{"http:\/\/dx.doi.org\/10.14288\/1.0059005":{"http:\/\/vivoweb.org\/ontology\/core#departmentOrSchool":[{"value":"Applied Science, Faculty of","type":"literal","lang":"en"},{"value":"Chemical and Biological Engineering, Department of","type":"literal","lang":"en"}],"http:\/\/www.europeana.eu\/schemas\/edm\/dataProvider":[{"value":"DSpace","type":"literal","lang":"en"}],"https:\/\/open.library.ubc.ca\/terms#degreeCampus":[{"value":"UBCV","type":"literal","lang":"en"}],"http:\/\/purl.org\/dc\/terms\/creator":[{"value":"Dawodu, Olukayode Fatai","type":"literal","lang":"en"}],"http:\/\/purl.org\/dc\/terms\/issued":[{"value":"2011-01-24T23:02:15Z","type":"literal","lang":"en"},{"value":"1991","type":"literal","lang":"en"}],"http:\/\/vivoweb.org\/ontology\/core#relatedDegree":[{"value":"Doctor of Philosophy - PhD","type":"literal","lang":"en"}],"https:\/\/open.library.ubc.ca\/terms#degreeGrantor":[{"value":"University of British Columbia","type":"literal","lang":"en"}],"http:\/\/purl.org\/dc\/terms\/description":[{"value":"The common industrial practice of using aqueous solutions of diethanolamine (DEA) for the removal of impurities such as carbon dioxide (CO\u2082), hydrogen sulphide (H\u2082S), carbonyl sulphide (COS) and carbon disulphide (CS\u2082) from natural, refinery and manufactured gases often entails irreversible reactions between the solvent and the impurities. This phenomenon is referred to as amine degradation and it not only constitutes a loss of the amine but may contribute to operational problems such as foaming, corrosion and fouling.\r\nDegradation of DEA by COS and CS\u2082 was studied by using a 600 mL stainless steel reactor under the following conditions: DEA concentration 10 - 40 wt%; temperature 120 - 195 \u00b0C; COS partial pressure 345 - 1172 kPa; CS\u2082 volume 2.5 - 10.5 mL (CS\u2082\/DEA mole ratio of 0.055 - 0.233). An analytical procedure consisting of gas chromatography (GC) and gas chromatography\/mass spectroscopy (GC\/MS) was used to identify over 20 compounds in the partially degraded DEA solutions. The major degradation products are monoethanolamine (MEA), bis hydroxyethyl ethylenediamine (BHEED), bis hydroxyethyl piperazine (BHEP), hydroxyethyl oxazolidone (HEOD), hydroxyethyl imidazolidone (HEI), tris hydroxyethyl ethylenediamine (THEED) and bis hydroxyethyl imidazolidone (BHEI); as well as a dithiocarbamate salt (in the case of the CS\u2082-DEA systems). In addition, both COS and CS\u2082 induced degradation formed solid products which were characterized on the basis of solubility, melting point, elemental composition, solid probe GC\/MS and infrared analysis. The number of degradation compounds in the COS-DEA\r\nand CS\u2082-DEA systems is large when compared with the three major degradation compounds found in CO\u2082-DEA systems; this demonstrates that the former systems are distinct and more complicated than the latter system.\r\nWhen COS or CS\u2082 was contacted with aqueous DEA solution, hydrolysis occurred and H\u2082S, CO\u2082, COS and, possibly, CS\u2082 together with their related ionic species were present in the system. Solubility and hydrolysis experiments were therefore conducted to establish the equilibrium composition of the COS-DEA system prior to the commencement of degradation. A modified Kent-Eisenberg (K\/E) model which was developed to correlate the experimental data, showed good agreement between the experimental results and model predictions. Since the K\/E and previous models were limited to amine-CO\u2082 and\/or H\u2082S systems, the present modified K\/E model which incorporates COS, is a significant improvement.\r\nThe rate of degradation of DEA was found to increase with temperature, DEA concentration, COS partial pressure and CS\u2082 volume. On the basis of the experiments conducted to evaluate the contributions of the various compounds in the partially degraded solutions, reaction schemes were developed for the formation of 18 degradation compounds in the COS-DEA and CS\u2082-DEA systems. Despite the complexity of the reactions, the overall degradation of DEA was well represented by a first order reaction for the present experimental conditions. A mathematical model based on the major reaction schemes was developed to estimate the concentrations of DEA and the major degradation compounds in the COS-DEA system.\r\nContrary to literature information, experiments conducted with gas mixtures of CO\u2082 and showed that H\u2082S enhanced the rate of DEA\r\ndegradation. A direct result of the combined effects of H\u2082S and CO\u2082 on alkanolamines was the production of the corresponding lower order alkanolamines from higher order ones. The resulting mixed amine solution increases the routes for degradation compared to single amine solutions. The study therefore provides an indication of what to expect in terms of degradation when mixtures of alkanolamines are used for gas sweetening.","type":"literal","lang":"en"}],"http:\/\/www.europeana.eu\/schemas\/edm\/aggregatedCHO":[{"value":"https:\/\/circle.library.ubc.ca\/rest\/handle\/2429\/30797?expand=metadata","type":"literal","lang":"en"}],"http:\/\/www.w3.org\/2009\/08\/skos-reference\/skos.html#note":[{"value":"DEGRADATION OF DIETHANOLAMINE SOLUTIONS BY CARBONYL SULPHIDE AND CARBON DISULPHIDE b y OLUKAYODE FATAI DAWODU B.Sc. (CHEM. ENG.), UNIVERSITY OF IFE, NIGERIA, 1982 M.Sc. (CHEM. ENG.), UNIVERSITY OF IFE, NIGERIA, 1985 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES DEPARTMENT OF CHEMICAL ENGINEERING We accept t h i s thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA SEPTEMBER 1991 (c) OLUKAYODE F. DAWODU, 1991 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of CH-\u00a3(V\\t Cfti\u2014 G rv\/G.HN\/ggfiJNt G The University of British Columbia Vancouver, Canada Date Q C T o f e g g - m\\ DE-6 (2\/88) i ABSTRACT The common i n d u s t r i a l p r a c t i c e of using aqueous solutions of diethanolamine (DEA) . f o r the removal of impurities such as carbon dioxide (CC^)\/ hydrogen sulphide (H2S)\/ carbonyl sulphide (COS) and carbon disulphide (CS2) from n a t u r a l , r e f i n e r y and manufactured gases often e n t a i l s i r r e v e r s i b l e reactions between the solvent and the impurities. This phenomenon i s r e f e r r e d to as amine degradation and i t not only , constitutes a loss of the amine but may contribute to operational problems such as foaming, corrosion and f o u l i n g . Degradation of DEA by COS and CS2 was studied by using- a 600 mL s t a i n l e s s s t e e l reactor under the following conditions: DEA concentration 10 - 40 wt%; temperature 120 - 195 \u00b0C; COS p a r t i a l pressure 345 - 1172 kPa; CS 2 volume 2.5 - 10.5 mL (CS2\/DEA mole r a t i o of 0.055 - 0.233). An a n a l y t i c a l procedure c o n s i s t i n g of gas chromatography (GC) and gas chromatography\/mass spectroscopy (GC\/MS) was used to i d e n t i f y over 20 compounds i n the p a r t i a l l y degraded DEA solutions. The major degradation products are monoethanolamine (MEA), bis hydroxyethyl ethylenediamine (BHEED), bis hydroxyethyl piperazine (BHEP), hydroxyethyl oxazolidone (HEOD), hydroxyethyl imidazolidone (HEI), t r i s hydroxyethyl ethylenediamine (THEED) and bis hydroxyethyl imidazolidone (BHEI); as well as a dithiocarbamate s a l t ( in the case of the CS2-DEA systems). In addition, both COS and CS2 induced degradation formed s o l i d products which were characterized on the basis of s o l u b i l i t y , melting point, elemental composition, s o l i d probe GC\/MS and i n f r a r e d a n a l y s i s . The number of degradation compounds i n the COS-DEA ii and CS2-DEA systems i s large when compared with the three major degradation compounds found i n CC^-DEA systems; t h i s demonstrates that the former systems are d i s t i n c t and more complicated than the l a t t e r system. When COS or CS2 was contacted with aqueous DEA s o l u t i o n , hydrolysis occurred and H2S, CO2, COS and, possibly, CS2 together with t h e i r r e l a t e d i o n i c species were present i n the system. S o l u b i l i t y and hydrolysis experiments were therefore conducted to e s t a b l i s h the equilibrium composition of the COS-DEA system p r i o r to the commencement of degradation. A modified Kent-Eisenberg (K\/E) model which was developed to c o r r e l a t e the experimental data, showed good agreement between the experimental r e s u l t s and model p r e d i c t i o n s . Since the K\/E and previous models were l i m i t e d to amine-C02 and\/or H2S systems, the present modified K\/E model which incorporates COS, i s a s i g n i f i c a n t improvement. The rate of degradation of DEA was found to increase with temperature, DEA concentration, COS p a r t i a l pressure and CS2 volume. On the basis of the experiments conducted to evaluate the contributions of the various compounds i n the p a r t i a l l y degraded solutions, reaction schemes were developed f o r the formation of 18 degradation compounds i n the COS-DEA and CS2-DEA systems. Despite the complexity of the reactions, the o v e r a l l degradation of DEA was well represented by a f i r s t order reaction f or the present experimental conditions. A mathematical model based on the major reaction schemes was developed to estimate the concentrations of DEA and the major degradation compounds in the COS-DEA system. iii Contrary to l i t e r a t u r e information, experiments conducted with gas mixtures of CO2 and showed that H2S enhanced the rate of DEA degradation. A d i r e c t r e s u l t of the combined e f f e c t s of H2S and CO2 on alkanolamines was the production of the corresponding lower order alkanolamines from higher order ones. The r e s u l t i n g mixed amine solution increases the routes for degradation compared to single amine solutions. The study therefore provides an i n d i c a t i o n of what to expect in terms of degradation when mixtures of alkanolamines are used for gas sweetening. i v TABLE OF CONTENTS ABSTRACT i i LIST OF TABLES x i LIST OF FIGURES x i i i ACKNOWLEDGEMENT S x x i i CHAPTER 1 INTRODUCTION 1 1.1 OBJECTIVES OF THE PRESENT STUDY ' 6 2 LITERATURE REVIEW 7 2.1 PROPERTIES OF CARBONYL SULPHIDE AND CARBON DISULPHIDE 7 2.2 ABSORPTION OF ACIDIC GASES IN AQUEOUS ALKANOLAMINE SOLUTIONS 8 2.2.1 CARBON DIOXIDE AND HYDROGEN SULPHIDE 8 2.2.2 CARBONYL SULPHIDE 9 2.2.3 CARBON DISULPHIDE 15 2.3 EQUILIBRIUM REACTIONS OF ACID GASES IN AMINE SOLUTIONS 15 2.3.1 REACTIONS OF HYDROGEN SULPHIDE 15 2.3.2 REACTIONS OF CARBON DIOXIDE 16 2.3.3 REACTIONS OF CARBONYL SULPHIDE 18 2.3.4 REACTIONS OF CARBON DISULPHIDE 20 2.4 DEGRADATION OF DIETHANOLAMINE SOLUTIONS 22 2.4.1 DEGRADATION OF DIETHANOLAMINE BY CARBON DIOXIDE 22 v 2.4.2 DEGRADATION OF DIETHANOLAMINE BY CARBONYL SULPHIDE 28 2.4.3 DEGRADATION OF DIETHANOLAMINE BY CARBON DISULPHIDE 2 9 2.5 LIMITATIONS OF THE PREVIOUS COS-DEA AND CS2-DEA DEGRADATION STUDIES 30 2.6 ANALYSIS OF DEGRADED ALKANOLAMINE SOLUTIONS 32 3 EXPERIMENTAL APPARATUS AND PROCEDURE 34 3.1 REACTOR 34 3.2 MATERIALS 35 3.3 EXPERIMENTAL PROCEDURE 37 3.4 SAMPLING 39 3.5 OPERATING CONDITIONS 40 3.6 ANALYTICAL PROCEDURE 41 3.7 TECHNIQUES USED TO IDENTIFY THE DEGRADATION PRODUCTS 42 3.7.1 GAS CHROMATOGRAPHIC (GC) ANALYSIS 43 3.7.2 GAS CHROMATOGRAPHIC\/MASS SPECTROMETRIC (GC\/MS) ANALYSIS 43 3.7.3 GC\/MS ANALYSIS OF SILYLATED DERIVATIVES 45 3.7.4 GC ANALYSIS OF DEGRADED MIXTURE SPIKED WITH SUSPECTED COMPOUNDS 46 3.8 EXPERIMENTAL DESIGN 47 4 IDENTIFICATION OF DEGRADATION PRODUCTS 50 4.1 PRELIMINARY EXPERIMENTS 50 4.1.1 EFFECTS OF ELEVATED TEMPERATURES 50 vi 4.1.2 SURFACE EFFECTS 51 4.1.3 EFFECTS OF STIRRER SPEED 53 4.1.4 REPRODUCIBILITY 53 4.2 DEGRADATION PRODUCTS RESULTING FROM COS-DEA INTERACTIONS 54 4.3 DEGRADATION PRODUCTS RESULTING FROM CS2-DEA INTERACTIONS 76 4.4 CHARACTERIZATION OF THE SOLID PRODUCTS 78 4.4.1 SOLUBILITY 79 4.4.2 MELTING POINT 79 4.4.3 ELEMENTAL ANALYSIS 80 4.4.4 MASS SPECTRAL ANALYSIS .82 4.4.5 INFRA-RED ANALYSIS 86 5 EFFECTS OF OPERATING VARIABLES 90 5.1 COS-DEA SYSTEM 90 5.1.1 EFECTS OF INITIAL DEA CONCENTRATION 90 5.1.2 EFFECTS OF TEMPERATURE 109 5.1.3 EFFECTS OF INITIAL COS PARTIAL PRESSURE 124 5.2 CS2\"DEA SYSTEM 135 5.1.1 EFECTS OF INITIAL DEA CONCENTRATION 135 5.1.2 EFFECTS OF TEMPERATURE 149 5.1.3 EFFECTS OF INITIAL VOLUME OF CS 2 161 6 EXPERIMENTS DESIGNED TO ELUCIDATE REACTION MECHANISMS 172 6.1 EFFECTS OF MIXED GASES 172 6.2 EFFECTS OF OXYGEN 183 6.3 EFFECTS OF DEGRADATION COMPOUNDS 185 vii 6.3.1 EFFECT OF ETHANOL 185 6.3.2 EFFECT OF ACETALDEHYDE 186 6.3.3 EFFECT OF ACETIC ACID 189 6.3.4 EFFECT OF ACETONE 191 6.3.5 EFFECT OF BUTANONE 191 6.3.6 EFFECT OF ETHYLENE GLYCOL 192 6.3.7 EFFECTS OF THE ALKYL ALKANOLAMINES 192 6.3.8 EFFECT OF WATER : 200 6.3.9 EFFECT OF MONOETHANOLAMINE 202 6.3.10 EFFECT OF BHEED 204 7 SOLUBILITY AND HYDROLYSIS OF CARBONYL SULPHIDE .....205 7.1 THEORY 206 7.2 EXPERIMENTAL EQUIPMENT AND PROCEDURE 212 7.2.1 PROCEDURE 212 7.2.2 ACID GAS LOADINGS . . . .' 214 7.2.3 GAS ANALYSIS 215 7.2.3 SOLUBILITY DETERMINATION AT LOW TEMPERATURES 217 7.2.5 HYDROLYSIS AT ELEVATED TEMPERATURES 219 7.3 RESULTS AND DISCUSSION OF SOLUBILITY AND HYDROLYSIS RUNS 220 7.3.1 SOLUBILITY OF COS IN DEA SOLUTIONS AT LOW TEMPERATURES 221 7.3.2 HYDROLYSIS OF COS IN DEA SOLUTIONS AT ELEVATED TEMPERATURES 226 7.3.3 MODEL PREDICTIONS 235 viii 7.3.4 REPRODUCIBILITY 238 7.3.5 COS BALANCE 238 8 REACTION MECHANISMS 240 8.1 COS-DEA DEGRADATION 240 8.1.1 FORMATION OF MONOETHANOLAMINE (MEA) 240 8.1.2 FORMATION OF ACETALDEHYDE AND KETONES 243 8.1.3 FORMATION OF ACETIC ACID 244 8.1.4 FORMATION OF ETHYL AMINOETHANOL (EAE) 246 8.1.5 FORMATION OF DIETHYL DISULPHIDE 246 8.1.6 FORMATION OF SUBSTITUTED PYRIDINES 247 8.1.7 FORMATION OF ETHYLDIETHANOLAMINE (EDEA) 248 8.1.8 FORMATION OF N,N,N -TRIS HYDROXYETHYL ETHYLENEDIAMINE (THEED) 24 9 8.1.9 FORMATION OF BIS HYDROXYETHYL ETHYLENEDIAMINE (BHEED) 250 8.1.10 FORMATION OF N,N -BIS HYDROXYETHYL PIPERAZINE (BHEP) AND N-HYDROXYETHYL PIPERAZINE (HEP) ....251 8.1.11 FORMATION OF N-HYDROXYETHYL OXAZOLIDONE (HEOD) 251 8.1.12 FORMATION OF N,N -BIS HYDROXYETHYL IMIDAZOLIDONE (BHEI) 252 8.1.13 FORMATION OF N-HYDROXYETHYL IMIDAZOLIDONE (HEI) 253 8.1.14 FORMATION OF N-HYDROXYETHYL ACETAMIDE (HEA) ...254 8.1.15 FORMATION OF ETHANETHIOIC ACID-(S-(HYDROXYETHYL) AMINO) METHYL ESTER (ETAHEAME) .254 8.2 CS2-DEA DEGRADATION 255 8.3 FORMATION OF THE SOLID PRODUCT 257 ix 9 KINETIC MODEL FOR DEA DEGRADATION 258 9.1 COS INDUCED DEGRADATION OF DEA 258 9.2 CS 2 INDUCED DEGRADATION OF DEA 267 10 CONCLUSIONS AND RECOMMENDATIONS 270 10.1 CONCLUSIONS 270 10.1.1 COS INDUCED DEGRADATION 270 10.1.2 CS 2 INDUCED DEGRADATION 273 10.2 RECOMMENDATIONS 274 NOMENCLATURE 277 REFERENCES 280 APPENDICES 287 A. l ALKANOLAMINES COMMONLY USED INDUSTRIALLY 287 A. 2 SYNTHESIS OF SELECT DEGRADATION COMPOUNDS 289 B. l CALIBRATION OF THE GAS CHROMATOGRAPH 2 93 B.2 MASS SPECTRA OF MINOR DEGRADATION COMPOUNDS 308 C EXPERIMENTAL CONCENTRATIONS 317 D COMPARISON BETWEEN EXPERIMENTAL AND PREDICTED CONCENTRATIONS 338 E ERROR AND SENSITIVITY ANALYSIS 358 F PROGRAM LISTINGS 365 x LIST OF TABLES TABLE 2.1 SELECTED PHYSICAL PROPERTIES OF COS AND CS 2 8 4.1' REPRODUCIBILITY AND EFFECT OF STIRRER SPEED IN COS-DEA SYSTEMS 54 4.2 DEGRADATION COMPOUNDS DETECTED IN COS-DEA SYSTEMS AND THEIR RETENTION TIMES IN THE GC 59 4.3 ELEMENTAL COMPOSITIONS OF THE SOLIDS FORMED IN THE COS-DEA SYSTEMS 81 4.4 ELEMENTAL COMPOSITIONS OF THE SOLIDS FORMED IN THE CS2-DEA SYSTEMS 81 4.5 FUNCTIONAL GROUPS ASSIGNMENTS IN THE SOLIDS FORMED IN THE COS-DEA AND CS2-DEA SYSTEMS 8 9 6.1 CONTRIBUTIONS OF OXYGEN TO DEGRADATION IN THE COS-DEA SYSTEM 184 6.2 CONCENTRATIONS OF DEA AND THE LOW BOILING DEGRADATION COMPOUNDS IN THE REGULAR AND EDEA-SPIKED RUNS 196 6.3 CONCENTRATIONS OF DEA AND THE LOW BOILING DEGRADATION COMPOUNDS IN THE REGULAR AND EAE-SPIKED RUNS 197 7.1 FITTING CONSTANTS IN THE HENRY'S LAW EXPRESSION FOR THE COS-DEA SYSTEM (T = 20 - 50 \u00b0C) 223 7.2 HENRY'S CONSTANTS FOR THE SOLUBILITY OF COS IN WATER 223 7.3 EQUILIBRIUM DATA FOR THE HYDROLYSIS OF COS IN AQUEOUS DEA SOLUTIONS (COMPOSITIONS ARE EXPRESSED IN MILLIMOLES) ...224 xi 7.4 EQUILIBRIUM DATA FOR THE HYDROLYSIS OF COS IN AQUEOUS DEA SOLUTIONS (LIQUID PHASE CONCENTRATIONS ARE EXPRESSED IN MOLE\/MOLE DEA) 225 7.5 PREDICTED AND EXPERIMENTAL ACID GAS LOADINGS 23 9 C.l - C.20 CONCENTRATIONS OF COMPOUNDS IN THE COS-DEA SYSTEMS ...317 C.21 - C.36 CONCENTRATIONS OF COMPOUNDS IN THE CS2-DEA SYSTEMS ...327 C. 37 - C.43 CONCENTRATIONS OF COMPOUNDS IN OTHER SYSTEMS 335 D. l - D.18 COMPARISON BETWEEN EXPERIMENTAL AND PREDICTED CONCENTRATIONS 338 D. 19 - D.20 RATE CONSTANTS OBTAINED FROM THE OPTIMIZATION ROUTINE 356 E. l NITROGEN BALANCE FOR THE DEGRADATION RUNS 359 E.2 MAXIMUM DEVIATIONS IN THE DEA CONCENTRATIONS REPORTED FOR THE DEGRADATION RUNS 361 E.3 DEVIATIONS BETWEEN THE EXPERIMENTAL AND FITTED VALUES OF THE PROTONATION, CARBAMATE AND THIOCARBAMATE EQUILIBRIUM CONSTANTS 362 E.4 COMPARISON OF PROTONATION (K^) AND CARBAMATE (K 2) CONSTANTS FROM THE KENT-EISENBERG AND MODIFIED KENT-EISENBERG MODELS 362 E.5 SENSITIVITY OF THE OBJECTIVE FUNCTION TO CHANGES IN THE RATE CONSTANTS FOR THE COS-DEA SYSTEMS (% CHANGE IN k i = + 20%) 363 E.6 SENSITIVITY OF THE OBJECTIVE FUNCTION TO CHANGES IN THE RATE CONSTANTS FOR THE COS-DEA SYSTEMS (% CHANGE IN k i = - 20%) 364 xn LIST OF FIGURES Figure 1.1 Typical alkanolamine sweetening unit 4 3.1 Sketch of the reactor 36 3.2 Set-up f o r the CS2-DEA degradation experiments 38 4.1 Chromatograms of p a r t i a l l y degraded DEA solutions of 4M i n i t i a l concentration (a: 180 \u00b0C, 0.34 MPa COS; b: 150 \u00b0C, 0.34 MPa COS; c: 120 \u00b0C, 0.68 MPa COS) 52 4.2 Chromatograms showing gradual formation of degradation products i n a COS-DEA system (4M DEA, 180 \u00b0C, 0.34 MPa COS) 55 4.3 Mass spectra of peak 1 i d e n t i f i e d as Acetone (a: EI spectrum; b: EI reference spectrum; c: CI spectrum) 60 4.4 Mass spectra of peak 2 i d e n t i f i e d as Butanone (a: EI spectrum; b: EI reference spectrum; c: CI spectrum) 61 4.5 