close

Вход

Забыли?

вход по аккаунту

?

j.ijggc.2018.08.004

код для вставкиСкачать
International Journal of Greenhouse Gas Control 78 (2018) 125–134
Contents lists available at ScienceDirect
International Journal of Greenhouse Gas Control
journal homepage: www.elsevier.com/locate/ijggc
Using sodium thiosulphate for carbon steel corrosion protection against
monoethanolamine and 2-amino-2-methyl-1-propanol
Samara A. Sadeek, Daryl R. Williams, Kyra L. Sedransk Campbell
T
⁎
Department of Chemical Engineering, Imperial College, South Kensington Campus, London, SW7 2AZ, United Kingdom
A R T I C LE I N FO
A B S T R A C T
Keywords:
Post-combustion CO2 capture
Amine
Siderite
Corrosion
Carbon steel
Corrosion inhibitor
The corrosion performance of carbon steel (C1018) with inhibitor sodium thiosulphate (STS) has been compared
to C1018 (without an inhibitor) and stainless steel (SS316 L) at 80 and 120 °C to determine the feasibility of use
in a post-combustion CO2 capture plant (PCCC). The corrosivity of variable ratio monoethanolamine (MEA) and
2-amino-2-methyl-1-propanol (AMP) aqueous solvent blends were assessed after seven days using a gravimetric
method for mass change, Fe ion solution concentration (ICP–OES), surface imaging (SEM) and analytical
techniques (EDX and XRD). At low concentrations of MEA (25%), the use of the corrosion inhibitor is ineffective
as it prevents the formation of protective films naturally developed by AMP. The performance of the inhibitor
with MEA–AMP blends was noteworthy at higher MEA concentrations of 50 and 75%. In these cases, reduced
corrosion rates were observed through gravimetric and ICP analyses. Imaging by SEM showed reduced surface
corrosion and adsorption of STS-derived species. Since these higher MEA concentration solutions offer better
CO2 loadings, but still some exhibit corrosive effects when used alone on carbon steel, the use of STS can
facilitate an economical usage of carbon steel for PCCC plants.
1. Introduction
Reduction in greenhouse gas emissions, specificallycarbon dioxide
(CO2), is paramount in mitigating continued climate change. Since
fossil fuel based power generation plants are one of the largest CO2
emissions sources (Freund, 2003), development of technologies to reduce CO2 concentrations from the flue gases emitted from these plants
is essential. Due to its applications in natural gas sweetening, amine
scrubbing is the most developed post-combustion CO2 capture (PCCC)
technology to date. Despite this experience, financial and technical
limitations, including corrosion, still exist. Stainless steel is therefore
the recommended material for PCCC pilot plants (Kittel and Gonzalez,
2014), but this is neither economically feasible nor a panacea due to
localised corrosion concerns. Consequently, some recent studies have
identified means of using carbon steel under specific process and solvent conditions (Campbell et al., 2016; Yu et al., 2016).
Importantly, amines are not inherently corrosive, only demonstrating this potential when exposed to CO2 (DuPart et al., 1993). Since
varied CO2 uptake mechanisms exist for different amine classes, corrosive behaviour and the subsequent reaction between an Fe substrate
and amines cannot be generalized. Both primary (1°) and secondary (2°)
amines react with CO2 to form carbamate ions (Eq. (1)) and increases
solution pH through formation of a protonated amine (Belarbi et al.,
⁎
2016). One such 1° amine, monoethanolamine (MEA), has been historically used (30% by weight) due to its fast CO2 uptake kinetics.
However, it exhibits low loading capacity and undergoes thermal and
oxidative degradation, all of which reduce its desirability as a solvent
for the PCCC process. The corrosive behaviour of 1° amine solvents, on
Fe substrates, is linked to the formation of carbamates by some studies
(DuPart et al., 1993; Tomoe et al., 1996) which is favoured at reduced
temperatures (Kittel and Gonzalez, 2014). This behaviour was observed
in a previous study where 5 M MEA at 80 °C, proved to be more corrosive than 120 °C (Sadeek et al., 2018). By contrast to 1° amines, 3°
amines react with CO2 indirectly, requiring CO2 hydrolysis (Eq. (2)).
RNH2(aq) + CO2(aq) ↔ RNH2 COO (aq)
RNH2(aq) + RNH2 COO (aq) ↔
″
RNH3+(aq)
RR′R N(aq) + H2 O (l) + CO2(aq) ↔
(1a)
+
RNHCOO−
R′R″R‴NH+
(aq)
+
(aq)
HCO3−(aq)
(1b)
(2)
A subset of 1° amines are those which contain bulky side groups
around the N atom, causing steric hindrance and are referred to as
sterically-hindered (SH) amines. Such structures cause slow and unstable carbamate formation (Bougie and Iliuta, 2012; Dubois and
Thomas, 2012; Mazari et al., 2014; Sartori and Savage, 1983; Zhao
et al., 2011).The capture mechanism employed by SH amines, like 2-
Corresponding author.
E-mail address: k.sedransk@imperial.ac.uk (K.L. Sedransk Campbell).
https://doi.org/10.1016/j.ijggc.2018.08.004
Received 8 May 2018; Received in revised form 18 July 2018; Accepted 7 August 2018
1750-5836/ © 2018 Elsevier Ltd. All rights reserved.
International Journal of Greenhouse Gas Control 78 (2018) 125–134
S.A. Sadeek et al.
Gonzalez, 2014). Recent studies on low-toxic inhibitors have included
copper carbonate (Soosaiprakasam and Veawab, 2009), 2-metaptobenzimidazole (Zheng et al., 2015) and sodium thiosulphate (STS)
(Sadeek et al., 2018; Srinivasan et al., 2013). Long term testing and
surface analyses were used to investigate the performance of STS in
reducing corrosion rates in the presence of both 5 M MEA and MEAMDEA blends at 120 and 80 °C. This inhibitor is inorganic and designed
to impede corrosion by adsorption of the negative anion (S2O32− (TS))
(Roberge, 2012). Adsorption of TS was detected by EDX on coupons
exposed to tested solvents with a high concentration of MEA. However,
reaction of TS was observed and resulted in the formation of such
species as Na2CO3, reducing the efficacy of the TS adsorption to the
surface. Additionally, the promotion of metal-oxide products was also
observed, though this proved to be less problematic than the formation
of Na2CO3. However, in the 25% by weight MEA blended systems, the
inhibitor presence altered the natural protective mechanism, siderite
formation, characteristic of MDEA. That is, without the inhibitor,
naturally growing siderite films offer good corrosion protection; by
contrast, when the siderite was not formed in the presence of STS then
corrosion was considerably worse (Sadeek et al., 2018).
