Abstract
Title of Dissertation Functionalization of Consumer Grade Fabric for Atmospheric Car-
bon Capture Proposal
Dissertation Directed by Professor Mohamad Al-Sheikhly, Department of Materials Science
and Engineering
Primary amines contribute to a well-studied mechanism for capturing carbon dioxide (CO2) from
the atmosphere. This thesis describes two approaches to grafting amine-containing monomers
to three commercial-grade fabrics: polyethylene terephthalate, high-density polyethylene, and
Nylon 6. Initially, two monomers, allylamine and butenylamine, were chosen and evaluated for
their sorbent capabilities. After confirming the selected monomers chemisorb CO2, six novel
copolymers, composed of each of the three fabrics grafted with one of each monomer, were
synthesized through two unique single-step fabrication processes, though both rely on free rad-
ical addition polymerization of the monomers. In the first synthesis method, electron beam
radiation created the free radicals necessary for grafting. In the second, nitroxide and perox-
ide chemical initiators created the free radicals. This chemically initiated graft polymerization
process resulted in qualitatively successful monomer attachment. All copolymers synthesized
via radiation-induced graft polymerization achieved greater grafting with butenylamine com-
pared to allylamine, likely given the closer proximity of the primary amine to the radical on the
latter?s structure. Characterization of CO2 capacity, the figure of merit for sorbency, revealed
not only that the majority of the grafted amines likely reacted to adsorb CO2, but secondary
physical mechanisms also contribute to CO2 abstraction. Additionally, economic and environ-
mental analyses were conducted to quantitatively determine the viability of the manufacture of
consumer-grade CO2 sorbents, both in terms of cost and CO2-capturing efficacy.
Functionalization of Consumer Grade Fabric for
Atmospheric Carbon Capture Proposal
Sean Cook, Pablo Dean, Eli Fastow, Patrick Ott, Jonathan Wilson, Hojin Yoon
May 1, 2020
Dissertation submitted to the faculty of the Honors College of the University of
Maryland in fulfillment of the requirements for a Citation from the Gemstone Program,
2020
Advisory Committee
Professor Mohamad Al-Sheikhly, Chair
Professor John Cumings
Dr. Zois Tsinas
Professor Luz Martinez-Miranda
Professor Robert Briber
Professor Alison Flatau
?by
Sean Cook, Pablo Dean, Eli Fastow, Patrick Ott, Jonathan Wilson, Hojin Yoon
2020
Acknowledgements
We would like to show our appreciation towards the contributions of Dr. Travis Dietz,
Dr. Zois Tsinas, Dr. Fred Bateman, and Mr. Drew Herring to the progress of this
project. We also greatly appreciate the unwavering support of the entire Gemstone staff
and Material Science and Engineering Department. We wish to acknowledge the financial
support of the Do Good Initiative and the University of Maryland Sustainability Fund.
We appreciate the knowledge and guidance of our mentor Professor Al-Sheikhly and the
insight of our discussants. Finally, we appreciate the support and love of our respective
families and friends during our Gemstone careers.
