Article pubs.acs.org/IECR
Physical and Chemical Resistance of Elastomers in Aqueous Monoethanolamine (MEA) and CO2‑Loaded MEA Solutions during Postcombustion Carbon Dioxide Capture Processes Wayuta Srisang, Teeradet Supap, Raphael Idem,* and Paitoon Tontiwachwuthikul International Test Centre for CO2 Capture, Process Systems Engineering, Faculty of Engineering & Applied Science, University of Regina, Regina, Saskatchewan, S4S 0A2 Canada ABSTRACT: The resistance of raw EPDM, natural rubber, isobutylene isoprene, and styrene butadiene to amines in terms of mass and chemical changes was studied using aqueous solutions of 3−7 M MEA with 0−0.5 mol CO2/mol MEA each for 30 days at 40 °C. The results showed that SBR and NR had poor chemical resistance leading to formation of amides on their surfaces and resulting in high mass change. In contrast, EPDM and IIR had insignificant mass and chemical structure changes. Commercial EPDM and IIR were then benchmarked against PTFE, using aqueous 5 M MEA and with 0.5 mol CO2/mol MEA at 40 and 120 °C each for 30 days. Resistance was measured in terms of mass, hardness, and tensile strength changes. The results showed that PTFE was compatible with the solution at both temperatures. For EPDM and IIR at 40 °C, the changes in mass, hardness, and tensile strength were negligible while at 120 °C, these changes were very significant, implying that PTFE can be used in any part of the process while EPDM and IIR can only be used in low-temperature sections.
1. INTRODUCTION The problem of global warming has generated interest in developing processes for removal of carbon dioxide (CO2) from exhaust gas streams of industrial processes, such as fossil fuel-based electric power generation before releasing into atmosphere. One of the mature methods for CO2 removal from low-pressure gas streams is absorption using a chemical solvent.1 In a typical process for CO2 absorption with chemical reaction, two main vessels are used, namely, an absorber and a regenerator. The absorption of gas into liquid occurs in the absorber at relatively low temperatures, while desorption occurs in the regenerator with a requirement of heat. Various absorbents are used for gas absorption, such as NaOH, K2CO3, amines, NH3, or amino-alcohols. Monoethanolamine (MEA), a primary alkanolamine, is the common amine used as an absorbent due to its high reactivity as compared to other amines, and can be regenerated simply by heating. In the amine-based absorption process, seals and gaskets are used to prevent leakage of gas or liquid to the environment. Seals and gaskets can be made from fibers, metals, and elastomers with elastomer being the most commonly used because of its low cost and wide variety of chemical structures. However, care needs to be taken to choose the right kind of elastomer because there may be incompatibility between elastomers and the liquid used as the absorbent in the process. Any incompatibility between the elastomer and liquid or gas when in contact can lead to degradation of the elastomer. Currently, the selection of elastomers used as seal and gasket materials in chemical plants are guided mostly by compatibility tables provided by elastomer commercial suppliers. These tables normally rate elastomer materials with various chemicals with which potentially they will come into contact. The data in compatibility tables are typically obtained from static immersion experiments in pure or aqueous solution within a small range of temperatures. Some of the common amine © 2014 American Chemical Society
solvents including MEA, diethanolamine (DEA), and triethanolamine (TEA), used for CO2 capture process are also found in these tables. Compatibility data of these amine solvents with various types of elastomers are widely accessible. However, the data are often too generic since they are mostly generated outside of CO2 capture conditions. In CO2 absorption by amine solutions such as MEA, the composition and temperature of the solution change along the process. In addition, the amines may be degraded during the absorption of CO2 unavoidably introducing degradation products and heat stable salts into the solution, according to Rennie.2 These impurities definitely change the amine properties (e.g., pH and concentration) potentially causing the rubber gaskets and seals in the CO2 plant equipment to be more susceptible to premature breakage, leakage, and failure than what is originally suggested in the compatibility tables. As a result, it is not sufficient to rely only on the information from the elastomer companies. Tests that are specifically designed to evaluate amine resistance under the actual CO2 absorption conditions of potential rubbers that would be used with the CO2 plant must be performed. Information regarding chemical resistance of rubber materials for a specific application of the amine−CO2 capture process is extremely rare. In this work, four potential elastomers consisting of ethylene propylene diene monomers (EPDM), styrene butadiene (SBR), isobutylene isoprene rubber (IIR), and natural rubber (NR) were selected to evaluate their chemical compatibility to aqueous MEA solutions under the actual CO2 capture process. The elastomers were chosen based on their literature Received: Revised: Accepted: Published: 5932
February 11, 2014 March 17, 2014 March 20, 2014 March 20, 2014 dx.doi.org/10.1021/ie5005288 | Ind. Eng. Chem. Res. 2014, 53, 5932−5940
Industrial & Engineering Chemistry Research
Article
Table 1. Chemical Structures of Elastomers
used for the mobile phase of high performance liquid chromatography (HPLC). A 600 mL stainless steel batch reactor model 5500 (Parr Instrument Co., Moline, IL) equipped with a programmable temperature controller model 4836 with a tachometer display module (Parr Instrument Company, IL, USA) was utilized in high temperature commercial elastomer testing. ATR-FTIR spectra of elastomer samples were recorded from Nicolet iS5 with diamond crystal (Thermo Fisher Scientific Inc., Canada). The nuclear magnetic resonance (NMR) used for analysis of elastomer gel was a model Agilent/Varian MercuryPlus 300 M Hz with an ATB (Automation Triple Resonance Broadband) probe. The HPLC used for analysis of the liquid samples was equipped with refractive index detector (RID), diode array detector (DAD), and an online degasser (model 1100/ G1362A/G1315B, Agilent Technologies Canada, Mississauga, Ontario, Canada). The column used was Nucleosil 100-5 SA containing a strong cationic exchanger of sulfonic acid (Macherey-Nagel, Germany) 250 mm in length and 4.6 mm in diameter. The samples were introduced to HPLC automatically by an autosampler (model G1313A Agilent Technologies Canada, Mississauga, Ontario, Canada). Density and viscosity of the solutions were measured using 2000 ME Microviscometer (Anton Paar, Graz, Australia) equipped with density meter (DMA 4100/4500/5000 M), viscosity meter (DSA 5000
information and compatibility data suggested by different suppliers.
