Corrosion in CO2 Capture Process Using Blended

Sep 29, 2009 - The system was computer automated with Agilent Chemstation software, which ... The corrosion cell was placed in a water bath to control...
5 downloads 0 Views 939KB Size
Download PDF
Ind. Eng. Chem. Res. 2009, 48, 9299–9306

9299

Corrosion in CO2 Capture Process Using Blended Monoethanolamine and Piperazine Manjula Nainar and Amornvadee Veawab* Faculty of Engineering, UniVersity of Regina, Regina, Saskatchewan, Canada S4S 0A2

This work explores the promise of aqueous solutions of blended monoethanolamine (MEA) and piperazine (PZ) as a cost-effective solvent for carbon dioxide (CO2) capture from industrial flue gas streams with respect to corrosion, which is regarded as one of the most severe operational problems in typical CO2 capture plants. Electrochemical corrosion experiments were carried out using the potentiodynamic polarization technique for corrosion measurements. The results show that the blended MEA/PZ solutions are more corrosive than the MEA solutions. The corrosion rate of carbon steel increases with concentration of PZ, total amine concentration, CO2 loading of solution, solution temperature, and the presence of heat stable salts. Among the tested heat-stable salts, formate is the most corrosive salt, followed by acetate, oxalate, and thiosulfate in the absence of oxygen (O2), while acetate is the most corrosive salt followed by formate, oxalate, and thiosulfate in the presence of O2. 1. Introduction

2. Research Motivation and Objectives

The acid gas absorption process has been used in industries for many decades for the removal of carbon dioxide (CO2) and hydrogen sulphide (H2S) from industrial gas streams. Its purposes are to enhance the quality of gas products and to prevent operational difficulties that may occur in downstream processes. This process can also potentially be used for capturing CO2 from the flue gas emitted from industries to reduce the emissions of CO2, the leading greenhouse gas (GHG) contributing to global warming and climate change. The industries requiring such flue gas clean up are coal-fired power plants, natural gas processing, cement plants, steel plants, and hydrogen and ammonia manufacturing units. Among these, the coal-fired power plants are the primary target industry, since they typically emit approximately 30% of the total CO2 emissions.1 A wide variety of absorption solvents have been applied to the acid gas absorption process. Such solvents include aqueous solutions of alkanolamines, potassium carbonate (K2CO3), blended solvents, and proprietary solvents. Many researchers have investigated piperazine (PZ)-based blends. Their results indicated that PZ is an effective promoter in monoethanolamine (MEA), methyldiethanolamine (MDEA), and potassium carbonate (K2CO3) solutions from both thermodynamic and kinetic perspectives.2-4 Since PZ is a cyclic secondary diamine, its efficiency is attributed to its cyclic diamine structure that may favor rapid formation of carbamates with CO2. Also, the molecule can theoretically absorb 2 mol of CO2 for every 1 mol of amine.4 It was also reported that the rate constant of PZ with CO2 is an order of magnitude higher than that of conventional MEA.5 Among the PZ-based solutions, MEA/PZ blends have been proven to be efficient for capturing CO2. The absorption rate of CO2 at 40 and 60 °C in aqueous MEA with 0.6-1.2 kmol/ m3 PZ is 1.5-2.5 times greater than that in MEA alone.3 The removal efficiency of CO2 from gases containing 10% CO2 by MEA/PZ solution was found to be superior to that of AMP/PZ, MDEA/PZ, and MEA.6

The aqueous solution of blended MEA and PZ has been demonstrated to be a promising solvent for CO2 capture from coal-fired power plant flue gas due to its capture performance and energy efficiency. Although there are extensive research data available on kinetics and solubility of MEA/PZ blends, no research has been done to investigate the corrosiveness of this solvent. It is, therefore, worthwhile to investigate the corrosivity of the MEA/PZ system since corrosion is one of the most severe operational problems in the acid gas absorption process and causes substantial expenditure in addition to process costs. This gives the motivation for this work to further explore the promise of this solvent with respect to corrosion. The objective of this work is to investigate corrosion of carbon steel in aqueous solutions of blended MEA/PZ in the presence and absence of solvent degradation products. The research involves comprehensive experimental evaluations in the following sequences. All tested parameters and conditions for corrosion tests are summarized in Table 1. 2.1. Evaluation of Corrosion in Base Solution. The objective of this task was to examine corrosiveness of the aqueous solutions of blended MEA/PZ when carbon steel is used as the material of plant construction. The investigation was carried out by varying the total amine concentration, MEA/PZ mixing ratio, CO2 loading, temperature, and oxygen (O2). The tests involved electrochemical experiments using various concentrations of MEA and MEA/PZ blends. Corrosion rates of MEA/ PZ blends were compared against those of MEA.

* To whom correspondence should be addressed. Tel.: (306) 5855665. Fax: (306) 585-4855. E-mail: [email protected].

Table 1. Tested Parameters and Conditions for Corrosion Experiments parameter total amine concentration (kmol/m3) CO2 loading of solution (mol CO2/mol amine) solution temperature (°C) partial pressure of O2 (kPa) mixing ratio of MEA and PZ (mol MEA: mol PZ) (total amine concentration ) 6.2 kmol/m3) heat stable salt (1.00 wt %)

10.1021/ie801802a CCC: $40.75  2009 American Chemical Society Published on Web 09/29/2009

condition 5.0, 6.2, 7.0, 8.7 0.20, 0.40, 0.58, 0.63 40, 80 0.00, 5.07, 10.13 1:0, 1:1, 2:1, 4:1

acetate, formate, oxalate, thiosulfate

9300

Ind. Eng. Chem. Res., Vol. 48, No. 20, 2009

Figure 1. Experimental setup for electrochemical corrosion tests.

