water wash samples, as presented in Table 3.

Figure 6 clearly demonstrates the effectiveness of two water wash sections.

Figure б: MEA concentrations in wash water 1 and 2.

The concentration of alkylamines, nitrosamines and nitramines in wash water samples are given in Tables 2 and 3. TONO were above detection limit only in the first water wash section, in one of two samples. This clearly indicates that nitrosamine volatility is low and that nitrosamines escaping from the solvent are efficiently captured in the first water wash section. No generic or solvent specific nitramines were found in either of the wash water sections.

Methylamine and minor amounts of ethylamine were found in low concentrations (^g/L) and it is seen that the water wash also has effect of these volatile compounds. No generic or solvent specific nitramines were found in either of the wash water sections.

Table 2. TONO (Total nitrosamines) measurement, measured by Ramboll (flue gas, wash water) and Sintef (lean MEA).

Date Sample TONO, ^mol/L Operational Conditions

11.02.2014 Wash water 2 <0.05 30 wt% MEA

11.02.2014 Wash water 1 0.13 30 wt% MEA

04.02.2014 Lean MEA 797 30 wt% MEA

Table 3. Degradation components in solvent and wash water measured by Ramboll

Compound Unit Wash water 1 Wash water 2 Lean MEA

11.02.2014 11.02.2014 04.02.2014

Methylamine (MA) ^g/L 3700 1600 -

Dimethylamine (DEA) ^g/L <50 <500 -

Ethylamine (EA) ^g/L 270 <500 -

Diethylamine (DiEA) ^g/L <50 <50 -

Ethylmethylamine (EMA) ^g/L <100 <1000 -

MEA mg/L 1600 37 -

DEA mg/L <0.05 <0.5 -

Morpholine mg/L 5.8 <1 -

MEA-NO2 ^g/L <1 <1 2120

Dimethylnitramine ^g/L <0.2 <0.2 <2

Diethylnitramine ^g/L <0.4 <0.4 <4

NDMA ^g/L <0.1 <0.1 <1

NMEA ^g/L <0.1 <0.1 <1

NDEA ^g/L <0.1 <0.1 <1

NDPA ^g/L <0.1 <0.1 <1

NPYR ^g/L <0.1 <0.1 <1

NMOR ^g/L <0.2 <0.2 <2

NPIP ^g/L <0.1 <0.1 <1

NDBA ^g/L <0.1 <0.1 <1

NDELA ^g/L <1 <1 4200

3.3 Analysis of gas emission samples

Thirteen manual isokinetic sampling emission campaigns were conducted during the MEA-campaign. All emission samples were collected by TCM DA, except one which was performed by FORCE Technology. The amine plant operating conditions and detailed emission results are given in Tables 4 to 6. All nitrosamine and nitramine emissions were below detection limits. Emissions of alkyl amines were limited and only methylamine is quantified in the low ppb range. Possible emission of unknown compounds has been investigated via PTR-TOF-MS. A list of identified or tentatively identified compounds is given in Table 7. No alkylamines, nitrosamines and nitramines were detected by PTR-TOF-MS.

Table 4. List of emission measurements during the MEA campaign

Date and time Start Stop Flue Gas volume, m3/h Temp. gas out, °C Operational Conditions Operational hours*

26.11.2013 09:14 11:14 58.000 46 30 wt% MEA 50

09.12.2013 10:33 12:33 50.000 25 30 wt% MEA 350

09.12.2013 13:33 15:33 50.000 25 30 wt% MEA 350

16.12.2013 10:38 12:38 47.000 43 30 wt% MEA 500

08.01.2014 12:11 14:11 49.000 25 30 wt% MEA 1000

08.01.2014 14:35 16:53 49.000 25 30 wt% MEA 1000

08.01.2014 17:12 19:12 49.000 25 30 wt% MEA 1000

09.01.2014 10:20 12:20 49.000 25 30 wt% MEA 1030

09.01.2014 12:40 14:40 49.000 25 30 wt% MEA 1030

27.01.2014 12:25 14:25 61.000 25 30 wt% MEA 1260

04.02.2014 11:53 13:53 62.000 27 30 wt% MEA 1390

11.02.2014 08:15 10:15 49.000 26 30 wt% MEA 1540

14.02.2014 10:50 12:50 62.000 25 40 wt% MEA 1600

* Operating hours counted as hours with CO2 capture

Table 5. Result from isokinetic gas emission measurements during the MEA campaign

