and H+ in Aqueous Diamine Solutions - ACS Publications - American

Subscriber access provided by READING UNIV

Article

Insights Into the Chemical Mechanism for CO2(aq) and H+ in Aqueous Di-amine Solutions - An Experimental Stopped-flow Kinetic and 1H/13C NMR Study of Aqueous solutions of N,Ndimethylethylenediamine (DMEDA) for Post Combustion CO2 Capture Bing Yu, Lichun Li, Hai Yu, Marcel Maeder, Graeme Puxty, Qi Yang, Paul H. M. Feron, William Owen Conway, and Zuliang Chen Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.7b05226 • Publication Date (Web): 07 Dec 2017 Downloaded from http://pubs.acs.org on December 13, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Environmental Science & Technology is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 31

Environmental Science & Technology

1

Insights Into the Chemical Mechanism for CO2(aq) and H+ in Aqueous Di-

2

amine Solutions - An Experimental Stopped-flow Kinetic and 1H/13C NMR

3

Study of Aqueous solutions of N,N-dimethylethylenediamine (DMEDA) for

4

Post Combustion CO2 Capture

5

Bing Yu1,2, Lichun Li2,3, Hai Yu2, Marcel Maeder3, Graeme Puxty2, Qi Yang4, Paul Feron2,

6

William Conway2,* and Zuliang Chen1,*

7

(1)

8

Global Centre for Environmental Remediation, University of Newcastle, Callaghan NSW 2308, Australia.

9

(2)

CSIRO Energy, Mayfield West NSW 2304, Australia.

10

(3)

Discipline of Chemistry, University of Newcastle, Callaghan NSW 2308, Australia.

11

(4)

CSIRO Manufacturing, Clayton VIC 3169, Australia.

12 13

* Corresponding authors

14

Dr William Conway, [email protected]

15

Dr Zuliang Chen, [email protected]

16

1 ACS Paragon Plus Environment


17

Page 2 of 31

TOC Graphic

18 19 20

Keywords Coal fired power station, global warming, sustainability, environmental chemistry,

21

chemical processes, engineering.

22

Abstract

23

In an effort to advance the understanding of multi-amine based CO2 capture process

24

absorbents we report here the determination of the kinetic and equilibrium constants for a

25

simple

26

spectrophotometric and 1H/13C NMR titrations at 25.0oC. From the kinetic data the formation

27

of mono-carbamic acid (DMEDACOOH) from the reaction of DMEDA with CO2(aq) is the

28

dominant reaction at high pH > 9.0 (k7 = 6.99 × 103 M-1.s-1). Below this pH, the formation of

29

protonated mono-carbamic acid (DMEDACOOH2) via the pathway involving DMEDAH+

30

and CO2(aq) becomes active and contributes to the kinetics despite the 107 fold decrease in

31

the rate constant between the two pathways. 1H and

32

decreasing pH (increasing HCl concentration) at 25.0 oC have been evaluated here to confirm

33

the protonation events in DMEDA. Calculations of the respective DMEDA nitrogen partial

34

charges have also been undertaken to support the NMR protonation study. A comparison of

35

the DMEDA kinetic constants with the corresponding data for piperazine (PZ) reveals that

linear

di-amine

N,N-dimethylethylenediamine

13

(DMEDA)

via

stopped-flow

C NMR spectra as a function of


Page 3 of 31


36

despite the larger basicity of DMEDA, the enhanced and superior kinetic performance of PZ

37

with CO2(aq) above its predicted Bronsted reactivity, is not observed in DMEDA.