Mass spectra of peak 3 i d e n t i f i e d as MEA (a: EI spectrum; b: EI reference spectrum; c: CI spectrum; d: CI spectrum of the s i l y l d e r i v a t i v e ) 62 4.6 Mass spectra of peak 4 i d e n t i f i e d as EAE (a: EI spectrum; b: EI reference spectrum; c: CI spectrum) 63 4.7 Mass spectra of peak 5 i d e n t i f i e d as DEA (a: EI spectrum; b: EI reference spectrum; c: CI spectrum; d: CI spectrum of the s i l y l d e r i v a t i v e ) 64 4.8 Mass spectra of peak 6 i d e n t i f i e d as EDEA (a: EI spectrum; b: EI reference spectrum; c: CI spectrum; d: CI spectrum of the s i l y l d e r i v a t i v e ) 65 4.9 Mass spectra of peak 7 i d e n t i f i e d as HEA (a: EI spectrum; b: EI reference spectrum; c: CI spectrum; d: CI spectrum of the s i l y l d e r i v a t i v e ) 66 4.10 Mass spectra of peak 8 i d e n t i f i e d as HEP (a: EI spectrum; b: EI reference spectrum; c: CI spectrum; d: CI spectrum of the s i l y l d e r i v a t i v e ) 67 4.11 Mass spectra of peak 9 i d e n t i f i e d as ETHEAME (a: EI spectrum; b: CI spectrum; c: CI spectrum of s i l y l derivative) 68 xiii 4.12 Mass spectra of peak 10 i d e n t i f i e d as BHEED (a: EI spectrum; b: EI reference spectrum; c: CI spectrum; d: CI spectrum of the s i l y l derivative) 69 4.13 Mass spectra of peak 11 i d e n t i f i e d as BHEP (a: EI spectrum; b: EI reference spectrum; c: CI spectrum; d: CI spectrum of the s i l y l derivative) 70 4.14 Mass spectra of\u2022peak 12 i d e n t i f i e d as HEOD (a: EI spectrum; b: EI reference spectrum; c: CI spectrum; d: CI spectrum of the s i l y l derivative) 71 4.15 Mass spectra of peak 13 i d e n t i f i e d as HEI (a: EI spectrum; b: EI reference spectrum; c: CI spectrum; d: CI spectrum of the s i l y l derivative) 72 4.16 Mass spectra of peak 14 i d e n t i f i e d as THEED (a: EI spectrum; b: EI reference spectrum; c: CI spectrum; d: CI spectrum of the s i l y l derivative) 73 4.17 Mass spectra of peak 15 i d e n t i f i e d as BHEI (a: EI spectrum; b: EI reference spectrum; c: CI spectrum; d: CI spectrum of the s i l y l derivative) ........74 4.18 Chromatograms of p a r t i a l l y degraded DEA solutions of 3M i n i t i a l concentration degraded with 10 mL of CS2 for 48 hours (a: 180 \u00b0C; b: 165 \u00b0C; c: 150 \u00b0C) 77 4.19 EI and CI mass spectra of the s o l i d formed i n the COS-DEA system 84 4.20 EI and CI mass spectra of the s o l i d formed i n the CS2-DEA system 85 4.21 Infra-red trace of the s o l i d formed i n the COS-DEA system ...87 4.22 Infra-red trace of the s o l i d formed i n the CS2-DEA system ...88 5.1 DEA concentration as a function of i n i t i a l DEA concentration and time ( P c o s = 0.34 MPa, T = 120 \u00b0C) 92 5.2 DEA concentration as a function of i n i t i a l DEA concentration and time ( P c o s = 0.34 MPa, T = 150 \u00b0C) 93 5.3 DEA concentration as a function of i n i t i a l DEA concentration and time ( P c o s = 0.34 MPa, T = 165 \u00b0C) 94 5.4 Overall degradation rate constant as a function of i n i t i a l DEA concentration and temperature (PQQS = \u00b0 - 3 4 MPa) 95 xiv 5.5 I n i t i a l degradation rate as a function of i n i t i a l DEA concentration and temperature (P^og = 0.34 MPa) 97 5.6 Acetone concentration as a function of i n i t i a l DEA concentration and time (PQQS = \u00b0 - 3 4 MPa, T = 165 \u00b0C) 98 5.7 Butanone concentration as a function of i n i t i a l DEA concentration and time ( P c o s = 0.34 MPa, T = 165 \u00b0C) 99 5.8 MEA concentration as a function of i n i t i a l DEA concentration and time ( P c o s = 0.34 MPa, T = 165 \u00b0C) 100 5.9 BHEED concentration as a function of i n i t i a l DEA concentration and time (P C O s = \u00b0 - 3 4 M P a ' T = 1 6 5 \u00b0 c ) 1 0 2 5.10 BHEP concentration as a function of i n i t i a l DEA concentration and time ( P C O g = \u00b0- 3 4 M P a ' T = 1 6 5 \u00b0c> 1 0 3 5.11 HEOD concentration as a function of i n i t i a l DEA concentration and time ( P C O g = 0.34 MPa, T = 165 \u00b0C) ...... 104 5.12 HEI concentration as a function of i n i t i a l DEA concentration and time (PQQS = 0.34 MPa, T = 165 \u00b0C) 105 5.13 THEED concentration as a function of i n i t i a l DEA concentration and time (P^QS = \u00b0 - 3 4 MPa, T = 165 \u00b0C) 106 5.14 BHEI concentration as a function of i n i t i a l DEA concentration and time (P C O s = 0.34 MPa, T = 165 \u00b0C) 107 5.15 DEA concentration as a function temperature and time ( P c o s = 0.34 MPa, DEA Q = 4M) 110 5.16 DEA concentration as a function temperature and time ( P c o s = 0.34 MPa, DEA Q = 3M) I l l 5.17 DEA concentration as a function temperature and time ( P c o s = 0.34 MPa, DEA Q = 2M) 112 5.18 Arrhenius plots of the o v e r a l l degradation rate constant
Na 2C0 3 + Na2S + 2 H20 2.1 One such method was described by Schultze et a l . (42) but the hydrolysis was slow and required long contact times to go to completion. Johnson et a l . (43) found that the hydrolysis could be accelerated by using aqueous mixtures of MEA and NaOH. The MEA acts as a c a t a l y s t ; i t forms a thiocarbamate with the COS, which i s then hydrolysed by NaOH thereby regenerating the MEA. These methods consumed NaOH due to the d i f f i c u l t y i n regenerating i t from sodium sulphide (Na2S) and sodium carbonate (Na 2C0 3). 10 The use of alkanolamines was regarded as a better a l t e r n a t i v e since they are e f f e c t i v e i n absorbing C0 2 and H2S and the amine i s e a s i l y regenerated. Schultze and Short (44) described a method u t i l i z i n g MEA impregnated on a bed of alumina for the absorption of COS from l i q u i d propane and butane. It was reported that the system removed COS from a stream o r i g i n a l l y containing 0.002 wt% COS, but the COS was i r r e v e r s i b l y bound to the MEA. Kearns and Beamer (45) also used aqueous solutions containing 10 - 60 wt% MEA to absorb COS from a gas stream and found that the MEA was i r r e v e r s i b l y transformed to diethanolurea. As a r e s u l t of the substantial amine losses that occur from the i r r e v e r s i b l e reactions of MEA with COS, DEA i s generally a preferred choice f o r tr e a t i n g gases containing COS. Nevertheless, the choice of DEA f o r processing COS bearing streams has been c o n t r o v e r s i a l i n the past. Kerns and Beamer (45) and Reed (46) claimed that DEA i s i n e r t to COS and cannot e f f e c t i v e l y remove COS from gas streams even though they d i d not provide d e t a i l s of t h e i r experimental conditions. On the other hand, Easthagen et a l . (47) reported that 99% of carbonyl sulphide in gaseous hydrocarbons can be removed by treatment with aqueous DEA and that the spent solution i s e a s i l y regenerated by steam s t r i p p i n g . Other studies have also shown that DEA absorbs COS from gas streams (19,20). Pearce et a l . (20) passed COS into aqueous DEA solutions and analyzed them by gas chromatography. Carbonyl sulphide was detected i n the solution and t h i s was taken as an i n d i c a t i o n that DEA absorbs COS d i r e c t l y and not i t s hydrolysis products. Orbach and Selleck (19) contacted a gas stream containing 1 mole% COS i n N 2 a l t e r n a t e l y with aqueous solutions containing 20% MEA and 35 wt% DEA at room temperature. The gas and 11 solvents flowed counter-currently i n a 25 mm I.D. glass column packed to a height of 40 cm with 6 mm O.D. glass Raschig rings. The gas leaving the column was analysed with an i n f r a - r e d spetrophotometer set to a wavelength of 4.87 microns where the absorption c o e f f i c i e n t of COS i s 9.18 X 10 \" 3 mm Hg\" 1 cm\"1. It was found that both MEA and DEA solutions absorbed COS from the gas stream but more COS was absorbed by the DEA so l u t i o n . The absorption of COS i n amine and a l k a l i solutions was reported by Sharma and Danckwerts (48). The experimental data obtained from various contacting devices such as a wetted wall column, s t i r r e d c e l l and j e t apparatus showed that most of the amines absorbed COS. At 25 \u00b0C, the second order reaction rate constants for the amine-COS reaction ( k A M . c o s ) was approximately 1% of the rate constant for the corresponding amine-CO2 reactions. The r a t i o was only 0.25% i n the case of MEA. Less common amines such as methyl aminoethanol (MAE) and ethyl aminoethanol (EAE) gave better r e s u l t s than e i t h e r MEA or DEA in the absorption of COS. The rate constants obtained for these amines were 250 and 220 L\/(mole s) re s p e c t i v e l y , as opposed to 16 and 11 L\/(mole s) f o r MEA and DEA, re s p e c t i v e l y . Rahman (49) has studied the absorption of a c i d gases i n anhydrous alkanolamines. CO2 ,H2S, COS and mercaptans were contacted with MEA, DEA, DGA, DIPA and MDEA in a Claisen d i s t i l l a t i o n f l a s k at i n i t i a l pressures up to 20 psig. The temperature i n the f l a s k was not co n t r o l l e d , but depended on the heats of reactions. Temperature and pressure var i a t i o n s were recorded during the experiments which lasted 1 hour. A l l alkanolamines were found to absorb COS as indicated by the 12 drop i n the t o t a l pressure within the reaction v e s s e l . Analysis of 1 T samples of the amine-COS systems by C NMR spectroscopy revealed the presence of the respective thiocarbamates. Although the study was able to e s t a b l i s h the species i n the amine solutions, the data pertained only to anhydrous systems and are therefore of l i t t l e i n d u s t r i a l relevance. The physical s o l u b i l i t y of a gas i n a solvent with which i t reacts cannot be determined d i r e c t l y . Al-Ghawas et a l (50) therefore used the ^ 0 analogy o r i g i n a l l y proposed by Clarke (51) to determine the physical s o l u b i l i t y of COS i n 0 - 30 wt% aqueous MDEA solutions at 20 to 40 \u00b0C and 1 atm pressure. The r a t i o of the Henry's constants f o r the s o l u b i l i t y of COS and N 20 < H o C 0 S \/ H \u00b0 N 2 0 ) i n water and 15.5 wt% ethylene g l y c o l solutions were f i r s t determined at 25\u00b0C. The difference i n the r a t i o s f o r both systems was found to be only 0.26%. It was then assumed that t h i s r a t i o was the same for a l l temperatures and amine solutions. The Henry' s constants f o r the s o l u b i l i t y of COS i n aqueous MDEA solutions were then determined from the expression: HCOS = H\u00b0COS \/ H\u00b0N20 X HN20 where H\u00b0^ and H^ r e f e r to the Henry' s constant for the s o l u b i l i t y of gas i i n water and aqueous MDEA solutions, r e s p e c t i v e l y . The Henry's constants were found to range from 3.941 kPa m \/mol for water at 20 \u00b0C to 7.554 kPa m3\/mol f o r 30.21 wt% MDEA solution at 40 \u00b0C. The temperature v a r i a t i o n of the Henry's constant was well correlated by the Arrhenius expression. Since the focus of the study was the k i n e t i c s of the COS-MDEA reaction, only physical s o l u b i l i t y and 13 d i f f u s i v i t y data were reported. No information was provided on the t o t a l s o l u b i l i t y of COS i n MDEA solutions. It was recently reported (52) that the absorption of COS int o a l k a l i n e solutions could be enhanced by using a second emulsified l i q u i d phase. A t h e o r e t i c a l enhancement factor was defined as the r a t i o of the s p e c i f i c rate of absorption i n the base, with and without the second emulsified l i q u i d phase. Enhancement factors of 2.5 and 4.0 r e s p e c t i v e l y were achieved using toluene (20% v\/v) and 2-propanol (50% v\/v) as the second emulsified l i q u i d phases i n sodium hydroxide so l u t i o n at 30 \u00b0C. There was no i n d i c a t i o n whether such enhancements could be achieved i n alkanolamine systems. The operational d i f f i c u l t i e s which may accompany the use of emulsions, were not discussed e i t h e r . R e i l l y et a l . (89) have also reported that the rate of absorption of COS in MDEA solutions was enhanced by he t e r o c y c l i c amine a d d i t i v e s . Enhancements factors up to 8 f o l d were obtained f o r the range of additives concentrations investigated. The additives produced l i t t l e or no enhancements i n the corresponding CO2 systems. Other alkanolamines such as DIPA (53) and DGA (54) are used f o r absorbing COS without in c u r r i n g appreciable amine loss but t h e i r i n d u s t r i a l acceptance i s s t i l l quite l i m i t e d . Singh and B u l l i n (55) studied the k i n e t i c s of the reaction between COS and aqueous DGA over a temperature range of 307 - 322 K and pressure range of 345 - 414 kPa. They used a p e r f e c t l y mixed flow reactor operated with continuous gas and l i q u i d feeds. The composition of the amine sol u t i o n leaving the absorber was determined by gas chromatography. Although the hydrolysis of COS produces equimolar amounts of CO2 and H2S, the GC analysis 14 recorded i n s i g n i f i c a n t l e v e l s of C0 2. The l a t t e r may be due to i t s consumption i n the degradation of DGA. The rates of reaction were ca l c u l a t e d by d i v i d i n g the sum of the molar flow rates of COS (and the H2S from the hydrolysis reaction) by the volume of the reactor; they were found to be much higher than those for the COS-H2O reaction (56). This l ed to the conclusion that DGA was c a t a l y s i n g the hydrolysis of COS. A s i m i l a r e f f e c t has also been reported f o r other amines (57). The COS-DGA reaction was second order; i t was f i r s t order with respect to the amine and COS concentrations. The second order reaction rate constants obeyed the Arrhenius expression. At 27 \u00b0C, the value was 0.0023 m3\/(mol s) compared to 0.166 m3\/(mol s) (49) and at most 0.016 mJ\/(mol s) (48,57). The l a t t e r was i n f e r r e d from the rate constant reported at 25 \u00b0C f o r the r e a c t i o n of COS with MEA, a primary amine that reacts f a s t e r than DGA. The reported values d i f f e r s i g n i f i c a n t l y but i t i s d i f f i c u l t to make a d i r e c t comparison of the rate constants for acid gas-amine reactions in aqueous and non-aqueous amine systems because of the di f f e r e n c e s in d i f f u s i v i t y , i o n i c strength, nature of g a s - l i q u i d i n t e r f a c e and physical s o l u b i l i t y . The combined e f f e c t s of these factors may explain the higher values found by Rahman. It i s c l e a r from the foregoing that COS i s absorbed d i r e c t l y into amine solutions and not i t s hydrolysis products. However, -compared to CO2 and H 2S, very l i t t l e work has been reported on the t o t a l s o l u b i l i t y of COS i n amine solutions. The few studies on COS-amine systems cover narrow temperature ranges below 50 \u00b0C. It i s not s u r p r i s i n g therefore that Vapour-Liquid-Equilibrium (VLE) models to date, are l i m i t e d to aqueous amine-CO2-H2S systems. 15 2.2.3 CARBON DISULPHIDE Osenton and Knight (21) contacted CS2 vapour with aqueous alkanolamine solutions and used changes i n system pressure to e s t a b l i s h the rates of absorption of CS2 into alkanolamine solutions. The rates of absorption followed the order S u l f i n o l > 25% DEA > 20% MEA. S u l f i n o l i s a mixture of DIPA and a physical solvent. As i n the case of COS, equilibrium data for the s o l u b i l i t y of CS2 i n amine solutions are lacking. 2.3 EQUILIBRIUM REACTIONS OF ACID GASES IN AMINE SOLUTIONS When ac i d gases are absorbed into alkanolamine solutions, i o n i c species are generated and various e q u i l i b r i a , which are discussed i n the following sections, are established. 2.3.1 REACTIONS OF HYDROGEN SULHIDE Hydrogen sulphide in aqueous amine solutions d i s s o c i a t e s i n t o H + and HS\": H 2S = H + + HS\" 2.2 The proton i s transferred to the amine v i r t u a l l y instantaneously (2,58): RR' NR\" + H + = RR' N + R\"H 2.3 16 where R and R represent H and \/ or -C2H4OH, depending on the class of amine, while R represents -C2H4OH. The bisulphide ion can di s s o c i a t e further: HS\" = H + + S\"\" 2.4 However, t h i s reaction occurs only i n highly basic solutions (pH > 12 ) and can usually be disregarded f o r alkanolamine solutions contacting a c i d gases (2,58). 2.3.2 REACTIONS OF CARBON DIOXIDE The e q u i l i b r i a i n C02-amine systems can be represented by the following series of equations: C0 2 + H 20 = H + + HCO3\" 2.5 HCO3\" = H + + CO3\"\" 2.6 Reaction 2.5 has been shown to be catalysed by bases (37,58,59). The protons (or hydrogen ions) react with the amines to form protonated amines according to equation 2.3. In addition, a d i r e c t reaction occurs between CO2 and primary and secondary amines to form carbamates: RR1 NH + C0 2 = RR'NCOO\" H + 2.7 17 It has been postulated that t h i s equation may involve two steps; a zwitterion intermediate i s f i r s t formed which by v i r t u e of i t s i n s t a b i l i t y , i s e a s i l y deprotonated i n the presence of a base (B) to form the carbamate (60): RR1 NH + C0 2 = RR'N +HCOO\" 2.8 RR*N+HCOO\" + B = BH + + RR'NCOO\" 2.9 Equation 2.8 i s the rate determining step since the deprotonation of the zwitterion (Eq. 2.9) i s instantaneous (59,60). Reaction 2.7 i s much f a s t e r than the hydration of CO2 (eq. 2.5) and represents the major mode of i n t e r a c t i o n between CO2 and primary or secondary amines. T e r t i a r y amines have no l a b i l e hydrogen atoms and are therefore unable to form carbamates. Absorption of CO2 i n such amines proceeds v i a Eqs. 2.5 and 2.6. They are able to absorb H2S at a f a s t e r rate and are thus used f o r the s e l e c t i v e absorption of hydrogen sulphide. Methyldiethanolamine (MDEA) i s widely used f o r t h i s purpose (61-63). 