17th International Conference on Greenhouse Gas Control Technologies, GHGT-17

20th -24th October 2024 Calgary, Canada

Screening study on the performance of different amine systems for

direct CO2 capture from air
Francesco Barzaglia,*, Maurizio Peruzzinia, Rui Zhangb,*
a
National Research Council, ICCOM Institute, via Madonna del Piano 10, 50019 Sesto F.no, Florence, Italy
b
College of Chemical Engineering, Xiangtan University, Xiangtan, Hunan 411105, P.R. China

Abstract

Direct Air Capture (DAC) of carbon dioxide from ambient air is recognized as a pivotal technology for reducing atmospheric CO2
levels and limiting global temperature rise to less than 2°C above pre-industrial levels. In pursuit of developing novel sorbents that
can efficiently capture ultra-diluted CO2 from the air while minimizing regeneration energy requirements, we present a
comprehensive screening study evaluating the performance of various aqueous amine-based solutions in DAC systems. This study
aims to correlate CO2 capture efficiency with the physical and chemical properties of the amines. A range of amine types were
investigated, including primary, secondary and tertiary amines, alkylamines, alkanolamines, sterically hindered amines, and
diamines. These amines were assessed both individually and in binary or ternary blends to combine the optimal properties of each
species. Additionally, selected amines were studied in non-aqueous solutions, where the higher solubility of CO2 in organic
diluents, coupled with their lower heat capacity and vapor pressure compared to water, offers the potential for enhanced CO2
absorption efficiency and reduced regeneration energy requirements. The CO2 capture efficiencies of the formulated amine sorbents
were measured over time in batch experiments at 25°C using a custom-built DAC system. These results were compared to those
obtained under identical conditions with aqueous NaOH, sodium carbonate, and potassium glycinate, which are among the most
commonly proposed sorbents for DAC processes. The carbonated species formed during the capture process were identified and
quantified using 13C NMR spectroscopy, offering key insights into the capture mechanisms across different liquid sorbents. This
screening study provides valuable correlations between CO2 absorption efficiency, amine chemical structure, and the species
formed during the process, offering guidelines for the formulation of optimized sorbents for DAC applications.

Keywords: Direct air capture; carbon dioxide capture; amines; 13C NMR speciation

1. Introduction

Human-induced climate change, driven by the accumulation of greenhouse gases (GHGs), particularly carbon
dioxide, is widely accepted as a critical factor in global warming. The Intergovernmental Panel on Climate Change
(IPCC) has unequivocally attributed the majority of this warming since the mid-20th century to human activities [1].
Urgent actions are required to mitigate the rising levels of CO2 in the atmosphere, as outlined in the Paris Agreement,
which aims to limit global temperature increases to between 1.5°C and 2°C above pre-industrial levels [2]. Strategies
to reduce anthropogenic CO2 emissions have focused primarily on transitioning to renewable energy, improving

*
Corresponding authors. Tel.: +39 0555225298, E-mail addresses: [email protected]; [email protected]