2.2. Evaluation of Corrosion in Base Solution Containing Degradation Products. The objective of this task was to examine corrosive behavior of carbon steel in the MEA/PZ solutions containing irreversible degradation products of amine (heatstable salts). Four types of heat-stable salts selected in this work were acetate, formate, oxalate, and thiosulfate. Their effects on corrosion were examined in both the presence and absence of O2, and the most corrosive heat-stable salt was determined. This salt was used to represent heat-stable salts in further experiments. 3. Experiments 3.1. Specimen Preparation. Two types of specimens were used in this work, stainless steel 430 (SS 430) and carbon steel 1018 (CS 1018). SS 430 was used for standardizing the experimental procedures and instrumentation, whereas CS 1018 was used as the working electrode for all electrochemical tests. The CS 1018 and SS 430 specimens were cylindrical in shape with length, outside diameter, and hole diameter of 0.500 in. (1.270 cm), 0.375 in. (0.952 cm), and 0.210 in. (0.533 cm), respectively. Prior to experiments, the specimens were prepared by wet grinding with 600 grit silicon carbide papers using deionized water in accordance with the ASTM standard G190.7 They were then degreased with high purity methanol and dried with hot air. The specimens were kept in a desiccator until used. 3.2. Solution Preparation and Analysis. An aqueous solution of blended MEA and PZ of a desired concentration was prepared from a 99% MEA reagent, 99% PZ, and deionized water. The solution was loaded with a desired loading of CO2 by passing the gas through the solution. The prepared MEA/ PZ blended solution was subsequently added with various heatstable salts and oxidative degradation inhibitor to examine the corrosion behavior. The heat-stable salt anions, including acetate, formate, and oxalate, were made up by dissolving the respective acid forms of the anions in the aqueous MEA/PZ blended solution. The solution of thiosulfate anion was prepared by dissolving its respective salt form in the aqueous MEA/PZ blended solution because of the unavailability of its acid form.

Two methods were used for solution analysis, volumetric titration, and gas chromatography/mass spectroscopy (GC/MS). The volumetric titration was used for determining the CO2 loading in the solution while the GC/MS was used for determining the ratio of blended MEA/PZ. The CO2 loading in amine solution was determined by a standard method given by the Association of Official Analytical Chemists (AOAC), namely, the official method of determining CO2.8 The amine concentration was measured by titration with 1 kmol/m3 hydrochloric acid (HCl), using a methyl orange indicator. The GC/MS used in this work was an HP 6890 with an HP 7673 auto injector. The injection volume was 0.2 µL and injected into a split/splitless inlet, with an inlet temperature set at 250 °C. The split/splitless injection was used for diluting the sample further for better analysis. Helium was used as the carrier gas with a flow rate of 44 cm/s. The split ratio was 10:1 with a split flow of 15 mL/min. The capillary column was HP-5 (model number-19091J-413), which was of length 30 m and inner diameter 320 µm. It had a coating with 5% phenyl methyl siloxane of thickness 0.25 µm. The temperature program used for this column started at 40 °C held for 3 min. The temperature ramped at 20 °C/min to a final temperature of 140 °C. The system then held this final temperature for 4 min. The detector used was HP-5973 mass selective detector. The scan parameters for MS were low mass 10, high mass 550, and a solvent delay of 1.5 min. The system was computer automated with Agilent Chemstation software, which fully controlled the injection and detection process. 3.3. Experimental Setup. The experimental setup for electrochemical corrosion tests (Figure 1) consisted of a corrosion cell, a potentiostat, a water bath equipped with a temperature controller, a gas supply set, a condenser, a data acquisition system, a pH meter, and a conductivity meter. The corrosion cell was a 1-L standard three-electrode cell (model K47 from EG&G instruments corporation, Princeton Applied Research, NJ) using a calomel reference electrode (SCE) (Hg/Hg2Cl2/ saturated KCl) and a graphite counter electrode. The potentiostat (model 273A EG&G Instruments Corporation, Princeton Applied Research, NJ) provided an accuracy of (0.2% of the