The work reported herein assesses the performance of STS inhibited
carbon steel, relative to both uninhibited carbon steel and stainless steel
coupons, in MEA-AMP blended solutions at process conditions found in
the high temperature units of PCCC plants. This industrially relevant
study is essential to understanding the performance of 1) MEA-AMP
blends relative to the MEA-MDEA blends and 5 M MEA, the current
industry standard and 2) the performance of STS as an inhibitor with
potentially usable blends under challenging PCCC process conditions.
With a wider range of applications employing amines, particularly
natural gas sweetening, the findings have even farther reaching implications.
amino-2-methyl-1-propanol (AMP), has been suggested to employ direct CO2 capture, producing an unstable carbamate which is then hydrolysed to form bicarbonate (Sedransk Campbell et al., 2017) (Eq. (3)).
The use of the SH amine AMP is popular due to its enhanced resistance
to thermal (Hatchell, 2015) and oxidative (Wang and Jens, 2013) degradation.
AMP(aq) + CO2(aq) ↔ AMPCOO(−aq)
AMPCOO(−aq)
unstable
unstable
+ H2 O(l) ↔ AMPH(+aq) + HCO3−(aq)
(3a)
(3b)
The production of bicarbonate associated with SH amine CO2 capture acts as a source of carbonate species (Eq. (4a)). These then react
with Fe ions present from Fe substrate oxidation (Eq. (4b)) and results
in Fe carbonate (FeCO3, siderite) formation (Eq. (4c)). Siderite formation can be advantageous as it has been demonstrated to provide corrosion protection for the underlying metal substrate (Fe or a carbon
steel with high Fe content) (Campbell et al., 2016; Farelas et al., 2013;
Guo and Tomoe, 1999; Han et al., 2011; Tanupabrungsun et al., 2013;
Yu et al., 2016).
H2 O(l) + HCO3− (aq) ↔ H3 O+(aq) + CO32 −(aq)
Fe(s) ↔
Fe 2 +
(aq)
+
2e−
Fe 2 +(aq) + CO32 −(aq) ↔ FeCO3(s)
(4a)
(4b)
(4c)
In addition to the improved solvent performance demonstrated by
AMP, relative to MEA, the reduced corrosion of Fe is highly beneficial.
Two reports investigated the corrosion behaviour of 5 M AMP (and
MEA) saturated with CO2 at 80 and 120 °C using long term immersion
testing and surface analytical techniques (SEM, EDX, XRD) (Campbell
et al., 2016; Yu et al., 2016). In these tests, significant siderite layers
formed on the surface with the concomitant corrosion reduction (as
measured by Fe ion content in solution) relative to MEA. In agreement,
using electrochemistry for 3 M AMP saturated with CO2 at 80 °C
(Veawab et al., 1999a), AMP was shown to be less corrosive than MEA.
However, it has also been suggested to be more corrosive than MDEA,
DEA and PZ (Gunasekaran et al., 2013). Some conflicting electrochemically determined corrosion rates have also shown that 3 M AMP
at 80 °C has corrosion rates an order of magnitude lower than that of
MEA for lean solutions (low CO2-loading), but comparable to MEA
under saturated CO2 conditions (Veawab et al., 1999b)
Although the use of SH amines (e.g. AMP) has been suggested as an
attractive solvent for amine scrubbing (Mandal et al., 2001), blending
with 1° amines has proven advantageous (Xiao et al., 2000). The
strategy of blending a 1° or 2° amine with a SH or 3° amine to improve
solvent performance has been employed (Conway et al., 2015; Freeman
et al., 2013; Kemper et al., 2011; Li et al., 2013; Mandal and
Bandyopadhyay, 2006; Wang and Jens, 2013; Zhao et al., 2011). These
mixtures take advantage of the fast reaction kinetics characteristic of 1°
and 2° amines, and the high loading capacities and low regenerative
requirements of SH and 3° amines. One such example which has shown
promise are blends containing the SH amine AMP (Kemper et al., 2011;
Mandal et al., 2001; Mandal and Bandyopadhyay, 2006; Xiao et al.,
2000). In an investigation of the corrosive behaviour of 5 M amine
blends containing MEA and AMP, Yu et al. (2016) observed: 1) reduced
corrosion with decreased MEA concentration (demonstrated by decreased bulk Fe ions in solution) and 2) siderite crystal growth on
coupons exposed to blends containing greater than 50% AMP. While
promising, not all cases offer sufficient corrosion protection from
siderite. As such, additional corrosion control methods may be required
for some blend compositions.
In addition to blends, a complementary approach to facilitate the
use of carbon steel in PCCC plants is inhibitor addition. Effective corrosion inhibitors for natural gas sweetening have been identified including arsenic, antimony and vanadium compounds, but these are
being phased out due to environmental and health concerns (Kittel and
2. Materials and method
2.1. Corrosion experiments
The chemicals monoethanolamine (MEA, Sigma ≥ 98%), 2-amino-2methyl-1-propanol (AMP, Sigma ≥ 99%), sodium thiosulphate (STS,
(Na2S2O3), Sigma ≥ 99%) (Fig. 1) were used as received. Three 250 mL
aqueous amine blends of solution composition 75, 50, 25 % by weight
of MEA, and an MEA control were prepared each with a total amine
content of 5 M (the balance being DI water). Two batches were prepared, where 0.625 g (2500 mg L−1) of STS (inhibitor) was added to
one batch. Round bottom double neck flasks (250 mL) with a reflux
condenser (to minimise vapour loss) were used on Dry-syn blocks (with
an aluminium heat transfer plate base) (Fig. 2). Purging with N2 was
conducted for two h (400 mL min−1) at ambient temperature and
pressure, thereby creating an O2-free environment; the solution was
subsequently loaded with CO2 at 400 mL min−1 for two h. After this
period the CO2 flow rate was reduced to 20 mL min−1 and bubbled
continuously for the duration of the experiment (i.e. maintaining a
constant CO2 loading). The addition of the inhibitor was by direct addition into the reaction vessel via single injection at the start of the
experiment. The temperature was controlled at either 80 or 120 °C to
mimic the conditions of a heat exchanger or stripper. All aqueous amine
solutions saturated with CO2 attained a pH of 7.81 ± 0.03.
Glass bead blast finished carbon (C1018) and stainless (SS316 L)
steel coupons (Dimensions: 76.20 × 12.70 x 1.59 mm, Type C1018: Fe:
98.85%, C: 0.17%, Mn:0.80%, Cr: 0.04%, Al: 0.04%, ≤0.02%: P, S, Si,
Cu, Ni, Sn, N, V,B, Ti, Co and Type SS316L: Fe: 67%, Cr:17%, Ni: 12%,
Mn: ≤2%, Mo: 2.5%, Si: ≤0.75%, C ≤ 0.03%, P: ≤0.045%, S:
≤0.03%, N: ≤0.1%) (Alabama Specialty Products) were prepared by
washing with DI water and acetone. The coupons were weighed and
then suspended into the prepared aqueous amine solutions with a PTFE
string through the top hole (radius of 6.35 mm) (Fig. 2).