Electronic copy available at: https://ssrn.com/abstract=5010492

 GHGT-17 Barzagli, Peruzzini and Zhang 2

energy efficiency, and capturing CO2 from large point sources through Carbon Capture and Storage (CCS)
technologies [3]. While amine-based CCS is a well-established method for capturing CO2 from concentrated emissions
at industrial sites, it is ineffective for more dispersed sources [4,5]. Given that a substantial fraction of CO2 emissions
originates from dispersed sources such as transportation and residential heating, Direct Air Capture (DAC) has
emerged as a potential solution for capturing CO2 directly from the atmosphere [6–8]. DAC operates by selectively
removing CO2, which is present in the air at an ultra-low concentration (around 0.044% v/v), using either solid or
liquid sorbents. These sorbents are then regenerated, releasing pure CO 2 for storage or reuse in various industrial
applications. DAC technologies offer the flexibility to be deployed in any location, including areas with access to
renewable energy or in proximity to CO2 storage or utilization sites, which can significantly reduce transportation
costs [9,10]. However, DAC faces significant energy demands, particularly due to the low concentration of CO 2 in
ambient air (approximately 0.04% by volume), making the process more energy-intensive and expensive than
traditional CCS methods [11–13]. Despite these challenges, DAC technology has made progress, and by 2022 there
were 19 DAC plants in operation worldwide, capturing an average of 10000 tCO2/year [14]. However, further research
is needed to develop sorbents that can efficiently capture CO2 at low concentrations while minimizing energy
consumption for regeneration [9]. Solid sorbents, such as amine-functionalized silica, hollow fiber sorbents, and metal-
organic frameworks (MOFs), have been widely studied for DAC applications due to their moderate capture efficiency
and relatively low operating costs [15–19]. Liquid sorbents, particularly aqueous alkaline hydroxides like NaOH and
KOH, have also demonstrated high reactivity with atmospheric CO2, although they require higher regeneration
temperatures, making them less energy-efficient [20,21]. In contrast, aqueous amine solutions may offer lower
regeneration temperatures (around 100°C) without compromising CO2 capture efficiency. While amines have been
extensively researched for CCS applications, their potential in DAC systems has been less explored [22–24].
Recently, our research group carried out a series of experimental studies to assess the performance of some
commercial alkanolamines as sorbents for atmospheric CO2 capture. As a generic result, it was found that the rapid
formation of amine carbamates is a critical factor for efficient CO2 sorption, with unhindered primary and secondary
amines outperforming tertiary and sterically hindered amines [25,26]. Building on this previous work, the current
study expands the screening to include a variety of liquid amine-based systems, in order to provide a comprehensive
overview of the potential of such sorbents for DAC processes, going on to combine the ability to capture CO2 from air
with the reaction mechanisms involved and the different species formed, closely related to the energy requirements
for the regeneration of the sorbent itself [27,28]. For this reason, we evaluated a large number of different amines,
including primary, secondary and tertiary amines, alkylamines, alkanolamines, sterically hindered amines and
diamines. These were tested in aqueous solutions either individually or in binary/ternary combinations to exploit the
best characteristics of each species [29,30]. Additionally, selected amines were also evaluated in non-aqueous
solutions. Organic diluents were used to explore the possibility of enhancing CO2 absorption efficiency due to the
higher CO2 solubility in these media, combined with lower heat capacity and vapor pressure compared to water, which
could reduce the energy needed for sorbent regeneration [31,32].
Performance comparisons of the different formulated amine sorbents (total amine concentration = 1.5 M) were
based on the CO2 capture efficiency from an air stream, measured over time in batch experiments at 25°C using a
custom-built DAC system. The results were also compared with those obtained in the same system with aqueous
NaOH, sodium carbonate and potassium glycinate, which are among the most commonly proposed sorbents for DAC
processes. The reaction products formed during CO2 absorption were identified and quantified using 13C NMR
spectroscopy, a powerful non-invasive analytical technique that enables speciation of the carbonated compounds
formed during aerial CO2 capture, providing essential insights into the capture mechanisms of different liquid sorbents
[33,34]. This comprehensive screening provides a detailed comparison between CO2 capture efficiencies, chemical
structures, and reaction products across a range of amine solutions, offering valuable guidelines for formulating
efficient sorbents for DAC applications.

Electronic copy available at: https://ssrn.com/abstract=5010492

 GHGT-17 Barzagli, Peruzzini and Zhang 3

2. Experimental section

2.1. Materials

All amines and organic diluents used were provided by Merck and used as received. The complete list of reagents
used are given in Table 1, along with their acronyms and CAS identification numbers

Table 1. List of amines and diluent used, acronyms, and CAS numbers.
name acronym CAS
2-aminoethanol MEA 141-43-5
2-(2-aminoethoxy)ethanol DGA 929-06-6
1-amino-2-propanol 1A2P 78-96-6
3-Amino-1-propanol 3A1P 156-87-6
2-amino-1-butanol 2A1B 96-20-8
2-amino-2-methyl-1-propanol AMP 124-68-5
2-amino-2-methyl-1-3-propandiol AMPD 115-69-5
2-amino-2-(hydroxymethyl)-1,3-propanediol TRIZMA 77-86-1
2-(methylamino)ethanol MAE 109-83-1
2-(ethylamino)ethanol EAE 110-73-6
2-(butylamino)ethanol BAE 111-75-1
2-(tertbutylamino)ethanol tBAE 4620-70-6
Diethanolamine DEA 111-42-2
2-(dimethylamino)ethanol DMMEA 108-01-0
2-(2-(dimethylamino)ethoxy)ethanol DMAEE 1704-62-7
1-dimethylamino-2-propanol DMAP 108-16-7
2-dimethylamino-2-methyl-1-propanol 2DMA2M1P 7005-47-2
N methyl diethanolamine MDEA 105-59-9
1,2-diaminopropane 1,2-DAP 78-90-0
1,3-diaminopropane 1,3-DAP 109-76-2
N-Methyl-1,3-diaminopropane M-1,3-DAP 6291-84-5
N,N-diethylethylenediamine DEEDA 100-36-7
piperazine PZ 110-85-0
1-methyl-piperazine MPZ 109-01-3
ethylene glycol EG 107-21-1
1-propanol PrOH 67-63-0
diethylene glycol monoethyl ether DEGMEE 111-90-0