Ind. Eng. Chem. Res., Vol. 48, No. 20, 2009

potential and current readings. The water bath was equipped with a temperature controller to maintain a constant water temperature within (0.1 °C throughout the experiment. To minimize heat loss (or gain) to or from the surroundings, the water surface was covered with hollow balls. The gas supply set consisted of gas cylinders of CO2, O2, and nitrogen (N2). These gases were introduced into the corrosion cell for simulating the test environment, through a series of gas flow meters that measured the gas flow rate within (2% accuracy. The Allihn condenser (jacket length of 300 mm and a height of 445 mm) was connected to the cell to prevent any change in solution concentrations due to evaporation. The data acquisition system, model 352 SoftCorr III (EG&G Instruments Corporation, Princeton Applied Research, NJ), was used to control the potentiostat and also to record and analyze the produced corrosion data. The pH meter, model pH 11 series (Oakton, USA) with an accuracy of (0.01, and the conductivity meter, model YSI 3200 with an accuracy of (0.10%, were also used. 3.4. Experimental Procedure. The validation of instrumentation and experimental results was carried out according to the ASTM standard G5-94 (1999)9 prior to the electrochemical experiments. For each experiment, 1 L of blended MEA/PZ solution with the desired composition and CO2 loading was prepared and transferred to a corrosion cell and a salt bridge. The corrosion cell was placed in a water bath to control the solution temperature at 80.0 ( 0.1 °C, connected to a condenser, and purged through a gas disperser with given flow rates of N2 and CO2 to maintain a desired CO2 loading of solution. The flow rates of N2 and CO2 were determined from the equilibrium partial pressure of CO2 (CO2 solubility into solution) required for a given CO2 loading of solution. To ensure the constant CO2 loading throughout the experiment, samples of solution were drawn for titration to measure the CO2 loading before and after the experiment. After the cell reached the set environment, which took approximately 30 min, a prepared specimen was mounted to the holder. The corrosion cell was then connected to a potentiostat equipped with the data acquisition software. The polarization scan began when the potential of the specimen reached equilibrium or was constant with time. Potentiodynamic polarization and a Tafel plot were used for corrosion measurements. A Tafel plot involves polarizing a specimen in both anodic and cathodic directions in the potential range of (200 mV from an equilibrium corrosion potential (Ecorr) and generating an anodic/cathodic polarization curve of the specimen. The obtained corrosion current density (icorr) was used to calculate the corrosion rate of the specimen by using the following equation: (0.13icorrEW) CR ) AD

(1)

where CR is corrosion rate in mils (thousandths of an inch) per year (mpy), icorr is corrosion current density in microamperes per squared centimeters, EW is equivalent weight of the specimen, A is area of specimen in squared centimeters, and D is density of the specimen in grams per cubic centimeter. Before and after each experiment, samples of the solution were collected for solution analysis and both pH and conductivity of the solution were measured. The experiment was replicated to ensure data reproducibility. 4. Results and Discussion 4.1. Polarization Behavior in the Blended MEA/PZ System. Figure 2 illustrates typical cyclic polarization behavior of carbon steel 1018 immersed in aqueous solutions of MEA and blended

9301

Figure 2. Cyclic polarization curves of carbon steel 1018 in aqueous solutions of (a) MEA (5.0 kmol/m3) and (b) blended MEA/PZ (5.0/1.2 kmol/ m3) containing 0.20 mol/mol CO2 loading at 80 °C.

MEA/PZ containing 0.20 mol/mol CO2 loading at 80 °C. It is apparent that the polarization behavior of blended MEA/PZ is similar to that of MEA. That is, the carbon steel exhibits active, passive, and transpassive states, depending on the system’s potential. Under the tested condition at the corrosion potential (Ecorr) of -0.948 V SCE, the carbon steel is in the active state where corrosion takes place on the metal surface, which is free of passive film, through iron dissolution producing ferrous ion (Fe2+) (reaction 1) and reduction of oxidizing agents (mainly bicarbonate (HCO3-) (reaction 2). These two electrochemical reactions lead to the formation of ferrous carbonate (FeCO3) as corrosion products (reaction 3). The existence of these FeCO3 molecules is evidenced in the Pourbaix diagram at Ecorr ) -0.948 V SCE and pH ) 9.35.10 Iron dissolution: Fe f Fe2+ + 2e-

(1)

Reduction of bicarbonate ion: 2HCO3- + 2e- T 2CO32- + H2(g)

(2)

Overall reaction:11,12 Fe2+ + HCO3- T FeCO3 + H+

(3)

In addition to the above information, Figure 2 also provides information on corrosion prevention and control. That is, if the corrosion inhibition is to be applied to this system to reduce corrosion rate by raising the system’s potential to passive region, the metal surface will be covered by passive film, thereby reducing the current density substantially to the passive current density (ipass) and in turn reducing corrosion rate. The passive film is postulated to be ferrous oxide (Fe2O3) according to the Pourbaix diagram (at E ) -0.777-0.291 V SCE and pH ) 9.35).10 The film formation reaction is shown below.13

MEA 5.0 kmol/m3 MEA 6.2 kmol/m3 MEA 7.0 kmol/m3 MEA 8.7 kmol/m3 MEA/PZ -5.0/1.2 kmol/m3 MEA/PZ -7.0/1.7 kmol/m3 R ) 0.63, MEA/PZ 5.0/1.2 kmol/m3 R ) 0.58, MEA/PZ 7.0/1.7 kmol/m3 R ) 0.20, MEA/PZ 5.0/1.2 kmol/m3 b MEA/PZ 2:1 mol ratio b MEA/PZ 1:1 mol ratio R ) 0.20, MEA/PZ 7.0/1.7 kmol/m3 R ) 0.20, MEA/PZ 7.0/1.7 kmol/m3 0.00 kPa O2, 1.00 wt % oxalate 10.13 kPa O2, 1.00 wt % oxalate 0.00 kPa O2, 1.00 wt % acetate 10.13 kPa O2 1.00 wt % acetate 0.00 kPa O2, 1.00 wt % formate 10.13 kPa O2, 1.00 wt % formate 0.00 kPa O2, 1.00 wt % thiosulfate 10.13 kPa O2, 1.00 wt % thiosulfate