Immersion of the coupons was for seven days (168 h), after which
126
International Journal of Greenhouse Gas Control 78 (2018) 125–134
S.A. Sadeek et al.
they were removed, weighed, washed and re-weighed. The net mass
change was determined by dividing the difference in coupon mass before and after immersion (minitial−mfinal) by the mass before immersion
(minitial) (Eq. (5)). A positive net mass change indicates net mass loss,
whereas a negative change indicates net mass gain. The inhibition efficiency (IE) was calculated by dividing the difference in the net mass
change of the uninhibited and the inhibited coupons by the uninhibited
net mass change (Talati and Gandhi, 1983).
Net Mass
IE (%) =
Change (%) = 100 x
minitial − mfinal
minitial
(5)
(minitial − mfinal )uninhibited−(minitial − mfinal )inhibited
(minitial − mfinal )uninhibited
(6)
2.2. Analytical techniques
The solutions were retained and samples (n = 3) were prepared in
nitric acid matrix solution (HNO3, 7.433 M) and the Fe ion concentration was measured using ICP-OES (Perkin Elmer OES Optima 2000DV).
Surface imaging used an SEM (Jeol JSM-6400), where samples were
coated with 10 nm Au (EMITECH K550). The EDX functionality was
used to determine the surface chemical composition where the detector
was calibrated with Co. Crystalline species were studied using XRD
(PAnalytical X-Pert X-Ray Diffractometer), with 2θ of 10 and 100° at a
step size of 0.0334 and 40 s scan time.
Fig. 1. Chemical structures of (A) STS (B) MEA (C) AMP.
2.3. Carbon dioxide loading experiments
Aqueous amine solutions (25 mL) were prepared by purging with N2
and subsequently bubbling CO2 through each solution at 40 mL min−1
for two h at 20 °C. Using GC-TCD, the concentration of CO2 was determined. As a note: CO2 loading is a function of temperature, therefore
the loading at 20 °C is representative of the efficacy of the blends.
3. Results and discussion
3.1. 5 M MEA and MEA-MDEA
The performance of SS316 L and C1018 with and without inhibitor,
in CO2-loaded 5 M MEA and MEA-MDEA solvent blends at 80 and
120 °C were previously investigated (Sadeek et al., 2018). The use of
STS inhibited C1018 as an economical alternative to SS316 L in high
temperature sections (≥80 °C) of the PCCC plant has been demonstrated. The inhibitor mechanism and performance was found to be
influenced by temperature and solution composition. Traditional STS
inhibition, through adsorption of TS, was determined using EDX detection of S. This mechanism significantly reduced corrosive attack for
lower temperature (80 °C) and higher MEA concentration cases (5 M
MEA, 75% MEA-MDEA). By contrast, at higher temperatures (120 °C)
the formation of Na2CO3 hampered STS performance (5 M MEA, 75%
MEA-MDEA) as a corrosion inhibitor. A further decrease in MEA concentration (50% MEA-MDEA, 120 °C) resulted in the formation of
metal-oxide species, which provide some apparent protection against
continued Fe oxidation at the surface. A comparison between the different tested blends MEA-MDEA (Sadeek et al., 2018) and MEA-AMP
will follow the discussion of MEA-AMP results.
3.2. MEA-AMP
Coupon immersion in CO2 saturated aqueous amine solutions resulted in Fe lost which was accounted for, though not entirely, using an
array of solution and surface based analyses (Fig. 3). Oxidation of the
coupon releases Fe ions resulting in the highest concentration at, or
near, the surface. A decreasing concentration moving away from the
coupon is due, in part, to diffusion. Experimentally, the bulk solution
Fig. 2. Corrosion vessel.
127
International Journal of Greenhouse Gas Control 78 (2018) 125–134
S.A. Sadeek et al.
Table 1
Inhibition efficiency of STS in tested systems.
MEA Composition (% by weight)
75 % MEA
50 % MEA
25 % MEA
Inhibition Efficiency
80 °C
120 °C
90 %
94 %
−14 %
68 %
92 %
17 %
tested at both 80 and 120 °C (Fig. 4A). By contrast, C1018 coupons
without inhibitor (referred to subsequently as uninhibited) showed the
highest mass loss, which was solution concentration dependent where
the mass loss decreases with reduced concentration of MEA. However,
for the same solution composition, no significant influence of temperature is observed. At both 75 and 50% by weight MEA, reduced mass
loss is observed for inhibited, as compared to uninhibited coupons (this
is not true for 25% MEA). The net mass change results of inhibited
coupons exposed to 25% MEA solutions are very similar to those exposed to 50% MEA blends for both temperatures. Although presented
gravimetric results suggest that corrosivity of the varied solutions is not
substantially influential by temperature, surface analytical results
(presented in subsequent sections) must be considered before accepting
such trends.
Solutions containing 50% MEA with STS, at both 80 and 120 °C,
showed the best IE overall > 90%. At a higher concentration of 75%
MEA similar success was observed at 80 °C, but not at 120 °C, where the
IE dropped to 68% (Table 1). Of significant interest is the apparent lack
of effect from the inhibitor for cases with 25% MEA. Importantly, net
mass change results represent a multi-faceted system. As explained
previously (Fig. 3), the loss of metal ions (Fe) can be: 1) returned to the
surface as product species, 2) remain in solution, or 3) deposit on another surface. The mass change not only indicates the loss of all metal
ion species but also the product species deposited back to the original
metal coupon source. Therefore, this metric does not close the mass
balance. In the case of an inhibitor the mass change is further obscured
due to addition of product to the surface not previously accounted for.
However, these qualitative assessments do serve as a rudimentary
method for assessing gross sample change without sample destruction.
Used with additional analyses, including metal ion solution concentrations and coupon surface analysis, a more complete picture of the
corrosion can be ascertained, the results of which are explored in this
Fig. 3. Schematic of corrosion phenomena present in reaction vessel and analytical techniques used to investigate corrosion behaviour (Sadeek et al., 2018).
sampled at the end of the experiment gives the final concentration of Fe
ions (using ICP-OES). However, the concentration of Fe ions is altered
by formation of species such as FeCO3, which can precipitate on the
solid surface. In the case of FeCO3, precipitation occurs if the concentration of Fe ion and dissolved carbon species (HCO3− or CO32−)
are sufficiently high such that the solubility limit of FeCO3 is exceeded
(Nordsveen et al., 2003). The net mass change represents Fe lost from
the surface due to oxidation as well as Fe re-deposited on the surface as
a corrosion product and in some cases, deposition of inhibitor based
species. However, deposition of Fe-based species on walls of the reaction vessel during the experiment due to scale formation, are not
quantified in this study. The Fe ion solution concentration measured by
ICP-OES is therefore not representative of the total Fe lost from the
coupon. Whilst apparent corrosion trends are suggested by both
gravimetric and solution analyses, surface characterisation analyses
(SEM, EDX, XRD) are needed for validation. As such, surface analyses
are presented after gravimetric and solution analytical results.