i
Contents
1 Introduction 1
2 Review of Literature 3
2.1 Reported Methods of Chemisorption . . . . . . . . . . . . . . . . . . . . 4
2.1.1 Resins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
2.1.2 Fiber-Based Sorbents . . . . . . . . . . . . . . . . . . . . . . . . . 6
2.2 Reported Methods of Physisorption . . . . . . . . . . . . . . . . . . . . . 7
2.2.1 Zeolites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.2.2 Monolithic Carbon Fiber . . . . . . . . . . . . . . . . . . . . . . . 9
3 Single-Step Synthesis of Atmospheric CO2 Sorbents through Radiation
Induced Graft Polymerization on Commercial-Grade Fabrics 9
3.1 Introduction to Radiation Induced Graft Polymerization (RIGP) . . . . . 9
3.2 Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
3.2.1 Monomer Selection . . . . . . . . . . . . . . . . . . . . . . . . . . 11
3.2.2 Radiation Grafting . . . . . . . . . . . . . . . . . . . . . . . . . . 14
3.2.3 Characterization by Fourier Transformation Infrared Spectroscopy 18
3.2.4 Characterization by Scanning Electron Microscope and Energy Dis-
persive Spectroscopy . . . . . . . . . . . . . . . . . . . . . . . . . 18
3.2.5 CO2 Capacity Characterization . . . . . . . . . . . . . . . . . . . 19
3.3 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
3.3.1 Confirmation of Monomer CO2 Capacity . . . . . . . . . . . . . . 21
3.3.2 Monomer Attachment . . . . . . . . . . . . . . . . . . . . . . . . 23
3.3.3 CO2 Capacity, Kinetics, and Mechanisms . . . . . . . . . . . . . . 31
4 Comparison of Novel CO2 Graft Copolymer Synthesis by Radiation and
Chemical Initiated Graft Polymerization 35
4.1 Introduction to Chemical-Induced Graft Polymerization (CIGP) . . . . . 35
4.2 Methodology of Peroxide and Nitroxide-Induced Grafting . . . . . . . . . 35
ii
4.3 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
4.4 Conclusions on Comparing Radiation Induced and Chemical-Initiated Graft
Polymerization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
5 Economic and Environmental Analysis 43
5.1 Economic Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
5.2 Environmental Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
6 Conclusion 47
6.1 Comparison Between CIGP and RIGP . . . . . . . . . . . . . . . . . . . 47
6.2 Uses for Captured CO2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
6.3 Future Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
iii
List of Tables
Table 1 Degree of Attachment for the wet impregnation of both Page 21
monomers.
Table 2 Degree of Grafting for the fabrics with FITR spectra displayed in Page 27
figure 13
Table 3 Rate constants for second order kinetic behavior for CO2 sorbency Page 32
(Fastow et al 2019).
Table 4 CO2 capacities of untreated fabrics and fabrics functionalized via Page 33
RIGP; the former did not exhibit statistically significant capacity
while the latter does (Fastow et al 2019).
Table 5 Degree of grafting values for samples analyzed via EDS and FTIR Page 42
Table 6 Economic and environmental analyses of one-time use of the Page 45
functionalized fabric at experimentally-derived CO2 capacities.
The best monomer-substrate combinations are evaluated at each
radicalization method, with radiation being on allylamine-PET,
peroxide on diallylamine-Nylon, and nitroxide also being on
diallylamine-Nylon.
Table 7 Economic and environmental analyses of extended use (50 cycles) Page 45
of the functionalized fabric at experimentally-derived CO2 capac-
ities. The best monomer-substrate combinations are evaluated at
each radicalization method, with radiation being on allylamine-
PET, peroxide on diallylamine-Nylon, and nitroxide also being on
diallylamine-Nylon.
Table 8 Economic and environmental analyses of one-time use of the func- Page 45
tionalized fabric at a theoretical CO2 capacity of 2 mmol/g. The
best monomer-substrate combination of allylamine-PET is shown.
Table 9 Economic and environmental analyses of extended use (50 cy- Page 46
cles) of the functionalized fabric at a theoretical CO2 capac-
ity of 2 mmol/g. The best monomer-substrate combinations of
allylamine-PET is shown.
iv
List of Figures
Figure 1 Radiation-induced graft polymerization Page 6
Figure 2 Surface features of Zeolite 13X powder under SEM with 8000x Page 8
magnification. (Cavenati et al 2004)
Figure 3 Arrow pushing mechanism between primary amine and CO2 Page 10
to form a zwitterion (?), then deprotonate zwitterion to form
charged species (?).
Figure 4 Bond line structure of (A) allylamine and (B) 3-butenylamine. Page 12
Figure 5 Bond line structure for diallylamine. Page 14
Figure 6 Mechanism of radiation-induced graft polymerization (RIGP), Page 15
including desired grafting and undesired crosslinking (Fastow
et al 2019).