Date MEA, MEA, NH3, NH3, Formaldehyde, Formaldehyde, Acetaldehyd, Acetaldehyd,

^g/m3 ppmv ^g/m3 ppmv ^g/m3 ppmv ^g/m3 ppmv

26.11.2013 848 0.323 6413 8.3 - - - -

09.12.2013 78 0.030 4907 6.3 - - - -

09.12.2013 59 0.022 5242 6.8 - - - -

16.12.2013 29 0.011 8907 11.5 - - - -

08.01.2014 14 0.005 6336 8.2 - - - -

08.01.2014 21 0.008 9611 12.4 - - - -

08.01.2014 36 0.014 6452 8.3 - - - -

09.01.2014 38 0.014 6729 8.7 - - - -

09.01.2014 3.5 0.001 6806 8.8 - - - -

27.01.2014 14 0.005 - - 3.1 0.002 18.1 0.009

04.02.2014 12 0.004 - - 4.4 0.003 31.7 0.017

11.02.2014 21 0.008 - - 4.3 0.003 31.7 0.016

14.02.2014 22 0.008 10031 13 - - - -

Figure 6 and Figure 7 display atmospheric emission results of MEA and NH3 from absorber outlet over the entire campaign. Figure 6 display emission results from the FTIR and PTR-TOF-MS analyzer in comparison with results from manual isokinetic sampling and analysis. The MEA FTIR results are not considered to be reliable in the low ppm range, since they are below/around the detection limit. The first measurement (performed on the 26th of November) showed emissions above 300 ppb. The reason for the higher amine emission in the first measurement is related to amine plant operating conditions. The NH3 emissions were reasonably low and as expected for MEA.

According to TCM DA experience the aldehyde concentrations were varying from low ppbv to several hundred ppbv during operations. Results found in this campaign and earlier campaigns are in agreement, and they are confirmed by third party. The FTIR is not measuring aldehydes below 1 ppm, but PTR TOF gives a good agreement to results found by isokinetic sampling and analysis, see tables 7 and 8. The PTR TOF is a good candidate for a reliable online analyzer of aldehydes in the ppb range.

Comparison of emission results from three sampling and analysis methods is somewhat tricky as there are some fundamental differences like; sampling point, sample extraction principles and sample transfer to the collecting or detection units. In this case manual samples are collected on the top of absorber using isokinetic extraction principles while the online methods are extracted non-isokinetic and switched in through a 101 meter long sampling line. Hence a comparison of MEA emission data can reflect differences in the sampling configuration. The analytical differences are first of all related to instrumental detection limits. Taking sampling and analytical differences into account the comparison of results is summarized and illustrated in figure 6. It is clear that the FTIR data is affected by high detection limit and by then increased uncertainty for this low ppm to ppb-level. Results from manual sampling and online PTR-TOF-MS are first of all according to both methods reported in a low concentration level (0,001 - 0,3 ppmv). The variation between the two data sets is significant and in general manual sampling reflects lower values than online PTR-TOF-MS results. Based on TCM-experience it is likely that the different sampling set-up explain this. TCM has experienced during this MEA campaign, that switch between different sampling points and long heated sampling lines are challenging and need to be tested more and further optimized to secure stable and representative gas composition.

The manual isokinetic sampling and analysis is considered to be a reference method for TCM DA. Isokinetic sampling and analysis is verified by two independent third party companies (during earlier campaigns by Kema/SGS and FORCE, and in this MEA campaign by FORCE).

Figure 6: MEA emissions determined by different analyzer techniques during the campaign. Results on the FTIR below detection limit (<1 ppmv) are colored lighter blue.

I I II I I II I I

Figure 7: NH3 emissions determined by different analyzer techniques during the campaign Table 6. Degradation components in Flue gas out of absorber from isokinetic gas emission measurements.

Compound

04.02.2014 ^g/m3

04.02.2014 ppbv

10.02.2014 ^g/m3

10.02.2014

Methylamine 2.6 2 3.6 3

Dimethylamine <1.1 <1 <1.1 <1

Ethylamine <1.1 <1 <1.1 <1

Diethylamine <1.1 <0.3 <1.1 <0.3

Ethylmethylamine <2.2 <1 <2.1 <1

MEA 13 5 17 6

DEA <1.1 <0.2 <1.1 <0.2

Morpholine <2.2 <1 <2.1 <1

MEA-NO2 <0.01 <0.002 <0.01 <0.002

Dimethylnitramine <0.002 <0.0006 <0.002 <0.0006

Diethylnitramine <0.004 <0.001 <0.004 <0.001

NDMA <0.001 <0.0003 0.001 0.0004

NMEA <0.001 <0.0003 <0.001 <0.0003

NDEA <0.001 <0.0002 <0.001 <0.0003

NDPA <0.001 <0.0002 <0.001 <0.0002

NPYR <0.001 <0.3 <0.001 <0.3

NMOR <0.002 <0.0004 <0.002 <0.0004

NPIP <0.001 <0.0002 <0.001 <0.0002

NDBA <0.001 <0.0002 <0.001 <0.0002

NDELA <0.01 <0.002 <0.01 <0.002

TONO* <0.2 <0.04 <0.2 <0.04

*Converted from molar to mass basis, using Mw 130 g/mol

PTR-TOF-MS was also used for screening of potential other emissions. A list of identified or tentatively identified compounds is given in Table 7. It is noted that PTR-TOF-MS did not detect any emissions of alkylamines, nitrosamines or nitramines.