3.2.1. Gravimetric and solution analyses
The SS316 L coupons in all MEA-AMP blends underwent insignificant changes in mass before and after exposure to the solutions when
Fig. 4. (A) Change in mass measured for coupons exposed to MEA-AMP blends at 80 °C and 120 °C (B) Concentration of Fe ion in solutions measured using ICP–OES
for MEA-AMP at 80 °C and 120 °C.
128
International Journal of Greenhouse Gas Control 78 (2018) 125–134
S.A. Sadeek et al.
solution at 80 °C is distressed and slightly darkened (Fig. 6C) while that
exposed at 120 °C is just darkened (Fig. 6F). Imaging of the sample
exposed at 80 °C (Fig. 7A) shows a non-uniform surface with layered
plate-like structures and patches exposing the underlying metal
substrate; XRD analysis confirms the presence of ferrite (Fe),
cementite (Fe3C), magnetite (Fe3O4) and chukanovite (Fe2(OH)2CO3)
(Fig. 7D). The exposure of cementite is as a result of oxidation of the
labile ferrite phase. The presence of oxide and chukanovite crystals may
be from highly localised nucleation with slow growth, resulting from
desirable local conditions created within the porous cementite layer
(Farelas et al., 2013). Further, the high bulk Fe ion concentration in
solution (Fig. 4B), significant scale build-up on vessel walls and mass
loss recorded for the metal substrate (Fig. 4A) resulting from exposure
of the coupon demonstrate that the removal of Fe ions, by oxidation,
and subsequent transport into the bulk solution, occur readily. By
contrast, analysis of the sample tested at 120 °C indicates nominal
corrosion with no product or cementite detection (Fig. 8A and D) and
an apparently roughened surface. Further, the absence of corrosion
product indicates that there is no physical barrier to prevent, or reduce,
Fe loss from the surface into the bulk solution. This results in relatively
high Fe ion solution concentrations (Fig. 4B) and noteable scale buildup on vessel walls. Correspondingly, a high mass loss is measured for
the metal substrate (Fig. 4A).
The coupons exposed to the 50% MEA system at both temperatures
appear to be darkened (Fig. 6B and E) but with distinct surface topographies (Figs. 7B and 8 B). The sample tested at 80 °C developed a
porous, mixed structure layer consisting predominantly of botryoidal
crystals with a few bladed habits (Fig. 7B). The botryoidal structure is
characteristic of siderite (Mineralogy, 2016), while the few dispersed
bladed habits are thought to be chukanovite (Fe2(OH)2CO3)
(Webmineral, 2016). Elemental analysis using EDX shows Fe: O: C% by
weight ratio of 4: 3: 1 (Supplementary material, Fig. C.1) which supports the hypothesised mix of products on an exposed Fe substrate.
Results from XRD further clarify this to confirm ferrite, chukanovite and
siderite peaks at 80 °C (Fig. 7E). The presence of these products explains
the reduced Fe ion concentration in the bulk solution (Fig. 4B) due to
both product formation on the surface and reduced Fe oxidation at the
surface. The coupon exposed to the 50% MEA solution at 120 °C demonstrated very different behaviour from that exposed to the lower
temperature solution. No product formation is observed through imaging (Fig. 8B) and was confirmed by the XRD detecting only ferrite (i.e.
no corrosion product). In fact, the corrosion behaviour of this solution is
very similar to that previously discussed for the 75% MEA solution at
the same temperature (120 °C). This lack of corrosion product species
contributes to the high bulk Fe ion solution concentration (Fig. 4B) and
mass loss (Fig. 4A) reported. The 75 and 50% MEA cases point to a
critical finding that reduced corrosion on C1018 metal coupons is
possible at elevated temperatures. Previous reports have suggested that
corrosion should increase with increased temperature in CO2 loaded
study.
Consistent with net mass change results (Fig. 4A), the Fe ion concentration in solutions used to immerse SS316 L coupons are very low at
all temperature and compositions tested (Fig. 4B). Since SS316 L has a
surface layer of CrO, initial Cr removal would occur and is found to be
of a minimal concentration in the bulk solution (Supplementary material, Table E.1). The Fe ion concentration of solutions used to corrode
C1018 coupons (uninhibited) at 75% MEA is the highest of all cases
tested and was nominally affected by temperature. Both vessels used to
house 75% MEA systems were completely coated in scale. Consequently, the Fe ion solution concentrations reported in Fig. 4B are an
underestimate of the total Fe loss from coupons under these conditions.
At reduced concentrations of MEA (50 and 25%) Fe ion concentration at
80 °C is significantly lower than at 120 °C. While the change in Fe ion
concentration is significant, the impact on net mass change is likely to
be very small. The similarities at 120 °C are important to consider in the
context of loss of Fe ions to deposition on the vessel wall, which is only
accounted for qualitatively. At 50% MEA a somewhat thick coating was
observed and at 25% no coating formed. This suggests that enhanced
oxidation of the surface with increasing MEA concentration results in
increased Fe ions available for deposition on the vessel walls. But this
also introduces the question of whether a saturation limit of Fe ions in
solution is being reached, and particularly the relationship between
solution saturation limits in such blends. At 25% MEA the absence of
scale indicates that Fe oxidised at the surface is almost exclusively redeposited on the coupon itself or in solution. This confirms that there is
reduced Fe oxidation on the coupon surface at this blend composition
for both temperatures tested. Additionally, it is suggested that the saturation limit of Fe ions in this blend has not been reached.
In strong contrast to the uninhibited C1018 coupons tested, all inhibited systems regardless of MEA concentration and temperature, have
no statistically significant difference in Fe ion concentration.
Additionally, the vessels all remained relatively uncoated.
3.2.2. Surface analyses
3.2.2.1. Stainless steel. Coupons (SS316 L) before (Fig. 5A) and after
(Fig. 5B) immersion in MEA-AMP solutions were indistinguishable for
all compositions and temperatures (Supplementary material, Section
B). Minimal changes to the surface composition were also confirmed by
EDX (Supplementary material, Section B). Although detection of taenite
(Fe, Ni) by XRD (Fig. 5C) could be interpreted as evidence of removal of
CrO layer, ICP-OES results indicate low Cr concentration in solution
(Supplementary material, Table E.1) suggesting minimal removal.
Instead, taenite detection is as a result of the XRD measurement
itself, where the angle of incidence results in X-ray penetration depth
exceeding the thickness of the CrO layer, known to be 3–5 nm thick
(Kerber and Tverberg, 2000).
3.2.2.2. Carbon steel. The C1018 coupon exposed to the 75% MEA
Fig. 5. SEM image at: (A) 250x, uncorroded SS316 L sample, (B) 500x SS316 L coupon exposed to MEA-AMP at 120 °C, (C) XRD spectrum with the standard for
taenite.
129
International Journal of Greenhouse Gas Control 78 (2018) 125–134
S.A. Sadeek et al.
