DC

SD

a

ARRA

KPMCDBN

1

(gtCo2aw2mdMta2e

h1

International Journal of Greenhouse Gas Control 31 (2014) 214–228

Contents lists available at ScienceDirect

International Journal of Greenhouse Gas Control

j ourna l ho me page: www.elsev ier .com/ locate / i jggc

egradation study of piperazine, its blends and structural analogs forO2 capture: A review

haukat A. Mazari, Brahim Si Ali ∗, Badrul M. Jan, Idris Mohamed Saeedepartment of Chemical Engineering, Faculty of Engineering, University of Malaya, 50603 Kuala Lumpur, Malaysia

r t i c l e i n f o

rticle history:eceived 23 December 2013eceived in revised form 8 October 2014ccepted 10 October 2014

eywords:iperazineonoethanolamine

O2 absorptionegradation

a b s t r a c t

Postcombustion carbon dioxide (CO2) capture using amine scrubbing is an emerging technology tomitigate CO2 emissions. Benchmark amines used in this technology are monoethanolamine (MEA),methyldiethanolamine (MDEA), and diethanolamine (DEA). Amines undergo irreversible reactions,resulting degradation of the solvent under process operating conditions. Degradation of solvent may cre-ate environmental concerns, increased costs as well as corrosion in the process equipment. Piperazine(PZ) has been investigated as a novel amine solvent for CO2 capture. It has high CO2 capture capacity andabsorption rate with thermal and oxidative degradation resistance. This study discusses the thermal andoxidative degradation of PZ under process operating conditions. A critical review of various parameters,such as the effect of temperature, amine concentration, CO2 loading (˛), partial pressure of oxygen (O2)

lendsitrosamine

and presence of metal ions on rate of degradation has been reported. Chemistry of degradation productsand mechanisms of degradation of PZ are reported for better understanding of PZ degradation kinetics.In addition, degradation of PZ blends, structural analogs, and diamines is focused, and behavior of theirdegradation is highlighted. Furthermore, kinetics of formation and degradation of nitrosamines throughPZ is also discussed.

© 2014 Elsevier Ltd. All rights reserved.

. Introduction

Amines usage for gas sweetening started in early 1930sBottoms, 1933). During last few decades, CO2 emissions increasedlobally which have enhanced the development of postcombus-ion CO2 capture (PCC) technology (Rochelle, 2009). Amine-basedO2 capture using amine scrubbing technology is known as onef the promising technologies for CO2 capture (MacDowell et al.,010; Mangalapally et al., 2009). Expected energy required forn advanced amine systems for CO2 capture is about 220 kilo-att hour per ton (kWh/ton) CO2 removed (Rochelle et al.,

011). The most conventional amines used in this technology areonoethanolamine (MEA), methyldiethanolamine (MDEA), and

iethanolamine (DEA) (Dumée et al., 2012; Lowe et al., 2009;acDowell et al., 2010). These amines undergo irreversible reac-

ions due to reactions with acidic gases (CO2, H2S, COS, CO, SOx,nd NOx), elevation of temperature, and oxidation (Freeman et al.,

010a; Goff and Rochelle, 2006; Islam et al., 2010, 2011; Lepaumiert al., 2009; Wang et al., 2011). Estimations show that the solvent

∗ Corresponding authors. Tel.: +60 3 7967 6896; Fax: +60 3 7967 5319.E-mail address: brahim@um.edu.my (B.S. Ali).

ttp://dx.doi.org/10.1016/j.ijggc.2014.10.003750-5836/© 2014 Elsevier Ltd. All rights reserved.

(MEA) make up accounts for around 10% of the total CO2 capturecost (Rao and Rubin, 2002).

Efficient use of the technology primarily depends on the selec-tion of viable solvent. The suitable solvent should have high CO2capture capacity, high CO2 absorption rate, low thermal and oxida-tive degradation rate, and low volatility with appropriate physicalcharacteristics (Boot-Handford et al., 2014; Freeman and Rochelle,2012b). A number of new amines and their blends have been exam-ined under different conditions to recognize a stable solvent systemwith promising CO2 capture properties (Hilliard, 2008; Idem et al.,2005; Namjoshi et al., 2013). PZ and its blends have been foundas effective and stable amine solvents as that of alkanolaminesand others (Dugas and Rochelle, 2009; Freeman, 2011; Freemanand Rochelle, 2011, 2012b; Li et al., 2013b; Namjoshi et al., 2013;Rochelle et al., 2011).

Previously, PZ has been used as a CO2 capture promoter inblends with different amines in small concentrations (Bishnoi andRochelle, 2002; Cullinane and Rochelle, 2006; Zhang et al., 2001).However, at present it is established that PZ alone can be usedas an alternative to MEA due its favorable CO2 capture proper-

ties and thermal and oxidative degradation resistance (Freemanet al., 2009, 2010a,b). PZ has exhibited thermal stability up to 150 ◦Cand oxidation stability in the presence of O2, CO2, and metal ions,like Fe2+, Cr3+, and Cu2+ (Alawode, 2005; Li et al., 2013b; Freeman,

of Gree

2aa22

re8clapCpfF2bM2

mttniboatTGhkitar

2

CbeotccluptR

tcioeiHaC

S.A. Mazari et al. / International Journal

011; Freeman et al., 2009). In addition, PZ has also possess favor-ble thermodynamic properties, such as CO2 solubility, volatility,nd heat capacity (Chen and Rochelle, 2011; Dugas and Rochelle,009; Ermatchkov and Maurer, 2011; Freeman et al., 2009; Li et al.,013b).

8 molal (m) PZ (40 wt.% PZ) has exhibited comparatively betteresults than 7 m MEA (30 wt.% MEA) for CO2 capture (Freemant al., 2009; Li et al., 2013b). Under process operating conditions,

m PZ has twice the absorption ratio of CO2 and higher CO2apture capacity than 7 m MEA (Li et al., 2013b). However, PZ hasimited solubility due to which precipitation may occur, eithert high or low CO2 loading (�) (Li et al., 2013b). It may be usedroperly if the CO2 loading is kept in a range of 0.26–0.42 molO2/mol alkalinity above 20 ◦C (Li et al., 2013b). The problem ofrecipitation may also be solved through using PZ in the blendedorms with other amines (Closmann et al., 2009; Closmann, 2011;reeman et al., 2013). PZ has been investigated in blends with-amino-2-methyl-1-propanol (AMP), 1,4-diaminobutane (DAB),is(aminoethyl)ether (BAE), aminoethylpiperazine (AEP), MDEA,EA, DEA, etc. (Closmann, 2011; Davis, 2009; Namjoshi et al.,

013; Wang and Jens, 2012, 2014).This paper reviews the recent advances and findings on ther-

al and oxidative degradation of PZ. Degradation products found,o date have been highlighted and some of the pathways ofheir formation are also discussed. Due to the raising concerns ofitrosamine through amine-based CO2 capture technology, kinet-

cs of their formation and decomposition is also reported. Effect oflending PZ with other amines is analyzed and reported in termsf degradation kinetics. Diamines and structural analogs of PZ arelso discussed to show the appropriateness of PZ family. In addi-ion, some of the suggestions are also made, based on conclusions.here are some other review articles and reports (Thong et al., 2012;ouedard et al., 2012; Fredriksen and Jens, 2013) which seem toave some relevance to the degradation of PZ. However, a vastnowledge has not yet been reviewed in those articles and reports,n form of recent knowledge, pilot plant advancements, degrada-ion kinetics of blends of PZ and structural analogs and formationnd decomposition of nitrosamines of PZ. This information has beeneviewed in this article.

. Degradation of PZ and its products

The main areas of degradation for any amine in postcombustionO2 capture process are absorber sump, cross heat exchanger, re-oiler, and re-claimer (Freeman, 2011). Degradation mainly occursither in form of thermal or oxidative degradation. The formerccurs due to the high temperature of the process, while the lat-er occurs because of the presence of dissolved oxygen and freeations. Freeman (2011) studied the effect of temperature, amineoncentration, and CO2 loading on thermal degradation of PZ. PZoss and degradation products were investigated at temperaturesp to 175 ◦C with CO2 loading ranged from 0 to 0.47 moles of CO2er mole alkalinity (mol CO2/mol alkalinity) and the molal concen-ration of PZ ranged from 4 to 20 m (Freeman, 2011; Freeman andochelle, 2012b).

Freeman (2011) and Freeman and Rochelle (2012b) conductedhermal degradation experiments of CO2 loaded PZ in thermalylinders for a maximum duration of 72 weeks. Oxidation exper-ments have been conducted in original oxidation and Teflonxidation reactors at 55 ◦C for 5 weeks (Freeman, 2011; Freemant al., 2010a). The studies reported that no significant degradation

s observed without CO2 loading but noticed with CO2 loading.owever, the effect of CO2 loading is expected to be complex,s degradation seems to increase up to the loading of 0.4 molO2/mol alkalinity, but decreases above 0.4 mol CO2/mol alkalinity

nhouse Gas Control 31 (2014) 214–228 215

(Freeman, 2011). However, the minimum degradation at 175 ◦Cfor rich PZ has been observed with 0.47 mol CO2/mol alkalinity(Freeman and Rochelle, 2012a).

The abundant thermal degradation products of 8 m PZ are N-formyl PZ (FPZ), ammonium (NH4

+), and AEP which accounted fortotal of 57% of nitrogen and 45% of carbon loss. Table 1 illustratesthe degradation products formed through thermal degradation ofPZ.

Identified oxidation products of PZ include EDA, carboxylateions, amides, and ammonium (NH4

+). Occasionally, formation ofsome minor products like glycolate, nitrite, and nitrate has alsobeen reported.

