Determination of Salt Impurities in MDEA Solution Used in Desulfurization of Highly Sulphurous Natural Gas

Liu Yucheng; Zhang Bo; Chen Mingyan; Wu Danni; Zhou Zheng

(1. School of Chemistry and Chemical Engineering, Southwest Petroleum University, Chengdu 610500; 2. Puguang Branch of Zhongyuan Oilfield Company, Dazhou 636156)

Determination of Salt Impurities in MDEA Solution Used in Desulfurization of Highly Sulphurous Natural Gas

Liu Yucheng1; Zhang Bo1; Chen Mingyan1; Wu Danni1; Zhou Zheng2

(1. School of Chemistry and Chemical Engineering, Southwest Petroleum University, Chengdu 610500; 2. Puguang Branch of Zhongyuan Oilfield Company, Dazhou 636156)

The foaming phenomenon of N-methyldiethanolamine (MDEA) solution used in desulfurization process occurs frequently in the natural-gas purification plant. The foaming phenomenon has a strong impact on operation of the process unit. The salt impurities are the main reason for causing the foaming of MDEA solution, so the full analysis of salt impurities is necessary. A method for comprehensive analysis of salt impurities in MDEA solution used in desulfurization process was established. Anions and non-metallic cations of MDEA solution were determined by different conditions of ion chromatograph, respectively. Metallic cations of the solution were detected by atomic absorption spectrophotometer with the N2O-C2H2flame absorption. The analytical results of salt impurities in the desulfurization solution can provide a theoretical basis for an accurate analysis of the factors affecting the foaming of MDEA to unveil further control measures.

natural-gas purification plant; MDEA solution; salt impurities; full analysis; ion chromatograph

1 Introduction

MDEA solution is widely used in gas purification process because it has good selectivity for H2S[1]. Foaming of amine solution is a problem that is frequently encountered in the amine absorption process, which usually results in serious consequences such as severe entrainment of MDEA solution, low yield of product gas, loss of desulfurization capacity, and failure in regeneration of amine solution[2-3]. In order to prevent the foaming of MDEA solution, the factors that induce foaming must be first found out. Generally speaking, the composition and content of impurities in MDEA desulfurization solution, which constitute the specific factors inducing the foaming of MDEA, should be analyzed. The impurities of MDEA solution can be generally divided into three types, namely: salts, organics and solid particles. The salt impurities also include conventional salts and heat-stable salts. And the heat-stable salts are the combined products of some anions and amine ions. Most studies focused on the organic components of MDEA. Dawodu and Meisen investigated all the organic degradation products of MDEA, DEA, and MEA[4]. However, studies involving the qualitative and quantitative analysis of salt impurities of MDEA solution have not been published, especially for the analysis of all types of salt impurities. Only Wang and Lu reported the determination of few heat-stable salts of MDEA solution[5]. Therefore, a systematic study of salt impurities in MDEA solution is necessary.

2 Experimental

2.1 Instrumentation and chemicals

A Metrohm (Herisan, Switzerland) ion chromatograph (Mod. 883 Basic IC Plus) consisting of a pump (IC liquid handling unit), a packed bed suppressor and a conductivity detector was used. The analytical columns comprised a Metrosep A Supp 5 column (250 mm×4.0 mm in ID, 6.1006.530) for anion separation and a Metrosep C4 cation column (150 mm×4.0 mm in ID, 6.1050.420). The guard column was a Metrosep RP 2 column (1.0 mm× 3.5 mm i. d., 6.1011.030), which functioned as a universal guard column. The results were analyzed using the Metrodata IC Net 2.3 software. In addition, an atomic absorption spectrophotometer (East & West AA-7020 AAS, Beijing, China) was also used.