Fig. 6. Photograph of C1018 coupons from the corrosion vessels containing MEA-AMP blends of variable ratio: (A) 25% MEA, (B) 50% MEA, (C) 75% MEA, at 80 °C
and: (D) 25% MEA, (E) 50% MEA, (F) 75% MEA at 120 °C.
Fig. 7. Surface analysis of C1018 coupon exposed to variable ratio MEA-AMP blends at 80 °C: using SEM imaging of: (A) 75% MEA 500x, (B) 50% MEA 500x, (C) 25%
MEA at 500x; and using XRD with the standards of detected crystals of: (D) 75% MEA, (E) 50% MEA, (F) 25% MEA.
(Fig. 7B). The relative intensity of the siderite peak is also notably
higher (Fig. 7E and F). Although increased surface coverage is attained
when the surface is exposed to the 25% MEA solution relative to the
50% solution (at 80 °C), and explains the significant improvement in
mass loss results (Fig. 4A) reported, only a minimal decrease in bulk Fe
ion concentration has been reported (Fig. 4B). The relative scale build
up on vessel walls provides an explanation; after immersion, the vessel
used to house the 50% MEA solution was completely coated in scale
(Fe-based product) while minimal scale formed on the vessel used to
test the 25% MEA solution. As such, the Fe ion solution concentration
detected in solution after the immersion experiment for the 50% MEA
system is much lower than that actually lost from the coupon due to Fe
oxidation.
The coupon exposed to the 25% MEA solution at 120 °C has a well
developed underlying bed of rhombohedral structures including some
protruding mixed sized hexagonal crystals topped with tiny botryoidal
habits (all of which are characteristic of siderite (Mineralogy, 2016))
(Fig. 8C), distinct from that at 80 °C. Observed crystal habits are
amine systems (Kittel and Gonzalez, 2014). By contrast, DuPart et al.
(1993) argues that carbamate formation, which is favoured by reduced
temperatures (Kittel and Gonzalez, 2014), is responsible (directly or
indirectly) for amine solvent corrosivity, and therefore results in higher
rates of corrosion at lower temperatures. Moreover, the reduced concentration of dissolved CO2 species (e.g. HCO3−) in the aqueous component of the solution may contribute to the reduced corrosivity at
increased temperature.The aforementioned contradictions in the literature, and the findings presented herein highlight the complexities of
carbamate, bicarbonate and corrosion product formation. Therefore, to
make such generalisations is inappropriate. Interestingly, similar results
were also observed when C1018 coupons were exposed to 5 M MEA
(Sadeek et al., 2018; Zheng et al., 2016) and 75% MEA – 25% MDEA
blends (5 M) (Sadeek et al., 2018) at 80 and 120 °C.
The C1018 coupon exposed to the 25% MEA blend at 80 °C attained
an orange hue (Fig. 6A) with apparent botryoidal structures (characteristic of siderite (Mineralogy, 2016)) (Fig. 7C) and somewhat better
surface coverage than the coupon tested in the 50% MEA blend at 80 °C
130
International Journal of Greenhouse Gas Control 78 (2018) 125–134
S.A. Sadeek et al.
Fig. 8. Surface analysis of C1018 coupon exposed to MEA-AMP blends at 120 °C: using SEM imaging of: (A) 75% MEA 500x, (B) 50% MEA 500x, (C) 25% MEA 500x,
inset 2000x; and using XRD with standards of detected crystals: (E) 75% MEA, (F) 50% MEA, (G) 25% MEA.
inhibition mechanisms; the primary mechanism reduces Fe oxidation
by surface adsorption of TS (S2O32−) (Roberge, 2012). A second step
results in an alternate inhibition mechanism. An electrochemical
reduction of TS results in the formation of melkovite (Eq. (7)), which
inhibits corrosion as a physical barrier (Brogan, 2011; Roberge, 2012).
Whilst both samples show the presence of adsorbed inhibitor, there is a
much larger quantity on the higher temperature sample. These
observations are similar to that previously reported when STS was
tested for CS corrosion inhibition in 5 M MEA at 80 °C (Srinivasan et al.,
2013).
supported by EDX (Supplementary material, Fig. C.2) and XRD (Fig. 8F)
results, confirming the presence of siderite. The variation in siderite
crystals has also been reported in other literature (though with different
amine solutions) (Campbell et al., 2016; Yu et al., 2016). Sedransk
Campbell 2016 (Campbell et al., 2016) explains that whilst crystals
produced may be the same, asymetrical growth of different facets can
occur as a result of non-uniform solute adsorption on facets of the
nascent crystal. This influences the growth kinetics of facets differently
thereby affecting the final crystal shape and size attained, i.e. crystal
habit. Although siderite films provide surface protection, the Fe ion
solution concentration detected is high (Fig. 4B) given the surface
coverage observed. It is likely that the slow formation of these crystals
leads to this high Fe concentration. This may be mitigated now that the
protective layer is formed, as has been reported previously (Campbell
et al., 2016; Yu et al., 2016).
Relative to the 25% MEA 80 °C system (Fig. 7C), the siderite crystals
formed at 120 °C are well developed (Fig. 8C) and provide apparently
superior surface coverage indicating that siderite growth and stability is
promoted at higher temperatures in agreement with literature
(Bénézeth et al., 2009; Sun et al., 2009).
S2 O32 −(aq) + Fe 2 +(aq) → FeS(s) + SO32 −(aq)
(7)
The low Fe bulk solution concentrations (Fig. 4B) and net mass
change (Fig. 4A) relative to the uninhibited systems, demonstrate that
the removal of Fe ions from the surface does not occur readily, highlighting the efficacy of the inhibitor. A high IE of 90% was attained at
80 °C (Table 1), where the primary inhibition mechanism seems predominant; by contrast, a reduced IE of 67% is observed at 120 °C
(Table 1), where the secondary mechanism appears prevalent. This
suggests that the surface adsorption of TS is more effective at impeding
corrosion than protection by melkovite in the case described.
The coupon exposed to the 50% MEA blend at 80 °C is not discoloured but distressed (Fig. 9B) and orange at 120 °C (Fig. 9E). The
coupon surface exposed at 80 °C is covered in a very thin, grainy layer
presumed to be TS, with the underlying surface seemingly uncorroded
(Fig. 10B). The 120 °C sample is covered in a similar grainy but apparently thicker layer (Fig. 11B). The TS film presence is corroborated
by EDX measurements (Supplementary material, Section D) showing a
small, but certain, S content (0.02% by weight at 80 °C and 0.3% by
weight at 120 °C). The only crystalline structure confirmed by XRD, at
both temperatures, is ferrite (Figs. 10E and 11 E). In comparison with
the 75% MEA systems, which were both incompletely covered in
3.2.2.3. Inhibited carbon steel. The coupon surface exposed to the 75%
MEA system at 80 °C is coated with a metallic-black film (Fig. 9C), and
somewhat more matte-black at 120 °C (Fig. 9F). The coupon surfaces (at
both temperatures) show some irregular, possibly amorphous
structures, from the inhibitor, adsorbed on the surface (Figs. 10A and
11 A). Using EDX, S content detected was 3% by weight at 80 °C
(Supplementary material, Fig. D.1) and 9% by weight at 120 °C
(Supplementary material, Fig. D.2). The only crystalline species
detected by XRD for the low temperature system is ferrite (Fig. 10D);
at 120 °C, the XRD spectrum shows weak signals from both siderite and
melkovite (FexS.H2O) in addition to a prominent ferrite fingerprint
(Fig. 11D). As an inorganic inhibitor, STS can exhibit multiple
131
International Journal of Greenhouse Gas Control 78 (2018) 125–134
S.A. Sadeek et al.