Furthermore, unidentified liquid phase degradation prod-ucts were also found in highly oxidized solutions (Freeman,2011). Unidentified products were classified as monoamines,polyamines, and amides but were significantly in smaller quan-tities to that of primary degradation products. It was assumedthat aldehydes (e.g. formaldehyde, acetaldehyde, and hydrox-yacetaldehyde), alcohols (e.g. methanol, ethanol, and ethyleneglycol), ureas of PZ, amino acids (e.g. N(2-aminoethyl)glycine, N,N-1,2-ethanediylbis-glycine), and oxidized PZ molecules (2-pipera-zinone, 2,5-piperazinedione, 2-piperazinol, 1,2,3,6-tetrahydropyrazine, 1,2,3,4-tetrahydropyrazine, 1-piperazineaceticacid) maybe the unidentified products based on total organic carbon (TOC)mass balance. Some identified oxidation products are listed inTable 2.

Nielsen et al. (2013) studied degradation of PZ in two pilotplants: Separation Research Program (SRP) and Pilot Plant 2 (PP2).The former used highly oxygenated synthetic flue gas containingambient level NOx, while the latter used a slipstream of real fluegas taken from a coal-fired power plant. All the experiments wereconducted with 8 m PZ with a CO2 loading of 0.3 mol CO2/mol alka-linity. Four campaigns were run in SRP, the first two campaignswere run at 135 ◦C for 1000 hours and the rest of the two campaignstested a two-stage flash configuration at 150 ◦C for 1300 hours. Realflue gas stream was treated with selective catalytic reduction andlimestone slurry to minimize the concentrations of SOx and NOxprior to PP2 entrance. Dominant degradation products of SRP andPP2 were found to be FPZ, EDA, and formate. Quantified degradationproducts of the SRP and PP2 are listed in Table 3.

The degradation products of PZ produced at bench scale exper-iments are listed in Tables 1 and 2. However, Table 3 representsthe degradation products formed at pilot plant. It can be observedthat roughly similar degradations products (EDA, FPZ, AEP, HEP,etc.) of PZ have been reported from the results of the pilot plantdegradation products analysis to that of discussed in bench scaleexperiments. In addition, there are many products which werenot commonly found at both experimental scales. At both scales,AEP, EDA, and FPZ were found as abundant degradation products.However, incidentally ammonia/ammonium was not reported asdegradation product in pilot plant results which is found as anabundant degradation product in bench scale experiments. Differ-ence in the bench scale and pilot plants degradation products assayand their concentrations may be due to the composition of flue gasused for degradation.

2.1. Nitrosamine formation and decomposition

One of the unresolved problems of postcombustion CO2 captureusing amine scrubbing technology is the formation of hazardousbyproducts (Freeman, 2011). Significant evidences for formationof nitrosamines and nitramines through CO2 capture using amine-

based solvents have been discovered. Nitrosamines are powerfulcarcinogen compounds (Bartsch and Montesano, 1984; Lijinskyand Epstein, 1970). The reaction of PZ and nitrite can form MNPZand N,N-dinitrosopiperazine (DNPZ). Carcinogenicity of DNPZ is

216 S.A. Mazari et al. / International Journal of Greenhouse Gas Control 31 (2014) 214–228

Table 1Thermal degradation products of PZ with CO2 loading of up to 0.47 mol CO2/mol alkalinity at temperatures up to 175 ◦C (Freeman, 2011; Freeman and Rochelle, 2012b).

Name of the degradation product Abbreviation Molecular weight (g/mol) Structure Statusa

2-Imidazolidone 2-Imid 86.09 I

Acetate ion/acetic acid Acetate 59.04/60.05 I

1-[2-[(2-Aminoethyl)amino]ethyl] piperazine AEAEPZ 172.27 S

1-[2-(Piperazinyl)ethyl]2-imidazolidinone AEAEPZ Urea 198.27 S

1-(2-Aminoethyl)-2imidazolidone AEI 129.16 INQ

N-(2-aminoethyl) piperazine AEP 129.20 I

Ethylenediamine EDA 60.10 I

N-ethylpiperazine EPZ 114.19 I

Formate ion/formic acid Format 45.02/46.04 I

N-formylpiperazine FPZ 114.15 I

N-(2-hydroxyethyl) piperazine HEP 130.19 I

Methylpiperazine MPZ 100.2 I

Ammonia/ammonium ion NH4+ 17.03/18.04 I

1,1′-(1,2-Ethanediyl)bis-piperazine PEP 198.31 S

– Sus

a2crapdAt

a Status: I – fully identified and quantified, INQ – identified, but not quantified, S

lmost 20 times higher compared to MNPZ (Ashouripashaki,012). MNPZ and DNPZ have tumor dose required for 50% ofontrol (TD50) values of 8.7 and 3.6 mg/kg body weight/dayespectively (Garcia et al., 1970; Pai et al., 1981). Nitrosaminend nitramines contamination of air and drinking water sup-

lies downwind of amine-based CO2 plants can cause significantamages (Fostås et al., 2011). Norwegian Climate and Pollutiongency (2011) recommended that the sum of the concentra-

ion of nitrosamines and nitramines should not exceed 0.3

pected, but no commercial standard available.

nano-gram (ng)/m3 for air concentrations and 4 ng/l for fresh watersources.

Ashouripashaki (2012) studied the kinetics of the nitrite solu-tion and PZ to form nitrosamine. Experiments were conducted attemperatures ranging from 20 to 150 ◦C. PZ concentrations ranged

from 2 to 8 mol/kg and CO2 loading was ranging from 0.1 to 0.3 molCO2/mol alkalinity. Results of the study exhibited that the nitritereacts with PZ at temperatures less than 75 ◦C to form MNPZand decomposition of MNPZ occurs at higher temperatures. The

S.A. Mazari et al. / International Journal of Greenhouse Gas Control 31 (2014) 214–228 217

Table 2Summary of identified degradation products through oxidation of PZ.

Products name(s)/formula(s) References

HCOO− , CH3COO− , HOCH2COO− , (C2O4)2− Freeman et al. (2009, 2010b),Sexton (2008)

NO2− , NO3

− Freeman (2011), Sexton (2008)CH3NO Freeman (2011), Freeman et al.

(2009), Freeman et al. (2010b),Sexton (2008)

N-formylpiperazine Freeman et al. (2010b)C2H4(NH2)2 Freeman (2011), Freeman et al.

(2009), Sexton (2008)

dlaotncstPiepb

sl(tatlsetni

TSa

1E-10

1E-9

1E-8

1E-7

1E-6

PZ MDE A AM P MEA

Fo

rmat

ion R

ates

(M

/s)

Nitr ite

Total Nitrosamine

Fig. 1. Formation rates of nitrite and total nitrosamines during purging of 2.5 Meach: PZ, MDEA, AMP, and MEA solutions at 44 ◦C and an initial pH of 12 with

N-formylethylenediamine Freeman et al. (2010b)NH3, NH4

+ Sexton (2008), Freeman (2011)

ecomposition of MNPZ was noticed to follow the similar trendine for all the CO2 loading ranges (0.1, 0.2, and 0.3 mol CO2/mollkalinity). Furthermore, no evidence was found for the formationf DNPZ as a reaction product under all conditions examined inhe study. Jackson and Attalla (2011) reported the formation ofitrosamines for 15% (w/w) PZ (99.8%). A synthesized gas streamonsisting of 81.4% N2, 12.8% CO2, 0.8% NO, and 5.0% O2 wasparged, flowing at 1.34 L/min through 200 g of aqueous PZ solu-ion at 60 ◦C. Peak through the protonated forms of nitrosamine ofZ (N-nitrosopiperazine m/z 116) was detected, and its structuraldentity was confirmed using mass spectrometry/mass spectrom-try (MS/MS) experiments. Furthermore, Jackson and Attala studyroposed that MNPZ is stable up to 150 ◦C and may be decomposedy using ultraviolet (UV) light.

Dai et al. (2012) conducted a comparative study for the mea-urement of formation of nitrosamines through different aminesike; morpholine (Mor), MDEA, PZ, AMP, and MEA. Nitrous oxideNO) and nitrogen dioxide (NO2), each at 25 ppmv were bubbledhrough 2.5 molar (M) solutions of each: PZ, MEA, AMP, and MDEAt 44 ◦C. Total nitrosamine accumulation rate for PZ was foundo be 27 times higher than other three amines. Higher accumu-ation rate of nitrosamine through PZ is due to the formation oftable nitrosamines through secondary amines (Ridd, 1961). How-ver, nitrite formation in PZ was almost equal to those of otherhree amines, MEA, AMP, and MDA. A detailed presentation of

itrosamine and nitrite formation rate with errors for each amine

s provided in Fig. 1.

able 3ummary of degradation products of 8 m PZ at pilot plant, with 0.3 mol CO2/mollkalinity up to 150 ◦C (Nielsen et al., 2013).

SRP (mmol/kg) PP2 (mmol/kg)

Operating time (h) 1350 N/ADegraded solvent alkalinity 4112 4194Degraded PZ solvent 3855 35461-Nitrosopiperazine (MNPZ) 0.09 1.22EDA 12.7 9.33FPZ 1.26 38.0AEP 3.58 2.94HEP 3.91 2.64EPZ 0.00 1.4AEAEPZ 0.3 1.2Formate 1.03 38.6Oxalate 0.38 8.62Acetate 0.28 11.2Nitrate 0.14 4.82Total formate 2.29 72.7Total oxalate 0.34 17.0Total acetate 0.29 14.6Cr3+ 0.016 2.21Fe2+ 0.023 1.13Ni2+ 0.021 1.86Cu2+ 0.00 0.02

25 ppmv NO and NO2 each.