Table 7. Results from PTR-TOF-MS measurements on 11.02.2014, 08:15-10:15. Estimated uncertainty in measurements is +20%.

Name Formula ppbv Structure m/z

2-aminoethanol H2NCH2CH2OH 8.9 62.060

Ammonia NH3 18265.7 18.034

Formaldehyde HCHO 43.1 31.018

Acetaldehyde CH3CHO 454.9 45.033

Acetone (CH3)2CO 88.2 59.049

Aceticacid CH3COOH 12 61.028

Formamide* CHONH2 13 46.028

Acetamide* NH2CH2CHO 14.1 60.044

Methane,nitro* CH3NO2 19.8 62.024

Ethane,nitro* CH3CH2NO2 0.8 73.039

Pyrrole* C4H4NH 5.2 ;::;>« 68.049

Pyrazine* C4H4N2 107.1 O 81.044

Pyrazinemethyl* C4H3N2CH3 23.2 X 95.060

Pyrazinedimethyl* C4H2N2(CH3)2 7.1 £ 109.079

* Tentative interpretation based on chemical formula, temporal profile or possible chemical pathway of formation.

3.4 Third party gas emission measurement

One third-party emission measurement was done on January 6. FORCE Technology carried out isokinetic sampling onto a solid sorbent (Thermosorb/N) in combination with condensate collection in an impinger. The condensate was analyzed separately. Analysis of collected samples was done by Isconlab GmbH. The results show that all nitrosamine and nitramine emissions were below detection limits.

Table 8. Results from third part measurement on 6th of January, done by FORCE Technology

Compound ^g/m3

Total nitramine <0.2

Total nitrosamine <0.08

NDMA (N-nitrosodimethylamine) <0.08 Formaldehyde <70 Acetaldehyde 310

4. Conclusion

Extensive atmospheric emission monitoring has been conducted at the CO2 Technology Centre Mongstad (TCM DA) during amine based post-combustion CO2 capture. The TCM DA amine plant was operated with the aqueous MEA solvent system treating flue gas from a combined heat and power plant (CHP). Emission monitoring was conducted by a Fourier Transform Infrared (FTIR) Spectrometry analyzer, a Proton Transfer Reaction Time-ofFlight Mass Spectrometry (PTR-TOF-MS) analyzer, and manual isokinetic sampling followed by off-line analysis in the laboratory.

Atmospheric emissions of monoethanolamine (MEA) were very low throughout the complete campaign, and determined to be in the parts per billion (ppb) range. Atmospheric emissions of MEA amine based degradation products such as nitrosamines and nitramines were below detectable levels. Atmospheric emissions of ammonia (NH3) were in the low ppm range, and alkyl amines in the low ppb range.

Absorber wash water sections were found to effectively reduce possible atmospheric emissions from amine based solvent system.

Acknowledgements

The authors gratefully acknowledge the staff of TCM DA, Gassnova, Statoil, Shell, Sasol, and Aker Solutions for their contribution and work at the TCM DA facility, and the staff of the Statoil CP laboratory for their contribution and work with the LC-MS measurements.

The authors also gratefully acknowledge Gassnova, Statoil, Shell, and Sasol as the owners of TCM DA and Aker Solutions for their financial support and contributions.

Appendix A. Abbreviations

AA Acetic acid

CLLE Continuous Liquid Liquid Extraction

DCC Direct Contact Cooler

DiEA Diethylamine

DMA Dimetylamine

DMNA N-nitro-N-methyl-methanamine

DMO 4,4-dimethyl-2-oxazolidinone

EA Ethylamine

FA Formic acid

GA Glycolic acid

GC-HRMS Gas Chromatography-High Resolution Mass Spectrometry

HEA N-(2-hydroxyethyl) acetamide

HEF N-(2-hydroxyethyl)formamide

HeGly N-(2-hydroxyethyl)glycine

HEI N-(2-hydroxyethyl)imidazole

HEIA N-(2-hydroxyethyl)imidazolidinone

HEPO 4-(2-hydroxyethyl)piperazin-2-one

HSS Heat Stable Salt

IC-ECD Ion Chromatography-Electric Conductivity Detection LC MS QQQ Liquid Chromatography Mass Spectrometry Triple Quadrupole