-754.00 ( 1.00 -743.00 -740.00 ( 4.00 -743.00 ( 1.00 -777.00 ( 3.00 -761.00 -618.00 -593.00 ( 15.00 -778.00 ( 5.00 -773.00 ( 5.00 -795.00 ( 1.00 -748.00 ( 2.00 -755.00 ( 4.00 -741.00 ( 2.00 -736.00 ( 2.00 -766.00 ( 3.00 -741.00 ( 24.00 -763.00 ( 1.00 -760.00 ( 3.00 -342.00 ( 3.00 -334.00 ( 4.00

00

7.19 × 10-06 ( 1.45 × 10-02 9.25 × 10-06 ( 1.04 × 1000 1.25 × 10-05 ( 3.45 × 10-02 1.22 × 10-05 ( 8.50 × 10-02 8.06 × 10-06 ( 1.45 × 10-02 1.09 × 10-05 ( 4.5 × 10-02 2.40 × 10-05 1.35 × 10-05 ( 6.00 × 10-02 1.59 × 10-06 ( 9.06 × 10-02 5.70 × 10-06 ( 5.50 × 10-02 5.32 × 10-06 ( 5.00 × 10-03 3.77 × 10-06 ( 6.50 × 10-02 5.18 × 10-06 ( 5.95 × 10-07 1.80 × 10-05 ( 1.50 × 10-02 6.47 × 10-06 ( 6.00 × 10-02 7.26 × 10-06 ( 3.00 × 10-02 3.95 × 10-06 ( 1.50 × 10-02 7.96 × 10-06 ( 2.00 × 10-02 3.99 × 10-06 ( 1.00 × 10-02 5.34 × 10-01 ( 7.00 × 10-03 5.68 × 10-01

1.89 × 10-03 ( 3.50 × 10-02 2.40 × 10-03 ( 4.00 × 10-02 2.96 × 10-03 ( 1.00 × 10-02 2.83 × 10-03 1.55 × 10-03 ( 6.00 × 10-02 1.87 × 10-03 ( 2.00 × 10-02 4.58 × 10-03 2.76 × 10-03 ( 1.30 × 10-01 1.66 × 10-03 ( 2.00 × 10-02 1.06 × 10-03 ( 5.00 × 10-03 6.62 × 10-04 ( 1.00 × 10-02 2.19 × 10-04 1.59 × 10-03 ( 1.08 × 10-04 2.83 × 10-03 ( 2.00 × 10-02 2.76 × 10-03 ( 1.00 × 10-02 1.42 × 10-03 ( 3.00 × 10-02 1.42 × 10-03 ( 1.00 × 10-02 1.40 × 10-03 ( 2.50 × 10-02 1.47 × 10-03 ( 3.50 × 10-02 3.95 × 10-04 ( 1.50 × 10-02 3.64 ( 1.00 × 10-02

429.00 ( 5.00 361.00 ( 3.00 327.00 344.00 ( 17.00 291.00 ( 3.00 302.00 ( 25.00 440.00 ( 35.00 526.00 ( 7.00 374.00 ( 2.00 376.00 ( 6.00 359.00 ( 6.00 481.00 ( 5.00 300.00 ( 2.00 327.00 334.00 ( 23.00 322.00 ( 21.00 350.00 ( 15.00 370.00 ( 6.00 350.00 ( 18.00 361.00 ( 23.00 392.00 ( 8.00

118.00 ( 5.80 137.95 ( 0.45 120.75 ( 3.95 133.25 ( 13.65 109.35 ( 1.05 136.85 ( 1.35 147.50 ( 2.05 181.70 ( 8.50 118.40 ( 1.50 150.45 ( 10.05 176.10 ( 4.30 143.85 ( 19.45 163.2 ( 0.90 153.95 ( 0.65 163.15 ( 4.25 175.70 ( 3.60 197.20 ( 45.20 178.35 ( 3.55 161.65 ( 1.35 138.60 ( 14.80 150.65 ( 10.75

βc (mV/decade)

16.44 ( 0.24 19.23 ( 2.19 22.01 ( 0.25 24.54 ( 1.85 21.79 ( 0.37 37.25 ( 2.03 60.20 ( 0.40 132.75 ( 3.45 28.36 ( 4.33 84.17 ( 18.44 181.10 ( 6.80 33.28 ( 1.29 63.01 ( 2.24 70.32 ( 2.98 65.04 ( 7.56 168.60 ( 3.90 171.35 ( 1.95 198.75 ( 30.25 111.25 ( 1.95 34.59 ( 0.36 34.19 ( 0.56

0.42 ( 0.01 0.49 ( 0.06 0.56 ( 0.01 0.62 ( 0.05 0.55 ( 0.01 0.95 ( 0.05 1.53 ( 0.01 3.37 ( 0.09 0.72 ( 0.11 2.14 ( 0.47 4.60 ( 0.17 0.85 ( 0.03 1.60 ( 0.07 1.79 ( 0.08 1.65 ( 0.19 4.28 ( 0.10 4.35 ( 0.05 5.05 ( 0.77 2.83 ( 0.05 0.88 ( 0.01 0.87 ( 0.01

(mmpy)

corrosion rate (mpy)

66.55 ( 1.79 71.07 ( 1.58 92.14 ( 3.62 89.58 ( 6.13 85.79 ( 5.85 99.78 ( 8.82 103.90 ( 3.05 180.25 ( 2.05 88.14 ( 5.49 170.35 ( 5.65 462.20 18.95 ( 2.74 108.55 ( 2.75 98.41 ( 0.30 93.81 ( 2.36 208.95 ( 27.45 176.80 ( 41.20 353.40 ( 50.30 138.50 ( 4.90 87.45 ( 3.77 87.07 ( 8.67