Adapted with permission from Dai et al. (2012). Copyright (2014) American Chem-ical Society.

Nielsen et al. (2013) found a low concentration of 0.09 mol/kg ofMNPZ in SRP campaigns using a synthetic flue gas stream withoutadditional concentration of NOx. The formation of MNPZ under thiscondition was speculated to take place due to the oxidation of PZ.Furthermore, this study brought to light that after highest concen-tration of 2.87 mmol/kg, the higher steady-state concentrations of0.9–1.2 mmol/kg were noticed for MNPZ formation in PP2.

Fine et al. (2013) studied formation and decomposition of MNPZat temperatures ranging from 50 to 135 ◦C. PZ concentration rangedfrom 0.1 to 8 m and CO2 loading ranging from 0.1 to 0.4 mol CO2/molalkalinity. MNPZ formation was the first order in the presenceof NO2

−, PZCOO−, and H+. Decomposition was also first order,depending on the concentration of PZ and CO2 loading. In 8 mPZ with 0.3 mol CO2/equiv. N, thermal degradation of MNPZ waspseudo first order that followed Arrhenius temperature depend-ence with a rate constant of 10.2 × 10−6 s−1 at 135 ◦C (Fine andRochelle, 2013). Fine et al. (2014) in another study, investigateddecomposition of nitrosamines formed in PZ, MEA, MDEA, and AMP.Concentration of amines varied from 0.1 to 6 M with a CO2 load-ing ranged from 0 to 0.4 mol CO2/mol alkalinity. Decompositionof nitrosamine was found to follow first order kinetics at tempera-tures 120–150 ◦C. Study revealed that CO2 loading and temperatureseems to influence the decomposition of MNPZ. Due to the acidicnature of CO2, pH of the solution decreases with CO2 loading, whichdecreases the decomposition rate of MNPZ. However, temperaturehas a direct effect on the decomposition of MNPZ. Table 4 presentsthe pseudo first order decomposition rates of MNPZ with differentCO2 loading and temperatures.

Formation of nitrosamines is a challenge to the viability ofpostcombustion CO2 capture using amine-based solvents. Benchscale to pilot plant experiments showed that nitrosamine; espe-cially MNPZ is formed during CO2 capture while using PZ as asolvent. However, it is speculated that MNPZ can be decomposedusing ultraviolet (UV) light. Furthermore, some of the studies havealso shown that under desorber conditions, nitrosamines degradealmost at a similar rate at which they are formed. This concept isstill not well established, as some of the lab scale and pilot plantstudies have shown that nitrosamines could be formed even underdesorber conditions (Dai et al., 2012; Fine et al., 2013).

3. Types of degradation

3.1. Thermal degradation

Thermal degradation is an important property for the solventselection, as solvent can spend more than one third of its residencetime at temperatures above 100 ◦C (Freeman and Rochelle, 2012b).Thermal degradation primarily occurs in stripper, re-claimer, and

218 S.A. Mazari et al. / International Journal of Greenhouse Gas Control 31 (2014) 214–228

Table 4Summary of first order rate constants k1, of degradation of MNPZ.

Temperature (◦C) Based (M) CO2 loading (mol CO2/mol N) k1, 10−6 (s−1) References

100 4.9 PZ 0.31 0.72 ± 0.03 Fine and Rochelle (2013)120 4.9 PZ 0.31 3.7 ± 0.1 Fine and Rochelle (2013)135 4.9 PZ 0.31 11.1 ± 0.5 Fine and Rochelle (2013)150 4.9 PZ 0.31 26.9 ± 1.6 Fine and Rochelle (2013)165 4.9 PZ 0.31 65.3 ± 0.7 Fine and Rochelle (2013)

ctws

sFLDatRmwcwa1orT

tD

TSC

150 4.9 PZ 0.01

150 4.9 PZ 0.1

150 4.7 PZ 0.39

ross heat exchanger. Due to the formation of thermal degrada-ion products, capacity of the solvent to capture CO2 decreases,hich causes an increase in the requirement of steam for stripping,

olvent makeup, reclaiming, and disposal costs.Thermal degradation studies for PZ have been carried out exten-

ively by researchers (Chen and Rochelle, 2011; Freeman, 2011;reeman et al., 2009, 2010a; Freeman and Rochelle, 2011, 2012b;i et al., 2013b; Namjoshi et al., 2013), under different conditions.avis (2009), conducted experiments on 5 m and 8 m PZ whilstmine screening tests and concluded that PZ did not degrade upo 135 ◦C. Freeman (2011), Freeman et al. (2010a), Freeman andochelle (2011, 2012a,b) carried out thermal degradation experi-ents using 5 inches (in.) long 316 stainless steel (316SS) cylindersith Swagelok end caps and cylinders were kept in mechanical

onvection ovens for thermal degradation. Thermal degradationas investigated for 4, 8, 12 and 20 m PZ for up to 72 weeks with

CO2 loading ranging from 0 to 0.47 mol CO2/mol alkalinity at35–175 ◦C. Thermal degradation of PZ followed first and secondrder kinetics (Freeman and Rochelle, 2012b; Freeman, 2011). Theate constants of first and second order kinetics are presented inables 5 and 6, respectively.

Thermal degradation of PZ increased with the increase in all thehree parameters: temperature, PZ concentration and CO2 loading.egradation of 8 m PZ increased linearly from 0 to 0.3 mol CO2/mol

able 5ummary of results of k1 for PZ, with respect to temperature, PZ concentration, andO2 loading (Freeman, 2011).

Temperature (◦C) PZ concentration(m)

CO2 loading (molCO2/mol alkalinity)

k1, 10−9 (s−1)

135 8 0.3 1.08 0.4 0.5

10 0.3 4.810 0.4 –

150 8 0.3 6.18 0.4 7.9

10 0.3 8.910 0.4 6.015 0.3 –20 0.3 24.2

165 4 0.3 35.58 0 0.88 0.1 18.48 0.3 31.48 0.4 40.9

12 0.3 50.3

175 4 0.3 1148 0 78 0.1 65.88 0.2 78.88 0.3 1328 0.4 1718 0.47 23.8

12 0.3 15620 0.1 44.720 0.3 168

31.3 ± 1.0 Fine et al. (2014)28.6 ± 0.4 Fine et al. (2014)17.9 ± 0.8 Fine et al. (2014)

alkalinity at 150–175 ◦C. However, it decreased when CO2 loadingwas above 0.4 mol CO2/mol alkalinity under the same conditions.At the meantime the maximum degradation noticed at 175 ◦C waswith 0.3–0.4 mol CO2/mol alkalinity (Freeman, 2011). Based on oneof the modeling studies (Freeman and Rochelle, 2012a) of 8 m PZat 175 ◦C, it was believed that the concentrations of free PZ andPZCOO− were low above 0.4 mol CO2/mol alkalinity while the con-centration of HCO3

− reached a significant level. The presence of CO2in form of HCO3

− neutralized the catalytic influence of CO2 whenexisting in the form of other species, at lower CO2 loading.

Pilot plant experiments conducted under conditions describedin Section 2 to investigate thermal degradation of PZ (Nielsenet al., 2013). In SRP campaigns, little degradation occurred after1350 hours of operation. However, oxidative degradation was thedominant part of the campaigns, mainly in PP2. Freeman et al.(2009) reported a comparative thermal degradation study for PZand MEA. At 175 ◦C, PZ seemed to lose up to 32% of its initial con-centration in 4 weeks. A weekly loss of 5.3% and 11%, 0.25% and 0.8%,was noticed for 7 m MEA with 0.4 mol CO2/mol alkalinity and 10 mPZ with 0.3 mol CO2/mol alkalinity at 135 and 150 ◦C respectively.Furthermore, 8% weekly loss was reported for 8 m PZ with a CO2loading of 0.3 mol CO2/mol alkalinity at 175 ◦C. Results of PZ andMEA under conditions indicated in Fig. 2 are assumed for 4 weeks,based on weekly loss.

3.2. Oxidative degradation

Oxidative degradation takes place due to two main reasons:presence of oxygen, and free metal ions (Goff and Rochelle, 2004).The products formed through amine (MEA and PZ) loss are for-mate (10–40%) or ammonia (70–100%), as reported in Table 2 (Liet al., 2013b). Freeman (2011) and Freeman et al. (2010a) con-ducted baseline oxidative degradation experiments of PZ in low gasflow (LGF) (100 mL/min) Teflon oxidation reactor (TOR) and orig-

inal oxidation reactor (OOR). Experiments for 8–10 m PZ, loadedwith CO2 ranging from 0.2 to 0.4 mol CO2/mol alkalinity were con-ducted at temperatures from 55 to 70 ◦C, allowing 2% CO2, and 98%

Table 6Summary of results of k2 for PZ, at different temperatures, PZ concentrations, andCO2 loading (Freeman, 2011; Freeman and Rochelle, 2012b).