2. EXPERIMENTS 2.1. Materials and Chemicals. The raw elastomers used in this study were EPDM, SBR, NR, and IIR. They were obtained from MDR International Co., Ltd., Bangkok, Thailand. The commercial elastomers used in the study were commercial EPDM, IIR, and PTFE. They were in the form of sheets with thickness of 1.57 mm and were purchased from Rubber Sheet Roll, Pennsylvania, USA. The chemical structures of elastomers are shown in Table 1. The aqueous MEA solutions (of molarities of 3, 5, and 7 M) as well as CO2-loaded aqueous 5 M MEA (with loadings of 0.16, 0.25, 0.5 mol CO2/mol MEA) solutions were prepared from laboratory grade MEA purchased from Fisher Scientific, Ontario, Canada. Pure research grade CO2 cylinder used for the CO2-loaded experiments was supplied by PRAXAIR, Regina, Canada. Hydrochloric acid of concentration of 1 kmol/m3 was purchased from VWR, USA, which, along with 0.1% methyl orange solution (obtained from Sigma-Aldrich, Canada), was used in the titration to determine the CO2 loading using the method provided by the Association of Official Analytical Chemists (AOAC).3 Potassium phosphate monobasic (KH2PO4) and phosphoric acid (H3PO4) solution reagent grade purchased from Sigma-Aldrich, Canada, were 5933
dx.doi.org/10.1021/ie5005288 | Ind. Eng. Chem. Res. 2014, 53, 5932−5940
Industrial & Engineering Chemistry Research
Article
the reference cell served the purpose of distinguishing between dissolution components from elastomers and degradation products of MEA solutions themselves. After 30 days of immersion, both the elastomer samples and the solutions were analyzed. Characterization of Materials. Prior to elastomer analysis, the remaining trace solutions in the elastomer samples was washed off using deionized water in an ultrasonic washer, and washed samples were left to dry at room temperature. The samples were weighed every 24 h to constant weight. Mass change of elastomers was calculated as in eq 1.
M), and an automatic sample changer (Xsample). The refractive index of the solutions was measured with Abbemat 550 (Anton Paar, Graz, Australia). 2.2. Method. The experiment was performed in two parts, namely, (i) the screening of raw elastomers and (ii) the testing of commercial elastomers. Raw elastomers were tested in preliminary experiments because, according to Schweitzer,4 the resistance to chemicals by seals and gaskets material depended mainly on the base polymers. Also, as reported by Schweitzer,4 additives or fillers are mostly responsible for improving their physical properties. The four selected raw elastomers were first tested at low temperature to identify elastomers with good resistance to the MEA solution while the elastomers in their commercial state were tested at both low and high temperatures. 2.2.1. Raw Elastomer Compatibility Tests. This test was conducted to screen selected raw elastomers (i.e., EPDM, SBR, NR, and IIR) identified as unfinished rubbers without any fillers and additives. The test was adapted from ASTM D 471-10, the standard test method for rubber property-effect of liquid. It consisted of immersion of the elastomer in aqueous MEA solution. Since the raw rubbers received from the suppliers were mostly nonuniform, the weight of each specimen rather than size was used as the testing parameter. The weight of each sample was approximately 2 g. The specimens were immersed in 100 mL of the desired concentration of aqueous MEA solution (i.e., 3, 5, 7 M) contained in a test cell as shown in Figure 1. For the CO2
⎛ M − M1 ⎞ %ΔM = ⎜ 2 ⎟ × 100 ⎝ M1 ⎠
(1)
where %ΔM, M1 (g), and M2 (g) are percentage of mass change, mass of specimen before immersion, and mass of specimen after immersion, respectively. In order to investigate the chemical change in the elastomers after immersion in the solutions, FTIR spectra of each immersed elastomer were recorded at room temperature in the wavenumber range of 550−4000 cm−1. The resolution and the number of scans were 6 cm−1 and 32, respectively. Also, prior to these measurements, the samples were washed to get rid of remaining trace of MEA solution and then dried. Analysis of Used Amine Samples. The solutions remaining after immersion were analyzed using HPLC. Physical properties of the solutions such as density, viscosity, and refractive index were also measured. The mobile phase used for HPLC was 0.05 M potassium hydrogen phosphate prepared from KH2PO4 and adjusted to a pH of 2.6 by adding 85% w/w H3PO4.5,6 The amine samples were diluted to 1:100 then introduced to HPLC with an injection volume of 20 μL. Density and viscosity of the solutions were measured. The results obtained from cells with elastomers were compared with those from reference cells. 