38

Introduction

39

Rising carbon dioxide (CO2) emissions since the beginning of the industrial revolution

40

are eliciting an enormous environmental challenge, climate change.1 Around 37% of

41

anthropogenic CO2 emissions arise from fossil-fuel-fired power plants for the generation of

42

electricity, and the capture and sequestration of CO2 emission from these power plants has

43

been suggested as an important strategy to ameliorate global climate change.2 Up to now,

44

chemical absorption using amine-based absorbents in a process referred as post combustion

45

capture (PCC) has been considered among the most mature technologies for CO2 capture

46

from coal-fired power plants.3, 4

47

For sterically unhindered primary and secondary amine absorbents, their reactions with

48

CO2 can be explained by the carbamic acid mechanism, which suggests that the amine is

49

firstly reacting with CO2 to form carbamic acid [Eqs. (1)], which instantaneously

50

deprotonates transferring a proton to another amine [Eqs. (2)].5 In contrast, tertiary amines

51

and sterically hindered amines only act as proton accepting bases from the formation of

52

carbonic acid and the subsequent formation of bicarbonates and carbonates [Eq. (3)]:6

53

CO2+RNH2 ↔ RNH2CO2H

(1)

54

RNH2CO2H+ RNH2 ↔ RNHCOO-+RNH3+

(2)

55

R3N+H2O+CO2 ↔ R3NH++HCO3-

(3)

56

Primary and secondary amines are characteristically utilised for CO2 absorption due to

57

their fast CO2 absorption kinetics but they suffer from limited CO2 absorption capacities (up

58

to 0.5 mole CO2/mole amine). On the contrary, tertiary and sterically hindered amines are

59

slow reacting but possess, high absorption capacities (up to 1 mole CO2/mole amine) and 3 ACS Paragon Plus Environment


Page 4 of 31

60

lower heat requirements for CO2 regeneration.7 For amine scrubbing technologies, the

61

balance between acceptable CO2 absorption rates and low regeneration energy performance

62

has hampered its industrial implementation due to the correlation of these properties with the

63

capital and operating costs of the process. Therefore, developing an ideal amine based solvent

64

with high CO2 absorption rate and capacity, as well as lower energy requirement for

65

regeneration, is regarded as a significant contribution to the advancement of PCC.8

66

To date, increasing attention has been concentrated on the development of blended

67

amine absorbents which has the potential to combine the advantages of several amines. For

68

example, Luo et al. demonstrated that a blended amine solution containing MEA and N, N-

69

diethylethanolamine has a much higher CO2 absorption rate, lower regeneration energy, and

70

improved cyclic capacity compared to that of conventional amine absorbents.7 Muchan and

71

co-workers conducted a screening test for a series of blended amine solutions and found

72

that all the solvent combinations showed better performance than 5.0 M MEA.9 More

73

recently, Nwaoha et al. investigated a tri–solvent blend (2–Amino–2–Methyl–1–

74

Propanol/PZ/MEA), which exhibited significantly lower heat duties than the standard

75

5.0M MEA.10 Narku-Tetteh et al recently developed a set of criteria to guide the selection of

76

components to improve the performance of an amine blend. Among the series of properties

77

evaluated the effect of steric hindrance, number of hydroxyl groups, and chemical properties

78

were considered to be important characteristic of amines and in their blends in terms of CO2

79

absorption rates, CO2 absorption heat, and desorption.11 According to the mechanism for the

80

enhancement of CO2 absorption in blended amine systems proposed by Kim et al, the

81

additional amine molecules in blends can act to lower the activation energy of the reaction

82

between CO2 and amine by enhancement of intermolecular interaction and withdrawing of a

83

proton from a reacting amino group.12 However, this intermolecular proton transfer process

84

also produces a strong and dense hydrogen-bonded network, resulting in increasing viscosity 4 ACS Paragon Plus Environment

Page 5 of 31


85

upon CO2 uptake.13 While blending amine absorbents can overcome limitations in terms of

86

the chemical behaviour, blended amine systems typically have the volatility, toxicity and

87

stability of the most limited component in the formulation.14 More pertinently, the

88

formulation of absorbents is non-ideal and may even combine unfavourable characteristics of

89

different solvents while suppressing their favourable characteristics.