Temperature (◦C) PZ concentration(m)

CO2 loading (molCO2/mol alkalinity)

k2, 10−13

(mmol/kg s)

135 8 0.3 2.39150 8 0.3 14.9165 8 0.3 128175 4 0.3 1120175 8 0 148175 8 0.1 195175 8 0.2 277175 8 0.3 565175 8 0.4 617175 8 0.47 134175 12 0.3 617175 20 0.3 821

S.A. Mazari et al. / International Journal of Greenhouse Gas Control 31 (2014) 214–228 219

0

20

40

60

80

100

120

0 1 2 3 4 5

Am

ine

conce

ntr

atio

n l

oss

(%

)

Time (Weeks)

7 m MEA at 13 5 °C(Davis and Roch elle,2009 )7 m MEA at 150 °C(Davis and Roch elle,2009 )10 m PZ at 135 °C(Freeman et al. 200 9)

10 m PZ at 150 °C(Freeman et al. 200 9)

8 m PZ at 175 °C(Freeman et al. 200 9)

Fig. 2. The percentage loss of 7 m MEA with 0.4 mol CO2/mol alkalinity degraded at1a4

Oatc(N

iswfi9Ti8taCoRoaaoFdcfrP

MOcT

Fig. 3. Oxidative degradation of 8 m PZ and 7 m MEA with 0.3 and 0.4 mol CO2/molalkalinity respectively in the presence of metals, such as: 0.4 mM Fe2+, 0.1 mM Cr3+,

TO

35 and 150 ◦C is compared with 10 m PZ with 0.3 mol CO2/mol alkalinity degradedt 135 and 150 ◦C and 8 m PZ with 0.3 mol CO2/mol alkalinity degraded at 175 ◦C for

weeks, on the basis of weekly loss results.

2 as input gases for 3–5 weeks. Reactors were equipped with angitator (1400 rpm) to maximize the mass transfer of O2 in the solu-ion. Metal salts, as aqueous sulfates were also added to know theiratalytic effect on oxidation of PZ. Metal catalysts used were iron1 mM Fe2+), stainless steel (0.4 mM Fe2+, 0.1 mM Cr3+, and 0.05 mMi2+) and copper (4 mM Cu2+) (Freeman, 2011).

Results revealed that the oxidation of PZ in presence of metalons and O2 followed first order kinetics. However, oxidationeemed to increase with the increase in partial pressure of O2 asell as temperature. An increase of 245% in O2 partial pressure

rom 40 to 98 kPa, enhanced PZ loss by 240% at 55 ◦C. The increasen temperatures from 55 to 70 ◦C increased oxidation by 230% at8 and 94 kPa O2 partial pressures respectively at 55 and 70 ◦C.he increase in CO2 loading from 0.2 to 0.3 mol CO2/mol alkalinityncreased oxidation, even so the effect was not well understood.

m PZ loss in presence of 1.0 mM Fe2+ was observed to be 8% ofhe original amine in 450 hours. Higher degradation of 10 m PZ,ccounted for 28% of the original amine in 220 hours when 4.0 mMu2+ was added as a catalyst (Freeman et al., 2010a). In the absencef any catalyst, a loss of 2–7% of PZ was noticed for 350 hours.esults exhibited that adding Cu2+ as a catalyst for the oxidationf PZ, increased the rate of degradation. An inhibitor A (named byuthor) was also used to minimize the catalytic effect of coppernd iron (Freeman et al., 2010a). In the presence, of Cu2+, 100 mMf inhibitor ‘A’ reduced oxidation by up to 67%. In the presence ofe2+, 100 mM inhibitor A decreased oxidation to a small extent. Oxi-ation of PZ was nearly four times less than MEA under the sameonditions, however it resulted a minute formation of formate andormamides. Table 7 represents the rate of amine loss as well as theate of formation of formate, formamide and EDA through MEA andZ oxidation.

Rochelle et al. (2011) reported oxidative degradation of PZ and

EA in presence of 0.4 mM Fe2+, 0.1 mM Cr3+, and 0.05 mM Ni2+.xidation experiments were conducted by sparging a gas streamonsisting of 98 mole percent (mol%) O2 and 2 mol% CO2 at 55 ◦C.he solution was agitated at 1400 rpm to achieve the maximum

able 7xidative degradation of PZ and MEA at 55 ◦C, in presence of 98% O2 and 2% CO2, in 350 m

Solvent Additives (mM) Rate of forma

Formate

7 m MEA 1. Fe 0.29

10 m PZ Fe2+, 0.25 Cr3+, 0.25 Ni2+ 0.005

10 m PZ Cu2+ 0.14

8 m PZ Fe2+, 0.1 V4+ 0.006

8 m PZ 4.0 Cu2+, 0,1 Fe2+, 100 ‘A’ 0.011

0.05 mM Ni2+ and 98% O2 and 2% CO2 degraded at 55 ◦C (Rochelle et al., 2011).

mass transfer. PZ did not degrade under provided conditions, how-ever MEA degraded up to 35% within 400 hours. Fig. 3 illustratesthe results of oxidative degradation of 8 m PZ and 7 m MEA with0.3 mol CO2/mol alkalinity in the presence of Fe2+, Cr3+, and Ni2+ at55 ◦C. The increase in concentration of PZ, higher than 100%, indi-cates the experimental error due to the water balance in the reactorchanging whilst the experiment and the cumulative error involvedin the analytical process (Freeman et al., 2010a).

Wang and Jens (2014) investigated oxidative degradation ofunloaded PZ. Experiments were carried out in a glass reactor attemperatures ranging from 80 to 120 ◦C, with O2 partial pressure of250 kPa. Degassed 140 mL aqueous solution was loaded in the reac-tor and agitated at 200 rpm. 1.5 m PZ was observed to degrade byup to 1.30 × 10−4 millimoles per kilogram per second (mmol/(kg s))in 912 hours at 80 ◦C and 3.33 × 10−4 mmol/(kg s) in 456 hours at100 ◦C. Oxidation of PZ followed first order and estimated rateconstants were found to be as 1.0 × 10−7 s−1 and 3.3 × 10−7 s−1

degraded at 80 and 100 ◦C respectively.Nielsen et al. (2013) investigated degradation of PZ in two pilot

plants: SRP and PP2, under conditions described previously in Sec-tion 2. PZ seemed to oxidize more in PP2 compared to SRP due tothe composition of flue gas used. Formate generation measured atthe end of campaigns was about 73 mmol/kg and 3 mmol/kg for PP2and SRP respectively.

4. Mechanisms of degradation

Mechanism for each degradation product has not yet beenwell understood. However, some of the pathways have been sug-gested in producing major degradation products of PZ. The mainreactions are: secondary nucleophilic substitution (S 2) reactions,

Nelimination reactions, urea generation and hydrogen-abstraction(Freeman, 2011; Wang and Jens, 2014).

L solution (Freeman et al., 2009, 2010b).

tion (mM/h)

Formamide EDA Amine

0.35 – −3.80.007 0 −1.10.24 0.43 −3.00.013 0 −0.80.016 0.009 −1.1

220 S.A. Mazari et al. / International Journal of Greenhouse Gas Control 31 (2014) 214–228

Scheme 1. Attack of H+PZ at the �-carbon by another PZ molecule producing AEAEPZ and H+.

Scheme 2. Formation of internal urea through the degradation of AEAEPZ in presence of CO2.

Scheme 3. Protonation of the terminal amino function on the urea of AEAEPZ, prodcing an ammonia (NH4+), CO2, and quintamine molecule.

t of th

4

4

ot2

Irr(

Scheme 4. The attack of PZ molecule on the carbon on righ

.1. Thermal degradation mechanism

.1.1. Secondary SN2 substitution reactionsBeginning of thermal degradation of PZ starts with the attack

f H+PZ at the �-carbon by another PZ molecule to proceedhrough an intermediate to give AEAEPZ (Scheme 1) (Freeman,011).

In the presence of CO2, EDA reacts rapidly with CO2 to form 2-

md at higher temperatures (150–175 ◦C). In addition, AEAEPZ mayeact with CO2 in the solution to produce internal urea and shouldemain in equilibrium as a result of behavior of EDA and 2-ImdScheme 2) (Freeman, 2011).

e secondary amino function produces, two AEP molecules.

After formation of AEAEPZ and its urea, more SN2 type reactionmay happen to form more degradation products. Resulting SN2 sub-stitution reaction, PZ may attack at multiple positions of AEAEPZ,as if terminal amino function get protonated, urea of AEAEPZ, CO2,NH4

+, and quintamine will be formed as presented in Scheme 3(Freeman, 2011).

When the inner secondary amino function is protonated orremains as cyclic urea, any of its �-carbons might be confronted to

generate different molecules. When the carbon on right is attacked,two AEP molecules are formed (Scheme 4), and if the carbon on leftis attacked EDA and 1,1-(1,2-ethanediyl)bis-piperazine are formed(Scheme 5) (Freeman, 2011).

S.A. Mazari et al. / International Journal of Greenhouse Gas Control 31 (2014) 214–228 221

Scheme 5. The attack of PZ molecule on the carbon on left of the secondary amino function produces, EDA and PEP.

Scheme 6. Reactants generation when either of the amino functions within the cyclic PZ structure is protonated.

Scheme 7. Long chain PZ hexamines formation if amino functions within cylic structure are protonated.

gh th

tti

(mdSA(f2

Scheme 8. Long chain PZ hexamine formation throu

Most likely, neither of amino functions gets protonated withhe cyclic structure of PZ, but if it happens then there is possibilityhat either reactants may be regenerated or long chain hexamines formed as presented in Schemes 6–8 (Freeman, 2011).

Along with PZ, some degraded products like EDA, AEP and N-2-hydroxyethyl)pierazine (HEP) may act as nucleophiles to form

ore degradation products (Freeman, 2011). Capability to attackepends on concentration of solution and pKa of the amino group.cheme 9 exhibits AEP as a nucleophile that may attack another

EP molecule to form PZ and N,N-dis(2-aminoethyl)piperazine

DAEP). PolyAEP and ammonia would be formed, if terminal aminounction on AEP is protonated as shown in Scheme 10 (Freeman,011).

e protonation of amino function in cyclic structure.