Figure 7 Schematic of process used in synthesizing and characterizing Page 16
novel graft copolymeric CO2 sorbents.
Figure 8 Rendering of turntable used to cycle fabric samples, depicted Page 17
as white rectangles in clear vials, through the electron beam
path, depicted as the translucent yellow line.
Figure 9 A generalized schematic for the (a) test chamber and a (b) Page 19
detailed rendering of the test chamber for fabric samples and
(c) activated carbon samples (Fastow et al 2019).
Figure 10 Calibration curve of input CO2 concentration and measure- Page 20
ments identified by CozIR sensor (Fastow et al 2019)
Figure 11 Adsorbance capacity of raw and monomer impregnated acti- Page 22
vated carbon.
Figure 12 Degree of grafting of allylamine (red) and butenylamine (blue) Page 24
as a function of dose for all three fabrics.
Figure 13 FTIR spectra of HDPE, PET, and Nylon highlighting the Page 26
growth in the peaks associated with primary amines (Fastow
et al 2019).
v
Figure 14 Nitrogen to carbon ratios obtained from EDS for both PET Page 28
and Nylon fabrics generated with allylamine and butenylamine
compared to ungrafted samples (Fastow et al 2019).
Figure 15 EDS map of PET grafted with butenylamine, DoG of 14.48 %. Page 29
Observed nitrogen is colored red and the secondary electron
image is colored red.
Figure 16 PET fibers before (A) and after (B) grafting of allylamine Page 30
monomer at 150 kGy. No significant structural damage or ho-
mopolymer buildup was observed after radiation.
Figure 17 A characteristic decay in CO2 concentration from adsorption on Page 31
allylamine-grafted PET as a function of time. The insert dis-
plays a second order kinetics fit with an effective rate constant
of 6.54 x 10-5 (M(s))?1 (Fastow et al 2019).
Figure 18 CO2 sorbent capacity reported in mmol(g
?1) (a) and Page 34
mmol(mmol?1) (b) (Fastow et al 2019).
Figure 19 Degree of grafting values for allylamine-grafted (orange) and Page 37
diallylamine-grafted (blue) PET and Nylon fabrics that have
undergone chemically initiated graft polymerization (CIGP)
treatment.
Figure 20 FTIR spectra of ungrafted, allylamine-grafted, and Page 39
diallylamine-grafted Nylon (top) and PET (bottom).
Figure 21 EDS-obtained ratios between nitrogen-attributed and carbon- Page 41
attributed X-ray counts on ungrafted, allylamine-grafted, and
diallylamine-grafted PET and Nylon.
vi
1 Introduction
Climate change remains one of the most daunting modern issues. With fossil fuel usage
beginning in the Industrial Revolution, atmospheric carbon dioxide (CO2) levels have
steadily risen. Currently, the global mean marine surface CO2 levels measure at roughly
410.88 parts per million (ppm) according to the National Oceanic and Atmospheric Ad-
ministration [42]. For comparison, before widespread industrialization in the 19th century,
CO2 concentrations remained fairly constant at 280 ppm [65]. While research continues
today to fully elucidate the effects of rising CO2 levels on human and natural systems,
the Intergovernmental Panel on Climate Change?s 4th assessment report directly tied
rising CO2 levels to global warming [40]. This mechanism for global warming is well
understood through the greenhouse gas effect. Greenhouse gases in the atmosphere, such
as CO2, absorb infrared radiation emitted from the earth. In turn, these gases raise the
temperature of the earth?s surface and the troposphere. However, the effects of rising
CO2 levels are not limited to the greenhouse gas effect. The current literature offers a
bleak assessment that implicates higher atmospheric concentrations of CO2 as the cause
of extreme weather patterns, rising ocean levels, ocean acidification, and ice cap melting.