Fig. 9. Photograph of inhibited C1018 coupons
from the corrosion vessels containing MEAAMP blends of variable ratio: (A) 25% MEA,
(B) 50% MEA, (C) 75% MEA at 80 °C and (D)
25% MEA, (E) 50% MEA, (F) 75% MEA at
120 °C. (For interpretation of the references to
colour in the text, the reader is referred to the
web version of this article.)
(Fig. 11C) is less grainy in appearance than the 50% MEA at 120 °C
(Fig. 11B), demonstrating (and corroborated by EDX (Supplementary
material, Fig. D.2) that there is negligible TS adsorption. Results from
XRD conclude only ferrite is present (Fig. 11F). The absence of protective TS explains the increased Fe ion concentration in solution
(Fig. 4B) and unchanged mass loss results (Fig. 4A) as compared to the
50% MEA 120 °C sample.
The primary inhibition mechanism employed by STS involving adsorption of TS was apparent in all tested solutions at 80 °C with outstanding performance demonstrated in the higher MEA blends (75 and
50%). Melkovite, formed from reduction of TS, occurred only in the
75% MEA 120 °C case. However, this species significantly reduced inhibitor performance efficiency from > 90% (observed in most high
concentration MEA samples) to 67%. Performance also dropped for the
25% MEA systems, particularly at 80 °C, due to surface coverage competition between TS and siderite. Siderite films, occurring in uninhibited cases, effectively impede Fe oxidation. The competitive surface
coverage hampered siderite film formation and inhibitor adsorption
inhibitor-derived product, there is a small decrease in Fe ion solution
concentration for the 50% MEA cases (Fig. 4B) at the two tested temperatures. This is consistent with reduced mass losses (Fig. 4A) and
therefore high IEs (> 90%) (Table 1). This distinction highlights that
the increased surface coverage by an inhibitor leads to an apparent
mass gain and enhanced protection against corrosive attack.
The coupons exposed to the 25% MEA blend appear to be dark at
80 °C (Fig. 9A) and coated with a green layer at 120 °C (Fig. 9D). The
surface topography of the coupons exposed to the 25% MEA solution at
80 °C (Fig. 10C) is similar to the 50% MEA sample tested at 80 °C
(Fig. 10B), also showing a few round structures. These surface features
are visual confirmation of siderite nucleation (confirmed by XRD,
Fig. 10F). Increased TS adsorption was detected using EDX (Supplementary material, Fig. D.1). The increased TS adsorption relative to the
50% MEA 80 °C case, is paired with an increase in Fe ion concentration
(Fig. 4B) and no reduction in mass loss (Fig. 4A). This suggests that less
surface area coverage occurs in the 25% MEA system due to siderite
formation. Surface imaging of the sample tested at 25% MEA at 120 °C
Fig. 10. Surface analysis of inhibited C1018 coupons exposed to variable ratio MEA-AMP blends at 80 °C: using SEM imagining of: (A) 75% MEA 150x, inset 500x, (B)
50% MEA 500x, (C) 25% MEA 150x; and using XRD with standards of detected crystals for: (D) 75% MEA, (E) 50% MEA, (F) 25% MEA. Magnified siderite signal and
siderite standard spectra inset included in (F).
132
International Journal of Greenhouse Gas Control 78 (2018) 125–134
S.A. Sadeek et al.
Fig. 11. Surface analysis of inhibited C1018 coupons exposed to variable ratio MEA-AMP blends at 120 °C: using SEM imaging of: (A) 75% MEA at 150x, (B) 50%
MEA 500x, (C) 25% MEA 500x; and using XRD with standards of detected crystals for: (D) 75% MEA, (E) 50% MEA, (F) 25% MEA. Magnified melkovite and siderite
signals and standard spectra inset included in (D).
inhibits the growth of naturally protective layers and therefore offers
less overall protection. However, the use of STS in solutions with higher
concentrations of MEA was extremely positive for both MDEA and AMP
blends. At 80 °C, surfaces appeared to be minimally corroded and only
ferrite was detected for both mixtures comprising 75% MEA with either
MDEA or AMP. At 120 °C, 75 and 50% MEA, sufficient Fe dissolution for
minimal corrosion species formation occurred in the MDEA system
while enhanced surface coverage by TS was attained on the AMP systems; crucially indicating the enhanced performance of this inhibitor in
the AMP system.
therefore highlighting the impracticality of using STS with this blend.
3.2.3. Comparison of corrosion behaviour of MEA-AMP and MEA-MDEA
blends
A comparison of MEA-AMP blends with MEA–MDEA (Sadeek et al.,
2018) demonstrate a significantly different impact on the corrosive
behaviours. The solution composition in blends does not allow explicit
predictions based on the behaviour of the individual components,
proportion of components, or temperature. For high and middle concentration aqueous MEA solutions, temperature appears to be influential in the apparent corrosion for both blends. For MEA-MDEA blends,
at 80 °C, ferrite is predominantly observed by contrast to oxides and
hydroxide in MEA-AMP. For MEA-AMP blends, increased corrosion is
apparent at reduced temperatures with only ferrite detection at 120 °C
while cementite is noted at 80 °C highlighting particularly aggressive
corrosion.
Both blends consisting of 50% MEA at 80 °C have developed siderite
crystals (but not at 120 °C). An adequate bicarbonate production and Fe
oxidation at the surface, both conditions are favoured by reduced
temperatures, result in siderite formation. Based on net mass change
results, Fe ion solution concentration and surface analysis, MEA-AMP
appears to be less aggressive than MEA-MDEA, highlighting the impact
of solvent components and temperature on corrosivity.
At 25% MEA for both temperatures and solution mixtures (i.e.
MDEA or AMP), the presence of siderite, and some hematite (MEAMDEA) was observed. With sufficient MDEA and AMP, the production
of bicarbonates clearly is adequate to produce a protective corrosion
product layer and reduce continued Fe oxidation at the surface. As the
surface coverage and XRD peaks suggest, AMP results in a higher
concentration of bicarbonates produced more quickly than MDEA.
Thus, a more complete and protective surface is formed.