Sodium carbonate, sodium bicarbonate, sodium oxalate,sodium acetate, sodium formate, sodium fluoride, sodium chloride, sodium bromide, sodium nitrite, sodium nitrate, sodium sulfate, sodium phosphate tribasic dodecahydrate, sodium thiosulfate, sodium sulfocyanate, and ammonium chloride were purchased from the Kelong Chemical Reagent Company (Chengdu, China), and all chemicals were of AR grade in purity. Acetone with a purity of HPLC grade was also purchased from Kelong. The stock solution of calcium ions, iron ions, potassium ions, magnesium ions, strontium ions, manganese ions and sodium ions all had the concentration of 1 000 mg/L and was purchased from the Standard Material Research Center (Beijing, China). The regenerated MDEA solution samples were obtained from the natural gas purification plant of Puguang (Dazhou, China).

2.2 Experimental methods

The anions and non-metallic cations of the regenerated MDEA solution were determined by ion chromatography (IC). The metallic cations of MDEA solution were detected by a flame atomic absorption spectrophotometer (AAS).

3 Results and Discussion

3.1 Determination of anions

3.1.1 Eluent condition

The standard samples containing F-, Cl-, Br-,CH3COO-, HCOO-,and SCN-ions were prepared with deionized water, respectively. Then, they were subject to filtration under vacuum through a 0.45 μm polycarbonate filter prior to being routed to ion chromatograph (IC) for analysis. The IC was run at a conductivity cell temperature of 25 ℃ and an eluent flow rate of 0.7 mL/min of eluent (which was composed of 3.2 mmol/L of Na2CO3and 1.0 mmol/L of NaHCO3), with the injected sample size equating to 20 μL. The test results showed that the peak shapes of the standard samples of anions were all good. However, the peaks ofand SCN-ions all appeared at 35.5 min. The mixed standard sample ofand SCN-ions was injected into IC for analysis, with the chromatogram presented in Figure 1. The mixed standard sample ofand SCN-ions only presented one peak at 35.5 min as shown in Figure 1. There were two factors to be taken into account when studying why the sample only showed one peak. One reason is that the S2O32-ions of this sample was oxidized in the course of pretreatment or testing procedure[6]. Another is that theand SCN-ions failed to separate because the ratio of eluent composition for IC use was unsuitable[7]. Therefore, in order to prevent the oxidation of S2O32-ions, a sulfide antioxidant buffer (SAOB, vitamin C compounded with a certain proportion of NaOH) was added to the mixed standard sample[6], with the IC analysis result shown in Figure 2. In addition, the eluent for IC was adjusted to a mixture comprising 3.0 mmol/L of Na2CO3+ 1.5 mmol/L of NaHCO3, with the IC analysis result presented in Figure 3.

It can be seen from Figure 2 that the mixed standard sample ofand SCN-ions still showed one peak when SAOB was added. The sample presented two peaks when the eluent for IC use was modified to contain 3.0 mmol/L of Na2CO3+ 1.5 mmol/L of NaHCO3as depicted in Figure 3. However, the effect for separation of two peaks was poor and the peak of SCN-ions revealed the tailing phenomena. Therefore, the ratio of eluent of IC should be readjusted, and an organic modifier (acetone) was added into the eluent to alter the tailing phenomena. Through trial and error, the two peaks separated completely and the peak shapes were good, when the eluent was composed of 3.0 mmol/L of Na2CO3+ 2.5 mmol/L of NaHCO3and 14% (volume fraction) of acetone was added to it. Accordingly, this eluent composition was used to analyze the regenerated MDEA solution.

Figure 1 Chromatogram of mixed standard sample ofand SCN-ions

Figure 2 Chromatogram of mixed standard sample after addition of SAOB added

Figure 3 Chromatogram of mixed standard sample using an eluent consisting of 3.0 mmol/L of Na2CO3+ 1.5 mmol/L of NaHCO3

3.1.2 Standard curves

The anions species of the sample solution were determined according to the specific retention time, and their concentration was obtained by comparing the standard curves using known concentration of the respective compounds. Eight mixed standard samples of F-, Cl-, Br-,, CH3COO-, HCOO-,and SCN-ions were injected into IC for analysis, the concentration of which was 0.5, 1.0, 2.0, 5.0, 10.0, 20.0, 40.0, and 80.0 mg/L, respectively. The standard curve of each anion was obtained according to the fact that the different peak area responded to the different concentration value, as shown in Figure 4.