Additionally, since the largest use of energy in the temperate latitudes can be attributed
to space heating and cooling, the rising Earth surface temperatures are responsible for an
additional energy burden. All these impacts severely impair the habitability of the earth,
and with the continued rise in carbon dioxide levels, these problems will only worsen and
reach irreversible levels.
Thus, climate change is a pressing issue that requires a multifaceted response. The
various approaches being considered include improved energy efficiency, development of
alternative energy sources, and CO2 capture and storage. With over 60 % of carbon
emissions stemming from power and industrial sectors in 2005, the demand for fossil fuel
alternatives has grown [40]. Similarly, carbon capture technologies have sought to mit-
igate carbon emissions from these large point sources. Even if carbon capture methods
were 90 % efficient and implemented at every large point source facility, over 50 % of
CO2 emitted globally would remain unaffected [63]. Additionally, few companies have
1
been willing to implement carbon capture solutions due to economic feasibility concerns.
Therefore, the direct capture of atmospheric carbon dioxide is appealing as it promises
the capture of CO2 from delocalized sources in addition to large point sources, yet the
technology remains relatively unexplored.
Carbon capture technologies generally rely on the development of specialized materi-
als with properties that allow the material to either absorb or adsorb CO2. Absorption
refers to the diffusion of carbon dioxide into the bulk of the material, while adsorption
refers to the binding of carbon dioxide to the surface of the material. Sequestration is
achieved through either chemical or physical mechanisms. The two mechanisms have
benefits and shortcomings that are highly dependent on the specific material. Chemical
carbon capture involves the formation of a chemical bond between the material and CO2,
usually achieved with a functional group that is reactive with CO2. On the other hand,
physical mechanisms are based on highly specific geometries that allow for the trapping
of CO2 within the bulk of the material [60].
For carbon capture technology, there exist specific criteria the technology must meet in
order to be effective. The main criteria are high carbon dioxide capacity and rapid kinet-
ics, which maximizes the material?s cost-effectiveness. Additionally, the material must
selectively react with or physically trap carbon dioxide rather than other compounds.
Otherwise, the efficiency of the system will be severely limited. Other considerations in-
clude the robustness of the material to withstand physical and chemical deterioration and
the cost-efficiency for sequestering CO2. This cost-efficiency of the material for removing
carbon dioxide should be around $ 100 per ton of carbon dioxide absorbed [24]. Lastly,
the material must allow for the removal of captured CO2 such that the functionalized
material is regenerated and available for further CO2 capture. In order to meet this cri-
teria, Team Capture proposed to functionalize commercial-grade fabrics to adsorb CO2
from the atmosphere.
In order to create a novel fabric with the potential of widespread use, Team Cap-
ture used Polyethylene Terephthalate (PET), High-Density Polyethylene (HDPE), and
Nylon-6. Since PET is the most commonly used polyester and HDPE and Nylon are also
2
commonly used, successful modifications to these fabrics that do not affect the morphol-
ogy of the fibers will allow carbon capture technology to be adapted to a wide range of
markets. Thus, the potential market for carbon capture technology will be drastically
extended. Similarly, by targeting commercial fabrics, consumers will be offered the ability
to participate in the fight against climate change. Unlike large industries, consumers are
more likely to pay the additional cost of the modified fabrics for the sake of combating
global warming. While large industries take on large expenses to implement new tech-
nology, the cost would also be split among all the consumers who purchase the product.
The literature review below will evaluate the feasibility of several of the most preva-
lent sorbents. The optimal sorbent for this project is one that has reasonable flexibility,
low cost, high regenerability, and high capacity and selectivity for CO2. This project
has focused on evaluating the carbon capture density of fabrics modified using radiation-
induced and chemically initiated graft polymerization with three amine-based monomers.