2.3.3 REACTIONS OF CARBONYL SULPHIDE Sharma (57) postulated that, because of the s t r u c t u r a l s i m i l a r i t i e s of CO2 and COS, t h e i r reactions with amines are s i m i l a r . The relevant equations f o r the COS-DEA system would therefore be: R2NCOS\" + H + R2NH + C0 2 + H2S 2 .10 2.11 18 Equation 2.10 represents the thiocarbamate formation. Primary and secondary amines undergo t h i s reaction. In the case of MEA, the thiocarbamate may be further transformed into degradation products from which the amine i s not e a s i l y recovered (19,20,64). It has been estimated that 10 to 20% of the MEA reacting with COS i s l o s t i n such i r r e v e r s i b l e reactions (20). Equation 2.11 i s the COS hydrolysis reaction. It has been represented i n t h i s form since most of the COS i n solut i o n was absorbed as thiocarbamate and also to indicate that the hydrolysis of COS i s catalyzed by the DEA. Rahman (49) and Rahman et a l . (65) confirmed through NMR spectroscopy, the presence of the respective carbamates, thiocarbamates and protons i n anhydrous .alkanolamine-acid gas systems. The H2S and CO2 produced v i a Eq. 2.11 then i n i t i a t e the reactions shown i n Eqs. 2.2 - 2.4 and 2.5 - 2.7, respectively. Al-Ghawas et a l . (50) have suggested that COS may be hydrated l i k e CO2 by forming a thio-bicarbonate ion: COS + H 20 = HC02S\" + H + 2.12 The existence of HCO2S\" i n sol u t i o n was not proven d i r e c t l y but was inf e r r e d from spectroscopic observations. A mixture of CO2 and H2S was bubbled through an aqueous sol u t i o n of MDEA and COS was bubbled through another aqueous MDEA sol u t i o n . Both solutions, when inspected spectrophotometrically, were found to absorb at d i f f e r e n t wavelengths; the so l u t i o n exposed to COS absorbed at 518 nm, whereas the other sol u t i o n absorbed at 503 nm. It i s known that CO2 and H2S i n solu t i o n give r i s e to HCO3\" and HS\", re s p e c t i v e l y . If hydrolysis of COS i s the 19 predominant reaction i n the COS-MDEA solution, the solution should absorb at the same wavelength as the MDEA solu t i o n exposed to CO2 and H2S. The observed differences i n absorption wavelengths were a t t r i b u t e d to the existence of another species, i . e . HCO2S\". The o v e r a l l reaction for the COS-MDEA system was postulated as: R 2NCH 3 + H20 + COS -\u00bb R 2N +HCH 3 + HC02S\" 2.13 The second order rate constant k^, was given by the expression: k 2 = 4198.74 exp (-4575.80\/T). The units of k2 and T are m \/mol s and Kelvin (K) r e s p e c t i v e l y . It i s therefore c l e a r that an aqueous COS-amine system at equilibrium, consists of the amine, C02, H2S, unhydrolysed COS and t h e i r derived species such as protonated amine, carbamates, thiocarbamates, hydrogen ions and bicarbonate ions. The s e l e c t i v i t y of the amine f o r these gases w i l l e s t a b l i s h the equilibrium composition of the system. 2.3.4 REACTIONS OF CARBON DISULPHIDE Osenton and Knight (21) used t h i n layer chromatography to confirm that CS 2 reacts with aqueous alkanolamine solutions. In t h e i r work, 7 mL of CS2 were added to 40 mL of aqueous amine solutions and the r e s u l t i n g mixture was s t i r r e d f o r a short time. Samples of the aqueous layer of 20 the mixture were then spotted on a s i l i c a gel TLC plate and developed i n methanol. Subsequent inspection of the plate i n iodine or U.V. l i g h t showed, i n a l l cases, the presence of a new compound i n addition to the s t a r t i n g materials. This compound was reported as amine dithiocarbamate, the product of the CS 2-amine reaction: RR NH + CS 2 = RR NCSS\"H+ 2.14 The amine dithiocarbamate reacts further i n the presence of excess amine to give the dithiocarbamate s a l t of the amine: RR'NH + RR'NCSS\" H + = RR'NCSS\" H 2N +R 2 2.15 Kothari and Sharma (66) have determined the k i n e t i c parameters of the r e a c t i o n of CS 2 with aqueous amine solutions using s t i r r e d c e l l s . The amine solutions were contacted with CS 2 i n the vapour and l i q u i d states at temperatures between 5 and 30 \u00b0C. The k i n e t i c parameters were estimated by applying the penetration theory equation to the experimental data, and were found to be independent of the state of the CS 2 and speed of a g i t a t i o n above 25 r.p.m. The rate constants obtained for the amine-CS 2 reactions obeyed the Arrhenius expression. Diethanolamine was the only alkanolamine investigated and a rate constant of 0.10 L\/g-mole-s was reported for the CS2-DEA reaction at 30 \u00b0C. By comparing the second order rate constants ^AM-CS2^ with published rate constants f o r the corresponding C0 2-amine i t was found that: kAM-C02 1 kAM-CS2 = 2 X 1 C | 4 \u2022 21 Since KAM-C02 ^ ^A!*I-COS = 1 X 1 (57), i t follows that KAM-COS 1 KAM-CS2 = 2 X 10 2. The r a t i o s of the reaction rate constants ind i c a t e that the rates of reaction follow the order: kAM-C02 > k A M _ c o s > k A M _ c s 2 -Carbon disulphide also undergoes hydrolysis i n aqueous solutions producing COS, C0 2 and H 2S (24). The o v e r a l l reactions are: CS 2 + H 20 = COS + H 2S 2.16 COS + H 20 = C0 2 + H 2S 2.17 CS 2 hydrolysis may also occur v i a the amine dithiocarbamate: RR NCSS\" H + + 2H 20 = RR NH + C0 2 + 2H2S 2.18 The C0 2\/H 2S r a t i o i s therefore 1\/2 compared with 1\/1 for COS hydrolysis. 2.4 DEGRADATION OF DIETHANOLAMINE SOLUTIONS 2.4.1 DEGRADATION OF DIETHANOLAMINE BY CARBON DIOXIDE The degradation of DEA by carbon dioxide was f i r s t reported by Polderman and Steele (10). Two compounds, b i s hydroxyethyl piperazine (BHEP) and hydroxyethyl oxazolidone (HEOD) were detected i n the degraded solution. It was suggested that these compounds were formed as follows: DEA + C0 2 = HEOD 2.19 2 HEOD = BHEP + 2CO, 2.20 22 Later, Hakka et a l . (11) i d e n t i f i e d another compound, t r i s hydroxyethyl ethylene diamine (THEED) i n a degraded DEA solution, but offered no mechanism for i t s formation. Other more de t a i l e d studies have since been published (9, 12-18). Choy (12) conducted degradation experiments using a s t a i n l e s s s t e e l reactor at temperatures between 165 and 180 \u00b0C and CO2 pressures of up to 4238 kPa. Analysis of the solutions by gas chromatography in d i c a t e d a number of products, but only BHEP was conclusively i d e n t i f i e d . It was reported that the i n i t i a l o v e r a l l DEA degradation could be represented by f i r s t order k i n e t i c s even though the rate constant was a function of the i n i t i a l DEA concentration. Kennard (16) as well as Kennard and Meisen (14) reported r e s u l t s of degradation experiments conducted i n a 600 mL s t a i n l e s s steel r eactor. Aqueous solutions containing 5 - 100 wt% of DEA were subjected to CO2 pressures up to 4.1 MPa at temperatures ranging from 90 - 205\u00b0C. A t o t a l of 12 products were i d e n t i f i e d i n the degraded solutions, the major ones being BHEP, HEOD and THEED. The following s i m p l i f i e d mechanism, supported by experimental observations, was suggested for the formation of the major degradation products: DEA + C0 2 = HEOD 2.21 DEA + DEACOO\" -> THEED 2.22 THEED \u2014\u00bb BHEP 2.23 The i n i t i a l rate of DEA degradation was found to increase with DEA concentration, temperature and C0 2 p a r t i a l pressure; these observations 23 agreed with those of Choy (12). A pseudo f i r s t order k i n e t i c model was developed to predict the rate of degradation and the formation of the major degradation compounds. The HEOD formation (Eq. 2.21) was assumed to be i r r e v e r s i b l e and the influence of C0 2 p a r t i a l pressure on the ki n e t i c s was eliminated by e s t a b l i s h i n g the pressure beyond which C0 2 loading was constant. Thus the k i n e t i c model was r e s t r i c t e d to the ranges: DEA concentration of 0 - 100 wt%; temperatures of 90 - 175 \u00b0C; C0 2 loadings > 0.2g \/ g DEA. C0 2 was neither consumed nor formed i n the degradation process and was therefore suggested as cat a l y s i n g the process. Kim and S a r t o r i (13) conducted degradation experiments at 100 and 120\u00b0C using 3.2M aqueous DEA solutions containing various amounts of C0 2 and\/or H 2S. The solutions were put i n sealed s t a i n l e s s s t e e l ampoules which were then immersed i n a temperature c o n t r o l l e d bath. The rate of degradation increased with increasing C0 2 loadings, producing BHEP, HEOD and THEED as the major degradation compounds. A scheme suggesting successive reactions was put f o r t h to account f o r the formation of the degradation compounds: DEA + C 0 2 = HEOD 2 . 2 4 DEA + HEOD -+ THEED 2 . 2 5 THEED -> BHEP 2 . 2 6 Rate constants f o r the reaction steps were determined at 100 and 120\u00b0C. Again, the c a t a l y t i c role of C0 2 was noted while H 2S exerted e s s e n t i a l l y no e f f e c t . It should be noted that the r a t i o of C0 2 to H 2S solution 24 Loading i n the only mixed gas run reported was 36.7. The extremely low H 2S loading might explain the si m i l a r i n i t i a l rate of degradation obtained i n the corresponding run conducted with C0 2 alone. Chakma (18) and Chakma and Meisen (9) have reported degradation of DEA by C0 2 i n a heat'transfer loop. The rate of degradation increased with temperature, DEA concentration and C0 2 p a r t i a l pressure, but decreased with solution flow rate. The mechanism suggested by Kennard and Meisen (14) was used to develop a mathematical model f o r the degradation. However, unlike i n the previous case (14), C0 2 was e x p l i c i t l y included i n the rate expressions so that the model could cover a wider range of operating conditions. In essence, the formation of HEOD and THEED was governed by second order k i n e t i c s while BHEP formation was f i r s t order as shown below: 2.27 2.28 2.29 The rate constants were reported as functions of temperature. Hsu and Kim (15) used a gas chromatograph coupled to a mass spectrometer (GC\/MS) to i d e n t i f y higher molecular weight triamino degradation compounds i n aqueous DEA solutions degraded by C0 2. The mixture was f i r s t d e r i v a t i z e d to make the compounds more v o l a t i l e and thus amenable to GC\/MS an a l y s i s . The procedure f o r the d e r i v a t i z a t i o n k l DEA + C0 2 -\u00bb HEOD k2 DEA + C0 2 -\u00bb THEED k3 THEED -> BHEP + C0 2 25 consisted of saturating the p a r t i a l l y degraded sol u t i o n with potassium carbonate to remove water, followed by extraction with isopropyl a l c o h o l . The alcohol was then evaporated to obtain a dehydrated o i l y material. 1 mL of anhydrous py r i d i n e , 0.2 mL of hexamethyl dis i l a z a n e and 0.1 mL of t r i m e t h y l c h l o r o s i l a n e were added to the o i l and the r e s u l t i n g mixture was vigorously shaken for 30 s and kept at room temperature f o r over 5 min u n t i l d e r i v a t i z a t i o n was' complete. In addition to DEA, HEOD, BHEP and THEED three other compounds, 3-(-(bis(2-hydroxyethyl)-amino)ethyl)-2-oxazolidone (HAO), N-(2-(N,N-bis(2-hydroxyethyl)amino)ethyl)-N -(2-hydroxyethyl)piperazine (HAP) and N,N,N ,N -tetrakis(2-hydroxyethyl)diethylenetriamine (THEDT) were i d e n t i f i e d and reaction mechanisms were proposed for t h e i r formation. Chakma (18) reported that the above procedure was not always s e l e c t i v e , sometimes r e s u l t i n g i n the d e r i v a t i s a t i o n of both-the alcohol and amino groups i n alkanolamines. Instead of a mixture of chemicals, Chakma used trimethyl s i l y l imidazole (TSIM) to d e r i v a t i z e p a r t i a l l y degraded MDEA solutions. Only the hydroxyl groups were d e r i v a t i z e d and d e r i v a t i z a t i o n was complete within 30 minutes. The s e l e c t i v e d e r i v a t i z a t i o n made i t possible to i n f e r the number of hydroxyl groups i n the underivatized compounds. Smith and Younger (17), i n a more p r a c t i c a l approach, offered some help f u l hints on the design and operation of gas plants to avoid or minimise problems such as foaming, corrosion and degradation. They reported that 10 of 19 plants investigated experienced some degradation, the products being BHEP, r e s i d u a l s a l t s , thiosulphates and probably TEA. 26 These studies have provided appreciable ins i g h t into alkanolamine degradation due to carbon dioxide. The degradation i s believed to occur p r i m a r i l y v i a amine carbamates which may be formed by the d i r e c t reaction of CO2 with amines. Since H2S i s incapable of forming carbamate-type compounds, i t i s generally agreed that H2S does not cause amine degradation. Choy (12) and Kim and S a r t o r i (13) have reported findings which suggest that H2S in the presence of C0 2 a c t u a l l y hinders amine degradation. The difference i n the mechanisms suggested by Kim and Sa r t o r i (13) and Kennard and Meisen (14) i s i n the formation of THEED. In order to determine which of the mechanisms better represents THEED formation, the equilibrium between HEOD and DEA under CX^-rich and C02-l i m i t i n g conditions as represented below, i s re-examined: co 2 DEA = DEACOO\" H + = HEOD + H20 2.30 N2 DEA + C0 2 = DEACOO\" H + = HEOD + H20 2.31 In the C02-rich system (Eq. 2.30), the equilibrium w i l l always favour HEOD production and can be maintained by the transformation of excess DEA carbamate (DEACOO\") or HEOD to other products such as THEED. However, the high CO2 loading reduces the b a s i c i t y of the so l u t i o n and the transformation of HEOD to THEED, a reaction which involves r i n g breakage, may be s l i g h t l y hindered. On the other hand, the reaction of DEA with DEACOO\" to form THEED as proposed by Kennard and Meisen w i l l proceed w e l l . In support of t h i s a s s e r tion i s the fi n d i n g of Kennard and Meisen that the HEOD concentration increased only s l i g h t l y i n an aqueous 27 sol u t i o n of DEA and HEOD that was saturated with CO2 and maintained at 175 \u00b0C for 8 hr. In the nitrogen or CO2-limiting case (Eq. 2.31), the equilibrium l i e s to the l e f t i n favour of DEA and DEACOO\" formation. The equilibrium i s maintained by the reac t i o n of excess DEACOO\" with DEA to form THEED. Data reported by Kim and S a r t o r i showed that i n an aqueous solution containing HEOD and DEA at 120 \u00b0C, the HEOD concentration decreased by 0.4 M while the DEA and THEED concentrations increased by 0.03 M and 0.20 M, re s p e c t i v e l y . If the mechanism of Kennard and Meisen i s correct the 0.2 M THEED could be produced from 0.2 moles\/L each of DEA and DEACOO\", both of which account f o r the 0.4 M drop i n HEOD concentration. Using the mechanism by Kim and S a r t o r i , the concentration data implies that the 0.2M THEED was produced from 0.2 moles of HEOD and DEA. The remaining 0.2 M drop i n HEOD i s that l o s t i n the reversal to carbamate. The l a t t e r keeps the DEA concentration approximately constant since GC analysis does not d i s t i n g u i s h between free amine and amine carbamate. Thus HEOD reversal to DEACOO\" and i t s reaction with DEA to form THEED would proceed at the same rates, a s i t u a t i o n that would disrupt the equilibrium i n Eq. 2.31. The mechanism of Kennard and Meisen therefore appears to be the more applicable one to the CO2 l i m i t i n g case. In summary, the mechanism postulated by Kennard and Meisen appears to represent both the CO2 l i m i t i n g and non-limiting systems whereas the mechanism by Kim and S a r t o r i i s p a r t i a l l y applicable to the l a t t e r but not the former case. Since amine degradation i n sweetening units occurs mostly at high temperatures, a condition that l i m i t s CO2 loadings, the 28 Kennard and Meisen mechanism i s therefore more appropriate for i n d u s t r i a l s i t u a t i o n s . 2.4.2 DEGRADATION OF DIETHANOLAMINE BY CARBONYL SULPHIDE Compared with CO2-induced degradation, l i t t l e work has been reported on the COS-alkanolamine degradation reactions (19,20). Orbach and Selleck (19) contacted pure COS with 20 wt% MEA and 35 wt% DEA solutions i n a bench-scale p i l o t plant simulating a t y p i c a l , continuous absorption-regeneration process. The absorber and regenerator were operated at 40 \u00b0C and 104 \u00b0C, r e s p e c t i v e l y . Periodic . analysis of the amine solutions using Kjeldahl analysis and a c i d t i t r a t i o n revealed that, while MEA was s u b s t a n t i a l l y degraded, no loss of a l k a l i n i t y occurred i n the DEA solution over 8 hours. Although the authors reported the formation of some extraneous products, they were \"too small for i s o l a t i o n and c h a r a c t e r i z a t i o n \" . B e l i e v i n g that these products could be formed from small quantities of MEA i n the i n i t i a l DEA s o l u t i o n , they concluded that COS does not degrade DEA. Pearce et a l . (20) contacted, i n a batch-mode, pure COS with 20 wt% DEA solutions at temperatures ranging from 40 to 120 \u00b0C. The solutions were subsequently analyzed by i n f r a r e d and mass spectroscopy. Minor quantities of ethanol and oxazolidone were detected. However, these quantities were i n s i g n i f i c a n t compared with those formed when MEA was subjected to COS under s i m i l a r operating conditions. Pearce et a l . (20) also contacted COS with DEA continuously using an approach analogous to that of Orbach and Selleck (19). The concentrations of the 29 DEA solutions at the s t a r t and end of the experiments were determined by-wet chemical a n a l y s i s and found to be e s s e n t i a l l y the same. This again led to the conclusion that COS does not degrade DEA. A further conclusion was that COS underwent s i g n i f i c a n t hydrolysis as revealed by the presence of CO2 and H2S i n the regenerator off gas and the DEA solution leaving the absorber. 