As exhibited in Figure 4, the correlation coefficient (R2) for each standard curve was 0.999. It demonstrated that the detection precision of this experiment was excellent and the results were stable and reliable when the concentration of each anion ranged from 0.5 mg/L to 80.0 mg/L.

Figure 4 Standard curves of different anions

3.1.3 Analysis of sample solutions

As it has been known from the previous section of this article that in order to obtain the reliable analytical result, the regenerated MDEA solution should be diluted with deionized water to make itself within the detection range. And then, the sample solutions were filtered under vacuum through a 0.45 μm polycarbonate filter prior to being injected to IC for analysis. After some experiments, each anion could meet the detection range requirements when the dilution ratio of the regenerated MDEA solution was equal to 20. The IC analysis result of anions in MDEA solution is presented in Figure 5 and Table 1.

It can be seen from Figure 5 that F-, CH3COO-, HCOO-, Cl-, Br-,and SCN-ions appeared in the regenerated MDEA solution, andandions were not identified. The chromatogram showed that the peak of each anion in MDEA solution had a good separating effect without any tailing phenomenon. Table 1 shows that the concentration of CH3COO-, HCOO-and Cl-ions was relatively high, which was equal to 1 221.456 mg/L, 1 103.729 mg/L, and 847.943 mg/L, respectively.

Figure 5 Chromatogram of anions of regenerated MDEA solution

Table 1 IC analysis results of anions of regenerated MDEA solution

3.2 Determination of cations

3.2.1 Metallic cations

For the quick and efficient determination of metallic cations, an AAS with N2O-C2H2flame absorption was used. The specific test conditions of different cations are shown in Table 2. The concentration of each cation of the regenerated MDEA solution could be obtained by means of the standard curve method. Standard curve was prepared by using the 1 000 mg/L standard stock solution, which was diluted with deionized water to comply with the detection range requirement of this cation, and thereby the standard solutions were obtained. However, different ions had different detection range requirements, as depicted in Table 3. Similarly, in order to detect one cation of the regener-ated MDEA solution, the sample solution must be diluted to comply with the detection range of this cation. For different cations, the dilution ratios of sample solutions are presented in Table 3. After being diluted, the sample solutions were filtered under vacuum through a 0.45 μm polycarbonate filter prior to being directed to AAS for analysis. The results relating to the AAS analysis of regenerated MDEA solution are shown in Table 4.

Table 2 Conditions for AAS test of different cations

Table 3 Detection ranges and dilution ratios of sample solutions for different cations

Table 4 Results on AAS analysis of metallic cations of regenerated MDEA solution

It can be seen from Table 4 that the concentration of sodium salt and potassium salt was 3 984.468 2 mg/L and 329.461 7 mg/L, respectively, and the rest of metallic salt concentration ranged from 0 to 1.5 mg/L. Calcium salt, potassium salt, magnesium salt, strontium salt, manganese salt, and sodium salt could be introduced through the make-up water or the formation water in the form of fluoride, chloride, and bromide. Because the acid gases were entrained in the feed gas and their aqueous solution could corrode the equipment, ferric salt could be the product of equipment corrosion[11-12].

3.2.2 Non-metallic cations

As it has been known from the previous work, the ammonium salt could be produced in the regenerated MDEA solution[13]. Thus, for the determination of non-metallic cations, this experiment was mainly focused on the determination of ammonium ions. Before the experiment, the anion analysis column (Metrosep A Supp 5) of IC was the first to be taken down, and then the packed bed suppressor was removed, with the pipeline of IC being reconnected again. The pipeline of IC was washed with deionized water for 2 h after the pump was manually started. Then, the pipeline was flushed with the eluent having a concentration of 2 mmol/L of HNO3for one hour. At the end, the cations analysis column (Metrosep C4) was connected to IC. In other words, the solution sample was directly introduced into the detector for testing after passing through the analysis column.