4.1.2. Elimination reactionsIn elimination reactions of amines, a couple of substituents

should be eliminated from a molecule, most likely to form analkene relying on the presence of leaving group (McMurry, 2000).Leaving group replaces NH2, which is mostly accomplishedthrough Hofmann’s elimination, that coverts an amino group intoa quaternary salt before � elimination (Smith, 2011). Amines reactas a nucleophile in SN2 reactions with the excess alkyl group toform quaternary ammonium salt so is in the case of PZ (Freeman,

2011; Smith, 2011). Elimination reactions in thermal degradationof PZ may occur at the tertiary amine. SN2 and E2 eliminationtakes place in the meantime as a result of elevated temperatureand the mix of nucleophiles and the leaving groups (Freeman,

222 S.A. Mazari et al. / International Journal of Greenhouse Gas Control 31 (2014) 214–228

Scheme 9. Formation of PZ and DAEP through the attack of one AEP molecule to another AEP.

Scheme 10. Formation of PolyAEP and ammonia by the protonation of terminal amino function.

S ’s elimr

2oH1

4

tsmpt

4

bcmroh

mOwh

Sw

cheme 11. Formation of 1-ethenylpiperazine and HEP from PEP through Hoffmanespectively.

011). PZ and 1-ethenylpiperazine can be produced as a resultf Hofmann’s elimination of protonated PEP (Scheme 11) andEP can be produced through Anti-Markovnikov hydration of-ethenylpiperazine (Scheme 11) (Freeman and Rochelle, 2012b).

.1.3. Urea generationUreas are believed to be formed under alkaline conditions, in

he presence of CO2. 2-Imid and AEI were found with AEAEP asuspected ureas in CO2 loaded PZ. One of the examples of urea for-ation is EDACOO− (EDA carbamate), which is formed by EDA in

resence of CO2. In addition, the terminal amino function may reacto form 2-Imd (Scheme 12) (Freeman, 2011).

.2. Oxidative degradation mechanism

Wang and Jens (2014) postulated that the oxidation of PZ mayegin with hydrogen-abstraction from one of the four methylenearbons in PZ molecule forming a free radical (1) centered onethylene carbon. In the presence of O2, a free radical is expected to

eact with oxygen to form peroxyl radical (2). Deterioration of per-xyl radical may take place either through inter or intramolecularydrogen-abstraction step.

Peroxyl radical can produce peroxide (3) in an alkaline environ-

ent through intermolecular hydrogen-abstraction (Scheme 13).xygen-centered radical and •OH are formed from peroxide, then,hich is followed by the formation of 2-hydroxyl-PZ (4). 2-ydroxyl-PZ may form [(2-aminoethyl)amino]acetaldehyde (5) by

cheme 12. Formation of 2-Imd through the reaction of terminal amino functionith hydroxyl group.

ination of protonated PEP and anti-Markovnikov hydration of 1-ethenylpiperazine

its oxidation through ring opening. Aldehyde oxidizes to give [(2-aminoethyl)amino]acetic acid (6) followed by ring opening closureforming OPZ. Alternatively, (5) may deteriorate to give oxalic acid(7) and EDA (8) through cleavage of N C bond; in addition to that,EDA may degrade to form glycine (9) and glycolic acid (10).

Intramolecular hydrogen-abstraction may take place(Scheme 14) for PZ peroxyl radical (2) through amine function.Once •OH is ejected, imine intermediate (11) is produced throughhemolytic cleavage of C C bond. Being reactive, imine inter-mediate produces [(2-aminoethyl)amino]acetaldehyde (12) andformaldehyde (13). The [(2-aminoethyl)amino]acetaldehyde isunstable and can form EDA (8) and formaldehyde. Formalde-hyde potentially may oxidize to produce formic acid (14).A detailed oxidative degradation mechanism is provided inSchemes 13 and 14.

5. Degradation of PZ blends, its structural analogs anddiamines

5.1. Degradation of PZ blends

5.1.1. AMP/PZ blendAMP/PZ blend provides higher CO2 loading compared to

AMP and PZ individually without precipitation as noticed in PZalone at higher CO2 loading (Brúder et al., 2011; Samanta andBandyopadhyay, 2009). However, the blend has higher volatilityat lean conditions (Li et al., 2013a). Thermal degradation of 5 mPZ/2.3 m AMP blend was investigated by Li et al. (2013a), at 135 and150 ◦C with a CO2 loading of 0.4 mol CO2/mol alkalinity for up to 6weeks. Two sets of experiments were run for AMP/PZ, and resultsare organized in percentage range, from lower to higher. In 6 weeks,

9–13% loss for PZ and 19–22% for AMP was noticed at 135 ◦C. How-ever, degradation loss was higher at 150 ◦C, which accounted for20–28% for PZ and 56–60% for AMP. The combination of AMP and PZpromoted the rate of degradation when blended together as higher

S.A. Mazari et al. / International Journal of Greenhouse Gas Control 31 (2014) 214–228 223

Scheme 13. An intermolecular H-abstraction decomposition pathway for peroxyl radical from PZ (Wang and Jens, 2014).

Scheme 14. An intramolecular hydrogen-abstraction degradation pathway for peroxyl radical from PZ (Wang and Jens, 2014).

224 S.A. Mazari et al. / International Journal of Gree

0

50

100

150

200

250

300

130 135 14 0 14 5 150 155

k1,1

0-9

(s-1

)

Temper ature (°C )

PZ

AMP

PZ in (PZ /AMP)

AMP in (PZ/AMP)

PZ+AMP in (PZ/AMP)

FaC

dDwmM

owaf(t

5

CMpCftecoeP

t

dHt1

i3iCat

5

aewsw

ig. 4. Thermal degradation first order rate constant (k1) comparison between PZnd AMP individually and in blend (5 m PZ/2.3 AMP) with a CO2 loading of 0.4 molO2/mol alkalinity at 135 and 150 ◦C (Freeman, 2011; Li et al., 2013a).

egradation was noticed in the blended form than AMP or PZ alone.egradation of PZ increased up to 10 folds in the blend of AMP/PZhile AMP degraded 3 times faster as exhibited in Fig. 4. Further-ore, results revealed that the blend degraded slowly compared toEA up to 150 ◦C.Wang and Jens (2012, 2014) investigated oxidative degradation

f AMP and AMP/PZ blend. Unloaded AMP/PZ blend experimentsere conducted at temperatures ranging from 80 to 120 ◦C, with

n O2 partial pressure of 250 kPa. Oxidative degradation observedor 13 days was 0.16, 0.62, and 2.5 milimole per kilogram per hourmmol/(kg h)) for AMP, and 0.45, 1.9, and 3.2 mmol/(kg h) for PZ atemperatures 80, 100, and 120 ◦C respectively.

.1.2. MDEA/PZ blendThe MDEA/PZ blend was investigated by Closmann et al. (2009),

losmann (2011), and Voice et al. (2013) in order to replace 30%EA. Experiments for 7 m MDEA/2 m PZ were conducted at tem-

eratures from 100 to 150 ◦C with CO2 loading from 0 to 0.43 molO2/mol alkalinity. Thermal degradation experiments were per-ormed in stainless steel cylinders, which were then placed in aemperature-controlled oven for up to 9 weeks. In addition, somexperiments were also run by adding 1 mM Fe2+, to examine itsatalytic effect on thermal degradation of the blend. A numberf degradation products were identified in thermal degradationxperiments. The identified degradation products of 7 m MDEA/2 mZ are presented in Table 8.

Degradation kinetics results of 7 m MDEA/2 m PZ exhibited thathe blend is stable up to 135 ◦C as reported in Table 9.

Magnitude of errors as reported in Table 9 illustrate that eitheregradation of PZ and MDEA was almost zero or immeasurable.owever, degradation rate increased by increasing the tempera-

ure as well as CO2 loading. Higher degradation was noticed from20 to 135 ◦C.

Closmann (2009, 2011) conducted oxidative degradation exper-ments of MDEA/PZ using an enclosed jacketed glass reactor. Nearly50 mL solvent was degraded for 14 days at 55 ◦C with a CO2 load-

ng ranging from 0.1 to 0.3 mol CO2/mol alkalinity, allowing 98%O2 and 2% O2 as input gases. MDEA individually formed formatend amide faster compared to MDEA/PZ blend. Table 10 illustrateshe rate of formation of formate and amide.

.1.3. Other PZ blendsOther than AMP and MDEA, PZ has also been investigated with

number of other amines under distinctive conditions (Freeman

t al., 2013; Li et al., 2013b; Namjoshi et al., 2013). PZ was blendedith MPZ and DMPZ, in six different blended forms to under-

tand the thermal degradation (Freeman et al., 2013). Experimentsere conducted at 150 ◦C in stainless steel cylinders. A single

nhouse Gas Control 31 (2014) 214–228

two-component blend of 4 m PZ/4 m MPZ, and five three-component blends 5 m PZ/2 m MPZ/1 m DMPZ, 5 m PZ/2.5 mMPZ/0.5 m DMPZ, 5 m PZ/1.5 m MPZ/1.5 m DMPZ, 3.9 m PZ/3.9 mMPZ/0.2 m DMPZ, and 3.75 m PZ/3.75 m MPZ/0.5 m DMPZ wereinvestigated. Degradation of the blends was first order. Change ininitial concentration of the each amine (PZ, MPZ, and DMPZ) in theblend seemed to influence the rate of degradation. Lower degra-dation was observed in 3.9 m PZ/3.9 m MPZ/0.2 m DMPZ blend andthe higher for 5 m PZ/1.5 m MPZ/1.5 m DMPZ. However, in gen-eral, the presence of DMPZ in the PZ blends seemed to catalyze therate of degradation of the system. Degradation rate of blends werefound competitive compared to 8 m PZ but in comparison to 7 mMEA, all of the blends were found as stable. Detailed first order rateconstants (k1) are provided in Table 11.