2.2. Aerial CO2 capture experiment

The screening of sorbents for DAC systems was carried out by measuring the CO2 capture efficiency from a gas
stream for different amine solutions. All tested solutions, both single amines and blends of different amines, had a
total amine molar concentration of 1.5 mol dm−3. Aqueous and nonaqueous solutions were prepared, using either
DEGMME or a mixture of EG and PrOH (in a 1:1 volume ratio) as organic diluents. The apparatus and method used
are the same as those reported in our previous work [25,26]. Fig. 1 shows the diagram and photo of the experimental
setup. Briefly, 0.100 dm3 of the tested solution was introduced into custom-made glass absorber, provided with an
external jacket for circulating a thermostatted liquid, maintained at a constant temperature of 25 °C using a Julabo
F33-MC bath. The column was packed with 5 mm diameter glass rings to maximize the contact area between the gas
and liquid. The system operates in a counter-current flow configuration. Compressed atmospheric air was introduced
at the bottom of the column with a constant flow rate of 46.5 dm3 h−1, monitored by a digital gas mass flow controller
(Aalborg) and a gas meter (Cole Parmer). The tested sorbent was pumped in from the top of the column via a peristaltic
pump (Masterflex), which continuously recirculated the solution from the base to the top of the absorber at a consistent
flow rate of 0.300 dm3 h−1. To minimize the loss of amine and solvent, a condenser set to -5 °C was positioned at the
top of the absorber. The outgoing air then passed through a gas purification tower filled with P2O5 to remove any
residual water and amine before entering in the gas chromatograph, a Varian CP-4900 equipped with a PoraPLOT U

Electronic copy available at: https://ssrn.com/abstract=5010492

 GHGT-17 Barzagli, Peruzzini and Zhang 4

column (Agilent) and a thermal conductivity detector (TCD). This setup enabled the continuous measurement of the
CO2 concentration in the gas stream. The CO2 absorption efficiency (ABS%), defined as the ratio between the amount
of captured CO2 and the CO2 in the air stream, was determined by comparing the GC signal intensity for CO2 in the
air before and after the capture process. The measured percentage of CO2 in the used compressed air was 0.044%.
Each experiment lasted 24 hours.

Fig. 1. (a) Schematic flow diagram and (b) picture of the experimental setup for the evaluation of the CO2 absorption efficiency.

At the end of the 24-hour absorption experiment, each sorbent was analyzed by 13C NMR spectroscopy in order to
identify and quantify all species present in solution. Spectra were obtained with a Bruker Avance III 400 spectrometer
operating at 100,613 MHz. The analysis and data processing procedures used were developed and verified in our
laboratories, and have been described in detail in previous work [35,36].

3. Results and discussion

3.1. Reaction between CO2 and amine-based sorbents

The possible reactions involved in capturing CO2 from air with amine-based liquid sorbents are the same as those
already known and studied for conventional CCS processes [27,28]. The actual equilibria will clearly be influenced
by the ultra-dilute CO2 present in the air stream to be treated (approx. 0.04%). By indicating with AmH a generic
primary or secondary amine, and with Am-CO2− and AmH2+ the relative amine carbamate and protonated amine,
respectively, the main reactions in aqueous solutions are:

2 AmH + CO2 ⇄ Am-CO2− + AmH2+ (1)

AmH + CO2 + H2O ⇄ HCO3− + AmH2+ (2)
Am-CO2− + CO2 + 2 H2O ⇄ 2 HCO3− + AmH2+ (3)

Electronic copy available at: https://ssrn.com/abstract=5010492

 GHGT-17 Barzagli, Peruzzini and Zhang 5

HCO3− + AmH ⇄ CO32− + AmH2+ (4)

CO2 + CO32− + H2O ⇄ 2 HCO3− (5)

Tertiary amines are unable to form carbamate (reaction 1) due to the lack of hydrogen atoms [33]; however, they
can react with CO2 in H2O to form bicarbonate (reaction 2), or they can take part in the CO2 capture process in amine
blendss by acting as a base for carbamate formation of primary or secondary amines by protonating themselves.
In polyamine solutions, when two or more amine groups react simultaneously, polycarbamate species can be
formed [38,39]. As a result, diamines containing primary and/or secondary amine groups have the potential to produce
dicarbamate ions.
In nonaqueous amine systems, the absence of water prevents the formation of bicarbonate and carbonate ions
(reactions 2-5). However, organic diluents with hydroxyl groups (ROH) can react with both CO2 and an amine (which
acts as a base) to form the corresponding alkyl carbonate (ROCO2−) [34]:

AmH + CO2 + ROH ⇄ AmH2+ + ROCO2− (6)