Figure 6 Standard curves of

The correlation coefficient (R2) for the standard curve ofspecies was 0.999 9, as shown in Figure 6. It suggested that the analysis results were accurate and reliable when the concentration ofspecies ranged from 0.05 mg/L to 5.0 mg/L. According to this detection range, the concentration of the regenerated MDEA solution should be adjusted to be within this range. Some experiments showed that the regenerated MDEA solution did not have to be diluted and could be used for determination directly. The IC analysis result ofspecies of the regenerated MDEA solution is presented in Figure 7 and Table 5. Ammonium salt was found in the regenerated MDEA solution as evidenced by Figure 7. However, the concentrationof it was relatively low, which was equal to 0.123 8 mg/L as shown in Table 5. The ammonium salt of this solution might be formed via the chemical degradation of MDEA solution or might come from the formation water[13].

Table 5 IC analysis result ofions of regenerated MDEA solution

Table 5 IC analysis result ofions of regenerated MDEA solution

Non-metallic anions Retention time, min Peak area, min·μS·cm-1Concentration, mg/L NH4+5.573 0.1015 0.1238

Figure 7 Chromatogram ofions of regenerated MDEA solution

4 Conclusions

(1) A full analysis method for salt impurities of MDEA solution used in the desulfurization process was established. The IC with Metrosep A Supp 5 was used as analytical columns, a mixture containing 3.0 mmol/L of Na2CO3, 2.5 mmol/L of NaHCO3, and 14% of (CH3)2CO functioned as the eluent operating at a flow rate of 0.7 mL/min, and the mode of suppressed conductivity detection was used to determine the anions of the regenerated MDEA solution. Under these conditions, the fluoride, acetate, formate, chloride, bromide, sulfate, oxalate, thiosulfate, and thiocyanate species were detected by IC, with their concentration verified.

(2) The calcium salt, potassium salt, magnesium salt, strontium salt, manganese salt, sodium salt and ferric salt and their concentration were determined by AAS equipped with a N2O-C2H2flame absorption device. The ammonium salt was tested through IC, which was operated by using 2.0 mmol/L HNO3as the eluent introduced at a flow rate of 0.9 mL/min according to the mode of direct conductivity detection.

(3) The type and concentration of salt impurities of the regenerated MDEA solution were quickly and comprehensively analyzed by the established analysis method. Meanwhile, the analysis results could provide an evidence for the study of factors affecting the foaming of MDEA solution.

[1] Chakma A, Meisen A. Methyl-diethanolamine degradation--mechanism and kinetics[J]. The Canadian Journal of Chemical Engineering, 1997, 75(5): 861-871

[2] Reza J, Trejo A. Degradation of aqueous solutions of alkanolamine blends at high temperature, under the presence of CO2and H2S[J]. Chemical Engineering Communications, 2006, 193(1): 129-138

[3] Wu Jinqiao, Zhang Ningsheng, Wu Xinmin, et al. Study on foaming reasons of MDEA desulfurization solution in the second gas purification plant of Changqing gas field (Ι): Analysis on causes of liquid blocking of desulfurization unit[J]. Natural Gas Industry, 2005, 25(4): 168-170 (in Chinese)

[4] Dawodu O F, Meisen A. Degradation of alkanolamine blends by carbon dioxide[J]. The Canadian Journal of Chemical Engineering, 1996, 74(6): 960-966

[5] Wang Wenqiang, Lu Keping. Analysis of heat stable salts in methyl-diethanolamine solution used in desulfurization[J]. Petrochemical Technology, 2007, 36(1): 88-91 (in Chinese)

[6] Keller-Lehmann B, Corrie S, Ravn R, et al. 2nd International IWA Conference on Sewer Operation and Maintenance[C]. Vienna, Austria, 2006

[7] O’Reilly J W, Dicinoski G W, Shaw M J, et al. Chromatographic and electrophoretic separation of inorganic sulfur and sulfur–oxygen species[J]. Analytica Chimica Acta, 2001, 432(2): 165-192

[8] Fredriksen S B, Jens K J. Oxidative degradation of aqueous amine solutions of MEA, AMP, MDEA, Pz: A review[J]. Energy Procedia, 2013, 37: 1770-1777