90

Recently, incorporation of two or more amino groups into a single molecule, termed di-

91

amines or multi-amines, has been demonstrated as a simple methodology for improving CO2

92

absorption and stripping.15-17 Re-visiting Eq. (1) and (2), the stoichiometric absorption of CO2

93

for primary/secondary amines is limited (0.5 mole CO2/mole amine), relating to the generated

94

carbamic acid which further reacts and consumes a second amine molecule in a simple proton

95

transfer reaction.18 In theory, by introducing additional amine group(s) to a monoamine, thus

96

forming a di-amine or multi-amine, following formation of the carbamic acid the proton (on

97

the carbamic acid) can transfer directly and intra-molecularly to the second amine group

98

within the same diamine molecule. As a result, this intra-molecular proton transfer reaction

99

contributes to enhanced absorption of CO2 by increasing the amount of reactive amine groups

100

available while minimising the overall mass of amine required to achieve CO2 absorption.

101

Furthermore, reduction in the overall mass of amine required compared to similar nominal

102

amine concentrations in monoamines, can potentially result in reduced viscosities with

103

increasing CO2 loadings.19 Overall lower requirements for material circulation and heating in

104

the stripper will lead to reduction of the sensible heat and overall pumping work.16 The most

105

notable di-amine is the cyclic piperazine which incorporates two secondary amine groups into

106

a six membered ring structure. Concentrated PZ has been suggested as a replacement for the

107

benchmark MEA process to which emerging absorbents are compared due to advancements

108

in the absorbent system including high temperature stripping up to 150.0oC, resistance to

109

thermal degradation at elevated temperatures, rapid CO2 absorption rates, and increased 5 ACS Paragon Plus Environment


21

Page 6 of 31

110

resistance to oxidative degradation.20,

111

characteristics of the cyclic amine 1-(2-aminoethyl)piperazine (AEP) containing a primary,

112

secondary, and tertiary amine group. It was found that AEP absorbed CO2 at similar rates to

113

triethanolamine (TEA) at similar temperatures and AEP -monocarbamate was the dominant

114

reaction product formed between AEP and CO2.22 Ongoing attempts to understand the

115

superior performance and characteristics of PZ and these multi-amines are highly valuable to

116

the advancement of general PCC knowledge. This type of work can be further supported by

117

modelling studies such as Machida et al who developed a correlation between the pKa of

118

amines with the CO2 absorption capacity.23

Choi et al also investigated the absorption

119

Inspired by the potential for enhanced CO2 absorption performance of blended primary

120

and tertiary amines absorbents,7, 24 and di-amine absorbents, a scenario involving a single

121

diamine molecule containing one primary and one tertiary amino group is attractive. Herein,

122

we define this class of diamines by the notation ‘N1RN3’. A number of primary and tertiary

123

amines have been demonstrated to be stable against degradation in the presence of oxygen at

124

elevated temperatures.25 In our previous studies, a designer diamine, 1-(2-hydroxyethyl)-4-

125

aminopiperidine was found to have superior characteristics for CO2 absorption while a

126

similar absorption enthalpy

127

intramolecular tertiary amino group of 3-(diethylamino)propylamine (DEAPA) can promote

128

the primary amino group to absorb CO2. In particular, DEAPA has a heat of absorption lower

129

than that of MEA, DEA, and MDEA.15 Despite the equilibrium and thermodynamics

130

involving combined primary/tertiary amine absorbents with CO2 having been extensively

131

studied, fundamental kinetic information governing the pathways for CO2 in di-amine

132

solutions is relatively scarce and largely remains unclear. This critical information is essential

133

for the optimization of PCC processes through fundamental chemical and advanced process

134

modelling.26 Additionally, in order to further advance the development the di-amine

to MEA.16,

17

Zhang et al. also demonstrated that the


Page 7 of 31


135

absorbents for PCC and design of higher efficiency amine processes, thorough investigation

136

of the reaction mechanism is of acute importance. Due to the chemical complexity of di-

137

amine systems, the series of chemical reactions is more complicated than that of monoamines

138

and includes additional species such as the mono-protonated diamine, di-protonated diamine