3.2. DAC performance

The performance of different amine systems as sorbents for DAC processes was evaluated and compared in terms
of CO2 adsorption efficiency from air (ABS%), using the instrumentation and following the procedure described in
Section 2.2. At the end of each experiment, which lasted 24 consecutive hours, all solutions were analysed by 13C
NMR spectroscopy to identify and quantify the species present. For comparison purposes, the performance of some
sorbents commonly proposed for DAC processes, such as aqueous solutions of NaOH, Na2CO3 and potassium
glycinate (K+Gly−), were tested in the same apparatus under the same operating conditions.
Table 2 shows all the results of the screening study. For the sake of clarity, amines/amine systems of similar type
(i.e., primary, primary hindered, secondary, tertiary, diamine and blend) have been grouped together. The average
ABS% over the 24-hour experiment was reported along with the actual percentages of absorbed CO 2 measured after
1 hour and at the end of the experiment, to evaluate the change in capture efficiency over time. The carbonated species
formed as a result of CO2 capture and identified by 13C NMR were also reported. When two or more species are
present, the relative percentage of each species formed relative to the other is reported between parentheses. These
percentages were obtained by comparison of the intensity of the relative 13C NMR signals. Carbonate and bicarbonate
are indicated to as a single species (HCO3−/CO32−) because they give a unique signal in 13C NMR due to rapid proton
exchange [35,36]. In addition, 13C NMR spectra of aqueous solutions of MEA, AMP, EAE, 1,2-DAP and of the blend
AMP+PZ are shown in Fig.2, to highlight the different reaction mechanisms for the different types of amines
investigated.
The results of the screening study summarized in Table 2 reveal clear performance trends that are related to the
chemical nature of the amines and the medium in which they are dissolved.
In aqueous solutions, unhindered primary amines, i.e. MEA, DGA, 1A2P, 3A1P and 2A1B, exhibited the highest
CO2 capture efficiencies, with average values ranging from 85.5% and 88.4%. These high efficiencies can be directly
correlated with the rapid formation of amine carbamates, as confirmed by 13C NMR analysis. In the case of MEA, for
example, the NMR spectrum (see Fig.2) revealed the formation of the corresponding MEA-carbamate as the sole
product of the capture process. The ability of unhindered primary amines to form stable carbamate intermediates is a
crucial factor in ensuring sustained CO2 capture over time, as demonstrated by the minimal decrease in performance
during the 24-hour experiment.
In contrast, aqueous sterically hindered primary amines, such as AMP and AMPD, showed significantly lower
capture efficiencies. AMP, for example, achieved an average ABS of 72.3%, and NMR analysis reveals the formation
of bicarbonate/carbonate species as the only reaction product (see Fig.2). This result reflects the steric hindrance
around the amine group, which prevents the formation of stable carbamates. In the case of sterically hindered amines,
the bicarbonate pathway becomes the main mechanism for CO2 capture, but its slower kinetics limits overall
performance. The inefficiency is even more pronounced in nonaqueous systems (discussed below), where
HCO3−/CO32− cannot be formed, and the only product is the unstable carbamate of AMP.

Electronic copy available at: https://ssrn.com/abstract=5010492

 GHGT-17 Barzagli, Peruzzini and Zhang 6

Table 2. CO2 absorption efficiency (ABS%) and 13C NMR speciation for each amine system tested and for reference sorbents. The asterisk *
indicates the carbamate of the relative amine, while HCO3−/CO32− refers to the sum of the amounts of carbonate and bicarbonate ions.
ABS%
type amine diluent Carbonated species formed after CO2 absorption (13C-NMR)
1h 24 h average
H2O 86.9% 87.4% 87.3% MEA*
MEA EG/PrOH 85.6% 82.4% 84.7% MEA*
DEGMME 89.3% 44.9% 67.0% MEA*
H2O 85.8% 84.8% 85.5% DGA*
primary

DGA EG/PrOH 87.9% 76.1% 83.1% DGA*

DEGMME 85.2% 32.7% 57.4% DGA*
1A2P H2O 88.8% 87.9% 88.4% 1A2P*
3A1P H2O 88.6% 88.0% 88.2% 3A1P*
2A1B H2O 87.7% 85.0% 87.1% 2A1B*
H2O 76.9% 65.4% 72.3% HCO3−/CO32−
hindered

AMP EG/PrOH 52.2% 7.6% 29.7% AMP*

primary

DEGMME 13.0% 7.4% 9.8% AMP*

AMPD H2O 45.7% 22.1% 32.5% HCO3−/CO32−
TRIZMA H2O 34.7% 10.6% 22.2% HCO3−/CO32−
MAE H2O 88.4% 89.1% 88.9% MAE* (80.3%) + HCO3−/CO32− (19.7%)
H2O 86.1% 85.1% 85.9% EAE* (50.3%) + HCO3−/CO32− (49.7%)
EAE EG/PrOH 82.4% 32.8% 57.7% EAE*
secondary

DEGMME 71.3% 3.1% 28.4% EAE*

H2O 85.9% 83.5% 85.2% BAE* (66.7%) + HCO3−/CO32− (33.3%)
BAE EG/PrOH 79.3% 23.7% 51.6% BAE*
DEGMME 65.4% 4.8% 26.8% BAE*
tBAE H2O 64.8% 48.8% 56.3% HCO3−/CO32−
DEA H2O 78.4% 70.2% 74.5% DEA* (81.3%) + HCO3−/CO32− (18.7%)
DMMEA H2O 48.3% 36.6% 40.8% HCO3−/CO32−
DMAEE H2O 48.7% 38.5% 41.9% HCO3−/CO32−
tertiary