βa (mV/decade)

Etrans (mV SCE)

1.80 × 10 ( 4.25 × 10 2.11 × 1002 ( 2.57 × 1001 2.39 × 1002 ( 2.70 × 1000 2.66 × 1002 ( 2.00 × 1001 2.37 × 1002 ( 4.05 × 1000 4.04 × 1002 ( 2.21 × 1001 6.54 × 1002 1.44 × 1003 ( 3.75 × 1001 3.07 × 1002 ( 4.71 × 1001 6.50 × 1002 ( 6.37 × 1001 1.97 × 1003 ( 7.35 × 1001 3.61 × 1002 ( 1.39 × 1001 6.84 × 1002 ( 2.42 × 1001 7.64 × 1002 ( 3.23 × 1001 7.06 × 1002 ( 8.21 × 1001 1.83 × 1003 ( 4.20 × 1001 1.37 × 1003 ( 4.70 × 1002 2.16 × 1003 ( 3.28 × 1002 1.21 × 1003 ( 2.15 × 1001 4.14 × 1002 ( 4.29 × 1001 3.71 × 1002 ( 6.05 × 1000 02

icorr (µA)

ipass (A/cm2)

-916.00 ( 3.00 -939.00 ( 8.25 -973.00 ( 1.65 -972.00 ( 9.40 -948.00 ( 8.00 -961.00 ( 2.50 -854.00 ( 0.10 -834.00 ( 11.20 -940.00 ( 0.10 -921.00 ( 3.65 -916.00 ( 0.55 -919.00 -948.35 -935.00 ( 1.60 -932.00 ( 2.90 -912.00 ( 0.55 -905.00 ( 0.90 -907.00 ( 0.50 -903.00 ( 1.00 -951.00 ( 0.50 -944.00 ( 0.50

Ecorr (mV SCE)

ic (A/cm2)

26.02 ( 0.11 25.16 ( 0.05 24.62 ( 0.05 20.79 ( 0.21 20.35 ( 0.23 14.31 ( 0.15 33.03 15.09 ( 0.27 19.57 ( 0.03 18.01 ( 0.01 14.30 ( 0.10 8.19 ( 0.01 14.57 ( 0.09 23.04 ( 0.02 23.06 ( 0.11 21.97 ( 0.03 2.50 ( 0.06 24.30 ( 0.04 24.39 ( 0.03 21.58 ( 0.01 21.12 ( 0.01

9.16 ( 0.04 9.18 9.22 ( 0.04 9.35 ( 0.03 9.35 ( 0.04 9.57 ( 0.01 7.94 8.02 ( 0.01 9.37 ( 0.04 9.16 ( 0.02 9.31 ( 0.03 10.49 ( 0.01 9.37 ( 0.06 9.24 ( 0.02 9.23 ( 0.01 9.26 ( 0.02 9.38 ( 0.02 9.47 ( 0.08 9.41 ( 0.03 9.93 ( 0.03 9.90 ( 0.12

Epp (mV SCE)

σ (mS/cm)

pH

a Where σ ) conductivity, βa ) anodic Tafel slope, βc ) cathodic Tafel slope, Ecorr ) corrosion potential, icorr ) corrosion current density, EPP ) primary passivation potential, ic ) critical current density, ipass ) passivation current density, Etrans ) transpassive potential, R ) CO2 loading (mol/mol). b Total amine concentration 6.2 kmol/m3.

80 °C, R ) 0.20, MEA/PZ 5.0/1.2 kmol/m3

10.13 kPa O2, 40 °C 10.13 kPa O2, 80 °C 80 °C, R ) 0.20, MEA/PZ 5.0/1.2 kmol/m3

5.07 kPa O2, 80 °C 0.00 kPa O2, 80 °C, R)0.20

0.00 kPa O2, 80 °C

0.00 kPa O2, 80 °C, R ) 0.20

0.00 kPa O2, 80 °C, R ) 0.20

3

MEA 5.0 kmol/m MEA 6.2 kmol/m3 MEA 7.0 kmol/m3 MEA 8.7 kmol/m3 MEA/PZ -5.0/1.2 kmol/m3 MEA/PZ -7.0/1.7 kmol/m3 R ) 0.63, MEA/PZ 5.0/1.2 kmol/m3 R ) 0.58, MEA/PZ 7.0/1.7 kmol/m3 R ) 0.20, MEA/PZ 5.0/1.2 kmol/m3 b MEA/PZ 2:1 mol ratio b MEA/PZ 1:1 mol ratio R ) 0.20, MEA/PZ 7.0/1.7 kmol/m3 R ) 0.20, MEA/PZ 7.0/1.7 kmol/m3 0.00 kPa O2, 1.00 wt % oxalate 10.13 kPa O2, 1.00 wt % oxalate 0.00 kPa O2, 1.00 wt % acetate 10.13 kPa O2 1.00 wt % acetate 0.00 kPa O2, 1.00 wt % formate 10.13 kPa O2, 1.00 wt % formate 0.00 kPa O2, 1.00 wt % thiosulfate 10.13 kPa O2, 1.00 wt % thiosulfate

experimental condition

80 °C, R ) 0.20, MEA/PZ 5.0/1.2 kmol/m3

10.13 kPa O2, 40 °C 10.13 kPa O2, 80 °C 80 °C, R ) 0.20, MEA/PZ 5.0/1.2 kmol/m3

5.07 kPa O2, 80 °C 0.00 kPa O2, 80 °C, R)0.20

0.00 kPa O2, 80 °C

0.00 kPa O2, 80 °C, R ) 0.20

0.00 kPa O2, 80 °C, R ) 0.20

experimental condition

Table 2. Summary of pH, Conductivity, and Electrochemical Parameters of the MEA/PZ-CO2 Systema