2.2.2. Immersion of Commercial Elastomers. The commercial elastomers EPDM, IIR, and PTFE were cut into dumbbell shapes (ASTM Die C) and tested for resistance in a solution of 5 M MEA with 0.5 mol CO2/mol MEA at the temperature of 40 and 120 °C. For the tests at 40 °C, which is the temperature typically used in the absorber, the immersion was carried out in glass test tubes as mentioned earlier in the raw elastomer testing. Three pieces of specimens and 120 mL of the solutions were loaded into the glass test tubes. The test tubes were put in the water bath to control the temperature at 40 °C. The temperature of 120 °C, which is the temperature used in the CO2 stripping section of the process, was used for these tests. The apparatus and procedure used for the raw elastomer testing is not suitable at 120 °C; therefore, the immersion tests of the commercial elastomers was carried out in a 600 mL stainless steel batch reactor model 5500 (Parr Instrument Co., Moline, IL). This served to ensure that the compositions of the solutions remained constant during the test. For each run, the three dumbbell specimens of known weight and 450 mL of the solutions were loaded by hanging to the reactor head. The experimental temperature was set at 120 °C and kept constant for the 30 days of immersion. At the end of the immersion period, the temperature controller was set to 25 °C to cool down the solvents and the specimens. This took approximately 24−72 h. Then, the specimens were removed from the reactor. Each specimen was quickly dipped in acetone and blotted lightly with filter paper free of foreign material as in ASTM
Figure 1. Immersion cell diagram (ASTM D 471-10).
loaded solution, research grade CO2 gas was fed into the solution by bubbling prior to immersion of the sample. The CO2 loading in each solution was obtained from titration against standard solution of 1 kmol/m3 HCl. A water bath capable of controlling temperature to an accuracy of ±5 °C was used to control the temperature of the MEA solution and test specimen at 40 °C. A reference test cell without any test specimens containing only MEA solution similar to that used in the sample test cell was also kept under the same test condition. During the immersion period, MEA might degrade in the presence of CO2 and generate degradation products in the solution. Therefore, 5934
dx.doi.org/10.1021/ie5005288 | Ind. Eng. Chem. Res. 2014, 53, 5932−5940
Industrial & Engineering Chemistry Research
Article
D471. The weights of each specimen were measured with a four decimal weight scale. Three pieces of the specimen were tested for the physical properties of tensile strength (ASTM D412) and hardness (Shore A durometer) at the University of Manitoba, Winnipeg, Canada.
3. RESULTS AND DISCUSSION 3.1. Raw Elastomer Testing. 3.1.1. Mass Change of Raw Elastomers. The percentage mass change of the elastomers (IIR, EPDM, SBR, and NR) after the immersions for 30 days at 40 °C was calculated and plotted against MEA concentration (3, 5, and 7 M) as shown in Figure 2. It shows that the
Figure 2. Effect of MEA concentration on percent mass change of elastomers (errors of the percent mass change are within 0.006− 1.369). Experimental conditions: MEA (3, 5, and 7 M), 30 days of immersion, 40 °C.
percentage mass change of the elastomers was positive at almost every MEA concentration, implying that the mass of the elastomers increased in almost every case during immersion. NR had the highest percentage mass change compared to the other elastomers. This was followed by SBR and EPDM. The mass changes of IIR were zero at every MEA concentration. The mass change of elastomer is one of the parameters used to indicate the compatibility between elastomers and liquids. According to Haseeb et al.,7 the incompatibility of elastomer in liquid aqueous MEA medium can result in either a decrease or an increase in mass. These authors have shown that the adsorption or chemical reaction to form new compounds on elastomer surfaces causes an increase in mass, while the dissolution of elastomers contributes to a loss in mass. These phenomena can occur simultaneously. Therefore, the measured mass is the net mass of elastomers from these processes. The increase in the mass of NR immersed in the MEA solutions was also apparent based on physical observation (Figure 3a), as the NR specimens were swollen with significant color change after immersion unlike IIR. NR is a homopolymer of cis-1,4-polyisoprene. The structure of the polymer has a low level of interlocking between the polymer chains resulting in high permeability of gas and liquid.8 It has also been reported that the double bonds in the isoprene structure lead to chemical instability.9,10 For these two reasons of high permeability of gas and liquid and low chemical stability, NR was very likely to interact with MEA by physical and chemical adsorption resulting in an increase in mass. SBR was another elastomer with a high content of double bonds in its elastomer chain. It is possible that MEA chemically reacted with the double bonds in the SBR elastomer chain due to their high chemical reactivity.