139

and mono-carbamate which can be generated during CO2 absorption.27

140

In this study, N,N-dimethylethylenediamine (DMEDA) with the structure of NH2-

141

(CH2)2-N(CH3)2, was selected as a simple representative of a combined primary and tertiary

142

amine absorbent. Firstly, 1H and

143

spectrophotometry was used to quantitatively investigate the reversible reactions of CO2(aq)

144

with DMEDA solutions at 25.0 oC. Analysis of the resulting kinetic measurement data was

145

accomplished using a comprehensive mechanism including pathways for carbamic acid

146

formation via the free amine at high pH (> 7.0) as well as the pathway via the protonated

147

amine at low pH (99%, Sigma-Aldrich), hydrochloric acid (Sigma

187

Aldrich), thymol blue sodium salt (Sigma-Aldrich), Alizarin Red S sodium salt (BDH), and

188

methyl orange sodium salt (SELBY) were all used as obtained without further purification.

189

Ultra-high-purity Milli-Q water was boiled to remove dissolved CO2 and was used to prepare

190

all solutions for the stopped-flow kinetic and 1H/13C NMR Measurements.

191

Stopped-flow spectrophotometric measurements

192

Kinetic measurements were performed using an Applied Photophysics DX-17

193

spectrophotometer equipped with a J&M Tidas MCS 500-3 diode-array detector. Absorption

194

changes of coloured acid base indicators was used to monitor the kinetic reactions via

195

solution pH changes in the absence of useful individual absorption spectra for the reaction 9 ACS Paragon Plus Environment


Page 10 of 31

196

species. A wavelength range from 400 - 700nm in 2.0nm increments was followed here and

197

the entire spectrum used in the kinetic analysis. The reaction temperature was maintained at

198

25.0oC using a circulating Julabo F20 water bath monitored by a thermocouple located within

199

the stopped-flow cell. The full setup and procedure has been described in detail in our

200

previous work 26.

201

Carbamic acid formation

202

A series of reactions over a range of pH conditions involving the free (un-protonated)

203

and partially protonated DMEDA (DMEDAH+) was performed in this work. Generally, the

204

reactions were performed by mixing a DMEDA solution (also containing amounts of HCl for

205

the partially protonated solutions) with a CO2(aq) solution in a 1:1 ratio in the stopped-flow.

206

CO2(aq) solutions were prepared by bubbling a gas stream containing CO2 and N2 into a

207

temperature controlled reservoir above the stopped flow. The composition of the solution was

208

varied by adjusting the flow rate of the respective gas mass flow controllers. The

209

concentration of CO2(aq) was determined from the CO2/N2 ratio and the published saturation

210

constants for CO2 in water.A new measurement was performed for a series of DMEDA

211

concentrations according to the following initial concentrations of the solutions: [DMEDA]0

212

= 1.0 – 10.0mM, [CO2]0 = 4.35mM, [Thymol blue]0 = 12.5uM. The initial concentrations of

213

the partially protonated amine solutions were as follows: [DMEDA]0 = 4.0mM, [HCl]0 = 1.0

214

– 3.0mM, [CO2]0 = 4.35mM, [thymol blue]0 = 12.5uM.


Page 11 of 31

215


Carbamic acid/carbamate decomposition

216

The reversibility of the DMEDA reactions was studied during the decomposition of an

217

equilibrated solution of DMEDA carbamate following the addition of hydrochloric acid. An

218

aqueous solution of carbamate containing 0.15 M HCO3– and 0.05 M DMEDA was prepared

219

and equilibrated for ~ 24 hours at 25.0 °C. The concentration of DMEDA carbamate in the

220

solution was determined by quantitative 1H NMR ([DMEDACO2-]0 = 21.5 mM). The

221

carbamate solution was used directly in the stopped-flow and reacted with equal volumes of a

222

range of HCl solutions ([H+]0 = 160.0 – 200.0 mM) also containing 0.025mM methyl orange

223

and 0.05mM alizarin Red S indicators.