[9] Davoudi M, Safadoust A R, Mansoori S A A, et al. The impurities effect on thermal degradation and corrosivity of amine solution in South Pars gas sweetening plants[J]. Journal of Natural Gas Science and Engineering, 2014, 19: 116-124

[10] Verma N, Verma A. Amine system problems arisingfrom heat stable salts and solutions to improve system performance[J]. Fuel Processing Technology, 2009, 90(4): 483-489

[11] DuPart M S, Bacon T R, Edwards D J. Understanding corrosion in alkanolamine gas treating plants: Part 1[J]. Hydrocarbon Processing, 1993, 72(4): 75-89

[12] Rooney P C, Bacon T R, DuPart M S. Effect of heat stable salts on MDEA solution corrosivity[J]. Hydrocarbon Processing, 1996(3): 95-102

[13] Kamiński M, Jastrz?bski D, Przyjazny A, et al. Determination of the amount of wash amines and ammonium ion in desulfurization products of process gases and results of related studies[J]. Journal of Chromatography A, 2002, 947(2): 217-225

Acceptance of Key Technology for Design of Integrated Petrochemical Processes

The Sinopec’s key science and technical project “Development and application of key technology for design of integrated petrochemical processes” implemented by the Sinopec Engineering Incorporation (SEI) has passed the acceptance checks by experts organized by the Sinopec Corp. The research team after defining and implementing the design of integrated petrochemical processes has systematically optimized the work flow scheme of process design to achieve transition from the file-based work flow scheme to the data-based integrated, coordinated and intelligent work flow scheme to fill the gap in the area of petrochemical process design. Furthermore, a process integration design platform based on the technical standards, calculation methods, process data and design documents that constitute the core has been set up in order to develop a connector to a lot of process software and design documents, which can effectively integrate and manage the computing data and final design documents to grapple with some key technical problems related with integrated and coordinated design so that a solid foundation can be established for the construction of digitalized and intelligent plants capable of providing advanced design technology with core competitive edge among the engineering construction institutions.

Till now, this technology has been applied in more than ten projects, including the gas fractionation unit at the Shanghai Petrochemical Company, the natural gas purification unit in the Yuanba gasfield, the CNOOC’s Taizhou hydrogen plant and the pyrolysis gasoline hydrotreating plant in Huizhou, Guangdong province.

Successful Pilot Scale Test for Manufacture of Acrylic Acid through Oxo-synthesis of Acetylene

A thousand-ton-class pilot unit for manufacture of acrylic acid through oxo-synthesis of acetylene was successfully put on stream at the Anhui Vinylon Group Company (AVGC) in May 2015. This pilot test project was jointly developed by the Shanghai Pujing Chemical Technology Co., Ltd., AVGC and the CAS Chengdu Organic Chemistry Co., Ltd. The construction of this unit was started in March 2013 and the mechanical completion was finalized in May 2014, with the design capacity of this unit equating to 1000 t/a. All facilities in the process flowsheet have been tested in 72 hours of stable operation to meet the requirements for premium acrylic acid product specified by GB/T 17529.1-2008.

The technology for manufacture of acrylic acid through oxo-synthesis intends to use acetylene originated from the chloralkali plant, CO in the calcium carbide offgas and water as the feedstocks to produce acrylic acid through one-step process as a green chemical technology. This technology has been assimilating and using the experience and merits of overseas technology of similar industrial sector to shorten the links of process scheme and decrease the manufacturing and operating cost of equipment, resulting in recycling of water and various materials in the flowsheet without discharge of waste liquids. In the meantime, the product cost is less than that produced by the propylene oxidation route to assume an apparent competitive advantage.

gements: Financial support

from the Major National Science and Technology Projects of China (No. 2011ZX05017) and SWPU Science & Technology Innovation Youth Team for Pollution Control of Oil & Gas Fields (No. 2013XJZT003) is gratefully acknowledged.

Received date: 2015-04-17; Accepted date: 2015-08-24.

Liu Yucheng, E-mail: rehuo2013@sina.cn.