Thermal and oxidative degradation of an equimolar diamineblend of 4 m 2-methylpiperazine (2-MPZ) with 4 m PZ was exam-ined by Sherman et al. (2013). Thermal degradation experiments of4 m PZ/4 m 2-MPZ were conducted at 150 ◦C with a CO2 loading of0.3 mol CO2/mol alkalinity for 30 weeks. Thermal degradation of theblend was found to be first order, with a k1 value of 1.59 × 10−8 s−1.Thermal degradation rate of PZ and 2-MPZ blend was intermediateto both, PZ and 2-MPZ with a weekly loss of 2% of the initial con-centration. For the same study, oxidative degradation of 2-MPZ/PZblend was analyzed using LGF reactor with 98% O2 and 2% CO2 at70 ◦C. Different metals, like iron (Fe), nickel (Ni), and chromium (Cr),were also added to know the catalytic effect of corrosion productson the rate of oxidation. In presence of Cr and Ni, amine loss wasnoticed as less than 0.3 mmol/(kg h) and total formate generationwas measured to be 0.019 mmol/(kg h).

Du et al. (2013) proposed blend of PZ with AEP and suggested itas a highly suitable solvent. The blend was speculated to mitigatethe problem of precipitation of concentrated PZ maintaining highCO2 absorption rate and thermal stability. Thermal degradationexperiments of 5 m PZ/2 m AEP were conducted in stainless steelcylinders. Solution was loaded with CO2 (0.2–0.3 mol CO2/mol alka-linity), degradation time was 20 weeks, and temperatures rangedfrom 150 to 175 ◦C. 10% of PZ and 30% of AEP loss was observed for20 weeks at 150 ◦C, however blend degraded almost completely at175 ◦C for the same duration. Thermal degradation of the blend fol-lowed first order kinetics with the rate constants of 1.52 × 10−8 s−1

and 20.1 × 10−8 s−1 at 150 ◦C and 175 ◦C respectively. Such resultsdemonstrate that this blend system can be used up to 150 ◦C. How-ever, this blend may not be suitable above 150 ◦C. Furthermore,thermal degradation also increased by increasing the CO2 loading.

Oxidative degradation of 5 m PZ/2 m AEP was conducted in thepresence of 0.05 mM Cr3+, 0.1 mM Ni2+, 0.4 mM Fe2+ and 0.1 mMMn2+ in a LGF reactor with 100 mL/min for 2 weeks at 70 ◦C. Totalformate generated from the PZ/AEP blend was found to be compa-rable to that of 8 m PZ and much lower than 7 m MEA.

Thermal degradation rates of PZ blends reported by variousauthors are presented in Table 11. It can be observed from Table 11that the least stable blends of PZ are with MEA and 2-MPZ. MEAand 2-MPZ seemed to catalyze the rate of degradation of totalamine concentration. At 175 ◦C, the combination of BAE and hex-amethylenediamine (HMDA) with PZ showed higher stability. Therate of thermal degradation of 6 m PZ/2 m BAE and 6 m PZ/2 mHMDA was almost equal to that of 8 m PZ alone under similar con-ditions. However, DAB and AEP degraded almost at the rate of twicecompared to 8 m PZ at 175 ◦C.

5.2. Degradation of structural analogs of PZ

Piperidine (PD), Mor, pyrrolidine (Pyr), hexamethyleneimine(HMI), homopiperazine (HomoPZ), 1-MPZ, and 2-MPZ were inves-tigated by Freeman (2011), to compare the amine loss with PZ.Thermal degradation was investigated up to 175 ◦C with a CO2

S.A. Mazari et al. / International Journal of Greenhouse Gas Control 31 (2014) 214–228 225

Table 8Identified thermal degradation products of 7 m MDEA/2 m PZ (Closmann, 2011).

Name of the degradation product Abbreviation Molecular weight (g/mol) Structure

2-Dimethylaminoethanol DMAE 89.1

1-Methylpiperazine 1-MPZ 100.1

N-(2-hydroxyethyl)-N-methyl formamide MAE-Amide 103.2

Diethanolamine DEA 105.1

1,4-Dimethylpiperazine 1,4-DMPZ 114.1

3-(Hydroxyethyl)-2-oxazolidone HEOD 131.1

4-Methyl-1-piperazine-ethanamine MPZEA 133.1

N,N-bis-(2-hydroxyethyl) formamide AEMP 143.1

1-(2-Hydroxyethyl)-4-methyl piperazine HMP 144.1

N,N-bis-(2-hydroxyethyl)-glycine Bicine 163.1

2-[[2-(1-Piperazinyl)ethyl]amino]ethanol PEAE 173.1

N,N-bis-(2-hydroxyethyl)–Piperazine bHEP 174.1

2-[[2-(4-Methyl-1-piperazinyl)ethyl]amino]ethanol MPZEA-OH 187.1

Methyl-N,N,N-tris-(hydroxyethyl) ethylenediamine MTHEED 206.1

226 S.A. Mazari et al. / International Journal of Greenhouse Gas Control 31 (2014) 214–228

Table 9Summary of thermal degradation of MDEA and PZ with uncertainties at 100, 120, and 135 ◦C with a CO2 loading from 0.1 to 0.2 mol CO2/mol alkalinity (Closmann et al., 2009).

Solvent Temperature (◦C) Time (weeks) MDEA degradation rate (mmolal/day) PZ degradation rate (mmolal/day)

CO2 loading (mol CO2/mol alkalinity) ̨ = 0.1 ̨ = 0.2 ̨ = 0.1 ̨ = 0.2

7 m MDEA 100 9 −6 ± 6 −18 ± 52 NA NA120 9 −0.3 ± 11 −31 ± 16 NA NA

7 m MDEA/2 m PZ 100 8 −3 ± 13 −19 ± 4 −2 ± 4 −6 ± 1120 8 −11 ± 11 −7 ± 20 −7 ± 3 −9 ± 5

7 m MDEA/2 m PZ 1 mM Fe2+ 100 6 NA −3 ± 13 NA −2 ± 5120 7 NA −18 ± 20 NA −11 ± 10

7 m MDEA/2 m PZ 135 6 −9 ± 8 −30 ± 15 −31 ± 3 −44 ± 2

NA, not available.Negative sign (−) of rate is indicating the loss of amine.

Table 10Rate of formation of formate and amide in presence of 98% O2, 2% CO2, with 1400 rpm agitation at 55 ◦C (Closmann et al., 2009).

Solvent CO2 loading (molCO2/mol alkalinity)

Time (days) Production rate (mmol/L h)

Formate Amideb

7 m MDEA, 1 mM Fe2+ 0.1 14 0.024 ± 0.007 0.165 ± 0.0957 m MDEA/2 m PZ, 1 mM Fe2+ 0.30 14 0.011 ± 0.001 0.010 ± 0.0017 m MDEA/2 m PZ, 0.1 mM Fe2+, 0.6 mM Cr3+, 0.1 mM Ni2+ 0.24 10 0.012 ± 0.003 0.027 ± 0.0097 m MDEA/2 m PZ, 0.1 mM Fe2+, 5 mM Cu2+a 0.23 10 0.0159 ± 0.006 0.018 ± 0.004

a The glycolate formation was <0.0001 mmol/L h.b Calculated as the difference between the formate formation with and without the hydrolysis of amide by NaOH.

Table 11First order rate constants k1, of PZ blends.

Amine system (molality) CO2 loading (molCO2/mol alkalinity)

k1, 10−7, (s−1) (135 ◦C) k1, 10−7 (s−1)(150 ◦C)

k1, 10−7 (s−1)(175 ◦C)

References

4 m PZ/4 m 2-MPZ 0.3 16 Freeman (2011)7 m MEA/2 m PZ 0.4 2.51 6.08 Davis (2009)6 m PZ/2 m BAE 0.35 1.3 Namjoshi et al. (2013)6 m PZ/2 m HMDA 0.4 1.4 Namjoshi et al. (2013)6 m PZ/2 m DAB 0.4 2.4 Namjoshi et al. (2013)5 m PZ/2 m AEP 0.3 2.0 Du et al. (2013)3.75 m PZ/3.75 m MPZ/0.5 m DMPZ 0.3 0.103 Freeman et al. (2013)3.9 m PZ/3.9 m MPZ/0.2 m DMPZ 0.3 0.084 Freeman et al. (2013)4 m PZ/4 m MPZ 0.3 0.139 Freeman et al. (2013)

llc(m

wdp(rft

tgta5r

ft

amino group (2-MPZ) enhanced degradation than PZ. More degra-dation was noticed in 1-MPZ than 2-MPZ when a methyl group wasattached at 1-MPZ. First order rate constants for each structuralanalog are reported in Table 12.

Table 12Summary of thermal degradation rates of structural analogs of PZ, reported byFreeman (2011).

Solvent Concentration(m)

Loading (mol CO2/molalkalinity)

k1, 10−9 (s−1),at 175 ◦C

PZ 8 0.3 132PD 8 0.3 84Pyr 8 0.3 1400Mor 8 0.3 50

5 m PZ/1.5 m MPZ/1.5 m DMPZ 0.3

5 m PZ/2 m MPZ/1 m DMPZ 0.3

oading of 0.3 mol CO2/mol alkalinity. Four assessments were estab-ished to evaluate the stability of structural analogs of PZ: effect ofhanging heteroatoms (PZ, PD, and Mor), ring size in monoaminesPyr, PD, and HMI), ring size in diamines (PZ and HomoPZ), and

ethyl substitution (PZ, 1-MPZ, and 2-MPZ).The effect of heteroatom in 6-membered rings (PD, PZ, and Mor)

as evaluated for PD, PZ, and Mor. This analysis compared theegradation rate obtained when the methylene group in the 4thosition on PD was replaced with an amino group (PZ) or oxygenMor). The presence of the second amino function in 6-membereding enhanced thermal degradation. Presence of multiple aminounctions increased nucleophilic reactions causing more degrada-ion.