The monomers were selected based on their ability to both capture CO2 and attach onto
the fabrics examined. Adsorption experiments were run to examine the ability of each
monomer to capture CO2. Monomers were then grafted onto fabrics both through irra-
diation and chemical activation. We examined the modified fabrics for the presence of
amine groups using spectroscopy and also evaluated their respective CO2 capacities with
adsorption experiments. The steam stripping of loaded fabrics was not evaluated because
it is beyond the scope of the project. A simplified economic and environmental analysis
was then carried out to determine where improvements need to be made in order for the
process to be feasible.
2 Review of Literature
This review of CO2 sorbents intends not only to detail existing methods reported in
literature, but also evaluate their efficacy. The primary figure of merit for carbon dioxide
sorbents is CO2 capacity, typically measured in units of moles of CO2 captured per either
mass of sorbent or mole of sorbent [24, 29, 17]. Another important metric of sorbent
3
quality is its reusability, measured in the decay in CO2 capacity as a function of sorption-
desorption cycles [33].
This literature review divides CO2 sorbents into two categories based on their primary
means of absorbance: either physisorption or chemisorption. Each of these two categories
are further divided into subcategories that detail additional literature describing the
development of specific material systems. Each of these material systems works best for
specific applications. While many of the systems reviewed in section 3 are developed for
large point CO2 sources, such as factories or power plants, our work seeks to develop a
novel CO2 sorbent and a simple fabrication process ideal for local point sources.
2.1 Reported Methods of Chemisorption
2.1.1 Resins
Recently, researchers have evaluated commercial adsorbent resins, such as HP2MGL, as a
method for economically viable air capture. While the resins do not directly capture CO2,
they exhibit high chemical and physical stability and therefore serve as a support for the
adsorbent. As aforementioned, studies focus on amine-based monomers as adsorbents be-
cause of their ability to reversibly attach to CO2. The resins are impregnated with amines
by dissolving the monomer, adding the resin, and drying the adsorbent-resin complex.
In a recent study, this process lasted over a day for 2 grams of resin to be generated;
however, multiple samples could be generated simultaneously. Additionally, diffusion ad-
ditives were added to increase the rate of impregnation by the amines. Further research
is required to determine if the amines would detach from the resin over time or under
certain conditions. Wang et al. tested the CO2 adsorption capacity of PEI-impregnated
HP2MGL under various conditions to determine the ideal loading capacity. The percent-
age of amines occupied by CO2 significantly decreased when the amines surpassed 60 %
of the total weight of the solid [54].
Furthermore, the study researched the impact of both moisture and O2 in the air
on CO2 adsorption capacity. Initially, the rise in humidity causes an increase in the
maximum CO2 adsorption capacity because the water vapor reacts with the amine-CO2
4
complex to regenerate primary amines for further reaction. However, as humidity further
increases, the capacity steadily decreases, likely due to water interfering with CO2 and
competing for the same adsorption sites. Similarly, O2 competes with CO2 for adsorp-
tion sites and additionally oxidizes several regions of the amine-based molecules. This
oxidation negatively impacts the adsorption capacity because it renders several of the
adsorption sites unusable for future trials after the first desorption, reducing the capacity
by approximately 13.5 % from 2.13 to 1.84 mmol CO2 adsorbed per gram of adsorbent
[54].
Research conducted by Akingbogun et al. investigated the adsorption capacity of
ion-exchange resins (IERs) functionalized with a primary amine group. IERs are organic
polymer substrates that behave as media for the exchange of ions due to their porous
surface area and purifying nature. Specifically, Akingbogun et al. tested a cross-linked
polymeric resin containing benzyl amine groups in several conditions, such as at varying
levels of humidity and concentrations of CO2. At CO2 levels comparable to that of the
atmosphere, the resin was able to adsorb CO2 at a rate of 1 mol/kg of resin, with the rate
of adsorption rising to 3.4 mol/kg at conditions of moderate humidity [1]. The primary
benefits of resins include their low cost, capability to desorb 99 % of adsorbed CO2 at
relatively low temperatures, and effectiveness at atmospheric CO2 concentrations. The
main weakness of resins for carbon capture lies in its relatively low peak CO2 capacity,
which is further reduced after the first adsorption cycle due to the presence of O2. The
effects of these external factors on CO2 adsorption, as well as the cost of fabricating the
adsorbent complex relative to maximum CO2 capacity and rate of adsorption, must be
evaluated and considered.