224

Data analysis

225

Analysis of the stopped-flow data using a chemical model including equations (4) – (13)

226

was performed using Reactlab kinetics (www.jplusconsulting.com) and in house extensions

227

of the program to incorporate species charges and activity co-efficient corrections. Global

228

analysis technique was employed to the entire series of kinetic data incorporating the

229

reactions involving carbamic acid formation from the free and partially protonated DMEDA

230

solutions, as well as the carbamic acid decomposition reactions at low pH ( 9.0) of the solutions here, it could be

300

postulated that the formation of the carbamic acid is the dominant reaction pathway.

301

Beginning with evaluation of this assumption, the simplest iteration of the chemical model

302

involving the addition of the reversible formation reactions of CO2(aq) with the free amine to

303

form the singly protonated carbamic acid, k7, k-7, equation (10), and the single protonation of

304

the carbamate to carbamic acid, K12, equation (12), resulted in poor agreement between the

29

. Initial



Page 16 of 31

305

calculated and measured traces. Visually, the calculated kinetic trace was similar in shape to

306

the measurement data but severely under-predicts the initial kinetic region of the

307

measurement for the reactions of CO2(aq) with the free amine. This observation strongly

308

indicates the initial iteration of the mechanism was deficient in a kinetic pathway for CO2.

309

However, agreement between the calculated and measured traces was noticeably better for

310

the measurements involving higher concentrations of DMEDA where [DMEDA]0 >

311

[CO2(aq)]0 indicating such a simple mechanism could potentially be used to determine kinetic

312

constants for the formation of carbamic acid at high amine concentrations. Surprisingly, the

313

values for the kinetic constant k7 during this initial model evaluation were similar to the final

314

values determined during the comprehensive model validation fitting. However, due to the

315

increased buffer activity of the amine at such conditions (higher concentrations) the

316

corresponding absorbance changes, and thus pH during the reactions, is reduced and may

317

begin to limit the sensitivity of the measurements. While the above might appear useful for an

318

initial and rapid evaluation of the kinetic constant k7 the validity of the resulting constants are

319

bound by conditions where the formation of the carbamic acid via reaction of CO2(aq) and

320

the free amine is the main contributor to the kinetics at high pH (>8.0). Given the range of pH

321

conditions encountered during CO2 capture processes, often down to pH 8 the above

322

assumptions are not sufficient to interpret data outside of this kinetic region. Predictably,

323

kinetic data for the reaction of CO2(aq) with partially protonated DMEDA were also poorly

324

fitted using this simplified chemical model due to the absence of chemical pathways for

325

CO2(aq) in the model at intermediate pH. Similarly, kinetic data for decomposition of

326

DMEDA carbamate/carbamic acid at low pH (< 6.5) were also poorly fitted despite the

327

inclusion of the decomposition pathway k-7, in the simple model. The main reason for the

328

poor fit of the low pH measurements relates to the equilibrium concentrations and omitted


Page 17 of 31


329

protonation reactions which are required to balance the total concentrations and solution pH

330

during the fitting.

331

Expansion of the initial chemical model to include an additional pathway for the

332

reversible formation of the doubly protonated carbamic acid via the reaction of CO2(aq) with

333

protonated DMEDAH+, k8, k-8, equation (11), and the second protonation of the carbamic

334

acid to form the doubly protonated carbamic acid, K10, equation (13), was evaluated. From

335

the final fitting results using this comprehensive mechanism presented in Figure 2, agreement

336

between the measured and calculated data for the three series of reactions investigated here is

337

excellent. Unlike the initial simplified model, no significant deviations of the calculated data

338

from the measured data were observed, ultimately confirming the validity of the

339

comprehensive chemical mechanism. Importantly, the comprehensive mechanism and the

340

corresponding kinetic and equilibrium constants are now applicable to all pH and reaction

341

conditions.