Effect of ring size on monoamine (Pyr, PD, and HMI) was inves-igated for Pyr, PD, and HMI at 175 ◦C for 10 weeks. Formateeneration was found as a dominant degradation product for all thehree amines. HMI lost its total initial concentration within a weeknd Pyr and PD were lost, up to 55% and 8% for the same duration.- and 7-membered rings were more unstable than 6 membered

ing amines.

Effect of ring size in diamine (PZ and HomoPZ) was investigatedor PZ and HomoPZ at 175 ◦C. HomoPZ lost its all initial concen-ration with within 4 weeks, compared to PZ which lost only 30%

0.753 Freeman et al. (2013)0.137 Freeman et al. (2013)

of its initial concentration for the same duration. Degradation rateexhibited that 7-membered heterocyclic is unstable compared to6-membered PZ.

Effect of methyl substitution in PZ (PZ, 1-MPZ, and 2-MPZ)was analyzed for PZ, 1-MPZ, and 2-MPZ at 150 ◦C. Effect revealedthat the substitution of a methyl group at the �-position to the

1-MPZ 8 0.32-MPZ 8 0.3HomoPZ 8 0.3 1850HMI 8 0.3 5950

S.A. Mazari et al. / International Journal of Greenhouse Gas Control 31 (2014) 214–228 227

Table 13Summary of first order rate constants (k1) for diamines at 150 and 175 ◦C.

Solvent Concentration (m) Loading (mol CO2/mol alkalinity) k1, 10−7 (s−1) References

150 ◦C 175 ◦C

HMDA 8 0.3 1.3 Namjoshi et al. (2013)BAE 8 0.4 2.0 Namjoshi et al. (2013)DAB 8 0.4 27.2 Namjoshi et al. (2013)PDA 5.0 0.4 4.08 Hatchell (2014)Putrescine 5.0 0.4 2.69 Hatchell (2014)

5

cda(wrTwiB

6

liCnaCoP

thdsilis

osPi

opMum

itsPta

EDA 5.0 0.4

.3. Degradation of diamines

Mainly diamines were investigated in form of blends as dis-ussed previously in this manuscript. Thermal degradation of lineariamines, propane-1,3-diamine (PDA), 1,2-diaminoethane (EDA),nd butane-1,4-diamine (putrescine) was investigated by Hatchell2014). 5 m each solution loaded with 0.3 mol CO2/mol alkalinityas degraded at 150 ◦C. Degradation was found to follow first order

ate law, the values of k1 for each amine is reported in Table 13.hermal degradation of each 8 m, HMDA, BAE, and DAB at 175 ◦Cas reported by Namjoshi et al. (2013) and results are presented

n Table 13. The most stable diamines including PZ are HMDA andAE.

. Conclusions

PZ is thermally stable up to 150 ◦C compared to common aminesike MEA, DEA, MDEA, and AMP. However, degradation rate of PZncreases with increase in temperature, concentration of PZ, andO2 loading. The effect of CO2 loading on thermal degradation is yetot vivid, as higher degradation is observed up to 0.4 mol CO2/mollkalinity but it reduces when CO2 loading reaches to 0.47 molO2/mol alkalinity. Thermal degradation of PZ follows first and sec-nd order kinetics and it initiates with the nucleophilic reaction ofZ and dominant reactions are SN2 type.

PZ is oxidative degradation resistance compared to conven-ional amines. However, partial pressure of O2 and temperatureas a direct effect on the oxidative degradation. PZ does notegrade considerably in presence of construction material liketainless steel under absorber conditions but the presence of Cu2+

ncreases the oxidative degradation of PZ. Oxidation of PZ fol-ows first order kinetics. Oxidative degradation mechanism of PZs not clear but degradation may begin through a radical formingtep.

The most abundant thermal and oxidative degradation productsf PZ are FPZ, NH4

+, AEP, and EDA. MNPZ has also been found inmall concentrations. MNPZ which is formed through oxidation ofZ under absorber conditions may degrade at higher temperaturesn stripper or using through UV light.

Combination of AMP and PZ catalyzes the degradation of eachther. Unlike MDEA and PZ blend, which degrade slowly com-ared to their individual degradation rates. On the other hand,DEA/PZ blend degrades faster with CO2 loading as that of

nloaded MDEA/PZ blend. HMDA/BAE blend has competitive ther-al degradation rate compared to PZ.PZ has limited CO2 solubility due to which it may not be effective

f used alone. Blended forms of PZ can be explored as alternatives

o PZ. Though the blend of PZ with MDEA is a viable option yet PZhould also be explored with other amines. Structural analogs ofZ like Mor and PD are thermally more stable amines comparedo PZ, which should be investigated extensively for volatility, CO2bsorption rate and oxidative degradation.

7.97 Hatchell (2014)

Acknowledgement

This research is supported by High Impact Research GrantUM.C/625/1/HIR/123 from University of Malaya, 50603 KualaLumpur, Malaysia.

References

Alawode, A., (Master thesis) 2005. Oxidative Degradation of Piperazine in theAbsorption of Carbon Dioxide. University of Texas, Department of ChemicalEngineering, Austin, TX, USA.

Ashouripashaki, M., 2012. Formation and Decomposition of 1-Nitrosopiperazine inthe CO2 Capture Process. University of Texas, Austin, TX, pp. 1, online resource(xvi, 107 leaves).

Bartsch, H., Montesano, R., 1984. Relevance of nitrosamines to human cancer. Car-cinogenesis 5 (11), 1381–1393.

Bishnoi, S., Rochelle, G.T., 2002. Absorption of carbon dioxide in aqueous piper-azine/methyldiethanolamine. AIChE J. 48 (12), 2788–2799.

Boot-Handford, M.E., et al., 2014. Carbon capture and storage update. Energy Envi-ron. Sci. 7 (1), 130–189.

Bottoms, R.R., 1933. Process Fob Separating Acidic Gases. Google Patents.Brúder, P., Grimstvedt, A., Mejdell, T., Svendsen, H.F., 2011. CO2 capture into aqueous

solutions of piperazine activated 2-amino-2-methyl-1-propanol. Chem. Eng. Sci.66 (23), 6193–6198.

Chen, X., Rochelle, G.T., 2011. Aqueous piperazine derivatives for CO2 capture: accu-rate screening by a wetted wall column. Chem. Eng. Res. Des. 89 (9), 1693–1710.

Closmann, F., Nguyen, T., Rochelle, G.T., 2009. MDEA/piperazine as a solvent for CO2

capture. Energy Proc. 1 (1), 1351–1357.Closmann, F.B., 2011. Oxidation and Thermal Degradation of Methyldithanolamine/

Piperazine in CO2 Capture. University of Texas, Austin, TX, pp. 1, online resource(xxxi, 458 leaves).

Cullinane, J.T., Rochelle, G.T., 2006. Kinetics of carbon dioxide absorption into aque-ous potassium carbonate and piperazine. Ind. Eng. Chem. Res. 45 (8), 2531–2545.

Dai, N., Shah, D., Hu, L., Plewa, McKague, B., Mitch, W.A., 2012. Measurement ofnitrosamine and nitramine formation from NOx reactions with amines duringamine-based carbon dioxide capture for postcombustion carbon sequestration.Environ. Sci. Technol. 46 (17), 9793–9801.

Davis, J.D., 2009. Thermal Degradation of Aqueous Amines Used for Carbon DioxideCapture. University of Texas Libraries, Austin, TX, pp. 1, online resource (xxix,278 leaves).

Du, Y., Li, L., Namjoshi, O., Voice, A.K., Fine, N.A., Rochelle, G.T., 2013. Aque-ous piperazine/N-(2-aminoethyl) piperazine for CO2 capture. Energy Proc. 37,1621–1638.

Dugas, R., Rochelle, G., 2009. Absorption and desorption rates of carbon dioxide withmonoethanolamine and piperazine. Energy Proc. 1 (1), 1163–1169.

Dumée, L., Scholes, C., Stevens, G., Kentish, S., 2012. Purification of aqueous aminesolvents used in post combustion CO2 capture: a review. Int. J. Greenh. GasControl 10, 443–455.

Ermatchkov, V., Maurer, G., 2011. Solubility of carbon dioxide in aqueous solutionsof N-methyldiethanolamine and piperazine: prediction and correlation. FluidPhase Equilib. 302 (1–2), 338–346.

Fine, N.A., Goldman, M.J., Nielsen, P.T., Rochelle, G.T., 2013. Managing N-nitrosopiperazine and dinitrosopiperazine. Energy Proc. 37, 273–284.

Fine, N.A., Nielsen, P.T., Rochelle, G.T., 2014. Decomposition of nitrosamines in CO2

capture by aqueous piperazine or monoethanolamine. Environ. Sci. Technol.Fine, N.A., Rochelle, G.T., 2013. Thermal decomposition of N-nitrosopiperazine.

Energy Proc. 37, 1678–1686.Fostås, B., Gangstad, A., Nenseter, B., Pedersen, S., Sjøvoll, M., Sørensen, A.L., 2011.