LLE Liquid Liquid Extraction

MA Methylamine

NO2-MEA NO-HeGly

N-Nitrosodi-n-butylamine

N-Nitrosodiethylamine

N-methyl,N-nitroso-methanamine

N-Nitrosodi-n-propylamine

N-Nitrosomethylethylamine

2-(Nitroamino)ethanol

N-Nitroso(2-hydroxyethyl)glycine

N-Nitrosopiperidine

N-Nitrosopyrollidine

Oxalic acid

Total Nitroso amines

UPLC-MS/MS Ultra Performance Liquid Chromatography - Mass Spectrometry / Mass Spectrometry

References

[1] Thimsen D, Maxson A, Smith V, Cents T, Falk-Pedersen O, Gorset O, Hamborg E S. Results from MEA testing at the CO2 Technology Centre Mongstad. Part I: Post-Combustion CO2 capture testing methodology. Energy Procedia; 2014.

[2] Hamborg E S, Smith V, Cents T, Brigman N, Falk-Pedersen O, De Cazenove T, Chhaganlal M, Feste J K, Ullestad 0, Ulvatn H, Gorset O, Askestad I, Gram L K, Fostas B F, Shah M I, Maxson A, Thimsen D. Results from MEA testing at the CO2 Technology Centre Mongstad. Part II: Verification of baseline results. Energy Procedia; 2014

[3] Brigman N, Shah M I, Falk-Pedersen O, Cents T, Smith V, De Cazenove T, Morken A K, Hvidsten O A, Chhaganlal M, Feste J K, Lombardo G, Bade O M, Knudsen J, Subramoney S C, Fostas B F, De Koeijer G, Hamborg E S. Results of amine plant operations from 30 wt% and 40 wt% aqueous MEA testing at the CO2 Technology Centre Mongstad. Energy Procedia; 2014.

[4] Eirik Falck da Silva, Karl Anders Hoff, Andy Booth; Energy Procedia 37 (2013) 784-790

[5] Gelein de Koeijer, Vibeke Randgaard Talstad, Sissel Nepstad, Dag Tonnesen, Olav Falk-Pedersen, Yolandi Maree, Claus Nielsen; International Journal of Greenhouse Gas Control 18 (2013) 200-207

[6] In: Norwegian Climate and Pollution Agency (Ed.), Tilllatelse til virksomhet etter forurensningsloven for TCM DA (Norwegian only), Norwegian Climate and Pollution Agency, Oslo, Norway (Now: Norwegian Environmental Agency (Miljodirektoratet))

[7] http://www.gassnova.no/en/ccs-projects/full-scale-mongstad/results-from-the-second-round-of-studies-of-the-health-and-environmental-issues-of-amines

[8] Toine Cents, 0yvind Ullestad, Anne Kolstad Morken, Olav Falk-Pedersen, Otto Bade, Gunnar Schade, Liang Zhu, Claus Nielsen. Measurement of emissions from the amine based CO2 capture plant at CO2 Technology Centre Mongstad. TWELFTH ANNUAL CONFERENCE ON CARBON CAPTURE, UTILIZATION AND SEQUESTRATION May 13 - 16, 2013 • David L. Lawrence Convention Center • Pittsburgh, Pennsylvania

[9] Jan Mertens, Marie-Laure Thielens, Jacob Knudsen and Jimmy Andersen. Monitoring and impacting gaseous emissions in post combustion carbon capture. IEAGHG PCCC1, 1st Post Combustion Capture Conference

[10] Jan Mertens, Dominique Desagher, Marie-Laure Thielens, Han Huynh Thi Ngoc, Helene Lepaumier, Purvil Khakaria and Earl Goetheer. On- and off-line ethanolamine and ammonia emission monitoring in PCCC. IEAGHG PCCC2, 2nd Post Combustion Capture Conference.

[11] Zhu, L., Schade, G. W. & Nielsen , C. J. Real time monitoring of emissions from monoethanolaminebased industrial scale carbon capture facilities. Environmental Science & Technology 47, 14306-14314 (2013)

[12] Zhu, L., Schade, G. W. & Nielsen , C. J. TCM emission measurements: results from PTR-ToF-MS measurements during use of MEA 2012. (2013).

[13] Aslak Einbu, Eirik DaSilva, Geir Haugen, Andreas Grimstvedt, Kristin Giske Lauritsen, Kolbjorn Zahlsen, Terje Vassbotn, Energy Procedia 37 (2013) 717-726.