The use of the STS is counter-productive in solutions with 25% MEA
with either MDEA or AMP. In both cases, it is apparent that STS in fact
3.3. Carbon dioxide uptake
The CO2 loading increases with an increase in AMP concentration,
replacing MEA (Table 2). However, the replacement of 25% MEA with
AMP shows no statistical difference with a 100% MEA solution.
The improved uptake with the presence of AMP is in agreement
with theory and literature (Mandal and Bandyopadhyay, 2006). With
the formation of somewhat protective product species at 80 °C, representative of the heat exchanger region, and only ferrite for the 75%
AMP solution, there is promise in this blend.
Table 2
The CO2 uptake of aqueous amine solutions tested in this study at 20 °C.
133
Solution Compositions (% by weight)
molCO2/molamine
75 % AMP/ 25 % MEA
50 % AMP/ 50 % MEA
25 % AMP/ 75 % MEA
100 % MEA (5 M MEA)
0.54
0.43
0.36
0.35
±
±
±
±
0.04
0.02
0.04
0.04
International Journal of Greenhouse Gas Control 78 (2018) 125–134
S.A. Sadeek et al.
4. Conclusion
amines in CO2 capture process. Energy Procedia 37, 2094–2099. https://doi.org/10.
1016/j.egypro.2013.06.088.
Guo, X.P., Tomoe, Y., 1999. The effect of corrosion product layers on the anodic and
cathodic reactions of carbon steel in CO2-saturated MDEA solutions at 100??C.
Corros. Sci. 41, 1391–1402. https://doi.org/10.1016/S0010-938X(98)00194-2.
Han, J., Nešić, S., Yang, Y., Brown, B.N., 2011. Spontaneous passivation observations
during scale formation on mild steel in CO2 brines. Electrochim. Acta 56, 5396–5404.
https://doi.org/10.1016/j.electacta.2011.03.053.
Hatchell, D., 2015. Thermal Degradation and Corrosion of Amines for CO2 Capture.
Kemper, J., Ewert, G., Grünewald, M., 2011. Absorption and regeneration performance of
novel reactive amine solvents for post-combustion CO2 capture. Energy Procedia 4,
232–239. https://doi.org/10.1016/j.egypro.2011.01.046.
Kerber, S.J., Tverberg, J., 2000. Stainless steel. Adv. Mater. Process. 33–36.
Kittel, J., Gonzalez, S., 2014. Corrosion in CO2 post-combustion capture with alkanolamines – a review. Oil Gas Sci. Technol. Rev. d’IFP Energies Nouv. 69, 915–929.
https://doi.org/10.2516/ogst/2013161.
Li, L., Voice, A.K., Li, H., Namjoshi, O., Nguyen, T., Du, Y., Rochelle, G.T., 2013. Amine
blends using concentrated piperazine. Energy Procedia 37, 353–369. https://doi.org/
10.1016/j.egypro.2013.05.121.
Mandal, B.P., Bandyopadhyay, S.S., 2006. Absorption of carbon dioxide into aqueous
blends of 2-amino-2-methyl-1-propanol and monoethanolamine. Chem. Eng. Sci. 61,
5440–5447. https://doi.org/10.1016/j.ces.2006.04.002.
Mandal, B.P., Guha, M., Biswas, A.K., Bandyopadhyay, S.S., 2001. Removal of carbon
dioxide by absorption in mixed amines: modelling of absorption in aqueous MDEA/
MEA and AMP/MEA solutions. Chem. Eng. Sci. 56, 6217–6224. https://doi.org/10.
1016/S0009-2509(01)00279-2.
Mazari, S.A., Ali, B.S., Jan, B.M., Saeed, I.M., 2014. Degradation study of piperazine, its
blends and structural analogs for CO2 capture: a review. Int. J. Greenh. Gas Control
31, 214–228. https://doi.org/10.1016/j.ijggc.2014.10.003.
Mineralogy, H.I. of, 2016. Mindat. URL https://www.mindat.org.
Nordsveen, M., Nešić, S., Nyborg, R., Stangeland, A., 2003. A mechanistic model for
carbon dioxide corrosion of mild steel in the presence of protective Iron carbonate
films—part 1: theory and verification. Corrosion 59, 443–456. https://doi.org/10.
5006/1.3277576.
Roberge, P.R., 2012. Handbook of Corrosion Engineering, second ed. McGraw Hill.
Sadeek, S.A., Williams, D.R., Sedransk Campbell, K.L., 2018. Using sodium thiosulphate
for carbon steel corrosion protection against monoethanolamine and methyldiethanolamine. Int. J. Greenh. Gas Control 74, 206–218. https://doi.org/10.1016/j.ijggc.
2018.05.005.
Sartori, G., Savage, D.W., 1983. Sterically Hindered Amines for CO2 Removal from Gases.
pp. 239–249.
Sedransk Campbell, K.L., Yu, L.C., Williams, D.R., 2017. Siderite corrosion protection for
carbon steel infrastructure in post-combustion capture plants. Int. J. Greenh. Gas
Control 58, 232–245. https://doi.org/10.1016/j.ijggc.2017.01.018.
Soosaiprakasam, I.R., Veawab, A., 2009. Corrosion inhibition performance of copper
carbonate in MEA- CO2 capture unit. Energy Procedia 1, 225–229. https://doi.org/
10.1016/j.egypro.2009.01.032.
Srinivasan, S., Veawab, A., Aroonwilas, A., 2013. Low toxic corrosion inhibitors for
amine-based CO2 capture process. Energy Procedia 37, 890–895. https://doi.org/10.
1016/j.egypro.2013.05.182.
Sun, W., Nešić, S., Woollam, R.C., 2009. The effect of temperature and ionic strength on
iron carbonate (FeCO3) solubility limit. Corros. Sci. 51, 1273–1276. https://doi.org/
10.1016/j.corsci.2009.03.009.
Talati, J.D., Gandhi, D.K., 1983. N-heterocyclic compounds as corrosion inhibitors for
aluminium-copper alloy in hydrochloric acid. Corros. Sci. 23, 1315–1332.
Tanupabrungsun, T., Brown, B., Nešić, S., 2013. Effect of pH on CO2 corrosion of mild
steel at elevated temperatures. NACE International Corrosion Conference and Expo
2013. pp. 1–11.
Tomoe, Y., Shimizu, M., Kaneta, H., 1996. Active dissolution and natural passivation of
carbon steel in carbon dioxide-loaded alkanolamine solutions. Corrosion96 The
NACE International Annual Conference and Exposition. p. 395/1-12.
Veawab, A., Tontiwachwuthikul, P., Chakma, A., 1999a. Corrosion behavior of carbon
steel in the CO2 absorption process using aqueous amine solutions. Ind. Eng. Chem.
Res. 38, 3917–3924. https://doi.org/10.1021/ie9901630.
Veawab, A., Tontiwachwuthikul, P., Chakma, A., 1999b. Influence of process parameters
on corrosion behavior in a sterically hindered amine - CO 2 system. Ind. Eng. Chem.
Res. 310–315.