DMAP H2O 46.9% 35.9% 40.0% HCO3−/CO32−

2DMA2M1P H2O 44.9% 26.7% 32.1% HCO3−/CO32−
MDEA H2O 41.8% 14.2% 26.6% HCO3−/CO32−
H2O 90.1% 89.6% 89.9% 2 different 1,2-DAP*, one for each N (87.3% and 16.3%)
1,2-DAP
EG/PrOH 89.2% 88.6% 89.1% 2 different 1,2-DAP*, one for each N (83.1% and 16.9%)
1,3-DAP H2O 90.2% 89.8% 89.8% 1,3-DAP* (only monocarbamate)
M-1,3-DAP H2O 89.5% 88.9% 89.3% M-1,3-DAP* (only on primary nitrogen)
diamine

H2O 90.0% 89.5% 89.7% DEEDA* (only on primary nitrogen)

DEEDA
EG/PrOH 88.3% 82.6% 85.7% DEEDA* (only on primary nitrogen)
H2O 88.5% 87.0% 87.3% PZ* (only monocarbamate)
PZ
EG/PrOH 88.1% 72.9% 81.2% PZ* (only monocarbamate)
H2O 89.3% 76.8% 84.0% MPZ* (89.6%) + HCO3−/CO32− (10.4%)
MPZ
EG/PrOH 88.5% 20.0% 54.3% MPZ*
AMP + PZ 1,75:1 H2O 86.3% 82.5% 84.5% PZ* (72.9%) + HCO3−/CO32− (27.1%)
EAE + DMAEE 1:1 H2O 86.1% 70.1% 81.6% EAE* (41.2%) + HCO3−/CO32− (58.8%)
blend

MEA + MDEA + AMP 1:2:2 H2O 81.8% 67.1% 77.8% MEA* (68.5%) + HCO3−/CO32− (31.5%)
EAE + DMAEE + PZ 2:2:1 H2O 85.8% 80.1% 85.1% PZ* (55.5%) + EAE* (11.1%) + HCO3−/CO32− (33.4%)
MEA + 2DMA2M1P + AMP 1:2:2 H2O 85.9% 85.4% 86.1% MEA* (77.1%) + HCO3−/CO32− (22.9%)
NaOH H2O 86.4% 88.3% 88.1% HCO3−/CO32−
ref Na2CO3 H2O 20.4% 11.8% 16.9% HCO3−/CO32−
K+Gly- H2O 89.3% 89.3% 89.6% Gly*

Electronic copy available at: https://ssrn.com/abstract=5010492

 GHGT-17 Barzagli, Peruzzini and Zhang 7

Similar considerations can be made for the aqueous tertiary amines tested, namely DMMEA, DMAEE, DMAP,
2DMA2M1P and MDEA: due to the lack of H in the amine functionality they cannot form carbamates, and CO2
capture occurs only through slower formation of bicarbonate/carbonate.
Secondary amines followed a similar trend to primary amines but displayed more complex behavior due to the
simultaneous formation of both carbamates and bicarbonates. MAE, with an average capture efficiency of 88.9%,
demonstrated a predominance of the carbamate (80.3%), while HCO3−/CO32− accounted for 19.7% of the species
formed. EAE showed a more balanced distribution between EAE-carbamate (50.3%) and HCO3−/CO32− (49.7%),
reflecting the slower contribution of bicarbonate formation to overall absorption (see Fig.2). The dual mechanism in
secondary amines, while effective, leads to slightly lower overall efficiencies than primary amines. Again, steric
hindrance near the amine function leads to a drastic drop in capture efficiency: comparison between BAE and its
sterically hindered form tBAE shows a sharp decrease in average ABS% from 85.2% to 56.3%, which can be
explained by NMR analysis revealing the formation exclusively of HCO3−/CO32− in the aqueous BAE.

Fig. 2. 13C NMR spectra of aqueous MEA, AMP, EAE, 1,2-DAP and of the dual blend AMP+PZ at the end of the absorption experiment. The
numbers indicate the carbon atoms referred to both free and protonated amine fast exchanging in the NMR scale. Asterisks denote the chemical
shifts of carbon backbones of amine carbamate. The double asterisk refers to a different second carbamate. C indicates the carbonyl atoms of
amine carbamate, while b/c refers to the signal of fast exchanging bicarbonate/carbonate ions. The intensity of the signals at 160−167 ppm is not
in scale.