2.4.3 DEGRADATION OF DIETHANOLAMINE BY CARBON DISULPHIDE The only substantive study on alkanolamine degradation by CS2 was conducted by Osenton and Knight (21). They contacted 40 mL of a 25 wt% aqueous DEA s o l u t i o n with 7 mL of CS2 for 3 hours i n a s t i r r e d vessel at 25 \u00b0C. T i t r a t i o n of the so l u t i o n f or free amine and dithiocarbamate contents revealed the complete conversion of amine to the amine s a l t of dithiocarbamic a c i d (see eq. 2.14). When the reaction mixture was b o i l e d f o r 1 hour and analyzed i n the same manner, only 20% of the DEA was recovered from the dithiocarbamate s a l t . Using MEA instead of DEA, Osenton and Knight found that about 40% of the i n i t i a l MEA was recovered upon b o i l i n g . They also detected oxazolidone-2-thione amongst the reaction products, contrary to the findings of Pearce et a l . (20) who di d not detect any degradation compounds i n MEA\/CS2 systems. It should be noted that while Osenton and Knight contacted CS2 d i r e c t l y with MEA, Pearce et a l . based t h e i r conclusions on a f i e l d test of an MEA plant which processed gas containing COS and CS2. The assumption was made that the plant's MEA loss was due to the MEA-COS reaction only. Since COS and CS2 e x i s t i n p.p.m concentrations i n sour i n d u s t r i a l gases, i t i s 30 possible that, i n the f i e l d t e s t , the CS2-MEA reactions resulted i n degradation product concentration too low to be detected by the mass spectroscopic a n a l y s i s . This may be the reason f o r the contradictory findings. 2.5 LIMITATIONS OF THE PREVIOUS COS-DEA AND CS2-DEA DEGRADATION STUDIES Conclusions of some e a r l i e r studies notwithstanding, there are three basic reasons to believe that degradation may occur i n COS-DEA and CS2-DEA systems. F i r s t , recent studies on the C02-MDEA system have shown that degradation i s possible v i a amine protonation as well as carbamate formation (18). Since COS and CS 2 hydrolyze i n aqueous systems to give H 2S and C02, s i g n i f i c a n t concentrations of H + and C0 2 may r e s u l t to induce the degradation of DEA v i a protonated DEA and DEA carbamate. Second, i t has been reported that the COS-DEA and CS2-DEA reactions forming thiocarbamate and dithiocarbamate r e s p e c t i v e l y , are two and f i v e orders of magnitude slower than the C02-DEA reaction y i e l d i n g DEA carbamate (57,66). Since C0 2-induced degradation of DEA occurs p r i m a r i l y v ia DEA carbamate, i t i s l i k e l y that the COS and CS 2 induced degradation of DEA v i a the thiocarbamates and dithiocarbamates respectively, are correspondingly slower. For instance, i t was found that at 120 \u00b0C, a C0 2 p a r t i a l pressure of 4.1 MPa and an i n i t i a l DEA concentration of 30 wt%, i t took almost 20 hours to obtain a 5% reduction i n the amine concentration and only one degradation compound was formed i n appreciable q u a n t i t i e s (16). When the degradation 31 experiments were conducted f o r extended periods of up to 30 days at low temperatures (11,13,14,16) or for eight hours at elevated temperatures (14,16), s i g n i f i c a n t decreases i n amine concentrations were recorded and more degradation compounds were formed. Since Orbach and Selleck (19) and Osenton and Knight (21) conducted t h e i r DEA degradation experiments over just 8 and 4 hours, r e s p e c t i v e l y , i t i s not s u r p r i s i n g that s i g n i f i c a n t amounts of degradation compounds could not be detected. (The durations of the experimental runs performed by Pearce et a l . (20) were not c l e a r l y indicated.) Third, the a n a l y t i c a l techniques used previously may not have been s u f f i c i e n t l y s e n s i t i v e to detect the compounds a r i s i n g from the degradation reactions. The difference between Kjedahl t o t a l nitrogen and free amine determinations were often used to e s t a b l i s h the extent of amine i n a c t i v a t i o n . Polderman et a l . (4) have noted some inconsistencies i n t h i s approach. Values obtained f o r t o t a l nitrogen includes nitrogen from the amine as well as i t s nitrogenous degradation compounds. Acid t i t r a t i o n s to determine free amine concentration provide erroneous r e s u l t s since the degradation compounds are basic and some of them are t i t r a t e d as we l l . Therefore, i t may be concluded that values obtained from the di f f e r e n c e of the two analyses may not be r e l i a b l e estimates of the degree of amine i n a c t i v a t i o n . 32 2.6 ANALYSIS OF DEGRADED ALKANOLAMINE SOLUTIONS E a r l i e r attempts to analyze degraded amine solutions used methods such as potentiometric t i t r a t i o n (67), a c i d t i t r a t i o n and Kjedahl t o t a l nitrogen determination (4,68), f r a c t i o n a l d i s t i l l a t i o n and c r y s t a l l i z a t i o n (10). These methods were generally unsuccessful f o r reasons such as lack of r e p r o d u c i b i l i t y , i n a b i l i t y to separate degradation compounds, decomposition of amines and degradation compounds at elevated temperatures. D e r i v a t i z a t i o n p r i o r to gas chromatographic analysis has also been t r i e d (69,70,71). Although f a i r l y successful, gas chromatographic analysis of d e r i v a t i z e d degraded amine solutions suffers some drawbacks. These were outlined by Saha et a l . (72) and include: time consuming procedures of preparing d e r i v a t i v e s , incomplete d e r i v a t i s a t i o n , i n s t a b i l i t y of derivatives over long periods and in the presence of water, and long period of a n a l y s i s . As a r e s u l t , they developed a d i r e c t gas chromatographic technique using a column packed with Tenax GC, a porous polymer based on 2,6-diphenyl paraphenylene oxide (73). This column s u c c e s s f u l l y separated a mixture of MEA, DEA and TEA i n about 8 minutes. The use of the Tenax GC column was extended to the analysis of degraded DEA solutions (13,74) and Kennard and Meisen (74) have reported the d e t a i l e d a n a l y t i c a l conditions f o r the ana l y s i s . However, while Kennard and Meisen (74) reported good r e s o l u t i o n of peaks, Kim and S a r t o r i (13) were unable to separate some degradation compounds with the Tenax GC column alone. To overcome t h i s problem, each analysis was repeated with a 5% SE 30\/Chrom-GHP column. The lack of separation could have been due to the shorter column length (2 f t ) used 33 i n the study, as opposed to the 6 f t column used by Kennard and Meisen (74). A gas chromatographic method involving a combination of columns has also been reported by Robbins and B u l l i n (75). Tenax GC and Porapak Q columns were connected i n series f o r the analysis of amine solutions containing a c i d gases, hydrocarbons and water. The Tenax column separated the amine from the other components which were further separated by the Porapak Q column and detected by a thermal conductivity detector (TCD). A switching valve i n the set up was used to bypass the Porapak column once the l i g h t e r components had eluted. This was done to protect the Porapak column from i r r e v e r s i b l e adsorption or deactivation by the amine. Good r e s o l u t i o n was obtained f o r CC^, ^ S , H^ O and MDEA within 10 minutes. The concentrations of the ac i d gases obtained from the GC analysis were i n good agreement with the values obtained from wet chemical methods. This method i s able to simultaneously determine the concentrations of ac i d gases as well as the amine and i t s degradation products. However, care should be taken i n the handling of samples from degradation runs conducted at high temperatures and pressure, to prevent f l a s h i n g of the ac i d gases and hence erroneous a c i d gas loadings. Chakma (18) as well as Chakma and Meisen (76) have also used Tenax GC columns f o r the analysis of degraded DEA and MDEA solutions. CHAPTER 3 E X P E R I M E N T A L A P P A R A T U S AND P R O C E D U R E 3.1 REACTOR The degradation experiments were c a r r i e d out using a 600 mL st a i n l e s s s t e e l reactor (model 4560, Parr Instrument Co. 111., U.S.A.), shown i n Figure 3.1. Its main features and accessories are described below: 1. A pressure gauge for monitoring the pressure within the reactor. 2. A J-type thermocouple i n a 316 s t a i n l e s s s t e e l well, to monitor the temperature within the reactor. 3. A close f i t t i n g , quartz f a b r i c heating mantle i n an insulated aluminium housing, attached to a stand to enable movement up or down as desired. 4. An automatic heater\/temperature c o n t r o l l e r (model 4813EB, Parr Instrument Co., 111., U.S.A.). The c o n t r o l l e r maintains the system temperature within \u00b1 0.5\u00b0C of the set point by regulating the power supply to the mantle. An i n d i c a t o r within the c o n t r o l l e r assembly displays the system temperature i n terms of the magnitude of the deviation from the set point. The temperature l i m i t s of the reactor are room temperature and 400\u00b0C. 5. A s t a i n l e s s s t e e l s t i r r e r , driven by a variable speed motor at speeds up to 600 r.p.m. 34 35 6. Li q u i d sampling, gas i n l e t and gas outlet valves. A 1\/8\" sample tube connected to the gas i n l e t \/ l i q u i d sampling port enables the supply of gas to and the withdrawal of l i q u i d from the reactor. 7. A rupture d i s c which breaks when the pressure within the reactor exceeds the r a t i n g of the d i s c . The r a t i n g of the disc on the reactor i s 13.78 MPa (2000 psi) at 400 \u00b0C. 8. A s t a i n l e s s s t e e l cooling c o i l f o r rapid cooling of the reactor when necessary. 9. A pyrex l i n e r to reduce contact between the reactants and the inner walls of the reactor. 3.2 MATERIALS DEA (>99% purity) was purchased from A l d r i c h Chemical Co., Inc. (Milwaukee, WI). CS 2 (>99% purity) was purchased from BDH Chemicals Inc. (Vancouver, BC). COS was supplied by Matheson Inc (Edmonton, AB) with the following p u r i t y expressed i n mole percent: COS - 97.7%, C0 2 - 1.4%, CS 2 - 0.19%, H2S - 0.01%, 0 2 - 0.1%, CO &\/or N 2 - 0.6%. Nitrogen (>99% purity) was purchased from Medigas Ltd. (Vancouver, BC). The compounds used for the c a l i b r a t i o n of the gas chromatograph were purchased from A l d r i c h Chemical Co., Inc. (Milwaukee, WI) with the exception of HEOD, THEED and BHEI which were unavailable and had to be synthesized i n the laboratory. Procedures for t h e i r synthesis are described i n Appendix A2. 36 Figure 3.1: Sketch of the reactor. 37 3.3 EXPERIMENTAL PROCEDURE T y p i c a l l y , 250 mL of an aqueous DEA so l u t i o n with the desired concentration was placed i n the reactor before sealing i t . The s t i r r e r was turned on at a speed of 120 r.p.m and a i r was purged from the reactor by passing nitrogen through i t f o r about 15 min. The gas i n l e t and outlet valves were closed and the heater was turned on to bring the reactor to the desired operating temperature. Once steady state had been achieved, the pressure within the reactor was recorded as the i n i t i a l pressure. This pressure i s the sum of the vapour pressure of the amine solu t i o n and the pressure due to the r e s i d u a l nitrogen i n the reactor. COS was then introduced through a s t a i n l e s s s t e e l pressure hose connected to the pressurized COS c y l i n d e r . It was necessary to warm the COS c y l i n d e r to about 45 \u00b0C to increase the pressure within the c y l i n d e r beyond the normal value of 1.1 MPa (160 p s i ) at room temperature. The regulator on the COS cylinder was set so that the difference between the f i n a l and i n i t i a l pressures i n the reactor corresponded to the desired p a r t i a l pressure of COS for the run. The COS l i n e was l e f t open and connected throughout the experiments to ensure a continuous supply of COS to the reactor. A check valve on the COS l i n e prevented backflow even when the pressure in the reactor rose beyond the d e l i v e r y pressure set with the regulator. In the case of CS2 runs, a known volume of CS2 was placed i n a 40 mL s t a i n l e s s steel c y l i n d e r which was connected to the reactor, and F i g u r e 3 . 2 : S e t - u p f o r t h e C S 2 - D E A d e g r a d a t i o n e x p e r i m e n t s . 39 forced into the reactor with nitrogen. The d e l i v e r y pressure of the nitrogen cylinder was regulated to achieve a constant t o t a l pressure i n a l l the runs. The experimental set-up i s shown i n Figure 3.2. 3.4 SAMPLING The gas sampling l i n e was a 1\/8\" s t a i n l e s s s t e e l tube f i t t e d with i n l e t and outlet valves. Gas from the reactor was f i r s t trapped between the i n l e t valve of the sampling l i n e and the gas outlet valve on the reactor. With the oulet valve of the sampling l i n e closed, the i n l e t valve was opened to tr a n s f e r the gas sample to the l i n e . The sample was subsequently withdrawn with a 10 mL \"pressure-lok\" syringe by i n s e r t i n g the syringe through a septum i n the \"swagelok\" f i t t i n g connected to the outlet valve of the sampling l i n e . This arrangement was s u f f i c i e n t to reduce the pressure of the sample and make i t possible to c o l l e c t the sample without d i f f i c u l t i e s . Expansion of the sample i n the syringe during sampling, further reduced the pressure. P r i o r to sampling, the sampling l i n e was purged with the sample gas. Liquid samples were c o l l e c t e d with a 2 mL s t a i n l e s s s t e e l c o i l f i t t e d with i n l e t and outlet valves. The pressure i n the reactor was s u f f i c i e n t to force l i q u i d samples into the sampling c o i l once the sampling valve on the reactor and the i n l e t valve of the sampling c o i l were opened, while the outlet valve of the c o i l was closed. About 3 mL of solution were purged to ensure that the sample i s representative of the reactor solution. With a l l valves closed, the sampling c o i l was disconnected and r a p i d l y cooled to room temperature by immersing i t i n 40 i c e - c o l d water. The sample was then transferred into a covered glass v i a l ready for a n a l y s i s . The sampling c o i l was then flushed with water and a i r d r i e d . I t was then ready for subsequent use. 3.5 OPERATING CONDITIONS The degradation experiments were conducted under the following operating conditions: 1. DEA concentration 10 - 40 wt% (appr. 1 - 4 M) This range was chosen to cover the 10 - 35 wt% range commonly used i n d u s t r i a l l y . A few runs were done at i n i t i a l DEA concentration of 60 wt% to explore the e f f e c t of high amine concentrations on the rate of degradation. 2. Temperature 120 - 195 \u00b0C I n d u s t r i a l DEA plants operate at regeneration temperatures of about 120 \u00b0C. Part of the amine solution i s exposed to temperatures as high as 140 \u00b0C, e s p e c i a l l y at the r e b o i l e r surfaces. Temperatures as high as 195 \u00b0C have been used i n t h i s study merely to increase the rate of the reactions. In a d d i t i o n , since the reaction mechanism may change with operating temperature, a s u f f i c i e n t l y wide range of temperature needs to be covered. 41 3. COS p a r t i a l pressure 0.3 - 1.17 MPa CS2 volume 2.5 - 10.5 mL (0.055 - 0.233 mole CS 2\/mole DEA) It i s recognised that these ranges are much higher than those normally encountered i n d u s t r i a l l y . They have been chosen to obtain measurable quantities of degradation compounds within a reasonable time. It should be noted that the CS2\/DEA mole r a t i o s are much lower than 1.195 which was used by Osenton and Knight (21). They therefore represent plant conditions more c l o s e l y . 4. Volume of DEA so l u t i o n 250 mL This volume was considered s u f f i c i e n t so that sample withdrawals (10 x 5 mL) w i l l have no s i g n i f i c a n t e f f e c t on the volume of the l i q u i d reactant.In addition, enough space i s l e f t i n the reactor f or the gas. Chakma (18) reported s i m i l a r MDEA concentrations i n two runs conducted with i n i t i a l solution volumes of 100 and 250 mL. 3.6 ANALYTICAL PROCEDURE A Hewlett-Packard Gas Chromatograph (Model 5830A) equipped with an integrator terminal, was used to separate the various constituents of the reactor samples. The concentrations of the constituent compounds were determined from previously prepared c a l i b r a t i o n curves (see 42 Appendix B). The p r i n c i p a l operating conditions used i n the present study were: Column Detector Detector Temperature Temperature Program C a r r i e r gas Inje c t i o n Temperature Sample volume Tenax (GC and TA)., 60\/80 mesh packed i n a 9'x 1\/8\" s t a i n l e s s steel column (supplied by SUPELCO Inc., Oakville, Ont.) H 2 flame i o n i z a t i o n (FID) 300 \u00b0C Isothermal at 150 \u00b0C for 0.5 minutes, then r a i s i n g i t to 300 \u00b0C at the rate of 8 \u00b0C\/min N 2 at 23 mL\/min 300 \u00b0C 0.001 mL These conditions are s i m i l a r to those used by Kennard and Meisen (74) and Chakma and Meisen (76). The i d e n t i t y of the degradation compounds cannot not be confirmed by the GC analysis alone. A combination of the techniques described below was employed. Other a n a l y t i c a l techniques were used i n order to i d e n t i f y the s o l i d reaction products. 3.7 TECHNIQUES USED TO IDENTIFY THE DEGRADATION PRODUCTS In order to i d e n t i f y the products of the degradation reactions, four successive methods were used: 43 3.7.1 GAS CHROMATOGRAPHIC (GC) ANALYSIS This analysis provides the concentrations and the retention times of the compounds. The l a t t e r should remain approximately constant under the same a n a l y t i c a l conditions, assuming l i t t l e or no column de t e r i o r a t i o n . 