Wang, T., Jens, K.-J., 2013. Oxidative degradation of AMP/MEA blends for post-combustion CO2 capture. Energy Procedia 37, 306–313. https://doi.org/10.1016/j.
egypro.2013.05.116.
Webmineral, 2016. Chukanovite Mineral Data. http://webmineral.com.
Xiao, J., Li, C.-W., Li, M.-H., 2000. Kinetics of absorption of carbon dioxide into aqueous
solutions of 2-amino-2-methyl-1-propanol+monoethanolamine. Chem. Eng. Sci. 55,
161–175. https://doi.org/10.1016/S0009-2509(99)00303-6.
Yu, L.C.Y., Sedransk Campbell, K.L., Williams, D.R., 2016. Using carbon steel in the
stripper and reboiler for post-Combustion CO2 capture with aqueous amine blends.
Int. J. Greenh. Gas Control 51, 380–393. https://doi.org/10.1016/j.ijggc.2016.04.
031.
Zhao, B., Sun, Y., Yuan, Y., Gao, J., Wang, S., Zhuo, Y., Chen, C., 2011. Study on corrosion
in CO2 chemical absorption process using amine solution. Energy Procedia 4,
93–100. https://doi.org/10.1016/j.egypro.2011.01.028.
Zheng, L., Landon, J., Neal, C., Koebcke, Payal, Chandan, Liu, K., 2015. Suitability and
stability of 2-mercaptobenzimidazole as a corrosion inhibitor in a post-combustion
CO2 capture system. Corrosion 71. https://doi.org/10.5006/1524.
Zheng, L., Matin, N.S., Landon, J., Thomas, G.A., Liu, K., 2016. CO2 loading-dependent
corrosion of carbon steel and formation of corrosion products in anoxic 30 wt.%
monoethanolamine-based solutions. Corros. Sci. 102, 44–54. https://doi.org/10.
1016/j.corsci.2015.09.015.
The potential for the use of STS as a corrosion inhibitor in systems
with carbon steel infrastructure with MEA-AMP solvent blends has been
demonstrated. While the corrosion resistance of SS316 L is generally
viewed to be very good, there are some nominal indications that under
some conditions deterioration of the CrO layer may be occurring. As
such, this material selection may not offer the full protection, which is
often presumed for amine solvent applications. By contrast, inhibited
C1018 coupons offered some improved protection over carbon steel
systems where no STS was added. At low concentrations of MEA (25%),
the use of a corrosion inhibitor is not useful; naturally, a siderite product is formed on the surface of the carbon steel coupon, creating a
protective barrier thereby reducing continued Fe oxidation. However,
the performance of the inhibitor with MEA–AMP blends was significantly beneficial at higher MEA concentrations of 50 and 75%. In
these cases, reduced corrosion rates were measured using mass loss and
Fe ion concentration. Additionally, SEM imaging showed reduced damage to the surface and in some instances presence of the inhibitor was
still evident. As these higher MEA concentration solutions offer better
CO2 loading, but still exhibit some corrosive effects when alone on
carbon steel, the use of an inhibitor such as STS offers expanding possibilities for cheaper infrastructure materials based on carbon steel.
Acknowledgements
The authors would like to acknowledge the Imperial College
Department of Materials X-ray Diffraction and Electron Microscopy
facilities, as well as the Department of Chemical Engineering’s
Analytical Laboratory and Mechanical Workshop. Dr Sedransk
Campbell would like to acknowledge the support of the EPSRC and
Royal Society for her Dorothy Hodgkin Research Fellowship.
Appendix A. Supplementary data
Supplementary material related to this article can be found, in the
online version, at doi:https://doi.org/10.1016/j.ijggc.2018.08.004.
References
Belarbi, Z., Farelas, F., Singer, M., Nešić, S., 2016. Role of amines in the mitigation of CO2
top of the line corrosion. Corrosion 1300–1310. https://doi.org/10.5006/2121.
Bénézeth, P., Dandurand, J.L., Harrichoury, J.C., 2009. Solubility product of siderite
(FeCO3) as a function of temperature (25-250°C). Chem. Geol. 265, 3–12. https://doi.
org/10.1016/j.chemgeo.2009.03.015.
Bougie, F., Iliuta, M.C., 2012. Sterically hindered amine-based absorbents for the removal
of CO2 from gas streams. J. Chem. Eng. Data 57, 635–669. https://doi.org/10.1021/
je200731v.
Brogan, L.J., 2011. Electrochemistry of FeSO4–Na2S2O3 and CuSO4–Na2S2O3 Systems
for Template-Assisted Nanowire Synthesis.
Campbell, K.L.S., Zhao, Y., Hall, J.J., Williams, D.R., 2016. The effect of CO2-loaded
amine solvents on the corrosion of a carbon steel stripper. Int. J. Greenh. Gas Control
47, 376–385. https://doi.org/10.1016/j.ijggc.2016.02.011.
Conway, W., Bruggink, S., Beyad, Y., Luo, W., Melin-Cabrera, I., Puxty, G., Feron, P.,
2015. CO2 absorption into aqueous amine blended solutions containing monoethanolamine (MEA), N,N-dimethylethanolamine (DMEA), N,N-diethylethanolamine
(DEEA) and 2-amino-2-methyl-1-propanol (AMP) for post-combustion capture processes. Chem. Eng. Sci. 126, 446–454. https://doi.org/10.1016/j.ces.2014.12.053.
Dubois, L., Thomas, D., 2012. Screening of aqueous amine-based solvents for postcombustion CO 2 capture by chemical absorption. Chem. Eng. Technol. 35, 513–524.
https://doi.org/10.1002/ceat.201100523.
DuPart, M.S., Bacon, T.R., Edwards, D.J., 1993. Understanding corrosion in alkanolamine
gas treating plants. Hydrocarb. Process. 72, 89–94.
Farelas, F., Brown, B., Nešić, S., 2013. Iron carbide and its influence on the formation of
protective iron carbonate in CO2 corrosion of mild steel. Corros.. 2013 Conf. Expo.
pp. 1–15.
Freeman, Sa., Chen, X., Nguyen, T., Rafique, H., Xu, Q., Rochelle, G.T., 2013. Piperazine/
N-methylpiperazine/N,N’-dimethylpiperazine as an aqueous solvent for carbon dioxide capture. Oil Gas Sci. Technol. Rev. d’IFP Energies Nouv. 69. https://doi.org/10.
2516/ogst/2012089.
Freund, P., 2003. Making deep reductions in CO2 emissions from coal-fired power plant
using capture and storage of CO2. J. Power Energy 217, 1–7. https://doi.org/10.
1243/095765003321148637.
Gunasekaran, P., Veawab, A., Aroonwilas, A., 2013. Corrosivity of single and blended
134
Документ
Категория
Без категории
Просмотров
0
Размер файла
5 408 Кб
Теги
ijggc, 2018, 004
1/--страниц
Пожаловаться на содержимое документа