9302 Ind. Eng. Chem. Res., Vol. 48, No. 20, 2009

Ind. Eng. Chem. Res., Vol. 48, No. 20, 2009 -

-

2FeCO3 + 4OH f Fe2O3 + 2HCO3 + H2O + 2e

9303

-

(4) It should be noted in Table 2 that the transpassive potential (Etrans) decreases from 0.361 to 0.291 V SCE in the presence of piperazine. This suggests that the protective film formed in the piperazine blend is not as stable as that of MEA. In addition, the cyclic polarization curve shows negative hysteresis of reverse scan, thereby showing no pitting tendency. 4.2. Comparison of Corrosion Rate in MEA and Blended MEA/PZ Systems. The blended MEA/PZ solution was found to be more corrosive than the MEA solution when the corrosion comparison was made at the identical total concentration of amine. As shown in Figure 3 (at 80 °C and 0.20 mol/ mol CO2 loading), the corrosion rate of carbon steel in the blended MEA/PZ (5.0/1.2 kmol/m3) is 0.55 mmpy (21.79 mpy), while that in the 6.2 kmol/m3 MEA is 0.49 mmpy (19.23 mpy). The higher corrosion rate in the blended MEA/PZ solution is more pronounced when the total amine concentration is increased to 8.7 kmol/m3. This may be due to changes in reduction of the oxidizing agent when PZ is present in the solution. On the basis of the principle of metal complexation, the iron dissolution (reaction 1) occurs to produce Fe2+ which then combines with protonated ions of MEA and PZ. The equilibrium potential of metal complexation with PZ may be greater than that with MEA, reflecting a greater corrosion rate in a PZ system. Such alteration in oxidizing agents in blended MEA/PZ is evidenced by changes in anodic (βa) and cathodic (βa) Tafel slopes (Table 2). 4.3. Effect of Mixing Ratio of Blended MEA/PZ on Corrosion. Further experiments were carried out to examine the effect of mixing ratio of blended MEA/PZ solutions at a given total amine concentration. Results in Figure 4 show that as concentration of PZ increases, the corrosion rate increases. For example, at the total amine concentration of 6.2 kmol/m3, the corrosion rate is the highest at the mixing ratio of 1:1, followed by the mixing ratio of 2:1, 4:1, and 1:0, respectively. The increasing corrosion rate is due to the increasing amount of PZ as oxidizing agent, which results in a greater rate of oxidizer reduction. This is evidenced by the increases in cathodic current densities in Figure 4b. This essentially confirms the findings in the previous section in that PZ is more corrosive than MEA and governs the corrosion effect in the blended MEA/PZ solution. 4.4. Effect of Total Concentration of Blended MEA/PZ Solution on Corrosion. Amine concentration has an apparent effect on corrosion. As seen from Figure 3, increasing amine concentration makes the MEA/PZ systems more corrosive, and, thus, accelerates the corrosion rates of carbon steel. A similar effect was also observed in the MEA system. Such increase in corrosion rate is due to the fact that the higher total amine concentration leads to a greater amount of the oxidizing agents including dissolved CO2 (HCO3-) and protonated ions of amines, inducing a greater rate of iron dissolution. 4.5. Effect of CO2 Loading on Corrosion. The blended MEA/ PZ solutions containing 0.20 mol/mol CO2 loading and CO2 saturation were used for examining the effect of CO2 loading on corrosion. The results in Figure 5 (0.20 mol/mol; solution temperature ) 80 °C; in the absence of O2) show that the CO2 loading has a significant effect on corrosion rate of carbon steel. A higher CO2 loading causes the solution to be more corrosive. For instance, in blended MEA/PZ (5.0/1.2 kmol/m3) solution at 80 °C, the corrosion rate of carbon steel increases from 0.55 to 1.53 mmpy (21.79 mpy to 60.20 mpy) when the CO2 loading is increased from 0.20 to 0.63 mol/mol (saturation). Such a CO2 loading effect was found to be more pronounced at a higher

Figure 3. Comparison of corrosion rates of carbon steel in MEA and blended MEA/PZ solutions containing 0.20 mol/mol CO2 loading at 80 °C (A ) MEA 5.0 kmol/m3; B ) MEA 6.2 kmol/m3; C ) MEA 7.0 kmol/m3; D ) MEA 8.7 kmol/m3; E ) MEA/PZ 5.0/1.2 kmol/m3; F ) MEA/PZ 7.0/1.7 kmol/m3).

Figure 4. Effect of mixing ratio of blended MEA/PZ solution on (a) corrosion rate and (b) polarization behavior of carbon steel (total amine concentration ) 6.2 kmol/m3; CO2 loading ) 0.20 mol/mol; solution temperature ) 80 °C; in the absence of O2).

amine concentration. For the blended MEA/PZ solution (7.0/ 1.7 kmol/m3), the corrosion rate increases from 0.95 to 3.37 mmpy (37.25 to 132.75 mpy). Such an increase in corrosion rate is due to the increase in dissolved CO2 (HCO3-), which induces more iron dissolution. This is evidenced by greater cathodic current densities in Figure 6. This could also be seen by the decrease in pH values of the system with higher CO2 loading and the increase in conductivity values (Table 2).