Figure 3. (A) Raw elastomers: (a) IIR, (b) EPDM, (c) SBR, and (d) NR. Experimental conditions: 30 days of immersion, 40 °C. (1) Specimen before immersion. (2−4) Specimens after immersion in MEA 3, 5, and 7 M. (5−7) Specimens after immersion in MEA 5 M with 0.16, 0.25, and 0.5 mol CO2/mol MEA. (B) Effect of CO2 loading on the percent mass change of elastomers (IIR, EPDM, SBR, and NR; errors of the percent mass change are within 0.005−0.656. Experimental conditions: 5 M MEA (0.16, 0.25, and 0.5 mol CO2/ mol MEA in), 30 days of immersion, 40 °C.
The percentage mass change for EPDM increased with MEA concentration. This was similar to the results for SBR. However, the percentage mass change was very small. According to Tan et al.,11 as the backbone of EPDM contains mainly the monomers of ethylene and propylene, which is saturated and highly chemically stable, the chances of EPDM reacting with MEA is very low. The increase of mass of EPDM was likely due to the remaining MEA in the specimens. IIR was the only elastomer in the experiment that had no mass change during immersion. According to Dong et al.,12 IIR is a polymer with a high content of isobutylene and a small amount of isoprene (i.e., approximately 3%). Also, according to Chandrasekaran,8 the methyl groups present in isobutylene 5935
dx.doi.org/10.1021/ie5005288 | Ind. Eng. Chem. Res. 2014, 53, 5932−5940
Industrial & Engineering Chemistry Research
Article
monomer act as an interlock in the polymer chain, giving IIR its outstanding low permeability of gas and liquid (MEA), thus making it difficult for MEA to adsorbed on the IIR elastomer, thereby reducing the chances of occurrence of any reactions. Figure 3b reveals the effect of CO2 loading on the percentage mass change of the elastomers. The result shows that increasing the CO2 loading led to a reduction in percentage mass change of NR. In MEA/CO2 solutions, MEA and CO2 typically interact to form a complex electrolyte system containing mainly free MEA, carbamate, protonated MEA, and a small amount of bicarbonate and carbonate ions. The concentration of these species is a function of the CO2 loading.13 The ion species in the solution contributes to an increase in solvent polarity. According to Georgoulis et al.,14 NR as a nonpolar elastomer is more swollen in nonpolar solvents than in polar solvents. Therefore, it can be concluded that it is the dissolved CO2 that contributed to the higher polarity of the solution which decreased the swelling of NR. SBR is the only elastomer for which the percentage mass change was found to be negative after immersion in the MEA/ CO2 solutions. Moreover, the percentage loss of mass became more pronounced with increasing CO2 loading. In other words, increased CO2 loading led to a higher loss of SBR mass into the solutions. This might be because CO2 acted as an oxidant.15 According to Williams, 16 the oxidation of elastomers contributes to chain scissoring and dissolution of the elastomer into the solution. Moreover, the higher CO2 loading of MEA solutions increased the degree of oxidation and contributed to more solvation of the SBR. For EPDM and IIR, the CO2 loaded into aqueous MEA did not have any significant effect. The percentage mass change of EPDM was very low and almost constant with the variation in CO2 loading. Furthermore, the percentage mass change for IIR at every CO2 loading in 5 M MEA was close to zero which could be regarded as being negligible. 3.1.2. Chemical Change of Raw Elastomers. Figure 4a shows the FTIR spectra of the raw elastomers before and after their immersion in 5 M aqueous MEA solution. After the immersion of EPDM and IIR in the solutions, only some small peaks occurred in their FTIR spectra; otherwise, there was no major difference observed in the FTIR spectra between elastomer specimens before and after their immersions even in CO2-loaded aqueous MEA solutions. In contrast, several new peaks appeared in the FTIR spectra of NR and SBR after their immersion in both the CO2-loaded and unloaded MEA solutions (Figure 4b and c, respectively). In the case of CO2loaded aqueous MEA solutions, by comparing the FTIR spectra of the original raw elastomers and their corresponding spectra after immersion, there are some small peaks that appeared at 1070 and 1052 cm−1 which usually represent the C−O stretching vibration.17,18 The small peak at 719 cm−1 indicates the N−H wagging vibration, and the sharp peak at 3296 cm−1 represents N−H stretching vibration. These two peaks usually appear in secondary amines or amides. The peak at 1641 cm−1 results from the amide I band which is the carbonyl stretching vibration. According to Sedghi et al.13 and Mosadegh-Sedghi et al.,19 the peaks at 1551 and 1515 cm−1 represent the amide II band which is mainly due to the N−H bending vibration. The interference peaks in this area occur due to the intermolecular and intramolecular hydrogen bond N−H.20 The new peaks at 1415 and 1466 cm−1 could be from some carboxylate compounds (COO−), according to Hedzelek et al.21 This is in agreement with the research of Sedghi et al.13 who also
Figure 4. (a) Comparison of FTIR before and after the immersion of raw elastomers in 5 M aqueous MEA solution. Experimental conditions: 30 days of immersion, 40 °C. (b) FTIR spectra of NR. Experimental conditions: 30 days of immersion, 40 °C. (c) FTIR spectra of SBR. Experimental conditions: 30 days of immersion, 40 °C.