3.7.2 GAS CHROMATOGRAPHIC\/MASS SPECTROMETRY (GC\/MS) ANALYSIS The GC\/MS system consists of a gas chromatograph coupled to a mass spectrometer. The unit used i n t h i s study was a Hewlett-Packard, Model 5985B. In the analysis of a mixture of compounds, the gas chromatograph vaporises and fr a c t i o n a t e s an injec t e d sample of the mixture. The vapour fra c t i o n s e l u t i n g from the GC column flow i n t o the ion source of the mass spectrometer. The ion source, maintained at high temperature and vacuum pressure, produces electrons from a hot tungsten filament which bombards the incoming vapour f r a c t i o n s , thereby causing i o n i s a t i o n and fragmentation of the molecules. The r e s u l t i n g mixture of ions i s accelerated through an e l e c t r i c f i e l d into the ion c o l l e c t o r system where they are separated according to t h e i r mass to charge (m\/e) r a t i o . The charge c a r r i e d by each ion produces an e l e c t r i c current that i s detected by an electrometer and then amplified and recorded. This record of the numbers of d i f f e r e n t kinds of ions i s c a l l e d the mass spectrum. The uniqueness of the molecular fragmentation a s s i s t s with the i d e n t i f i c a t i o n of compounds because no two compounds ionise and fragment 44 i n exactly the same manner. The mass spectrum i s therefore, a \"finger p r i n t \" of each compound. The mass spectrometer can be operated i n the Electron Impact (EI) or Chemical Ionisation (CI) modes. In the EI mode, the ion source voltage i s about 70 eV and the system pressure i s between 10 and 10 mm Hg. Each molecule flowing across the ion source i s fragmented by d i r e c t bombardment with electrons. This bombardment generates numerous ions as the voltage applied i s s u f f i c i e n t to rupture many bonds i n the molecules. In the CI mode, the molecules to be fragmented are d i l u t e d with excess reagent gas, usually methane. At the normal system pressure, methane i s bombarded with a stream of electrons which r e s u l t s i n i o n i s a t i o n and fragmentation (77): e\" CH 4 -> CH 4 + + CH 3 + + CH 2 + + CH + + C + 3.1 By maintaining the ion source at a higher pressure (usually 0 . 5 - 1 mm Hg) than the rest of the system, ion-molecule reactions occur by c o l l i s i o n : CH 4 + + CH 4 -> CH 5 + + CH 3 3.2 CH 4 + CH 3 + -+ C 2 H 5 + + H 2 3.3 CH 4 + C 2 H 5 + - C 3 H 5 + + H 2 3.4 CH 5 + i s usually the most abundant of these ions. These p o s i t i v e ions are then used to ionize the molecules of a sample. Due to the high methane 45 to sample r a t i o ( t y p i c a l l y 10 ), i o n i z a t i o n of the molecule occurs via ion - molecule c o l l i s i o n s : M + CH 5 + (M+l) + + CH 4 3.5 M + C 2 H 5 + -> (M+29)+ 3.6 M + C 3 H 5 + -\u00bb (M+41)+ 3.7 Other reactions causing proton a b s t r a c t i o n ( M - l ) + and loss of water (M+l-18) + may also occur. Since these c o l l i s i o n s occur at lower energies than d i r e c t e lectron bombardment, less fragmentation of the molecules occurs. Thus the CI spectrum displays a prominent (usually the most abundant) protonated molecular ion (M+l) +, from which the molecular weight of the compound may be i n f e r r e d . 3.7.3 GC\/MS ANALYSIS OF SILYLATED DERIVATIVES To gain i n s i g h t into the structure of the degradation products, the mass spectra of the compounds i n a . s i l y l a t e d sample of the degraded solution were also obtained. S i l y l a t i o n i s a form of d e r i v a t i z a t i o n whereby s i l y l groups are introduced i n t o molecules to replace active hydrogen atoms. The s i l y l groups can a l s o replace metals i n s a l t s (78). The s i l y l a t i o n was performed with trimethyl s i l y l imidazole (TSIM) following Chakma's procedure (18). 5 mL of a degraded so l u t i o n were placed i n a glass v i a l and saturated with potassium carbonate to dehydrate the sample. Isopropyl alcohol was then added to extract the degradation compounds from the mixture. The extract was transferred to another v i a l where the alcohol was removed by evaporation, leaving a 46 viscous o i l . An excess\u2022 of TSIM was added to the o i l , the mixture was thoroughly shaken and l e f t , f o r at least 1 hour at room temperature to ensure complete d e r i v a t i z a t i o n . The d e r i v a t i z e d sample was then analysed by GC\/MS. A fused s i l i c a megabore column (50% phenyl methyl s i l i c o n e , 0.53 mm ID x 10 m) was used to separate the mixture because Tenax G.C. proved to be unsuitable. Since TSIM attacks p r i m a r i l y hydroxyl groups, the CI spectra could be used to determine the molecular masses of the s i l y l a t e d compounds and hence the number of hydroxyl groups per molecule. The s i l y l a t i n g trimethyl s i l i c o n ion (TMS) has a molecular mass of 73 and replaces the hydrogen in -OH groups. Thus s i l y l a t i o n of a compound increases i t s molecular mass by 72n, where n i s the number of hydroxyl groups in the compound. 3.7.4 GC ANALYSIS OF DEGRADED MIXTURES SPIKED WITH SUSPECTED COMPOUNDS GC\/MS analysis provided information on the possible i d e n t i t i e s of the unknown compounds i n degraded solutions. Pure forms of the suspected compounds were purchased or synthesized and then used to spike the sample. The spiked sample of the degraded sol u t i o n was then analysed by gas chromatography. An increase i n the peak area of the corresponding peak provided strong proof that the peak represents the unknown compound. In summary, the steps followed to i d e n t i f y the degradation compounds were: 47 1. Determination of the re t e n t i o n times under standard GC operating conditions. 2. Obtaining the EI and CI mass spectra from the GC\/MS analysis. The CI spectra provided the molecular weights of the unknowns, while the EI spectra were compared with l i t e r a t u r e spectra (15,16,18,79,80) . 3. Determination of the number of hyroxyl groups i n the unknown compounds by comparing the CI spectra of the s i l y l a t e d derivatives and u n s i l y l a t e d compounds. 4. Comparing the retention times of the suspected and the unknown compounds under the same GC operating conditions. 3.8 EXPERIMENTAL DESIGN Experiments were designed i n stages that r e f l e c t the stated objectives of the study: 1. I d e n t i f i c a t i o n of reaction products Two runs each were conducted f o r the COS-DEA and CS2-DEA systems at the boundary temperatures ind i c a t e d i n section 3.5, to ascertain that the reaction products at the high temperature are s i m i l a r to those produced at the low temperature. The runs w i l l also serve to v e r i f y the fact that the high temperature merely increases the rates of the reactions. However, differences observed i n the product spectra f o r the low and 48 high temperature runs f o r the CS2-DEA system necessitated another run at 165 \u00b0C. 2. E f f e c t s of operating conditions Runs were conducted as shown below, to determine the e f f e c t s of temperature, solution concentration, i n i t i a l COS p a r t i a l pressure and i n i t i a l CS 2 volume on the degradation reactions: DEA CONC. (wt%) 195 40 30 20 * 10 where * and + denote COS-DEA and CS2-DEA runs, r e s p e c t i v e l y . The runs were conducted with an i n i t i a l COS p a r t i a l pressure of 345 kPa or an i n i t i a l CS 2 volume of 6 or 10.5 mL. It should be noted that some of the runs were conducted with a glass l i n e d reactor and are only sui t a b l e f o r q u a l i t a t i v e explanations. The above runs were only able to show the e f f e c t s of temperature and DEA concentration. A separate set of runs was conducted to determine the e f f e c t s of COS p a r t i a l pressure and CS 2 volume on the reactions. The conditions used f o r these runs are: TEMPERATURE (\u00b0C) 190 180 175 170 165 160 150 135 130 127 120 49 DEA concentration COS p a r t i a l pressure CS 2 volume Temperature 30 wt% 345, 759, 1172 kPa 2.5, 6.0, 10.5 mL 150 \u00b0C for the COS runs 165 \u00b0C for the CS 2 runs 3. E f f e c t s of mixed gas phase The o r i g i n a l i n t e n t i o n was to conduct runs using a mixture of C0 2, COS and CS 2 i n the gas phase. While i t was possible to obtain a gas mixture of COS and C0 2, there was no way to include CS 2 in the mixture and maintain i t s concentration throughout the runs. Since COS and CS 2 are eventually hydrolysed to C0 2 and H2S, a better choice was to use a mixture con s i s t i n g of the two l a t t e r gases. The re s u l t s from the use-of a C0 2\/H 2S gas mixture a l s o augment the scanty data reported i n the l i t e r a t u r e on the degradation e f f e c t s of such mixtures. The operating conditions and compositions of the gas mixtures used f or these runs are discussed i n Chapter 6. 4. Additional and sp e c i a l runs Some duplicate runs were conducted to ascertain r e p r o d u c i b i l i t y . It was also found necessary to conduct more runs to elucidate the reaction mechanisms and determine the equilibrium d i s t r i b u t i o n of the ac i d gases i n the reactor. These runs are discussed i n d e t a i l i n Chapters 6 and 7, r e s p e c t i v e l y . CHAPTER 4 IDENTIFICATION OF DEGRADATION PRODUCTS 4.1 PRELIMINARY EXPERIMENTS Before proceeding with the main experimental programme, i t was necessary to conduct preliminary experiments aimed at evaluating the e f f e c t s of c e r t a i n operating variables on the degradation reactions. 4.1.1 EFFECTS OF ELEVATED TEMPERATURES Elevated temperatures were used i n t h i s study to speed up the degradation reactions. The r e s u l t s obtained under such conditions w i l l only have i n d u s t r i a l relevance i f the alkanolamines are not thermally degraded and the reaction products are s i m i l a r to those obtained at the lower temperatures commonly used i n industry. Thermal degradation experiments, conducted by heating aqueous solutions of DEA under a blanket of nitrogen, revealed the following: at 150 \u00b0C no change i n so l u t i o n composition was observed over a period of 220 hours; at 165 \u00b0C no change occurred over 60 hours; at 180 \u00b0C thermal degradation was n e g l i g i b l e up to 80 hours. Since most of the high temperature degradation runs were conducted over 48 hour periods, the influence of thermal degradation on the r e s u l t s can be disregarded. Kennard (16) also found thermal degradation to be n e g l i g i b l e when aqueous DEA 50 51 solutions were maintained at 205 \u00b0C for 8 hours under a blanket of nitrogen. Typical chromatograms of aqueous DEA solutions degraded i n the presence of COS are shown i n Figs. 4.1a to 4.1c. The q u a l i t a t i v e s i m i l a r i t y of the figures i s obvious. This suggests that the basic reaction mechanism i s not affe c t e d by temperature. Since i n d u s t r i a l DEA regenerators operate at r e b o i l e r temperatures of up to 140 \u00b0C, the products obtained i n t h i s study should also be formed under i n d u s t r i a l conditions. 4.1.2 SURFACE EFFECTS Two runs were performed with and without the pyrex l i n e r i n the reactor. The r e s u l t s indicated very s i m i l a r products, suggesting that the l i n e r d i d not a f f e c t the reaction mechanism. Since the sol u t i o n was i n contact with the s t a i n l e s s s t e e l s t i r r e r i n both cases, i t i s also c l e a r that the reactions were not influenced by the difference i n the s o l i d surface areas. However, the l i n e r caused a temperature gradient within the reactor. The r e s u l t of t h i s was the tr a n s f e r of water from the amine solu t i o n into the annulus, thereby increasing the concentration of the amine. The use of the l i n e r was therefore discontinued and a new set of runs (C.l - C.43 i n appendix C) was conducted. 14 10 15 t = 30h (a) 52 14 10 12 13 15 t = 50h (b) 3 5 14 10 12 15 t = ZI5h (c) 1: Chromatograms of p a r t i a l l y degraded DEA solutions of 4M i n i t i a l concentration (a: 180 \u00b0C, 0.34 MPa COS; b: 150 C, 0.34 MPa COS; c: 120 \u00b0C, 0.68 MPa COS). 53 4.1.3 EFFECTS OF STIRRER SPEED S t i r r i n g speed influences the rate of mass transf e r between the gas and l i q u i d phases. It was reported previously (16,18) that the rate of degradation i s slow and that changes i n s t i r r e r speed do not a f f e c t degradation rates i n CC>2-amine systems. Since the present study involves COS and CS 2 which hydrolyse to produce H2S and C0 2, i t was decided to re-examine the e f f e c t of s t i r r e r speed and hence the rates of mass tr a n s f e r on degradation. The DEA concentrations i n two runs performed at s t i r r e r speeds of 120 and 180 r.p.m. are shown i n columns 2 and 3 of Table 4.1 . The deviations i n the concentrations are generally less than \u00b14%. This confirms that the change i n s t i r r e r speed d i d not a f f e c t the rate of degradation. The run at 180 r.p.m was terminated a f t e r 30 hours because of a leakage from the reactor. 4.1.4 REPRODUCIBILITY Columns 2 and 4 of Table 4.1 also show DEA concentrations f o r two runs performed two weeks apart but under the same operating conditions. The deviations between concentrations are generally less than 2% and therefore lend confidence to the r e p r o d u c i b i l i t y of the experimental and a n a l y t i c a l procedures. 54 Table 4.1: Re p r o d u c i b i l i t y and e f f e c t of s t i r r e r speed i n COS-DEA degradation (DEA Q = 3 M, T = 150 \u00b0C, P c o s = 345 kPa). Time DEA CONCENTRATION AT DIFFERENT STIRRER SPEEDS (h) (moles\/L) 120 rpm 180 rpm 120 rpm 0 3.01 2.98 3.00 2 3.06 2.96 2.98 4 2.91 2.91 2.86 8 2.83 2.84 2.77 12 2.76 2.78 2.69 24 2.47 2.53 2.45 30 2.31 2.31 2.30 36 2.18 2.17 48 1.93 1.91 4.2 DEGRADATION PRODUCTS RESULTING FROM COS-DEA INTERACTIONS Figure 4.2 contains chromatograms showing the gradual formation of reaction products i n a t y p i c a l run conducted with a 40 wt% DEA solut i o n , at a COS p a r t i a l pressure of 0.34 MPa and a temperature of 180 \u00b0C i n the glass l i n e d reactor. The gradual formation of reaction products i s obvious. In addition, i t was noted that the samples became more pungent and viscous as the degradation -progressed. Some p a r t i c u l a t e matter was also found i n the samples. Figure 4.2: Chromatograms showing gradual formation of degradation products i n a COS-DEA system (4M DEA, 180 \u00b0C, 0.34 MPa COS). 56 By following the techniques described in section 3.7, the compounds responsible f or the peaks l a b e l l e d i n F i g . 4.1 were i d e n t i f i e d ic and are l i s t e d i n Table 4.2. The symbols M and n OH i n Table 4.2 r e f e r to the molecular weight of the d e r i v a t i z e d compound and the number of hydroxyl groups i n the compound. It should be noted that the retention times depend, to some extent, on the concentrations of the compounds and the age of the column. The corresponding mass spectra are shown i n Figures 4.3 to 4.17. For compounds whose mass spectra were not i n the computer's data base, the l i b r a r y EI spectra r e f e r to the spectra of the pure compounds purchased or synthesized in the laboratory, depending on t h e i r commercial a v a i l a b i l i t y . A consistent pattern in the fragmentation of these hydroxyl-amino compounds i s the loss of hydroxymethyl r a d i c a l s (m\/e 31) from the parent compounds to produce, i n general, the most abundant ions. Ions of mass 30 for MEA, 58 f o r EAE, 74 for DEA, 72 for HEA, 102 for EDEA, 143 for BHEP and BHEI, 100 f o r HEOD, 99 f o r HEI and HEP were generated in t h i s manner. In the case of BHEED and THEED, hydroxymethyl r a d i c a l s were also l o s t , but the p r i n c i p a l fragmentation r e s u l t e d from the cleavage of the C-C bond between two nitrogen atoms gi v i n g the most abundant ions with mass 74 f o r BHEED (N,N' isomer) and 118 f o r THEED. Water molecules were also l o s t i n the fragmentation. For example, ions of mass 74 l o s t water to give ions with mass 56 i n the case of DEA. The molecular ion peaks are not prominent i n most of the EI spectra because of the ease with which the hydroxymethyl groups break from the molecules. The c h a r a c t e r i s t i c peaks i n the CI spectra of the hydroxyl amino compounds are produced by the [M+H]+, [M+C 2H 5] + and [M+H-H20]+ ions. For the s i l y l 57 d e r i v a t i v e s , the CI spectra show prominent [M+H]+, [M+C2H5]+ and [M-CH3] + ions. The [M-CH3]+ ion i s c h a r a c t e r i s t i c of the s i l y l a t i n g agent (TSIM). The previous, tentative i d e n t i f i c a t i o n (100) of peak 9 as b i s -hydroxyethyl amino ethanol (BHEAE) now appears to be i n v a l i d . The compound has a molecular mass of 149 as indicated by i t s CI spectrum (Figure 4.11). Its s i l y l d e rivative has a molecular mass of 221 which suggests that i t contains one hydroxyl group. The EI spectrum shows the ion with mass 74 as the most abundant. Loss of the c h a r a c t e r i s t i c hydroxymethyl r a d i c a l does not produce a prominent peak even though the compound appears to have an hydroxyethyl attachment (deduced from ion 74). This, i n addition to the prominent ion of mass 89 i n the CI spectrum, suggests that the compound i s not very stable and fragments on electron bombardment to give ions with mass 89 or, more l i k e l y , 74. Based on the presently available information, the most l i k e l y structure f o r t h i s compound i s : 0 C 2H 4OH CH3-C-S-CH2 N \\ H The name of the compound i s ethanethioic a c i d S - [(2-hydroxyethyl) amino] methyl ester. Its abbreviation i s ETAHEAME. This compound i s not a v a i l a b l e commercially and i t was therefore not possible to compare i t s retention time under the present GC conditions with that, of peak 9. Time and resources d i d not permit i t s synthesis during the present study. 58 It also appears that triethanolamine (TEA) was formed as a degradation compound but could not be c l e a r l y separated from BHEED under the a n a l y t i c a l conditions used. This supposition arises from the fact that the GC analysis of the s i l y l d e r i v a t i v e s showed a peak before BHEED, having a molecular weight of 365. This would suggest an underivatized hydroxyl amino compound with a molecular weight of 149 and three hydroxyl groups. Triethanolamine f i t s t h i s structure. 