9304

Ind. Eng. Chem. Res., Vol. 48, No. 20, 2009

Figure 5. Comparison of corrosion rates of carbon steel in blended MEA/ PZ solution containing different CO2 loadings at 80 °C (A ) MEA/PZ 5.0/1.2 kmol/m3 0.2 mol/mol CO2 loading; B ) MEA/PZ 5.0/1.2 kmol/m3 0.63 mol/mol CO2 loading; C ) MEA/PZ 7.0/1.7 kmol/m3 0.2 mol/mol CO2 loading; D ) MEA/PZ 7.0/1.7 kmol/m3 0.58 mol/mol CO2 loading).

Figure 7. Effect of solution temperature on (a) corrosion rate and (b) polarization behavior of carbon steel in the blended MEA/PZ (7.0/1.7 kmol/ m3) solution containing 0.20 mol/mol CO2 loading under 10.13 kPa O2.

Figure 6. Effect of CO2 loading on polarization behavior of carbon steel in blended MEA/PZ solution with the blended concentration of (a) 5.0/1.2 and (b) 7.0/1.7 kmol/m3 at 80 °C.

4.6. Effect of Solution Temperature on Corrosion. The effect of solution temperature was studied at 40 and 80 °C in the presence of O2. Results in Figure 7 show that the solution temperature has a considerable effect on corrosion rate. The corrosion rate increases from 0.85 to 1.60 mmpy (33.28 to 63.01 mpy) as the temperature is increased from 40 to 80 °C. This can be explained by the dependence of reaction kinetics on temperature. It is well-known that the reaction rate increases with temperature. Therefore, the increase in temperature increases rates of metal dissolution and oxidizer reduction, thereby accelerating the corrosion process. This is shown by the shifts of both anodic and cathodic current densities toward greater

Figure 8. Cyclic polarization curves of carbon steel in an aqueous solution of blended MEA/PZ (5.0/1.2 kmol/m3) containing 0.20 mol/mol CO2 loading under 5.07 kPa O2 at 80 °C.

values (Figure 7b). It should also be noted that the increase in temperature leads to the decreasing pH from 10.49 to 9.37 and increasing conductivity from 8.19 to 14.57 (Table 2). 4.7. Effect of O2 on Corrosion. The effect of O2 on corrosion was studied using an aqueous solution of blended MEA/PZ (5.0/ 1.2 kmol/m3) containing 0.20 mol/mol CO2 loading under 5.07 kPa O2 at 80 °C. Results in Figure 8 show that the polarization behavior of carbon steel in the presence of O2 is similar to that in the absence of O2 (Figure 2b). That is, the carbon steel manifests active, passive, and transpassive behavior and shows negative hysteresis indicating no pitting tendency.

Ind. Eng. Chem. Res., Vol. 48, No. 20, 2009

Figure 9. Effect of O2 partial pressure on (a) polarization behavior and (b) corrosion rates of carbon steel in aqueous solutions of blended MEA/PZ (5.0/1.2 kmol/m3) containing 0.20 mol/mol CO2 loading at 80 °C.

Under the tested condition, the carbon steel is in an active state and corroded with the rate of 0.72 mmpy (28.36 mpy) (Figure 9), which is higher than the rate in the absence of O2 (0.55 mmpy or 21.79 mpy). The increase in corrosion rate in the presence of O2 is attributed to the change in oxidizer reduction reactions as demonstrated by an increase in βc value from 109.35 to 118.40 mV/decade (Table 2). As O2 is dissolved in the aqueous solution of MEA/PZ, it becomes an additional oxidizing agent, which reacts with undissociated water (H2O) to form hydroxyl ion (OH-) as shown below.14 Reduction of dissolved O2: O2 + 2H2O + 4e- f 4OH-

(5)

This additional reduction consequently induces a greater rate of corrosion. Moreover, the dissolved O2 is a cathodic depolarizer that reacts with and removes the corrosion products from the cathode, thereby permitting the attack to continue.14 This means that the cathodic reaction can be accelerated in the presence of O2, thereby increasing the corrosion rate. From Figure 9a, it is interesting to note that in the passive region (-0.778-0.374 V SCE), the passive current density of the system containing dissolved O2 is generally lower than that containing no O2. This implies that if the carbon steel is to be protected from corrosion by raising the system’s potential to the passive region, the presence of dissolved O2 is beneficial to the formation of a passive film and helps reduce the corrosion rate to a greater extent compared to the system without O2. Such a passive film is speculated to be hematite (Fe2O3) according to the Pourbaix diagram.10

9305

Figure 10. Comparison of corrosion rates of carbon steel in a blended MEA/ PZ (5.0/1.2 kmol/m3) solution containing 0.20 mol/mol CO2 loading, 1 wt % heat stable salts under (a) 0.00 and (b) 10.13 kPa O2 at 80 °C.