found an amide group on the LDPE surface when these researchers studied the degradation of LDPE in MEA solution. The formation of these amide groups was as a result of the auto-oxidation reaction which contributed to the formation of −COOH groups, followed by reaction of −COOH groups and 5936
dx.doi.org/10.1021/ie5005288 | Ind. Eng. Chem. Res. 2014, 53, 5932−5940
Industrial & Engineering Chemistry Research
Article
MEA resulting in the formation of secondary amides that was detected on the material’s surface.13 It is well-known that commercial NR and SBR are crosslinked and compounded with processing aids and fillers. This means that the double bond sites in the elastomer chains are reduced by cross-linking and the properties of the elastomers will improve. However, it has been indicated8 that, in general, the amount of cross-linking is limited to 1−5%.8 Therefore, a large fraction of active sites still remain in the elastomer chain which will not be cross-linked and which are still prone to the risks of the occurrence of reactions with the amine environment. Thus, NR and SBR are still capable to react with both the CO2-loaded and unloaded MEA solutions. Also, processing aids and fillers are added to elastomers to make commercially useable products. These components usually affect the mechanical properties of the commercial elastomers such as tensile strength, hardness, life expectancy, and color.4 In summary, even though after NR and SBR are processed by cross-linking and by adding fillers and processing aids which can contribute to lowering double bond sites and improving their physical properties, the high amount of double bond sites in the elastomers’ chain still remaining are still prone to chemical reaction with CO2-loaded and unloaded MEA solutions. For these reasons, it was decided not to include NR and SBR in subsequent tests. 3.1.3. Solution Analysis. After the immersion, the changes in all the solutions were first physically observed. It was found that most solutions did not have any significant color change after the immersion. However, in one condition, the immersion of SBR in the solutions of MEA with CO2 resulted in the solutions becoming turbid (Figure 5) and the turbidity of the solutions
Figure 6. (a) 1H NMR spectrum of the sludge from 5 M MEA with 0.5 mol CO2/mol MEA after immersion of SBR. Experimental conditions: 30 days of immersion, 40 °C. (b) 13C NMR spectrum of the sludge from 5 M MEA with 0.5 mol CO2/mol MEA after immersion of SBR. Experimental conditions: 30 days of immersion, 40 °C.
and styrene monomers caused the reduction in mass of SBR during immersion as shown in the results of mass change, in which it was observed that the increase of CO2 concentration resulted in increased solvation of butadiene and styrene monomers. All the results starting from mass change, FTIR which indicated the formation of amide compounds, as well as the NMR results of the suspended gel were further interpreted in order to explain the possible interaction during the immersion of SBR in MEA/CO2 solutions (5 M MEA with 0.16, 0.25, and 0.5 mol CO2/mol MEA). The NMR showed that the dissolved CO2 might have caused the oxidation at the double bonds in the SBR chain and cause chain scissoring. The chain scissoring of the SBR resulted in the possible dissolution of butadiene and styrene monomers, as was obviously seen in the form of suspended gel and confirmed by NMR results. The oxidation reaction not only caused chain scissoring but also caused the formation of a carbonyl group on the surface of SBR. This carbonyl group reacted further with MEA in the solution to form amide on the surface of the SBR as confirmed by the results from FTIR. Beside these observations, HPLC with a diode array detector (DAD) was used to analyze the liquid samples after the immersion of elastomers. It was not possible from the results to identify any differences between the MEA solutions after the immersion of IIR, EPDM, and NR and those solutions without the immersions. Also, the solutions of MEA/CO2 after the
Figure 5. MEA solutions after the immersion of SBR. Experimental conditions: 30 days of immersion, 40 °C. (a) 5 M MEA and 5 M with 0.16, 0.25, and 0.5 mol CO2/mol MEA. (b) 5 M MEA with 0.5 mol CO2/mol MEA after the immersion of SBR with suspended gel.