342

Following establishment of the chemical mechanism the specific kinetic behaviour of

343

DMEDA can begin to be interpreted and compared to other di-amines for which similar

344

kinetic data is available. Initial insight can be gleaned from the rate and equilibrium constants

345

in Table 1. Corresponding values for PZ have been included in the table for comparison.

346

Considering the rate constants alone the reactivity of free DMEDA toward CO2, k7, is

347

significantly larger than that of DMEDAH+, k8 , with values of 6.99 × 103 and 65.0 M-1s-1,

348

respectively. This agrees with the corresponding rate constants for PZ which also

349

demonstrate a much stronger reactivity of free PZ with CO2 compared to the singly

350

protonated PZH+. Repulsion effects induced by the positive charge on the protonated amine

351

group and reduced Lewis basicity of the free amine group are simple explanations for the

352

reduced kinetic reactivity of the protonated DMEDAH+ with CO2.27

353

equilibrium constants for the free and protonated amine pathways with CO2, K7 and K8

Similarly, the



Page 18 of 31

354

respectively, follow a similar trend. Considering the carbamic acid decomposition pathways,

355

DMEDA carbamate/carbamic acid(s) can decompose via the k–7 and k-8 pathways or a

356

combination of the two. On the basis of the rate constants the latter k-8 pathway is kinetically

357

preferred. However, it should be noted the decomposition reactions are influenced by the

358

protonation constants for the singly and doubly protonated carbamates/carbamic acids, K9 and

359

K10 respectively, and strongly dependent on the solution pH.

360

In addition to the rate constants, concentrations profiles for the species during the

361

reactions generated as part of the fitting procedure together with calculated pH profiles for

362

the reactions can aid in the interpretation of the kinetic behaviour.

363

concentration profiles for the reactions of 4.35mM CO2(aq) with 4.0mM and 10.0 DMEDA

364

together with calculated pH curves are shown in Figure 3(a) and (b), respectively.

365

Corresponding profiles for the reaction of 4.35mM CO2(aq) with 4.0mM DMEDA/1.0mM H+

366

and during the decomposition of a DMEDA carbamate solution in the presence of 100.0 mM

367

HCl are shown in Figure 3(c) and (d) respectively. (b)

(a) 5.0E-03

11

11

1.0E-02

pH

DMEDA

CO2

3.0E-03

9 DMEDA DMEDACO2H

2.0E-03

8

DMEDACO2-

1.0E-03

HCO3-

DMEDAH+

8.0E-03 pH

10.5

6.0E-03

pH

10

Concentration (M)

4.0E-03

pH

Concentration (M)

Representative

CO2

4.0E-03

DMEDACO2-

7

2.0E-03

6

0.0E+00 0.00

DMEDACO2H

OH-

0.0E+00 0.00

0.02

0.20

2.00

OH-

9.5

0.02

Time (s)

368

10

DMEDAH+

0.20

Time (s) (d)

(c) 5.0E-03

6.0E-02

10

6

CO2

9

8

DMEDAH+

2.0E-03 DMEDA

1.0E-03

5 pH

H2CO3

4 DMEDACO2H2+

2.0E-02

DMEDAH22+

3

7

DMEDACO2H HCO3-

OH-

pH

3.0E-03

CO2

4.0E-02

HCO3-

DMEDAH22+

DMEDACO2H

DMEDACO2-

0.0E+00 0.00

369

6 0.02

pH

4.0E-03

Concentration (M)

Concentration (M)

pH

0.20

0.0E+00 0.00

Time (s)

2 0.02

0.20

Time (s)


Page 19 of 31


370

Figure 3. Species concentration profiles: (a) and (b) reaction of 4.35mM CO2(aq) with

371

4.0mM and 10.0mM DMEDA respectively (c) reaction of 4.35mM CO2(aq) with 4.0mM

372

DMEDA in the presence of 1.0mM HCl (d) decomposition of DMEDA carbamates/carbamic

373

acids at low pH (

and H+ in Aqueous Diamine Solutions - ACS Publications - American

Recommend Documents