Diamines, particularly 1,2-DAP, 1,3-DAP, M-1,3-DAP and DEEDA, showed the highest CO2 capture efficiencies
among all tested sorbents, averaging around 90%. The presence of two reactive amino groups in diamines allows for
the formation of multiple carbamate species simultaneously, as confirmed by the NMR spectra (see Fig.2). However,
in the chosen experimental system, diamines with a primary and a secondary amine function (i.e., M-1,3-DAP and
DEEDA) produced only the carbamate of the primary function at the end of 24 h of capture. The structural advantage

Electronic copy available at: https://ssrn.com/abstract=5010492

 GHGT-17 Barzagli, Peruzzini and Zhang 8

of diamines results in more efficient CO2 capture, with minimal loss in performance over time. Diamines that form
carbamates on both nitrogen atoms offer a particularly efficient pathway for CO2 absorption, demonstrating the
potential of these compounds in DAC applications. In contrast, if a tertiary amine functionality is present, as in the
case of MPZ, the performance is similar to monoamines of comparable structure.
Blended systems devised in conventional CCS processes for the synergistic potential of the different amines used,
were also evaluated. The blend of AMP and PZ in a 1.75:1 ratio is a notable example, achieving an average efficiency
of 84.5%. In this mixture, PZ, a diamine known for its fast carbamate formation, compensates for the slower
bicarbonate production by AMP. The NMR analysis (see Fig.2) confirmed the dominant presence of PZ-carbamate
(71.9%) along with HCO3−/CO32− (28.1%), illustrating the balance between rapid and delayed capture mechanisms.
Similar results were also obtained with the EAE+DMAEE+PZ mixture (2:2:1 ratio), taking advantage of the
carbamate formation of EAE and PZ and the strong base properties of DMAEE, which further supports the formation
of bicarbonates.
Analysis of the performance of amines in organic diluents, particularly EG/PrOH and DEGMME, showed the
significant impact of solvent on CO2 capture. In all amine categories, switching from aqueous to nonaqueous systems
results in a significant decrease in the capture efficiency of ultra-diluted aerial CO2. This decrease is less pronounced
for unhindered primary amines and diamines, which are species that form very stable carbamates. In contrast, the
sterically hindered amine AMP suffered the most significant performance drops in organic solvents. Another result
was that the use of EG/PrOH yielded higher CO2 capture efficiencies than DEGMME in all tests performed. This
discrepancy may be attributed to the higher reactivity of CO2 with EG and PrOH in the presence of amines, which
facilitates the formation of alkyl carbonates, enhancing CO2 uptake.

4. Conclusions

This screening study of amine-based sorbents for DAC demonstrates that aqueous solutions of unhindered primary
amines are generally the most efficient sorbents for capturing CO2 from air, with performance comparable to that of
aqueous alkali hydroxides. The key to their high absorption efficiency lies in the rapid formation of amine carbamates,
which consistently enables effective CO2 capture over extended periods. Secondary amines and appropriate blends,
while slightly less efficient than primary amines, also perform well due to their ability to form both carbamates and
HCO3−/CO32− ions. The presence of a significant amount of bicarbonate, though slower to form, contributes to
achieving high overall CO2 capture efficiency. This dual formation mechanism allows secondary amines to be
competitive sorbents in DAC processes. On the other hand, amines that are unable to form carbamates, such as
sterically hindered and tertiary amines, were shown to be poorly suited for the capture of ultra-diluted CO2. The slow
kinetics of HCO3−/CO32− formation in these systems hampers their effectiveness, making them less viable candidates
for DAC, where rapid and efficient CO2 capture is essential. Diamines, especially those with two unhindered primary
amino groups, have proven advantageous due to their ability to form multiple carbamate species, which leads to high
CO2 capture efficiency. However, diamines with sterically hindered groups or secondary amine structures tend to
perform similarly to alkanolamines with comparable chemical structures, limiting their potential advantage. The
results also highlight that aqueous amines are consistently more efficient than the same amines dissolved in organic
solvents such as EG/PrOH and, to a greater extent, DEGMME. While organic solvents may offer some advantages in
CO2 solubility, they have proven to be less suited for DAC systems due to lower carbamate formation and decreased
overall capture efficiency. The findings of this study offer critical guidelines for the design of optimized sorbents for
DAC applications, emphasizing the need for sorbents that balance high reactivity with efficient regeneration. These
insights can help guide future research and development of next-generation sorbents aimed at reducing atmospheric
CO2 concentrations and addressing climate change mitigation.

Acknowledgements

The work was supported by the National Natural Science Foundation of China (22008204), and the ICCOM
Institute of the Italian National Research Council (project PARCO2, code DCM.AD004.254).