increased with increasing CO2 loading. The suspended gel in this solution was separated and analyzed by NMR. Figure 6a shows the 1H NMR spectrum of the sludge obtained from the solution of 5 M aqueous MEA solution with 0.5 mol CO2/mol MEA after the immersions of SBR for 30 days at 25 °C. The chemical shift at 3.678 ppm comes from free MEA,22 while the small peaks near 7 ppm come from aromatic proton and the chemical shift at 2.4−0.8 ppm comes from aliphatic proton.23 From the 13C NMR spectra shown in Figure 6b, the peaks at 61 and 42 ppm indicate the chemical shift of C from CH2OH and CH2NH2 in free MEA.22 The chemical shift around 29−22 ppm is the carbon long chain from butadiene including cis, trans, and vinyl units.24 The suspended gel in the solution may be monomers of butadiene and styrene based on the results of 1 H NMR and 13C NMR analyses. The solvation of butadiene 5937
dx.doi.org/10.1021/ie5005288 | Ind. Eng. Chem. Res. 2014, 53, 5932−5940
Industrial & Engineering Chemistry Research
Article
the solution of 5 M with 0.5 mol CO2/mol MEA at 40 and 120 °C for 30 days. The results show that the mass of EPDM and IIR increased after the immersion at both 40 and 120 °C while the mass of PTFE did not change significantly at either temperature. As illustrated by Haseeb et al.,7 commercial elastomers are basically a complex mixture of base polymer, as well as fillers and additives such as plasticizer, curing agents, antioxidants, antiozonants, and processing aids. The increase in mass can be as a result of higher liquid absorption compared to the extraction of soluble components from the elastomers. The absorption of liquid into the elastomers could be either chemical absorption or purely physical absorption. In the case of EPDM and IIR, the chemical resistance of base polymers (raw EPDM and IIR) to MEA and MEA/CO2 solutions at 40 °C were previously tested in which the results showed that both elastomers did not react with MEA and MEA/CO2 solutions. Therefore, the increase of mass of commercial EPDM and IIR at 40 °C might be more of physical absorption or chemical reaction between the solution with fillers and additives and less of the extraction of soluble components that might also occur. According to Schweitzer,4 PTFE is considered as a thermosetting plastic with a crystalline structure. As such, MEA solution can hardly diffuse into the packed structure of PTFE. Moreover, it has the outstanding ability of high chemical resistance due to the bonds between carbon atoms and fluorine atoms; thus, the chemical reaction between PTFE and the MEA solution is difficult. These might be the reasons why the mass of PTFE remained constant after the immersion. Even when the temperature was increased, the mass of PTFE remained almost constant. On the other hand, with an increase in temperature, the change in mass increased for EPDM and IIR. For an increase in temperature from 40 to 120 °C, the percentage mass change increased from 2.1% to 44.3% for EPDM and 2.3% to 43.1% for IIR. This can be because the increase of temperature facilitated the diffusion rate of the solution in commercial EPDM and IIR, as pointed out by Haseeb et al.7 Figures 10 and 11 show comparisons of hardness and tensile strengths of commercial EPDM, IIR, and PTFE obtained
immersion of EPDM, IIR, NR, and SBR did not produce any significant differences from those of the virgin solutions. Only one odd result was obtained in the MEA solutions after the immersion of SBR. New components which were labeled as Unknown A and Unknown B, occurred in the solutions. Figures 7 and 8 are the plot of relative concentrations of Unknown A
Figure 7. Effect of MEA concentration (3, 5, and 7 M) on the concentration of Unknown A. Experimental conditions: 30 days of immersion of SBR, 40 °C.
Figure 8. Effect of MEA concentration (3, 5, and 7 M) on the concentration of Unknown B. Experimental conditions: 30 days of immersion of SBR, 40 °C.
and B versus the MEA concentrations. From the results, the relative concentration of Unknown A and B increased with MEA concentration. It is important to note that the chromatograms provided from DAD have the limitation that it only responds to a specific range of compounds such as compounds with double bonds in their structure. The lack of differentiation in the chromatograms of the solutions from the elastomer immersion and the reference solutions might either indicate that even though some components dissolved from the elastomers into the solutions, these were undetectable by DAD. 3.2. Tests for Commercial Elastomers. Figure 9 shows the change in mass for IIR, EPDM, and PTFE after exposure to
Figure 10. Hardness of commercial EPDM, IIR, and PTFE. 5 M with 0.5 mol CO2/mol MEA at 40 and 120 °C for 30 days.
before immersion and after immersion. At 40 °C, the hardness of commercial EPDM, IIR, and PTFE had no significant changes. With the increase in immersion temperature to 120 °C, the hardness of PTFE still remained almost constant, while the hardness of EPDM and IIR decreased sharply. After exposure to the solution at 40 °C, the tensile strength of commercial EPDM decreased slightly as shown in Figure 12. There was a drastic decrease in tensile strength with an increase
Figure 9. Mass change of commercial EPDM, IIR, and PTFE after immersion. Errors of the percentage mass change are within 0.017− 0.571. 5 M with 0.5 mol CO2/mol MEA at 40 and 120 °C for 30 days. 5938
dx.doi.org/10.1021/ie5005288 | Ind. Eng. Chem. Res. 2014, 53, 5932−5940
Industrial & Engineering Chemistry Research
Article
Figure 11. Tensile strength of commercial EPDM, IIR, and PTFE. 5 M with 0.5 mol CO2/mol MEA at 40 and 120 °C for 30 days.