Effects of NOx in the flue gas degradation of MEA. Energy Proc. 4, 1566–1573.Fredriksen, S.B., Jens, K.-J., 2013. Oxidative degradation of aqueous amine solutions

of MEA, AMP, MDEA, PZ: a review. Energy Proc. 37, 1770–1777.Freeman, S.A., (PhD thesis) 2011. Thermal Degradation and Oxidation of Aqueous

Piperazine for Carbon Dioxide Capture. University of Texas at Austin.Freeman, S.A., Chen, X., Nguyen, T., Rafique, H., Xu, Q., Rochelle, G.T., 2013.

Piperazine/N-methylpiperazine/N,N′-dimethylpiperazine as an aqueous sol-vent for carbon dioxide capture. Oil Gas Sci. Technol. – Revue d’IFP Energiesnouvelles.

2 of Gree

F

F

F

F

F

F

G

G

G

G

H

H

I

I

I

J

L

L

L

L

28 S.A. Mazari et al. / International Journal

reeman, S.A., Davis, J., Rochelle, G.T., 2010a. Degradation of aqueous piperazine incarbon dioxide capture. Int. J. Greenh. Gas Control 4 (5), 756–761.

reeman, S.A., Dugas, R., Van Wagener, D., Nguyen, T., Rochelle, G.T., 2009. Car-bon dioxide capture with concentrated, aqueous piperazine. Energy Proc. 1 (1),1489–1496.

reeman, S.A., Dugas, R., Van Wagener, D.H., Nguyen, T., Rochelle, G.T., 2010b. Car-bon dioxide capture with concentrated, aqueous piperazine. Int. J. Greenh. GasControl 4 (2), 119–124.

reeman, S.A., Rochelle, G.T., 2011. Thermal degradation of piperazine and its struc-tural analogs. Energy Proc. 4, 43–50.

reeman, S.A., Rochelle, G.T., 2012a. Thermal degradation of aqueous piperazine forCO2 capture. 1. Effect of process conditions and comparison of thermal stabilityof CO2 capture amines. Ind. Eng. Chem. Res. 51 (22), 7719–7725.

reeman, S.A., Rochelle, G.T., 2012b. Thermal degradation of aqueous piperazine forCO2 capture: 2. Product types and generation rates. Ind. Eng. Chem. Res. 51 (22),7726–7735.

arcia, H., Keefer, L., Lijinsky, W., Wenyon, C.M., 1970. Carcinogenicity of nitrosoth-iomorpholine and 1-nitrosopiperazine in rats. Zeitschrift für Krebsforschung 74(2), 179–184.

off, G.S., Rochelle, G.T., 2004. Monoethanolamine degradation: O2 mass transfereffects under CO2 capture conditions. Ind. Eng. Chem. Res. 43 (20), 6400–6408.

off, G.S., Rochelle, G.T., 2006. Oxidation inhibitors for copper and iron catalyzeddegradation of monoethanolamine in CO2 capture processes. Ind. Eng. Chem.Res. 45 (8), 2513–2521.

ouedard, C., Picq, D., Launay, F., Carrette, P.-L., 2012. Amine degradation in CO2

capture. I. A review. Int. J. Greenh. Gas Control 10, 244–270.atchell, G.R., 2014. Thermal Degradation of CO2 Loaded Linear Diamines. McKetta

Department of Chemical Engineering, 4th Quarterly.illiard, M.D., 2008. A Predictive Thermodynamic Model for an Aqueous Blend of

Potassium Carbonate, Piperazine, and Monoethanolamine for Carbon DioxideCapture from Flue Gas. University of Texas, Austin, TX, pp. 1, online resource(lvii, 1004 leaves).

dem, R., Wilson, M., Tontiwachwuthikul, P., Chakma, A., Veawab, A., Aroonwilas,A., Gelowitz, D., 2005. Pilot plant studies of the CO2 capture performance ofaqueous MEA and mixed MEA/MDEA solvents at the University of Regina CO2

Capture Technology Development Plant and the Boundary Dam CO2 CaptureDemonstration Plant. Ind. Eng. Chem. Res. 45 (8), 2414–2420.

slam, M., Yusoff, R., Ali, B., 2010. Role of acid and heat stable salt in the degradationprocess of alkanolamine during sour gas sweetening process. Eng. e-Trans. 5 (2),53–60, ISSN 1823-6379.

slam, M., Yusoff, R., Ali, B., Islam, M., Chakrabarti, M., 2011. Degradation studies ofamines and alkanolamines during sour gas treatment process. Int. J. Phys. Sci. 6(25), 5883–5896.

ackson, P., Attalla, M., 2011. Environmental impacts of post-combustion capture –new insights. Energy Proc. 4, 2277–2284.

epaumier, H., Picq, D., Carrette, P.L., 2009. Degradation study of new solvents forCO2 capture in post-combustion. Energy Proc. 1 (1), 893–900.

i, H., Li, L., Nguyen, T., Rochelle, G.T., Chen, J., 2013a. Characterization ofpiperazine/2-aminomethylpropanol for carbon dioxide capture. Energy Proc. 37,

340–352.

i, L., Voice, A.K., Li, H., Namjoshi, O., Nguyen, T., Du, Y., Rochelle, G.T., 2013b. Amineblends using concentrated piperazine. Energy Proc. 37, 353–369.

ijinsky, W., Epstein, S.S., 1970. Nitrosamines as environmental carcinogens. Nature225, 21–23.

nhouse Gas Control 31 (2014) 214–228

Lowe, A., Beasley, B., Berly, T., 2009. Chapter 3 carbon capture and storage (CCS) inAustralia, carbon capture: sequestration and storage. R. Soc. Chem., 65–101.

MacDowell, N., Florin, N., Buchard, A., Hallett, J., Galindo, A., Jackson, G., Adjiman,C.S., Williams, C.K., Shah, N., Fennell, P., 2010. An overview of CO2 capture tech-nologies. Energy Environ. Sci. 3 (11), 1645–1669.

Mangalapally, H.P., Notz, R., Hoch, S., Asprion, N., Sieder, G., Garcia, H., Hasse, H.,2009. Pilot plant experimental studies of post combustion CO2 capture by reac-tive absorption with MEA and new solvents. Energy Proc. 1 (1), 963–970.

McMurry, J., 2000. Organic Chemistry, 5th ed. Brooks/Cole, Pacific Groove, CA.Namjoshi, O., Li, L., Du, Y., Rochelle, G., 2013. Thermal degradation of piperazine

blends with diamines. Energy Proc. 37, 1904–1911.Nielsen, P.T., Li, L., Rochelle, G.T., 2013. Piperazine degradation in pilot plants. Energy

Proc. 37, 1912–1923.Norwegian Climate and Pollution Agency, 2011. Permit for Activities Pursuant to the

Pollution Control Act. CO2 Technology Centre, Mongstad,DA.Pai, S., Shirke, A., Gothoskar, S., 1981. Long-term feeding study in C17 mice admin-

istered saccharin coated betel nut and 1,4-dinitrosopiperazine in combination.Carcinogenesis 2 (3), 175–177.

Rao, A.B., Rubin, E.S., 2002. A technical, economic, and environmental assessment ofamine-based CO2 capture technology for power plant greenhouse gas control.Environ. Sci. Technol. 36 (20), 4467–4475.

Ridd, J.H., 1961. Nitrosation, diazotisation, and deamination. Q. Rev. Chem. Soc. 15(4), 418–441.

Rochelle, G., Chen, E., Freeman, S., Van Wagener, D., Xu, Q., Voice, A., 2011. Aqueouspiperazine as the new standard for CO2 capture technology. Chem. Eng. J. 171(3), 725–733.

Rochelle, G.T., 2009. Amine scrubbing for CO2 capture. Science 325 (5948),1652–1654.

Samanta, A., Bandyopadhyay, S.S., 2009. Absorption of carbon dioxide into aqueoussolutions of piperazine activated 2-amino-2-methyl-1-propanol. Chem. Eng. Sci.64 (6), 1185–1194.

Sexton, A.J., 2008. Amine Oxidation in CO2 Capture Processes. University of Texas atAustin.

Sherman, B., et al., 2013. Carbon capture with 4 m piperazine/4 m 2-methylpiperazine. Energy Proc. 37, 436–447.

Smith, J.G., 2011. Organic Chemistry, 3rd ed. McGraw-Hill.Thong, D., Dave, N., Feron, P., Azzi, M., 2012. Environmental Impacts of Amine-based

CO2 Post Combustion Capture (PCC) Process. Activity 3: Process Modelling forAmine-based Post-combustion Capture Plant, Deliverable 3.2 Progress Report.CSIRO.

Voice, A.K., Closmann, F., Rochelle, G.T., 2013. Oxidative degradation of amines withhigh-temperature cycling. Energy Proc. 37, 2118–2132.

Wang, M., Lawal, A., Stephenson, P., Sidders, J., Ramshaw, C., 2011. Post-combustionCO2 capture with chemical absorption: a state-of-the-art review. Chem. Eng.Res. Des. 89 (9), 1609–1624.

Wang, T., Jens, K.-J., 2012. Oxidative degradation of aqueous 2-amino-2-methyl-1-propanol solvent for postcombustion CO2 capture. Ind. Eng. Chem. Res. 51 (18),6529–6536.

Wang, T., Jens, K.-J., 2014. Oxidative degradation of aqueous PZ solution and AMP/PZ

blends for post-combustion carbon dioxide capture. Int. J. Greenh. Gas Control24, 98–105.

Zhang, X., Zhang, C.-F., Qin, S.J., Zheng, Z.-S., 2001. A kinetics study on the absorptionof carbon dioxide into a mixed aqueous solution of methyldiethanolamine andpiperazine. Ind. Eng. Chem. Res. 40 (17), 3785–3791.