59 Table 4.2: Degradation compounds detected i n the COS-DEA system. Peak Retention C h a r a c t e r i s t i c Molecular Weight n OH Ident i t y Time (min) EI ions M M* (M*-M)\/72 1 1. 4 - 1.5 43, 58 58 ACETONE 2 2. 2 - 2.3 29, 43, 57, 72 72 BUTANONE 3 3. 1 - 3.3 30, 42, 61 61 133 1 MEA 4 5. 2 - 5.3 30, 42, 56, 58, 74 89 89 161 1 EAE 5 9. 2 -\u2022 10.0 45, 56, 74 105 249 2 DEA 6 10. 9 -\u2022 11.1 30, 45, 56, 58, 74, 102 88, 133 133 277 2 EDEA 7 11. 5 -\u2022 11.6 30, 43, 60, 72, 73 85 103 175 1 HEA 8 12 . 0 -\u2022 12.1 42, 56, 70, 88, 112 130 130 202 1 HEP 9 13. 4 \u2022 \u2022 13.5 56, 61, 74, 89, 118 149 149 221 1 ETAHEAME 10 15. 2 -- 15.5 44, 56, 74, 88, 100, 118, 127 148 292 2 BHEED 11 16. 2 \u2022 - 16.6 42, 56, 156 70, 100, 125 88, 113, 143 , 174 174 318 2 BHEP 12 16. ,7 \u2022 - 16.9 42, 56, 74, 88, 100 131 131 203 1 HEOD 13 18. ,2 \u2022 - 18.4 42, 56, 70, 85, 99 130 130 202 1 HEI 14 19. .7 - 19.8 42, 56, 174 70, 88, 100, 130 118, 143 192 408 3 THEED 15 21. .0 - 21.3 42, 56, 70, 99, 114, 143 130, 174 174 318 2 BHEI 60 \u2014 to 1 00 120 140 30.5-: \u00bb 1 L F R I l 3 0 0 2 SPECT . 9 ~ 1 4 2 - P r o p a n o n e ( 9 C I ) ( a ) r-i..i = C3H60 ice so -o 100 5 0 -0 -40 [M+H] I 5 9 87 9 ? 109 123 139 151 165 \u20141\u2014 60 80 100 ' I \" 120 140 160 \u2022 7 3 . 6 183 211 223 180 2 0 0 2 2 0 -i 1 1 \u2014 2 4 0 2 6 0 ( C ) Figure 4.3: Mass spectra of peak 1 i d e n t i f i e d as Acetone (a: EI spectrum; b: EI reference spectrum; c: CI spectrum). \u00ab 1 L F R M 3002 SFECT 233 MO= 72 C4H80 . 9 7 6 6 2 - E ' j t a n o n e ( 3 C I 9 C I ) (b) Figure 4.4: Mass spectra of peak 2 i d e n t i f i e d as Butanone (a: EI spectrum; b: EI reference spectrum; c: CI spectrum). 100 -80 -60 -40 -20 - 1 0 - I 1 ..II .1 i i \u2022 20 i ( i | 40 1 i 60 i SO T 100 120 i 140 62 [ (a) 2 9 . ^ , , ,1 1, I \" 2 LFRM 3002 SPECT 125 MM\" 61 C2H7M0 .3100 E t h a n o l , 2-aitiitio- (3CI9CI) 100 -80 -60 -40 -20 -0 - , ,1 1. i i. 2Q i i 40 60 i 80 100 i ' 120 \u2022 i 140 (b) 100 -80 -[ M + H ] + I 62 70 .8 60 -40 -0 Jr 84 20 30 40 50 60 70 80 \u202290 too (c) (d) Figure 4.5: Mass spectra of peak 3 i d e n t i f i e d as MEA (a: EI spectrum; b: EI reference spectrum; c: CI spectrum; d: CI spectrum of the s i l y l d e r i v a t i v e ) . 63 (a) (b) (c) Figure 4.6: Mass spectra of peak 4 i d e n t i f i e d as EAE (a: EI spectrum; b: EI reference spectrum; c: CI spectrum). \u00ab 1 LFRH 3002 SPECT 1457 MM\u00ab 1\u00a9S C4H111102 .9754 E t h a n o l , 2 ,2' - i m i r,ob i s- (?CI > (b) ( M + H ] + I 1 06 8 S 70 , I 1 . > 1 116 1 134 146 \" 1 162 183 21 1 \u2014 ' \u2014 I \u2014 1 1 I \u2014 I \u2014 1 \u2014 I \u2014 \u2022 \u2014 I I \u2014 I \u2014 ' \u2014 I \u2014 \u2022 \u2014 I \u2014 \u2022 \u2014 I \u2014 ' \u2014 I \u2014 I \u2014 I \u2014 - \u2014 I \u2014 \u2022 \u2014 1 \u2014 60 80 100 120 140 160 180 200 (C) (d) Figure 4.7: Mass spectra of peak 5 i d e n t i f i e d as DEA (a: EI spectrum; b: EI reference spectrum; c: CI spectrum; d: CI spectrum of the s i l y l d e r i v a t i v e ) . 65 lee -89-66 \u202249 -29 9 I 192 58 45 74 \u2022\u2022\u2022<\u2022',\u2022 \u2022\u2022 88 \u2014i-L* , r 1 18 133 174 297 42.3 69 89 199 129 149 169 189 299 229 (a) (b) 199 88 69-49 26-9 [ M + H ] + I 134 116 192 58 72 99 162 I 174 189 211 223 6 8 88 189 128 - r - * \u2014 i \u2014 \u2022 \u2014 i \u2014 \u2022 \u2014 r 149 169 189 - 1 \u2014 ' \u2014 l \u2014 \u2022 298 58 .9 229 ( C ) i ea - r29 . 4 188 se-57 \" 77 86 9S 163 U6 131 144 9 - 1 i I I \u2022 t. \u20141 '\u2014\"T\u2014 7 \u2022\"\"\u20141 * i i 199-. 69 S9 168 129 148 1 169 139 r 20 . 4 2 - \u20223 (K+HJ* ?~? se- 396 229 234 250 1 1 313 . i 1. 1. \u2014 i \u2022 i \u2014 299 ' t \" 229. 249 260 2 3*3 300 (d) Figure 4.8: Mass spectra of peak 6 i d e n t i f i e d as EDEA (a: EI spectrum; b: EI reference spectrum; c: CI spectrum; d: CI spectrum of the s i l y l d e r i v a t i v e ) . 66 (a) (b) (c) (d) F i g u r e 4.9: Mass s p e c t r a of peak 7 i d e n t i f i e d as HEA ( a : EI s p e c t r u m ; b : EI r e f e r e n c e s p e c t r u m ; c: CI s p e c t r u m ; d : CI s p e c t r u m of the s i l y l d e r i v a t i v e ) . (b) (c) (d) Figure 4.10: Mass spectra of peak 8 i d e n t i f i e d as HEP (a: EI spectrum; b: EI reference spectrum; c: CI spectrum; d: CI spectrum of the s i l y l d e r i v a t i v e ) . 68 100 -, 80 60 H 40 20 0 \u2022> r40 .6 5 45 U S 131 | '-f 14? 161 \u2022Mil, , , Uij , 1 tl 60 1 00 120 140 160 180 200 ( a ) (b) (c) Figure 4.11: Mass spectra of peak 9 i d e n t i f i e d as ETAHEAME (a: EI spectrum; b: CI spectrum; c: CI spectrum of s i l y l d e r i v a t i v e ) . (a) (b) (c) tee 80 46 \u2022 28 146 136 1 1 \u2022!\u00bb | . 1 146 166 263 138 186 266 \u2014 i \u2014 \u2022 \u2014 i \u2014 \u2022 \u2014 t \u2014 229 24e 266 277 r21 .6 2?3(M+HJ+ 30S 286 Ml , I 366 321 326 (d) Figure 4.12: Mass spectra of peak 10 i d e n t i f i e d as BHEED (a: EI spectrum; b: EI reference spectrum; c: CI spectrum; d: CI spectrum of the s i l y l d e r i v a t i v e ) 70 (b) (c) tee-i se 69 11? t46 i*e 187 202 21S 229 2S6 262 27S tee se 60 I i \u2022 ! ... , 1 I ' 1 88 100 12? 140 160 180 206 * I ' 'I \u2022 I \u2022 I \u2022 I ' r 220 246 266 3 9t*+H] + 363 e J i \u2014 i \u2022 280 306 :47 33S 3S9 I 320 3 4 0 \u2022 ' i 1 I 1 i \u2014 \u00bb 360 3S0 4 9 6 42Q 449 466 i 1 I 1 \u2014r-4 * 0 21 . 21 .9 (d) Figure 4.13: Mass spectra of peak 11 i d e n t i f i e d as BHEP (a: E I spectrum; b: E I reference spectrum; c: C I spectrum; d: C I spectrum of the s i l y l d e r i v a t i v e ) . 71 (b ) [M+H] + r S 6 .8 l e a -1 \u2022\u2022z s e - 114 1 S 9 e -6 1 7 6 8 1 8 9 1 9 1 j 1 2 3 1 4 1 1, ' T \u2014 \u2022 \u2022 i ~* 1 \u2014 - \u2014 I 1 i 1 i 1 0 9 - ] 6 e 8 8 t e e 1 2 0 1 4 0 1 6 9 5 6 . 8 s e -e -1 7 2 1 8 9 1 8 ? 1 9 9 2 8 6 2 2 3 2 6 3 > \" t \u2014 \u2022 1 \" i \u2014 \u2022 \u2022 ' r i \u2014 \u2022 \u2014 i \u2022 i \u2022 i 1 3 9 2 e e 2 2 0 2 4 0 2 6 0 2 8 0 (c) (d) Figure 4.14: Mass spectra of peak 12 i d e n t i f i e d as HEOD (a: EI spectrum; b: EI reference spectrum; c: CI spectrum; d: CI spectrum of the s i l y l d e r i v a t i v e ) . (a) (b) l e e -[M+H] + I 1 3 1 r 4 3 . 6 se-6 1 7 0 3 8 1 1 3 9 9 i e6 1 1 4 5 IS? e-iee-1 . f 1 i 1 i 6 8 1 I 1 I 8 8 1 0 0 1 2 6 1 4 0 ' 1 1 1 166 4 8 . 6 se-1 7 1 1 8 3 \u2022 2 0 3 2 1 1 2 3 3 2 6 1 o- 1 1 I 1 8 0 2 0 8 2 2 0 2 4 0 260 2 8 0 < c ) (d) F i g u r e 4.15: M a s s s p e c t r a o f p e a k 13 i d e n t i f i e d a s HE-I ( a : EI s p e c t r u m ; b : EI r e f e r e n c e s p e c t r u m ; c : CI s p e c t r u m ; d : CI s p e c t r u m o f t h e s i l y l d e r i v a t i v e ) . (a) lee se -68-48 -2 0 -8 1 18 74 56 45 38 f 48 88 iee I ^ 138 ' 4 3 159 169 187 68 89 109 129 11 I 1 \u2014 I \u2014 ' ' i \u2014 I \u2014 \u2022 \" i ' 140 160 180 290 r 2 1 .3 (b) (c) 199 59 -0 -91 103 115 30 >\u00ab\u00ab > \u2022 , I,. \u2022 160 174 188 591 215 262 \u2014i\u2014\u2022\u2014r-^-\u2014:\u2014\u2022-100 -I 50 -60 88 100 129 149 168 189 299 220 249 269 9 J 3 * 3 4 9 9 t \u00ab + H ] + 19.1 \u202219.1 299 393 319 i 1 i \" i 1 \u2022 \u2022 * > \u2014 \u2014 i \u2022 i \u2022 289 399 329 340 423 43T 4 0 0 -120 4 4 9 4 6 0 4 3 Q (d) Figure 4.16: Mass spectra of peak 14 i d e n t i f i e d as THEED (a: EI spectrum; b: EI reference spectrum; c: CI spectrum; d: CI spectrum of the s i l y l d erivative) 74 (a) ( b ) lee- {M+HJ* i rSS .2 1 88-60-40-1S7 20-I M .31 M 3 283 76 88 1 189 t 2 t s 236 0- i , i i - 1 I., I 1 88 168 128 146 166 186 206 220 246 ( c ) 166 50-0 160 50-0 69 97 117 146 160 187 202 215 229 250 262 275 I 1 I ' I i 1 i 1 i \u00bb I \u2022 \"I\"' i i \" i i \u2022 i 1 i 1 i 1 \u2014 i \u2022 v 1 \u2014 i * i \u2022 60 86 166 126 146 160 180 260 220 246 266 , 3i9(H+H)* 383 ?47 I 335 359 I\u2014 i ' \" i \u2022\u2022 i \u00bb ' ' ' \u2022 i \u2022 <\u2014t, \u201e. ., , , \u2022 , ., , 238 308 320 340 360 330 4 00 -I\u2014' ,\u2014'\u2014\u2014I\u2014\u2022\u2014I\u2014'\u2014I\u2014\u2022\u2014I 20 < j j 460 430 21 .9 21 . ? (d) Figure 4.17: Mass spectra of peak 15 i d e n t i f i e d as BHEI (a: EI spectrum; b: EI reference spectrum; c: CI spectrum; d: CI spectrum of the s i l y l d e r i v a t i v e ) . 75 The degradation compounds may be conveniently grouped into two categories: low-boiling degradation compounds which elute before DEA and h i g h - b o i l i n g degradation compounds which elute a f t e r DEA. Other low-b o i l i n g degradation compounds include methanol, ethanol, acetaldehyde and\/or ethylene oxide, a c e t i c acid, methyl pyridine, d i e t h y l disulphide, ethyl methyl pyridine and 1,2 dithiane. These compounds are not included i n Table 4.2, but t h e i r EI spectra are shown in Appendix B2. Methanol, ethanol, acetaldehyde being very v o l a t i l e , were not properly separated during the GC analyses and they were barely v i s i b l e i n the chromatograms. Their presence i n the degraded solutions was deduced from the GC\/MS an a l y s i s . Acetic a c i d eluted from the Tenax GC column just before butanone. In most of the analyses, i t was embedded i n the butanone peak. Methyl pyridine, ethyl methyl pyridine, d i e t h y l disulphide and 1,2 dithiane peaks were v i s i b l e and well separated but were formed in very low concentrations. Q u a l i t a t i v e analysis of the gas phase showed ethanol, acetone, butanone, pentanone, hexanone and some of the low b o i l i n g degradation compounds found i n the l i q u i d phase. In addition to the water soluble degradation compounds, an in s o l u b l e , l i g h t brown s o l i d product was also formed. Tests conducted to i d e n t i f y the s o l i d are discussed below. The present r e s u l t s are at variance with some previous findings (19,20). However, i t i s important to note that the degradation reactions proceed slowly. Even under the most severe conditions used i n t h i s study (40% DEA, 190 \u00b0C), only the low b o i l i n g degradation compounds were formed i n appreciable quantities within the f i r s t 6 hours. At 120 \u00b0C, 76 only the low b o i l i n g 'degradation compounds were detected within the f i r s t 24 hours. As pointed out e a r l i e r , perhaps the i n a b i l i t y of the previous investigators to detect degradation compounds i n COS\/DEA systems was due to the short experimental durations and t h e i r a n a l y t i c a l techniques. 4.3 DEGRADATION PRODUCTS RESULTING FROM CS2-DEA INTERACTIONS Figures 4.18a-c show chromatograms of runs conducted with 10.5 mL CS 2 and 250 mL of 3M aqueous DEA solutions at 180, 165 and 150 \u00b0C, r e s p e c t i v e l y . Runs conducted at temperatures between 165 and 120 \u00b0C (e.g. Figs. 4.18b and 4.18c) gave s i m i l a r products and samples of the solutions had a l i g h t e r colour, less pungent odour and greater s o l i d s content due to the formation of the dithiocarbamate s a l t , than the corresponding COS-DEA samples. It was necessary to centrifuge the samples from the CS2-DEA runs before i n j e c t i n g them into the gas chromatograph. At 180 \u00b0C, the reaction mechanism appears to be more complex because more products were formed i n s i g n i f i c a n t q u a n t i t i e s . Furthermore, the samples were deep brown and possessed a very pungent odour. The pressure i n the reactor was a l s o found to increase by about 0.45 to 0.55 MPa (70 to 80 psi) within 3 hours, compared with approximately 0.07 to 0.15 MPa f o r the corresponding COS-DEA runs. These r e s u l t s indicate that, unlike the COS-DEA system, some reaction steps in the CS 2-induced degradation of DEA are a f f e c t e d by temperature. The resemblance between Figs. 4.18a and 4.1a-c suggests 8 V (a) (b) 34 8 (c) Figure 4.18: Chromatograms of p a r t i a l l y degraded DEA solutions of 3M i n i t i a l concentration degraded with 10 mL of CS 2 for 48 hours (a: 180 \u00b0C; b: 165 \u00b0C; c: 150 \u00b0C). 78 that degradation i n the CS2-DEA system at 180 \u00b0C i s preceded by hydrolysis of CS2 to COS with the l a t t e r causing COS-induced degradation. Following the a n a l y t i c a l procedure described e a r l i e r , the peaks l a b e l l e d 1 to 8 i n F i g . 4.18 were i d e n t i f i e d as MEA, DEA, BHEED, BHEP, HEOD, HEI, THEED and BHEI, r e s p e c t i v e l y . MEA was the only low b o i l i n g compound of s i g n i f i c a n c e detected i n CS2-DEA systems at temperatures below 165\u00b0C. A s o l i d material was also produced by the CS2-DEA systems. Solids produced i n runs conducted at temperatures of 165 \u00b0C and below, had a beige colour whereas the s o l i d s recovered from runs conducted at 180 \u00b0C were yellowish brown and s i m i l a r to those of the COS runs. 4.4 CHARACTERIZATION OF THE SOLID PRODUCTS The s o l i d s recovered from a l l the COS runs and the CS2 runs conducted at 180 \u00b0C were yellowish brown i n colour. L i q u i d samples withdrawn from both sets of runs also had the same yellowish brown appearance. GC\/MS analysis of the centrifuged l i q u i d samples revealed that they contain the same compounds. In view of these s i m i l a r i t i e s , both sytems are believed to be undergoing s i m i l a r reactions. As for the CS2 runs conducted at temperatures of 165 \u00b0C and below, the recovered s o l i d had a beige colour while the l i q u i d samples were colourless and did not contain any ketones. The s o l i d formed fin e powders- on grinding, unlike the s o l i d s generated from the COS runs and CS2 runs at 180 \u00b0C 79 which were s t i c k y . In order to characterize the s o l i d materials from the CS2-DEA and COS-DEA systems, the following t e s t s were conducted. 4.4.1 SOLUBILITY The s o l u b i l i t y of the so l i d s i n a v a r i e t y of solvents such as water, ethanol, methanol, toluene, d i e t h y l ether, acetone and carbon disulphide was examined. Both the COS and CS 2 generated s o l i d s were found to be insoluble i n these solvents even when the solvents were heated to b o i l i n g . However, dimethyl formamide at b o i l i n g temperature (153 \u00b0C) was able to dissolve the s o l i d s but p r e c i p i t a t i o n occurred as soon as the solutions were cooled. The insoluble nature of the s o l i d suggests that i t i s a polymeric material. 4.4.2 MELTING POINT The melting points of the so l i d s were determined with a K o f l e r Hot Stage Microscope. The set up consisted of the Kofler hot bench mounted on a Wild Heitz microscope, a thermometer to monitor the temperature of the hot bench and a regulating transformer which c o n t r o l l e d the rate of heating. Controlled i l l u m i n a t i o n of the microscope was achieved with an on-line voltage regulator. A t i n y chip of s o l i d to be analysed was placed on a s l i d e which was then centred on the hot bench. The heating rate and hence temperature of the hot bench was c o n t r o l l e d by a series resistance connecting the hot bench to the main power supply. The resistance was 80 set i n such a manner that the temperature of the hot bench rose by 4 \u00b0C\/min. While the bench was heated, the state of the s o l i d was monitored by viewing the sample at a magnification of 100. At the outset, the sample appeared as a black dot i n the f i e l d of the microscope. At the melting temperature, the sample became f l u i d and the f i e l d became illuminated. The temperature indicated on the thermometer was recorded as the melting point of the sample. Four determinations were made f o r each sample and the average temperatures were recorded. The COS generated s o l i d melted i n the range 124 \u00b0C to 138 \u00b0C with most of i t melting above 135 \u00b0C. The CS 2 s o l i d generated at temperatures below 165 \u00b0C had a narrower melting range of 138 \u00b0C and 144 \u00b0C. The size of the melting ranges suggest that the CS 2 generated s o l i d i s less contaminated than the COS generated s o l i d . Since the b o i l i n g temperature of dimethyl formamide exceeds the melting points of the s o l i d s , t h e i r apparent d i s s o l u t i o n i n the b o i l i n g solvent was due to melting. Therefore the s o l i d s can be considered insoluble i n dimethyl formamide. 4.4.3 ELEMENTAL ANALYSIS The elemental composition of the s o l i d s was determined by the Canadian M i c r o a n a l y t i c a l Laboratory, New Westminster, BC, and the r e s u l t s are shown i n Tables 4.3 and 4.4. The compositions of the s o l i d s seem to depend on the conditions used f o r the degradation runs. The percentage of sulphur decreased with increasing temperature while the reverse was the case for the other 81 Table 4.3: Elemental compositions of s o l i d s formed i n the COS-DEA systems ELEMENT COMPOSITIONS (wt%) 150 \u00b0C 180 \u00b0C wt% mole% wt% mole% c 41. . 95 28. .31 44. .11 29, ,07 H 7. .08 57. .34 7. .26 57. .41 N 1. .76 1. .02 2. .94 1. .66 0 2. .87 1. .45 3. . 94 1. . 95 s 46. . 92 11. .88 40. .14 9, .92 TOTAL 100. .58 100. .00 98. .39 100, .00 Table 4.4: Elemental compositions of s o l i d s formed i n the CS2-DEA systems. ELEMENT COMPOSITIONS (wt%) 150 \u00b0C 180 \u00b0C wt% mole% wt% mole% c 38. 97 26. .88 48. 00 29.32 H 6. 95 57. .53 7. 89 57.84 N 3.09 1. .83 4. 44 2.32 O 4.92 2. .55 6. 19 2.84 S 43.35 11. .21 33. 49 7.67 TOTAL 97.28 100. .00 100. 01 100.00 82 elements. The changes were generally more pronounced f o r the s o l i d s generated i n the CS 2 runs. This trend could be due to the fact that, except f o r sulphur, a l l other elements are contained i n the amine whose i n i t i a l concentration increases with operating temperature (due to increased evaporation of water). In the same vein, increasing temperature l i m i t s the s o l u b i l i t y of COS and CS 2 and consequently the amount of sulphur a v a i l a b l e f o r reactions i n the l i q u i d phase. The r a t i o s of the elements could be ca l c u l a t e d from the elemental analysis shown i n Tables 4.3 and 4.4. For example, i n the s o l i d generated i n the COS runs at 180 \u00b0C, the C\/S, C\/H, C\/N, S\/O, N\/O, C\/O, H\/N and H\/O r a t i o s are 2.93, 0.50, 17.50, 5.09, 0.85, 14.93, 34.57 and 29.48, re s p e c t i v e l y . These r a t i o s transform approximately to an empirical formula of C 1 5 H 3 0 N O S 5 ( E- w 400). Using the same c a l c u l a t i o n s , the formula c14 H30 N O^6 ( E > w 420) can be derived f o r the so l i d s recovered from the CS 2 system at 150 \u00b0C. The higher molecular weight of the CS 2 s o l i d would imply a higher melting point as confirmed by the melting point determinations. 