Electronic copy available at: https://ssrn.com/abstract=5010492

 GHGT-17 Barzagli, Peruzzini and Zhang 9

References

[1] Pathak M, Slade R, Shukla PR, Skea J, Pichs-Madruga R, Ürge-Vorsatz D. Technical Summary. In: Climate Change 2022: Mitigation
of Climate Change. Contribution of Working Group III to the Sixth Assessment Report of the Intergovernmental Panel on Climate
Change. Cambridge Univ. Press, Cambridge, UK and New York, NY, USA: 2022, p. 51–147.
[2] UNFCCC - United Nations Framework Convention on ClimateChange. The Paris Agreement 2015.
[3] Barzagli F, Mani F. The increased anthropogenic gas emissions in the atmosphere and the rising of the Earth’s temperature: are there
actions to mitigate the global warming? Substantia 2019;3:101–11.
[4] Bui M, Adjiman CS, Bardow A, Anthony EJ, Boston A, Brown S, Fennell PS, Fuss S, Galindo A, Hackett LA, Hallett JP, Herzog HJ,
Jackson G, Kemper J, Krevor S, Maitland GC, Matuszewski M, Metcalfe IS, Petit C, Puxty G, Reimer J, Reiner DM, Rubin ES, Scott
SA, Shah N, Smit B, Trusler JPM, Webley P, Wilcox J, Mac Dowell N. Carbon capture and storage (CCS): The way forward. Energy
Environ Sci 2018;11:1062–176.
[5] Boot-Handford ME, Abanades JC, Anthony EJ, Blunt MJ, Brandani S, Mac Dowell N, Fernández JR, Ferrari MC, Gross R, Hallett JP,
Haszeldine RS, Heptonstall P, Lyngfelt A, Makuch Z, Mangano E, Porter RTJ, Pourkashanian M, Rochelle GT, Shah N, Yao JG,
Fennell PS. Carbon capture and storage update. Energy Environ Sci 2014;7:130–89.
[6] Marcucci A, Kypreos S, Panos E. The road to achieving the long-term Paris targets: energy transition and the role of direct air capture.
Clim Change 2017;144:181–93.
[7] Sanz-Pérez ES, Murdock CR, Didas SA, Jones CW. Direct Capture of CO2 from Ambient Air. Chem Rev 2016;116:11840–76.
[8] Hanna R, Abdulla A, Xu Y, Victor DG. Emergency deployment of direct air capture as a response to the climate crisis. Nat Commun
2021;12:1–13.
[9] Sodiq A, Abdullatif Y, Aissa B, Ostovar A, Nassar N, El-Naas M, Amhamed A. A review on progress made in direct air capture of
CO2. Environ Technol Innov 2023;29:102991.
[10] Wilcox J, Psarras PC, Liguori S. Assessment of reasonable opportunities for direct air capture. Environ Res Lett 2017;12.
[11] National Academies of Sciences Engineering and Medicine (NASEM). Negative Emissions Technologies and Reliable Sequestration:
A Research Agenda. 2019.
[12] Keith DW, Holmes G, St. Angelo D, Heidel K. A Process for Capturing CO2 from the Atmosphere. Joule 2018;2:1573–94.
[13] Sinha A, Realff MJ. A parametric study of the techno-economics of direct CO2 air capture systems using solid adsorbents. AIChE J
2019;65.
[14] Ozkan M, Nayak SP, Ruiz AD, Jiang W. Current status and pillars of direct air capture technologies. IScience 2022;25:103990.
[15] Darunte LA, Oetomo AD, Walton KS, Sholl DS, Jones CW. Direct Air Capture of CO2 Using Amine Functionalized MIL-101(Cr).
ACS Sustain Chem Eng 2016;4:5761–8.
[16] Lee JJ, Yoo CJ, Chen CH, Hayes SE, Sievers C, Jones CW. Silica-Supported Sterically Hindered Amines for CO2 Capture. Langmuir
2018;34:12279–92.
[17] Sujan AR, Pang SH, Zhu G, Jones CW, Lively RP. Direct CO2 Capture from Air using Poly(ethylenimine)-Loaded Polymer/Silica
Fiber Sorbents. ACS Sustain Chem Eng 2019;7:5264–73.
[18] Sadiq MM, Batten MP, Mulet X, Freeman C, Konstas K, Mardel JI, Tanner J, Ng D, Wang X, Howard S, Hill MR, Thornton AW. A
Pilot-Scale Demonstration of Mobile Direct Air Capture Using Metal-Organic Frameworks. Adv Sustain Syst 2020;4:1–8.
[19] Lee JJ, Sievers C, Jones CW. Silica-Supported Hindered Aminopolymers for CO2 Capture. Ind Eng Chem Res 2019;58:22551–60.
[20] Stolaroff JK, Keith DW, Lowry G V. Carbon Dioxide Capture from Atmospheric Air Using Sodium Hydroxide Spray. Environ Sci
Technol 2008;42:2728–35.
[21] Sen R, Goeppert A, Kar S, Prakash GKS. Hydroxide Based Integrated CO2 Capture from Air and Conversion to Methanol. J Am Chem
Soc 2020;142:4544–9.
[22] Kothandaraman J, Goeppert A, Czaun M, Olah GA, Prakash GKS. Conversion of CO2 from Air into Methanol Using a Polyamine and
a Homogeneous Ruthenium Catalyst. J Am Chem Soc 2016;138:778–81.
[23] Hanusch JM, Kerschgens IP, Huber F, Neuburger M, Gademann K. Pyrrolizidines for direct air capture and CO2 conversion. Chem
Commun 2019;55:949–52.
[24] Barzagli F, Mani F. Direct CO2 air capture with aqueous 2-(ethylamino)ethanol and 2-(2-aminoethoxy)ethanol: 13C NMR speciation
of the absorbed solutions and study of the sorbent regeneration improved by a transition metal oxide catalyst. Inorganica Chim Acta
2021;518:120256.
[25] Barzagli F, Giorgi C, Mani F, Peruzzini M. Screening Study of Different Amine-Based Solutions as Sorbents for Direct CO2 Capture
from Air. ACS Sustain Chem Eng 2020;8:14013–21.
[26] Barzagli F, Peruzzini M, Zhang R. Direct CO2 capture from air with aqueous and nonaqueous diamine solutions: a comparative
investigation based on 13C NMR analysis. Carbon Capture Sci Technol 2022;3:100049.
[27] Barzagli F, Giorgi C, Mani F, Peruzzini M. Reversible carbon dioxide capture by aqueous and non-aqueous amine-based absorbents: A
comparative analysis carried out by 13C NMR spectroscopy. Appl Energy 2018;220:208–19.
[28] Zhang R, Liu H, Liu R, Niu Y, Yang L, Barzagli F, Li C, Xiao M. Speciation and gas-liquid equilibrium study of CO2 absorption in
aqueous MEA-DEEA blends. Gas Sci Eng 2023;119:205135.
[29] Liu H, Qu J, Bhatti AH, Barzagli F, Li C, Bi J, Zhang R. A generic machine learning model for CO2 equilibrium solubility into blended
amine solutions. Sep Purif Technol 2024;334:126100.
[30] Chen G, Chen G, Peruzzini M, Barzagli F, Zhang R. Investigating the Performance of Ethanolamine and Benzylamine Blends as
Promising Sorbents for Postcombustion CO 2 Capture through 13 C NMR Speciation and Heat of CO 2 Absorption Analysis. Energy &
Fuels 2022;36:9203–12.
[31] Barzagli F, Lai S, Mani F. Novel non-aqueous amine solvents for reversible CO2 capture. Energy Procedia 2014;63:1795–804.