4.8. Effect of Heat-Stable Salts on Corrosion. Heat-stable salts are irreversible degradation products of amine solution. The presence of heat stable salts affects the corrosion behavior significantly. The corrosion behavior of carbon steel immersed in the aqueous solution of blended MEA/PZ containing four types of heat-stable salts, acetate, formate, oxalate, and thiosulfate, was investigated. Results in Figure 10 show that the presence of heat-stable salts increases the corrosion rate of the system in the presence and absence of O2. Formate is the most corrosive salt, followed by acetate, oxalate, and thiosulfate in the absence of O2. Note that the differences in corrosion rate between formate and acetate are very small. Pitting tendency was observed in the presence of oxalate. In the presence of O2, a different trend of corrosion is exhibited. Acetate is the most corrosive salt followed by formate, oxalate, and thiosulfate. The increase in corrosion rate can be explained by considering polarization behavior. It is apparent from Figure 11 and Table 2 that anodic and cathodic Tafel slopes change when heat-stable salts were added to the solutions. This indicates that the heatstable salts alter the corrosion mechanism on both anode and cathode sites (i.e. iron dissolution and reduction of oxidizing agent, respectively). The presence of heat-stable salts essentially introduces an additional oxidizing agent, heat-stable salt anions, to the corrosion process. This induces higher rates of iron dissolution and oxidizer reduction as evidenced by the increasing anodic and cathodic current densities. The presence of heatstable salts also increases the conductivity of solution (Table 2), indicating greater corrosion rates. 5. Conclusions The blended MEA/PZ solutions are more corrosive than the MEA solutions when the corrosion comparison was made at

9306

Ind. Eng. Chem. Res., Vol. 48, No. 20, 2009

required during plant operation to suppress the corrosion rate of carbon steel to below an acceptable level. Acknowledgment The authors gratefully acknowledge financial support from the U.S. Department of Energy (a subaward under the research grant of Dr. Gary T. Rochelle). Literature Cited

Figure 11. Effect of heat-stable salts on polarization behavior of carbon steel in blended MEA/PZ (5.0/1.2 kmol/m3) solutions containing 0.20 mol/ mol CO2 loading, 1 wt % heat-stable salts at 80 °C under (a) 0.00 and (b) 10.13 kPa O2.

identical total molar concentrations of amine. Increasing the PZ concentration, total concentration of blended solution, CO2 loading, and solution temperature increases the corrosion rate. The presence of heat-stable salts increases the corrosion rate of the system in both the presence and absence of dissolved O2. Among the tested salts, formate was the most corrosive salt, followed by acetate, oxalate, and thiosulfate in the absence of O2. Pitting may occur in the deareated solution containing oxalate. In the presence of O2, a different trend of corrosion was exhibited. Acetate was the most corrosive salt, followed by formate, oxalate, and thiosulfate. Corrosion control is

(1) Aaron, D.; Tsouris, C. Separation of CO2 From Flue Gas: A Review. Sep. Sci. Technol. 2005, 40, 321–348. (2) Bishnoi, S.; Rochelle, G. Absorption of Carbon Dioxide in Aqueous Piperazine/ Methyldiethanolamine. AIChE J. 2002, 48 (12), 2788–2799. (3) Dang, H.; Rochelle, G. T. CO2 Absorption Rate and Solubility in Monoethanolamine/ Piperazine/ Water. Sep. Sci. Technol. 2003, 38 (2), 337– 357. (4) Cullinane, J.; Rochelle, G. Carbon Dioxide Absorption with Aqueous Potassium Carbonate Promoted by Piperazine. Chem. Eng. Sci. 2004, 59, 3619–3630. (5) Bishnoi, S.; Rochelle, G. Absorption of Carbon Dioxide into Aqueous Piperazine: Reaction Kinetics, Solubility and Mass-Transfer. Chem. Eng. Sci. 2000, 55, 5531–5543. (6) Tan, C.; Chen, J. Absorption of Carbon Dioxide with Piperazine and Its Mixtures in a Rotating Packed Bed. Sep. Purif. Technol. 2006, 49, 174–180. (7) ASTM standard G1-90 (Re-approVed 1999). Standard Practice for Preparing, Cleaning and EValuating Corrosion Test Specimens; American Society of Testing and Materials: West Conshohocken, PA, 1999. (8) Horowitz, W. Association of Official Analytical Chemists (AOAC) Methods, 12th ed.; George Banta: Wisconsin, 1975. (9) ASTM standard G5-94 (ReapproVed 1999). Standard reference test method for making potentiostatic and potentiodynamic anodic polarization measurements; American Society of Testing and Materials: West Conshohocken, PA, 1999. (10) Nainar, M. Corrosion in CO2 Absorption Process Using Aqueous Monoethanolamine/Piperazine Solution. M.Sc. Thesis, University of Regina, Canada, 2007. (11) McIntire, G.; Lippet, J.; Yudelson, J. The Effect of Dissolved CO2 and O2 on the Corrosion of Iron. Corrosion. 1990, 46 (2), 91–95. (12) Parkins, R. N.; Forouslis, Z. Stress Corrosion Cracking of Mild Steel in Monoethanolamine Solutions. Mater. Performance 1988, (Jan 13), 19–29. (13) Frolova, L. V.; Fokin, M. N.; Zorina, V. E. Corrosion - Electrochemical behavior of carbon steels in carbonate- bicarbonate solutions. Protect. Met. 1997, 33 (3), 249–252. (14) Roberge, P. Corrosion basics - An introduction, 2nd ed.; NACE International: Texas, 2006.

ReceiVed for reView November 24, 2008 ReVised manuscript receiVed September 3, 2009 Accepted September 9, 2009 IE801802A