Due to the similar capture mechanism of amine-impregnated resins and amine-grafted
fabrics, knowledge of resins can guide the optimization of grafting methods. Specifically,
the critical amine loading (found to be 60 % for HP2MGL) must be determined for the
relevant fabrics and monomers to accurately determine maximum adsorption capacities.
The grafted amines are likely to have much lower loading due to the lack of bulk absorp-
tion; however the number of attached amines would have a more immediate impact on the
5
adsorption capacity. Furthermore, the effect of oxygen and humidity should be considered
when evaluating adsorption capacity due to their drastic impact. In order to be viable,
the cost of the graft polymerization must be less than that of making adsorbent-resin
complexes or the product must have a superior capacity per gram of adsorbent.
2.1.2 Fiber-Based Sorbents
For the ultimate purpose of synthesizing CO2 sorbents, the use of synthetic fibers as
the main substrate for polymerization has also been investigated in prior studies. A
common mechanism by which to attach monomers onto fibers involves radiation-induced
graft polymerization. The procedure consists of irradiating the substrate with ionizing
radiation so as to break a covalent bond, thus forming a molecule with a lone, unpaired
electron in one or more of its atoms? outer shells; these highly unstable atoms are called
free radicals, as shown in Figure 1. When in close contact with the monomer of choice, a
free radical formed on the substrate is expected to spontaneously react with the double
bond, thus beginning a chain mechanism through which the monomer becomes radical-
ized, thereby attaching to other monomers of its own kind in the process [38]. Only when
two free radicals form a covalent bond is the polymerization terminated.
Figure 1: Radiation-induced graft polymerization
The studies conducted by Yang et al. and Wu et al. both employed a straightfor-
ward grafting technique that is both quick and regenerative (Yang, Li, Chen 2010; Wu,
Chen, Liu 2014). When preparing to graft 2-propen-1-amine (allylamine) onto polyacry-
lonitrile (PAN), the substrate was irradiated, washed with DI water, and mixed with
two initiators, ammonium persulfate ((NH4)2S2O8) and sodium bisulfite (NaHSO3) [58].
The purpose of these initiators, which form free radicals under extreme heat, is to rad-
6
icalize the monomer so that it can form a covalent bond with the radicalized substrate.
Similarly, in the case of grafting polyethylenimine (PEI) onto polypropylene fiber, am-
monium ferrous sulfate ((NH4)2SO4)(FeSO4)(6 H2O) was used as a free radical initiator
[56]. Because both of these processes require the use of an initiator, a smaller margin for
error exists for efficiently grafting the desired monomer onto the substrate of choice. Due
to potential unwanted free radical reactions that may occur when introducing a set of
unstable initiators, other monomers must be researched for their spontaneity in reacting
with radicalized substrate. Grafting aside, the CO2 adsorption capacities exhibited by the
aforementioned fiber-based amine monomers show promise. Allylamine grafted onto PAN
showed a 6.22 mmol CO2/g capacity while PEI grafted onto polypropylene adsorbed 4.8
mmol CO2/g [58, 56]. In addition, another study that did not employ radiation-induced
grafting but instead incorporated a wet impregnation method found that PEI-coated glass
fibers adsorbed 6.29 mmol CO2/g [32]. However, this study used a cross-linking resin to
prepare the substrate and monomer, which would likely render the material too stiff for
a commercial-grade fabric. Grafting and impregnating CO2 adsorbents onto fibers is a
promising method by which to manufacture materials that renewably capture ambient
CO2.