4.4.4 MASS SPECTRAL ANALYSIS S o l i d probe EI and CI mass spectral analyses were performed on the s o l i d products to determine t h e i r fragmentation patterns as well as t h e i r molecular weights. Both s o l i d s produced s i m i l a r traces as shown i n Figures 4.19 and 4.20. The CS 2 s o l i d trace i s f o r the s o l i d generated i n the CS 2 run at 150 \u00b0C. Although the EI spectra resemble the l i b r a r y spectrum of a c e t i c a c i d - mercapto - 1,2 ethanediyl ester (CgH 1 Q0 4S 2) 83 with a molecular weight of 210, the elemental composition of t h i s compound i s at variance with the r e s u l t s of the elemental a n a l y s i s . Hence the i d e n t i t y of the s o l i d s i s d i f f e r e n t from that suggested by the l i b r a r y of mass spectra. The fragmentation pattern i n the CI spectra show successive losses of ions of masses 28 and 32. Since the so l i d s are r i c h i n sulphur, the ion of mass 32 i s most l i k e l y due to elemental sulphur. The ion of mass 28 could e i t h e r be a carbonyl group (C=0) or an ethenyl (C 2H 4) group. However, the very low oxygen to sulphur r a t i o i n the s o l i d s , the almost equal number of losses of masses 28 and 32 and the high carbon content of the s o l i d s , point to an ethenyl group as the fragmenting group with mass 28. It i s d i f f i c u l t to i d e n t i f y a molecular ion peak from the CI traces because of the low abundances of the high molecular weight ions. The most abundant peak has a mass of 121, suggesting a molecular weight of 120. The high melting point of the s o l i d s i s inconsistent with t h i s molecular weight. Furthemore, the absence of ions of mass 149 (M+29) makes a molecular mass of 120 very u n l i k e l y . The pattern of successive losses of masses 28 and 32 i n the CI spectra suggests a f r a g i l e l i n e a r structure containing several covalent bonds with sulphur interspersed between the ethenyl groups. The ease of bond breakage i s , perhaps, the reason f o r not having a prominent molecular ion peak. The s i m i l a r i t i e s i n the mass spectrum of the s o l i d s i s another i n d i c a t i o n that the s o l i d s have s i m i l a r structures. 84 (a) 100 5 0 0 100 50 0 J 61 7 5 I 89 I 121 105 153 181 191 2 0 9 2 2 4 80 100 120 \u2022I I ni | -1 I.I..).. \u2014<~i\u2014' \u2022\u2022\u2022 I 1 2 . 8 140 160 180 2 0 0 1\u2014'\u2014r 2 2 0 2 4 r , 24 251 2 6 9 2 8 6 3 0 2 3 1 2 3 3 9 3 4 6 3 5 8 1 2 . 8 3 9 0 4 0 0 4 1 9 \"I I\u2014 I I\u2014 1 2 6 0 2 8 0 3 0 0 \u2022 1\u2014 1 \"I' \u2022 \u2014 \u2014 I ' I ' 3 2 0 3 4 0 3 6 0 i \u2014 i \u2014 ' \u2014 l \u2014 ' \u2014 l \u2014 ' \u2014 r 3 8 0 4 0 0 (b) Figure 4.19: Mass spectra of the s o l i d formed i n the COS-DEA system (a: EI spectrum; b: CI spectrum). 85 l O O - - 1 6 . 1 S e - 4 5 69 9 2 1 1 1 7 4 1 105 t 152 184 0- J . J 1 I J . i i l , ,. III, .1 .... . ,...1 1 1 ' 1\"' I 1 \u2022 i \u2022 i i 1 i I l l 1 0 0 -60 8 9 100 120 140 160 180 r 1 6 . 1 5 0 -2 1 2 2 2 4 2 4 0 2 7 3 3 3 2 0 - I. 1 1 \u2022 | \" ' \u2022 i > i i \u2022 i 1 i i \u2022 i 2 0 0 2 2 0 2 4 0 2 6 0 2 8 0 3 6 0 3 2 0 (a) 50 241 2 7 3 301 4 1 4 2 4 0 2 6 0 2 8 0 3 0 0 3 2 0 3 4 6 3 6 0 3 8 0 400 (b) Figure 4.20: Mass spectra of the s o l i d formed in the CS2-DEA system (a: EI spectrum; b: CI spectrum). 86 4.4.5 INFRA-RED ANALYSIS To gain in s i g h t into the functional groups of the s o l i d compounds, i n f r a red (IR) absorption traces were obtained using KBr p e l l e t s i n a Bomem - Michelson 100 spectrophotometer. The r e s u l t i n g traces are shown in Figures 4.21 and 4 . 22 . Again there are obvious s i m i l a r i t i e s i n the absorption wavelenghts of both s o l i d s . It should be noted that the absorbances i n the traces are of weak to medium i n t e n s i t i e s since the transmittance scale i s from 78.7% to 102.75%, a range of only 24%. Table 4.5 shows a possible assignment of functional groups to the absorption bands i n the IR traces. The dependency of the composition of the s o l i d on the operating temperature suggests that the s o l i d s are probably not pure compounds but a mixture. If t h i s i s the case, then i t i s more appropriate to consider the s o l i d s as chemically derived organic f o u l i n g deposits. It could also be that, i n spite of the s i m i l a r i t i e s i n the r e s u l t s from the other analyses, the s o l i d s are d i f f e r e n t compounds belonging to the same homologous s e r i e s . The analyses conducted have enabled the i d e n t i f i c a t i o n of the fragments or functional groups that occur i n the s o l i d products. The insolub l e nature of the s o l i d s as well as t h e i r l e v e l of p u r i t y , posed problems that prevented further analysis and conclusive i d e n t i f i c a t i o n . Figure 4 . 2 1 : Infra-red trace of the s o l i d formed in the COS-DEA system. oo Figure 4.22: Infra-red trace of the s o l i d formed in the CS 2 -DEA system. 89 Table 4.5: Functional group assignments f or the sol i d s formed i n the COS-DEA and CS2-DEA systems. LITERATURE IR (81,82) GROUP BAND (cm\" h BAND SAMPLE (cm\"1) IR ASSIGNMENT ~S~CH2 670 - 760 (m) 672, 698 (m) CH 2 ~ S ~CH 2 -SH 815 - 930 (w) 910 (w) -SH 2420 - 2600(w) 2500 (w) -SH CH2-NH-CH2 3100 - 3500 (m) 3422 (w-m) CH2-NH-CH2 1480 - 1580 (w) 1492 (w) 1100 - 1200 (m) 1187 (m) 1080 - 1150 (m) 1143 (m) \" baL N 2 12.0 24.0 TIME 36.0 48.0 60.0 (h) F i g u r e 6 . 5 : MEA c o n c e n t r a t i o n s as a f u n c t i o n of t i m e and gas c o m p o s i t i o n ( D E A Q = 3M, T = 165 \u00b0 C ) . 180 0 . 1 0 A - 15. 27 CO, ; 15 .27. H,S ; baL N 2 X - 30. 07 15 .07 H,S ; baL N2 \u2022 - 15. 57 C 0 2 ; 29 .97 H ? ; baL \"I 3 0 . 0 8 o B 2 O i\u2014i K E-55 w a Z o u w w \u00ab \u00ab 0 . 0 6 0 . 0 4 0 . 0 2 x x g 0 . 0 0 0 . 0 1 2 . 0 2 4 . 0 3 6 . 0 TIME (h) 4 8 . 0 6 0 . 0 Figure 6.6: BHEED concentrations as a function of time and gas composition (DEAQ = 3M, T = 165 \u00b0C). 181 0.05 0.04 A - 15.27. CO, ; 15. 27. H,S ; baL N2 X - 30.07. CO, ; 15. 07. H,S ; baL N2 \u2022 - 15.57 CO* ; 29 . 97 ; baL N2 0.03 0.02 0.01 0.00 J L . 0.0 12.0 24.0 36.0 TIME (h) 48.0 60.0 Figure 6.7: HEI concentrations as a function of time and gas composition (DEAQ = 3M, T = 165 \u00b0C). 182 0.10 o a 0.08 A - 15.27. CO, } 15.27 H2S ; baL N2 X - 30.07 CO, } 15.07. H,S ; baL N2 \u2022 - 15.57 CO* r 29.97. H*S ; baL N2 55 o \u00ab H 55 W CJ 55 o CJ W w PQ 0.06 0.04 0.02 h A 0.00 2 _ i _ 0.0 12.0 24.0 36.0 TIME (h) 48.0 60.0 Figure 6.8: BHEI concentrations as a function of time and gas composition (DEA Q = 3M, T = 165 \u00b0C). 183 The only mixed gas run conducted by Choi (12) was performed by f i r s t saturating the so l u t i o n with 1480 kPa of H2S at room temperature followed by contact with 4238 kPa of C0 2 at 175 \u00b0C f o r 4 hr. The procedure s i g n i f i c a n t l y hindered the s o l u b i l i t y of C0 2 i n the amine sol u t i o n . Therefore, the reduced rate of DEA degradation compared to a C0 2 run can be linked to the low concentration of C0 2 i n the amine sol u t i o n . Since most of the degradation compounds formed i n the study were not i d e n t i f i e d , a comparison of product spectra with the present study cannot be made. 6.2 EFFECT OF OXYGEN The formation of ketones was i n i t i a l l y thought to be due to oxidation by the 0.1 mole% oxygen impurity i n the COS feed. However, the r e l a t i v e l y lower s o l u b i l i t y of oxygen i n water (1) and, by inference, i n amine so l u t i o n s , coupled with the fa c t that the ketones were formed within h a l f hour of each run make t h i s u n l i k e l y . Further, the absence of well known amine oxidative degradation products ( v i z . formic a c i d propionoic a c i d and o x a l i c a c i d (84)), demonstrates that oxidative degradation was not occuring i n the COS-DEA system. In order to remove any doubts i n t h i s regard, a run was conducted following the usual procedure, except that the DEA solution used was i n i t i a l l y exposed to a i r f o r about 16 hours. In addition, the reactor was not purged p r i o r to the commencement of the run. These actions were taken to ensure that oxygen was present i n the reactor. Table 6.1 shows the concentrations of acetone, butanone and DEA i n t h i s run and i n a corresponding run 184 conducted i n a thoroughly purged environment. The s i m i l a r concentrations i n both runs provide conclusive proof that DEA oxidation by molecular oxygen was not responsible f o r the formation of the ketones. Table 6.1: Contribution of oxygen to degradation i n COS-DEA system (DEAQ = 30 wt%, T = 150 \u00b0C, P C Q S = 345 kPa). TIME CONCENTRATIONS (MOLES\/L) PURGED SYSTEM UNPURGED SYSTEM HOURS ACET BUT DEA ACET BUT DEA 0 0 .000 0.000 3.01 0.000 0 .000 3.00 2 0 .006 0.000 3.06 0.005 0.000 2.98 6 0 .007 0.002 2.91 0.007 0.002 2.86 9 0 .010 0.006 2.83 0 .010 0.007 2.77 12 0 .013 0.011 2.76 0.011 0.010 2.69 24 0 .016 0.014 2.47 0.019 0.016 2.45 30 0 .019 0.016 2.31 0.020 0.015 2.30 36 0 .021 0.017 2.18 0 .021 0.015 2.17 48 0 .015 0.016 1.93 0.022 0.017 1.91 185 6.3 EFFECTS OF DEGRADATION COMPOUNDS The e f f e c t s of the various degradation products on the degradation of DEA were studied by conducting degradation runs with 30% aqueous DEA solutions spiked with the degradation compound of i n t e r e s t . In the runs described in sections 6.3.1 to 6.3.6 below, 0.25 moles of the degradation compound of i n t e r e s t was added to the aqueous DEA solution and the mixture was contacted with 10.5 mL of CS 2 at 180 \u00b0C. CS 2 was chosen as the degrading agent because the CS2-DEA system at 180 \u00b0C behaves l i k e the COS-DEA systems at a l l temperatures used i n t h i s study. Therefore, i n terms of the water soluble products, the r e s u l t s f or a run should be applicable to a l l the COS-DEA systems investigated and the CS2-DEA systems at 180 \u00b0C. The r e s u l t s from each of these runs were then compared with the r e s u l t s from the corresponding non-spiked (regular) run. Ethanol and acetaldehyde were c l a s s i f i e d as minor degradation compounds because of t h e i r low concentrations i n the degraded DEA s o l u t i o n s . Nevertheless, since they are oxygenated compounds, i t was decided to check whether t h e i r low concentrations were due to fast transformations to ketones under the reaction conditions. 6.3.1 EFFECT OF ETHANOL The degradation of the ethanol-spiked solution r e s u l t e d i n higher concentrations of acetone, butanone and a c e t i c a c i d , but lower concentrations of MEA within the f i r s t four hours (see Figs. 6.9a and 186 6.9b). However, a f t e r rrine hours the concentrations of the ketones and ac e t i c a c i d were comparable with t h e i r respective concentrations i n the regular run, but the MEA concentration remained lower. This r e s u l t suggests that ethanol enhances the production of a c e t i c a c i d and ketones, but i n h i b i t s MEA production. Considering the fact that ethanol has two carbon atoms as opposed to acetone and butanone which have three and four carbon atoms res p e c t i v e l y , i t i s u n l i k e l y that these compounds were produced d i r e c t l y from ethanol. I t i s probable that ethanol was dehydrogenated to acetaldehyde which, i n turn, produced the ketones and ace t i c a c i d . The equal concentrations recorded f o r the ketones and ac e t i c a c i d i n both runs at 9 hr suggest an approach to equilibrium. 6.2.2 EFFECT OF ACETALDEHYDE As shown by Figs. 6.10a and 6.10b, degradation of the acetaldehyde-spiked solution r e s u l t e d i n s i g n i f i c a n t increases i n the rates of production of acetone, a c e t i c a c i d and butanone within the f i r s t four hours. The increases were three f o l d i n the f i r s t two hours. The increase i n a c e t i c a c i d concentration was accompanied by an increase i n the rate of production of HEA. By the 25 t^ hour, the rates of production of these compounds had decreased such that t h e i r concentrations i n the spiked and regular runs were comparable. The concentrations of MEA were higher i n the spiked run while the DEA concentrations were lower. This run a l s o produced higher concentrations of ETAHEAME. These observations i n d i c a t e that acetaldehyde i s involved i n the production of a c e t i c acid, acetone and butanone. 187 Figure 6.9: Chromatograms of p a r t i a l l y degraded DEA solutions of 3M i n i t i a l concentrations degraded with 10.5 mL of CS 2 at 180 \u00b0C (a: regular run; b: ethanol-spiked run). 188 Figure 6.10: Chromatograms of p a r t i a l l y degraded DEA solutions of 3M i n i t i a l concentrations degraded with 10.5 mL of CS 2 at 180 \u00b0C (a: regular run; b: acetaldehyde-spiked run). 189 The comparable concentrations of these compounds at the l a t e r stages of both runs may also be due to an approach to equilibrium brought about by some l i m i t i n g reactants. There was extensive polymerisation of acetaldehyde i n the spiked run, thereby reducing the amount a v a i l a b l e f o r the reactions. This also contributed to the decline i n the production of the ketones at the l a t e r stages of the run. 6.3.3 EFFECT OF ACETIC ACID The r e s u l t s of the ac e t i c acid-spiked run indicate that the ketones were not formed v i a a c e t i c a c i d . In f a c t , t h e i r formation was e n t i r e l y suppressed. However, the rate of degradation of DEA was increased s i g n i f i c a n t l y , r e s u l t i n g i n much higher concentrations of the high b o i l i n g degradation compounds than i n the regular run. The MEA concentration was lower due to increased conversion to HEA (see Figs. 6.11a and 6.11b). The formation of MEA implies that the hydroxyethyl group was released from DEA, yet no ketones were formed. This could be due to the fac t that under the a c i d i c condition of the run, transformation of the hydroxyethyl r a d i c a l was i n h i b i t e d . It probably reacted with a c e t i c a c i d since i t i s impossible to remain as such i n the l i q u i d phase. A l o t of s o l i d product was formed i n t h i s run. Kennard (14) observed the rate of DEA degradation by CC^ to increase with so l u t i o n pH. Since the ac e t i c a c i d spiked s o l u t i o n had a lower pH than the s o l u t i o n used f o r the regular run, the increase i n the rate of DEA degradation cannot be at t r i b u t e d to the decrease i n solution pH. 190 Figure 6.11: Chromatograms of p a r t i a l l y degraded DEA solutions of 3M i n i t i a l concentration degraded with 10.5 mL of CS 2 at 180 \u00b0C (a: regular run; b: a c e t i c acid-spiked run). 191 It i s not cl e a r at t h i s stage why the a c e t i c a c i d spiked run resu l t e d i n an increased rate of DEA degradation. 6.3.4 EFFECT OF ACETONE The acetone-spiked run resu l t e d i n s l i g h t l y reduced concentrations of butanone and DEA. In spite of the acetone production expected, the acetone peak in the chromatogram decreased progressively. This suggests that the formation of acetone may involve an equilibrium r e a c t i o n , such that the presence of acetone at the i n i t i a l stages s h i f t s the equilibrium against further production of acetone. The rates of production of ac e t i c ac i d and MEA increased within the f i r s t s ix hours, but by the 2 5 t h hour, t h e i r concentrations were s i m i l a r to those obtained i n the regular run. Acetone thus seems to a f f e c t the i n i t i a l rates of formation\/depletion of MEA and a c e t i c a c i d only s l i g h t l y . 6 . 3 . 5 EFFECT OF BUTANONE Higher concentrations of acetone and a c e t i c a c i d were recorded within the f i r s t four hours of the run. By the 1 2 t n hour, t h e i r rates of production had declined such that acetone concentration was lower than i n the regular run while the a c e t i c a c i d concentration was s i m i l a r to the concentration i n the regular run. Except at 2 hr when the MEA concentration was lower than i n the regular run, other samples taken up t i l l the 1 2 t h hour showed higher MEA concentration than i n the regular run. Similar DEA concentrations were recorded i n both runs. 192 6.3.6 EFFECT OF ETHYLENE GLYCOL The production of MEA suggested that protonated DEA molecules were los i n g hydroxyethyl groups. Chakma (18) who had studied the CO2-MDEA system at elevated temperatures, suggested that the hydroxyethyl group released from MDEA produced ethylene g l y c o l v i a ethylene oxide. In t h i s study however, ethylene g l y c o l was not detected. It i s known that aldehydes and ketones could be produced from the transformations of substituted d i o l s (85). In order to check whether the apparent absence of ethylene g l y c o l i n t h i s study was due to the fast transformation to acetaldehyde or ketones, an ethylene glycol-spiked run was conducted. No increases were observed i n the concentrations of the ketones. This indicated that acetaldehyde and ketones were not formed by rapid tranformations i n v o l v i n g ethylene g l y c o l . 