in temperature. Similarly, there was no significant difference in tensile strength for commercial IIR between before and after it was exposed to the solution at 40 °C; however, the tensile strength of IIR rapidly dropped after being exposed to the solution at 120 °C. Unlike those of EPDM and IIR, the tensile strength of PTFE did not have any significant change after the immersion in MEA solution at both low and high temperatures. The changes of tensile strength and hardness of commercial EPDM and IIR are in agreement with the results of mass changes. The absorption of solution reduced the mechanical properties of the material.25
■
6. At low CO2 concentration, the formation of amides on SBR and NR surfaces increased because CO2 accelerates the oxidation reaction. At high CO2 concentration, the formation of amide started to decrease because MEA in the solution was consumed in the competing carbamate formation reaction. 7. Out of the commercial EPDM, IIR, and PTFE elastomers, only PTFE showed excellent chemical compatibility with the MEA solution used in the CO2 capture process at both high and low temperatures. It is therefore the recommended material for use in both the high-temperature sections (e.g., regenerator and reboiler) and low-temperature sections (e.g., absorber) of the CO2 capture plant. 8. EPDM and IIR were only resistant at about 40 °C. Therefore, these materials might be used in the absorber as only small changes on their masses and physical properties were observed at the absorber temperature. They are not recommended for use in the regenerator.
AUTHOR INFORMATION
Corresponding Author
*Phone: 1(306)585-4470. Fax: 1(306)585-4855. E-mail:Ra
[email protected]. Notes
The authors declare no competing financial interest.
4. CONCLUSIONS 1. Out of the raw EPDM, IIR, SBR, and NR elastomers selected from compatibility tables and studied for their chemical resistance to aqueous solutions of MEA and MEA/CO2 at 40 °C, only EPDM and IIR exhibited compatibility with the aqueous CO2-loaded and unloaded amine solutions. 2. After the immersion of EPDM and IIR in aqueous MEA and MEA/CO2 solutions, the mass of EPDM slightly increased due to physical absorption while the mass of IIR remained constant. Moreover, from FTIR interpretation, there was no significant change in chemical structure of EPDM and IIR. Therefore, raw EPDM and IIR can be recommended to be the base polymers used in sealing material in an amine-based CO2 capture process. 3. Raw SBR and NR have poor chemical resistance to MEA and MEA/CO2 solutions because they react with the MEA solutions to form amides on their surface that is observable from FTIR spectra analysis. The amides occur from the oxidation of the elastomers. The intermediate products from the oxidation reaction then react further with MEA to form amides on the elastomer surface as the final product. 4. A higher MEA concentration contributed to an increase in the formation of amides resulting in an increase in mass of SBR and NR after their immersion in CO2loaded and non-aqueous MEA solution. The higher the MEA concentration, the more MEA molecules are available for the amide formation reactions. 5. With the addition of CO2 into MEA solution, there was the dissolution of styrene and butadiene monomers for SBR which resulted in a decrease in SBR mass after its immersion in the amine solution. Solvation of monomer was not observed for NR.
■
ACKNOWLEDGMENTS
■
REFERENCES
The financial support provided by the Natural Science and Engineering Research Council of Canada (NSERC) is gratefully acknowledged.