Electronic copy available at: https://ssrn.com/abstract=5010492

 GHGT-17 Barzagli, Peruzzini and Zhang 10

[32] Wanderley RR, Pinto DDD, Knuutila HK. From hybrid solvents to water-lean solvents – A critical and historical review. Sep Purif
Technol 2021;260:118193.
[33] Hu XE, Yu Q, Barzagli F, Li C, Fan M, Gasem KAM, Zhang X, Shiko E, Tian M, Luo X, Zeng Z, Liu Y, Zhang R. NMR Techniques
and Prediction Models for the Analysis of Species Formed in CO2 Capture Processes with Amine-Based Sorbents: A Critical Review.
ACS Sustain Chem Eng 2020;8:6173–93.
[34] Karlsson HK, Makhool H, Karlsson M, Svensson H. Chemical absorption of carbon dioxide in non-aqueous systems using the amine 2-
amino-2-methyl-1-propanol in dimethyl sulfoxide and N-methyl-2-pyrrolidone. Sep Purif Technol 2021;256:117789.
[35] Barzagli F, Lai S, Mani F. CO2 capture by liquid solvents and their regeneration by thermal decomposition of the solid carbonated
derivatives. Chem Eng Technol 2013;36:1847–52.
[36] Barzagli F, Mani F, Peruzzini M. Carbon dioxide uptake as ammonia and amine carbamates and their efficient conversion into urea and
1,3-disubstituted ureas. J CO2 Util 2016;13:81–9.
[37] Hartono A, Da Silva EF, Grasdalen H, Svendsen HF. Qualitative determination of species in DETA-H2O-CO2 system using 13C NMR
spectra. Ind Eng Chem Res 2007;46:249–54.
[38] Muchan P, Narku-Tetteh J, Saiwan C, Idem R, Supap T. Effect of number of amine groups in aqueous polyamine solution on carbon
dioxide (CO2) capture activities. Sep Purif Technol 2017;184:128–34.
[39] Barzagli F, Di Vaira M, Mani F, Peruzzini M. Improved solvent formulations for efficient CO2 absorption and low-temperature
desorption. ChemSusChem 2012;5:1724–31.

Electronic copy available at: https://ssrn.com/abstract=5010492