6.3.7 EFECTS OF ETHYL DIETHANOLAMINE AND ETHYL AMINOETHANOL Methanol, ethanol, acetaldehyde, acetone, a c e t i c a c i d and butanone a l l contain a l k y l groups, whereas DEA which was the s t a r t i n g material does not. Examination of the product spectrum of the degraded solutions shows the presence of two a l k y l alkanolamines, v i z . EDEA and EAE. Therefore, i t was decided to check whether or not these compounds could produce the low b o i l i n g degradation compounds with a l k y l groups. The tests were done by degrading IM solutions of EDEA and EAE at 180 \u00b0C with 10.5 mL of CS 2. Figures 6.12 and 6.13, which are chromatograms of 193 la < Id Figure 6.12: Chromatogram of p a r t i a l l y degraded EDEA solu t i o n of IM i n i t i a l concentration degraded with 10.5 mL of CS 2 at 180 \u00b0C for 30 hours. Figure 6.13: Chromatogram of p a r t i a l l y degraded EAE sol u t i o n of IM i n i t i a l concentration degraded with 10.5 mL of CS2 at 180 \u00b0C f o r 24 hours. 195 samples withdrawn from these runs, show that EDEA and EAE are degraded by CS2 and by inference COS, to form acetone, a c e t i c a c i d and butanone among other products. The runs also showed the production of ethyl amine from EAE which i n turn was formed from EDEA. This trend i s consistent with the loss of a hydroxyethyl group from the parent compound as previously observed i n the formation of MEA from DEA. The fact that the hydroxyethyl group was released rather than the a l k y l group, coupled with the production of a c e t i c a c i d and the ketones, demonstrate that these low b o i l i n g degradation compounds were formed through complex transformations of the hydroxyethyl r a d i c a l . Two further runs were then performed using DEA solutions spiked with EDEA and EAE to check the contributions of these compounds to DEA degradation. As shown by Tables 6.2 and 6.3, spiking with EDEA and EAE caused increases i n the rates of production of acetone, butanone and also a c e t i c a c i d . Acetic a c i d was not included i n the table because the amount formed was only i n f e r r e d from the GC peak areas. Beyond 12 hours, s i m i l a r concentrations were recorded for acetone and butanone i n the spiked and regular runs. The approach to equilibrium suggests that acetone and butanone may have been produced from some equilibrium reactions. The MEA concentration was i n i t i a l l y lower i n the spiked runs, but approached the same maximum value as i n the regular run. The increased conversion of MEA to HEA i n the spiked run, due to the increased production of a c e t i c a c i d , r e s u l t e d in a lower f i n a l concentration of MEA. The concentrations of DEA are comparable at the i n i t i a l stages, but at the l a t t e r stages of the EDEA spiked run, lower DEA concentrations were recorded than i n the regular run. 196 Table 6.2: Concentratiens of DEA and the low b o i l i n g degradation compounds i n the regular and the EDEA-spiked runs* (DEA Q = 3M, EDEA0 = 0.25M, T = 180 \u00b0C, CS 2 volume = 10.5 mL). TIME REGULAR RUN SPIKED RUN HOUR ACET BUT MEA DEA ACET BUT MEA DEA 0 0 .00 0. 00 0. .00 3.10 0 .00 0.00 0.00 3.10 2 0 .01 0. 00 0. .11 3.08 0 .04 0.01 0.00 3.05 4 0 .02 0. 02 0. .23 2.85 0 .06 0.03 0.17 2.82 6 0 .06 0. 03 0. .35 2.54 0 .08 0.04 0.33 2.63 9 0 .08 0. 04 0, .48 2.15 0 .08 0.05 0.46 2.29 12 0 .07 0. 04 0, .57 1.87 0 .08 0.05 0.56 1.53 25 0 .03 0. 03 0. .46 0.88 0 .06 0.05 0.45 0.85 30 0 .04 0 . 04 0 .48 0 .76 0 .05 0 .04 0.43 0.59 48 0 .03 0. 04 0 .35 0 .45 0 .03 0.04 0.23 0.36 * The experiments reported i n Tables 6.2 and 6.3 were performed i n a glass l i n e d reactor. . The reported concentrations are higher than what would be obtained i n a reactor without the l i n e r . 197 Table 6.3: Concentrations of DEA and the low b o i l i n g degradation compounds i n the regular and the EAE-spiked runs (DEAQ = 3M, EAE Q = 0.25M, T =180 \u00b0C, CS 2 volume = 10.5 mL). TIME REGULAR RUN SPIKED RUN HOUR ACET BUT MEA DEA ACET BUT MEA DEA 0 0 .00 0 .00 0.00 3 .10 0.00 0 .00 0 .00 3.08 2 0 .01 0 .00 0.11 3 .08 0.04 0 .02 0 .06 3.00 4 0 .02 0 .02 0.23 2 .85 0.05 0 .03 0 .18 2.69 6 0 .06 0 .03 0.35 . 2 .54 0.06 0 .04 0 .33 2.49 9 0 .08 0 .04 0.48 2 . 15 0.07 0 .04 0 .46 2.24 12 0 .07 0 .04 0 .57 1 .87 0.06 0 .04 0 .56 2.00 25 0 .03 0 .03 0.46 0 .88 0.04 0 .04 0 .46 1.03 30 0 .04 0 .04 0.48 0 .76 0.04 0 .04 0 .44 0.75 48 0 .03 0 .04 0.35 0 .45 0 .02 0 .04 0 .25 0.43 The EAE run also provides an opportunity to evaluate the gas t r e a t i n g p o t e n t i a l of t h i s amine. Sharma and Danckwerts (48) had found EAE to be over 10 times more e f f e c t i v e than MEA and DEA i n absorbing COS. However, since there were no data on the resistance of EAE to degradation, no d e f i n i t e conclusions could be reached on i t s gas t r e a t i n g p o t e n t i a l . 198 Figure 6.14: EI mass spectrum of the compound i d e n t i f i e d as Ethyl amine in the p a r t i a l l y degraded EAE so l u t i o n . Figure 6.15: EI mass spectrum of the compound i d e n t i f i e d as Ethyl acetamide i n the p a r t i a l l y degraded EAE solu t i o n . 199 Figure 6.16: EI mass spectrum of the compound i d e n t i f i e d as Ethyl thiazolidone i n the p a r t i a l l y degraded EAE s o l u t i o n . Figure 6.17: EI mass spectrum of the compound i d e n t i f i e d as Ethyl thiazolidone-2-thione i n the p a r t i a l l y degraded EAE solution. 200 The present r e s u l t s show that EAE i s degraded by COS and CS2 just l i k e DEA. In a d i t i o n to acetone, butanone and a c e t i c a c i d , other degradation products t e n t a t i v e l y i d e n t i f i e d from the EI mass spectra were ethyl amine, ethyl acetamide, ethyl thiazolidone and ethyl thiazolidone-2-thione. The EI spectra of these compounds are shown i n Figs. 6.14 to 6.17. No s o l i d s were formed i n the EAE system, hence less f o u l i n g occurs. EAE degrades fa s t e r than DEA since 65% of the amine was l o s t i n 24 hours compared to a 55% DEA loss under s i m i l a r conditions. Since the major degradation compounds have r i n g structures, the amine could be recovered by the addition of a base, as i s done i n the case of HEOD (18). Therefore degraded EAE solutions may be amenable to the same chemical p u r i f i c a t i o n methods as those used f o r degraded DEA so l u t i o n s . 6.3.8 EFFECT OF WATER The r o l e played by water i n the degradation of DEA by COS and CS2 was investigated by conducting a run with a solu t i o n c o n s i s t i n g of 30% DEA and 70% MDEA at 150 \u00b0C and 345 kPa of COS for 48 hours. The temperature of 150 \u00b0C was chosen because i t has been shown that MDEA degradation at t h i s temperature i s n e g l i g i b l e within 48 hours (18). The samples from t h i s run were very viscous and d i f f i c u l t to draw int o the syringe. This affected the r e p r o d u c i b i l i t y of the GC a n a l y s i s . As shown by F i g . 6.18, degradation i n the non-aqueous system was s i g n i f i c a n t l y reduced, the concentration of the amine being f a i r l y 201 10 A = NON-AQUEOUS x = AQUEOUS $-X A A * x A X _L 0.0 12.0 24.0 TIME 36.0 48.0 60.0 (h) Figure 6.18: Concentrations of DEA as a function of time i n aqueous and non-aqueous sytems i n contact with 345 kPa of COS at 150 \u00b0C. 202 constant. Degradation compounds produced i n very low concentrations include BHEP, HEOD, HEI, THEED and ETAHEAME. THEED could have been formed by the d i r e c t reaction of DEA with DEA thiocarbamate. Dehydration of THEED to produce BHEP would provide some water for hydrolysis of COS to CO2 and hence the formation of HEOD. Since MEA was not detected, i t i s c l e a r that the presence of water i s e s s e n t i a l f or i t s formation. DEA usually contains some MEA as an impurity. This may be responsible for the small amounts of HEI formed. Degradation products such as BHEED and BHEI were also i n s i g n i f i c a n t i n the non-aqueous system and t h i s could be li n k e d to the absence of substantial amounts of MEA. A low b o i l i n g compound having the same retention time as acetone, but i d e n t i f i e d as ethane t h i o l , was also produced. In summary, water plays a s i g n i f i c a n t role not only i n terms of the rate of degradation, but also i n i n i t i a t i n g the hydrolysis reactions and consequently the formation of MEA and the other low b o i l i n g degradation compounds. 6.3.9 EFFECT OF MONOETHANOLAMINE Three runs were conducted to study the e f f e c t s of MEA on DEA degradation. In the f i r s t run, an aqueous mixture containing approximately IM MEA and 3M DEA, was heated and maintained at 165 \u00b0C under a blanket of nitrogen for 48 hours. No degradation compounds were formed i n t h i s run; t h i s provides an i n d i c a t i o n that DEA and MEA do not react d i r e c t l y and are not thermally degraded at 165 \u00b0C. 203 The second run was conducted f o r 32 hours under the same conditions as i n the f i r s t run except that 758 kPa. of carbon dioxide was used i n place of nitrogen. Degradation products such as BHEED, HEOD, BHEP, HEI, THEED and BHEI were formed. As shown in Table C.42 i n appendix C, the concentrations of DEA and MEA decreased while those of BHEI, HEI and BHEP increased with time. The HEOD concentration increased to a maximum value and then f e l l s l i g h t l y . Both THEED and BHEED concentrations a l s o passed through maxima. Since the degradation of aqueous DEA by produces mainly BHEP, HEOD and THEED, the formation of BHEI, BHEED and HEI i n aqueous MEA\/DEA\/C02 system can be a t t r i b u t e d to reactions i n v o l v i n g MEA, DEA and t h e i r respective carbamates. The behaviour of MEA when subjected to a gas mixture containing CO2 and H2S, as was the case i n the p a r t i a l l y degraded DEA-COS systems, was investigated i n the t h i r d run. A 2M MEA solution was contacted with a gas mixture containing 15.5% CO2, 29.9% H2S i n nitrogen, f o r 48 h at 165 \u00b0C and a p a r t i a l pressure of 1.55 MPa. Acetone and butanone were detected i n the p a r t i a l l y degraded solution. The formation of the ketones would imply the presence of ammonia, i n accord with the formation of MEA from DEA, EAE from EDEA, and ethyl amine from EAE. The non-detection of ammonia may be due to i t s high v o l a t i l i t y as well as the i n a b i l i t y of the FID to detect the gas. Ammonia was probably involved i n the formation of the pyridines. HEI was the only high b o i l i n g degradation compound formed, a l b e i t at low concentration. This may be as a r e s u l t of the p r e f e r e n t i a l protonation of MEA thereby l i m i t i n g the amount of MEA carbamate a v a i l a b l e to form appreciable 204 quantities of hydroxyethyl ethylenediamine (HEED) and oxazolidone (OZD), the other degradation products of MEA-CO2 reactions. 6.3.10 EFFECT OF BHEED The trends in BHEED and THEED concentrations suggest that they are intermediate compounds. THEED i s known to dehydrate to BHEP (13,14). T h e o r e t i c a l l y , BHEED could dehydrate in a s i m i l a r manner to produce HEP. BHEED could also react with CO2 to form BHEI. In order to determine the degradation compounds o r i g i n a t i n g from BHEED, an aqueous sol u t i o n of 0.2M BHEED was contacted with C0 2 at 180\u00b0C. Analysis of reaction samples showed the gradual formation of BHEI and by the 6 t h hour, the concentration of BHEI was 0.17M. HEP was not detected, but trace amounts of HEOD were formed. I t appears that CO2 reacted with BHEED to form BHEED carbamate, which was then dehydrated to form BHEI. Perhaps i n a CO2 l i m i t i n g environment, dehydration of BHEED would have produced HEP. BHEED carbamate i n aqueous s o l u t i o n may also e s t a b l i s h an equilibrium with MEA and DEA carbamate, the l a t t e r forming HEOD. The very low concentration of HEOD and the absence of HEP i n t h i s run suggests that BHEED, i n the presence of excess CO2, i s more l i k e l y transformed to BHEI than to HEOD or HEP. The behaviour of BHEP, HEOD and THEED i s well documented by Kennard (16) and the mechanisms for t h e i r formation are known. Therefore, no runs were conducted i n respect to these compounds. CHAPTER 7 SOLUBILITY AND HYDROLYSIS OF CARBONYL SULPHIDE When COS i s absorbed into aqueous solutions, i t hydrolyzes to CO2 and H2S according to the o v e r a l l r e a c t i o n : COS + H20 = C0 2 + H2S 7.1 Thompson et a l . (56) have obtained rate constants f o r the hydrolysis at temperatures between 15 and 47 \u00b0C. Sharma (57) has shown that the hydrolysis i s catalysed by bases. Thus an aqueous solution o r i g i n a l l y containing DEA and COS, i s eventually made up of the hydrolysis products, unreacted COS and i o n i c species derived from t h e i r i n t e r a c t i o n s with DEA and water. S o l u b i l i t y and hydrolysis proceed r e l a t i v e l y f a s t e r than amine degradation. Therefore, the extent of degradation i s l a r g e l y determined by the equilibrium composition of the sol u t i o n p r i o r to the commencement of degradation. This chapter describes experiments conducted to e s t a b l i s h the equilibrium composition of the COS-DEA system before the commencement of s i g n i f i c a n t degradation. The r e s u l t s of the experiments are analysed and an equ i l i b r i u m model i s developed to predict the equilibrium concentrations of COS, H 2S and C0 2. These r e s u l t s could be used i n the development of a k i n e t i c model f o r DEA degradation by COS, as well as i n the design and modelling of COS absorption processes u t i l i z i n g DEA. 205 206 7.1 THEORY The attainment of equilibrium i n a system c o n s i s t i n g i n i t i a l l y of gaseous COS and an aqueous DEA solution may be regarded as involving the following steps: ( i ) Physical absorption of COS into water, ( i i ) Reaction between DEA and absorbed COS to form DEA thiocarbamate, ( i i i ) Hydrolysis of absorbed COS and DEA thiocarbamate to y i e l d CO2 and H 2S, (iv) R e d i s t r i b u t i o n of COS, C0 2, H 2S and t h e i r associated compounds between the l i q u i d and gas phases. At low temperatures, steps ( i ) and ( i i ) are very much f a s t e r than ( i i i ) and i t i s therefore possible to measure absorption without s i g n i f i c a n t hydrolysis during the early stages of contact between COS and aqueous DEA solutions. By contrast, at elevated temperatures, steps ( i i i ) and (iv) are rapid and d i s t i n c t i o n between absorption and hydrolysis i s not e a s i l y achieved experimentally. 1. Absorption Regime (Steps i and i i ) The physical absorption of gas i into water may be described by Henry 1 s law, i . e . H i = p i \/ c i 7 - 2 where P^ and c^ denote the p a r t i a l pressure and l i q u i d phase concentration of p h y s i c a l l y absorbed species i , r e s p e c t i v e l y . When component i reacts to form i o n i c complexes, such as DEA thiocarbamate, 207 ^COS R2NCOS\" H + = R2NH + COS 7 . 3 then Henry' s Law may be written as Hi* = P j \/ C i * 7 . 4 * where Cj_ now denotes the t o t a l ( i . e . p h y s i c a l l y absorbed and chemically bound) concentration of component i i n solution. The temperature v a r i a t i o n of the Henry's constants may be represented by the Arrhenius expression: In {HL*) = Ai + Bi\/T 7 . 5 where the constants A^ and can be obtained from l e a s t square f i t s of the semilogarithmic plots of vs 1\/T. 2. Hydrolysis Regime (Steps i i i and iv) The DEA thiocarbamate formation i s a rapid reaction and the o v e r a l l COS hydrolysis i s therefore governed, i n e f f e c t , by the hydroly s i s of DEA thiocarbamate: R2NCOS\" H + + H 2 0 = R2NH + C 0 2 + H 2S 7 . 6 208 The newly formed CO2 and p a r t i c i p a t e i n the well known reactions for aqueous DEA solutions (36-41): K l R 2NH 2 + = H + + R2NH 7.7 K2 R2NCOO\" + H20 = R2NH + HCO3\" 7.8 K3 + H20 + C0 2 = H + + HCO3 7.9 K 4 \u2022 H20 = H + + OH 7.10 - K 5 + HCO3 = H + + CO3 7.11 K6 + H2S = H + + HS 7.12 K 7 + HS = H + + S 7.13 Reactions 7.3, 7.7 to 7.13 can be described by a model of the type f i r s t proposed by Kent and Eisenberg ( 3 9 ) : K-L = [H +] [R 2NH] \/ [ R 2 N H 2 + ] 7.14 K 2 = [R 2NH] [HCO3\"] \/ [R 2NCOO\"] 7.15 209 and K 3 = [H + ] [HC03\"] \/ tco 2] K 4 = [H + ] [OH\"] K5 = (H + ] [CO3\"] \/ [HCO3\"] K6 = [H+] [HS\"] \/ [H2S] K 7 = [H + ] [S\"\"] \/ [HS\"] Kcos = [R2NH ] [COS] \/ [R2NCOS'] [H + HC02 = = PC02 \/ [co 2] HH2S = = PH2S \/ [H2S] 7.16 7.17 7.18 7.19 7.20 7.21 7.22 7.23 HCOS = PCOS \/ [ C 0 S ] 7 - 2 4 where [i] denotes the concentration of component i i n solu t i o n . As was the case f o r the Kent-Eisenberg model, Eqs. 7.14 to 7.21 assume an excess of water, unit a c t i v i t y c o e f f i c i e n t s and incorporate system non-i d e a l i t i e s i n t o the equilibrium constants governing the carbamate, thiocarbamate and amine protonation reactions. In addition to the mass action equations and Henry1 s law, the following charge and material balances apply: Charge: [R 2NH 2 + ] + [H + ] = [HCO3\"] + [R2NCOO\"] + 2 [CO3\"] + [OH'] + [HS\"] + 2[S\"\"] + [R2NCOS\"] 7.25 DEA: m = [R2NH] + [R2NCOO\"] + [R2NCOS\"] + [R 2NH 2 +] 7.26 H2S: m 0 ^ 2 2 = [H2S] + [HS\"] + [S\"\"] 7.27 2 1 0 C O o : m OQQ2 =. [C02] + [HC03\"] + [R2NC00\"] + [CO3\"\"] 7.28 C O S ; m o Q^g = [COS] + [R2NCOS\"] 7.29 where m and a.^ represent the t o t a l DEA concentration and the DEA loading by a c i d gas i , r e s p e c t i v e l y . The above 16 equations contain 20 unknown operating variables (14 concentrations: m, [H +], [R2NH], [R 2NH +], [HCO3\"], [R2NCOO\"], [C02], [OH\"], [C03\"\"], [HS\"], [H2S] , [S\"\"], [R2NCOS\"], [COS]; 3 p a r t i a l pressures: PQQ2' PH2S' pCOS ; ^ loadings: O^.Q2' o^s' \u00b0COS' ' T n e s Y s t e m i s therefore f u l l y s p e c i f i e d provided the equilibrium and Henry1 s constants are known and four operating variables are given. For most design c a l c u l a t i o n s the l a t t e r are the three p a r t i a l pressures (pc02' I?H2S' P^os * a n c * t^ i e t o t a-'- D E A concentration (m). Equations 7.14 to 7.29 were combined to y i e l d four independent model equations: PH2S = (HH2S 1 K 6 K 7 ) ( A [ H + ] 2 1 ( 1 + [ H + ] 1 K 7 n 7.30 7.31 211 [H+] P C O S where: A B C K' A (1 + K ? \/