(1) Chakma, A. An Energy Efficient Mixed Solvent for The Separation of CO2. Energy Convers. Manage. 1995, 36 (6), 427−430. (2) Rennie, S. Corrosion and Materials Selection for Amine Service. Mater. Forum. 2006, 30, 126−130. (3) Horwitz, W. Official Methods of Analysis of The Association of Official Analytical Chemists; Association of Official Analytical Chemists, 1975. (4) Schweitzer, P. A. Corrosion of Polymers and Elastomers; Taylor & Francis, 2010. (5) Supap, T.; Idem, R.; Tontiwachwuthikul, P.; Saiwan, C. Analysis of Monoethanolamine and Its Oxidative Degradation Products During CO2 Absorption from Flue Gases: A Comparative Study of GC-MS, HPLC-RID, and CE-DAD Analytical Techniques and Possible Optimum Combinations. Ind. Eng. Chem. Res. 2006, 45 (8), 2437− 2451. (6) Wattanaphan, P. Studies and Prevention of Carbon Steel Corrosion and Solvent Degradation During Amine-Based CO2 Capture from Industrial Gas Streams. Ph.D. Thesis, University of Regina, Canada, 2012. (7) Haseeb, A. S. M. A.; Masjuki, H. H.; Siang, C. T.; Fazal, M. A. Compatibility of Elastomers in Palm Biodiesel. Renewable Energy 2010, 35 (10), 2356−2361. (8) Chandrasekaran, C. Rubber Seals for Fluid and Hydraulic Systems; Elsevier Science, 2009. (9) Chaikumpollert, O.; Sae-Heng, K.; Wakisaka, O.; Mase, A.; Yamamoto, Y.; Kawahara, S. Low Temperature Degradation and Characterization of Natural Rubber. Polym. Degrad. Stab. 2011, 96 (11), 1989−1995. (10) Ciesielsky, A. An Introduction to Rubber Technology; iSmithers Rapra Publishing, 1999. 5939
dx.doi.org/10.1021/ie5005288 | Ind. Eng. Chem. Res. 2014, 53, 5932−5940
Industrial & Engineering Chemistry Research
Article
(11) Tan, J.; Chao, Y. J.; Wang, H.; Gong, J.; Van Zee, J. W. Chemical and Mechanical Stability of EPDM in a PEM Fuel Cell Environment. Polym. Degrad. Stab. 2009, 94 (11), 2072−2078. (12) Dong, X. C.; Zhang, Q. L.; Feng, Y.; Xing, Z.; Zhao, J. R. Preparation and Properties of Isobutylene-isoprene Rubber Containing Multifunctional Groups. Iran Polym J. 2010, 19 (10), 771−779. (13) Sedghi, S. M.; Brisson, J. E.; Rodrigue, D.; Iliuta, M. C. Chemical Alteration of LDPE Hollow Fibers Exposed to Monoethanolamine Solutions Used as Absorbent for CO2 Capture Process. Sep. Purif. Technol. 2011, 80 (2), 338−344. (14) Georgoulis, L. B.; Morgan, M. S.; Andrianopoulos, N.; Seferis, J. C. Swelling of Polymeric Glove Materials During Permeation by Solvent . J. Appl. Polym. Sci. 2005, 97 (3), 775−783. (15) Raju, G.; Reddy, B. M.; Park, S.-E. Utilization of Carbon Dioxide in Oxidative Dehydrogenation Reactions. Ind. J. Chem. Sect. A. 2012, 51 (9), 1315. (16) Williams, I. Swelling and Solvation of Rubber in Different Solvents. Ind. Eng. Chem. 1937, 29 (2), 172−174. (17) Ho, C.; Khew, M. Surface Characterisation of Chlorinated Unvulcanised Natural Rubber Latex Films. Int. J. Adhes. Adhes. 1999, 19 (5), 387−398. (18) Ratnam, C. T.; Nasir, M.; Baharin, A.; Zaman, K. Electron Beam Irradiation of Epoxidized Natural Rubber: FTIR Studies. Polym. Int. 2000, 49 (12), 1693−1701. (19) Mosadegh-Sedghi, S.; Brisson, J. E.; Rodrigue, D.; Iliuta, M. C. Morphological, Chemical and Thermal Stability of Microporous LDPE Hollow Fiber Membranes in Contact with Single and Mixed Amine Based CO2 Absorbents. Sep. Purif. Technol. 2012, 96 (0), 117−123. (20) Lin, S.-Y.; Chen, K.-S.; Liang, R.-C. Thermal Micro ATR/FT-IR Spectroscopic System for Quantitative Study of The Molecular Structure of Poly(N-isopropylacrylamide) in Water. Polymer 1999, 40 (10), 2619−2624. (21) Hedzelek, W.; Wachowiak, R.; Marcinkowska, A.; Domka, L. Infrared Spectroscopic Identification of Chosen Dental Materials and Natural Teeth. Acta Phys. Pol. A. 2008, 114 (2), 471. (22) Fan, G.-j.; Wee, A. G. H.; Idem, R.; Tontiwachwuthikul, P. NMR Studies of Amine Species in MEA-CO2-H2O System: Modification of the Model of Vapor-Liquid Equilibrium (VLE). Ind. Eng. Chem. Res. 2009, 48 (5), 2717−2720. (23) Jukic, A.; Rogosic, M.; Franjic, I.; Soljic, I. Molecular Interaction in Some Polymeric Additive Solutions Containing Styrene-Hydrogenated Butadiene Copolymer. Eur. Polym. J. 2009, 45 (9), 2594− 2599. (24) Arantes, T. M.; Leao, K. V.; Tavares, M. I. S. B.; Ferreira, A. G.; Longo, E.; Camargo, E. R. NMR Study of Styrene-Butadiene Rubber (SBR) and TiO2 Nanocomposites. Polym. Test. 2009, 28 (5), 490− 494. (25) Alves, S. M.; Mello, V. S.; Medeiros, J. S. Palm and Soybean Biodiesel Compatibility with Fuel System Elastomers. Tribol Int. 2013, 65 (0), 74−80.
5940
dx.doi.org/10.1021/ie5005288 | Ind. Eng. Chem. Res. 2014, 53, 5932−5940