Additionally, the CO2 adsorption capacity of amine-grafted activated carbon fiber has
been tested at various levels of humidity [6]. However, said sorbent made use of a highly
microporous structure not commonly found in consumer-grade fabrics. The adsorption
capacity of amine-grafted consumer-grade materials remains an unexplored topic that
will offer new knowledge regarding the viability of consumer-scale carbon capture.
2.2 Reported Methods of Physisorption
2.2.1 Zeolites
Zeolites comprise a broad category of aluminosilicate ceramics typically used in catalysis
and separation science [37]. In a zeolite sorbent, aluminum, silicon, and oxygen form a
tetragonal cage where the partial negative charge on the aluminum atoms enables the
stabilization of positively charged groups that attract CO2 through coulombic interactions
7
with the partial negative charge on the oxygen atoms [45, 59, 7]. Synthetic zeolites
typically exhibit well controlled pore sizes on the order of single nanometers, enabling
size-based filtration of molecules in atmosphere [37].
Utilizing zeolites for the purpose of CO2 sequestration has remained an active area
of research since the 1970s through today. In a notable advance, Dr. Youngmin Cho
of the Korea Railroad Research Institute modified zeolites to capture CO2 at ambient
temperatures with the addition of lithium hydroxide to the aluminosilicate frameworks
[7]. Adding the lithium hydroxide provided the positively charged group that engages in
physical interactions with atmospheric CO2. Cho et al modified 13X zeolite pictured in
figure 2. In a similar vein, Dr. Arnost Zuka achieved similar adsorption properties by
introducing magnesium oxide to thin film zeolites [64].
Figure 2: Surface features of Zeolite 13X powder under SEM with 8000x magnification
[5].
Further work by Dr. Hauchhum mapped chemical properties and kinetics of CO2
8
capture by zeolite sorbents. That work found that the CO2 adsorption process matches
Le-Chatlier?s principles with respect to partial pressure, where an increase in pressure
increased the amount of CO2 adsorbed. However, raising the temperature decreased
the amount of CO2 adsorbed, potentially indicating an entropic driving force opposing
adsorption [21].
2.2.2 Monolithic Carbon Fiber
In response to a demand for structurally resilient sorbents for large-point source CO2
sequestration several industries adopted highly porous monolithic carbon fiber sorbents, a
well developed technology. These sorbents capture CO2 through intermolecular dispersion
interactions between the carbon fibers and the CO2 molecules trapped in small pores
[51, 48]. Monolithic carbon fiber consists of continuous carbon fiber?synthesized through
the pyrolysis of an ultra high molecular weight polymer fiber?secured in a polymer
matrix [51]. The most effective carbon fiber sorbents exhibit a 12 wt. % CO2 capacity.
3 Single-Step Synthesis of Atmospheric CO2 Sorbents
through Radiation Induced Graft Polymerization
on Commercial-Grade Fabrics
3.1 Introduction to Radiation Induced Graft Polymerization
(RIGP)
Primary or secondary amines enable a well studied reaction that binds atmospheric CO2
to a substrate. These functional groups are stable at high temperatures, humidity, and
pressure [36, 16, 11, 52]. Thus, chemisorbent systems commonly use the reaction driven
by primary or secondary amines to capture CO2. This mechanism exhibits an enthalpy
of reaction of ?80 kJ(mol?1) at 40 ?C and is therefore energetically favorable [39]. The
primary or secondary amine adsorption reaction with CO2 follows two steps: CO2 binds
to an amine to form a zwitterion, then another amine deprotonates the zwitterion, as
9
indicated in equation 1 and 2 [39].
RR?NH + CO ? RR?N+2 HCOO? (1)
RR?N+HCOO? + RR?NH? RR?NCOO? + RR?N+H2 (2)
Figure 3 depicts the arrow pushing mechanism of this reaction. As shown in Figure 3,
the reaction begins with a nucleophilic attack on the carbon in CO2 and concludes with
a deprotonation of the resulting zwitterion.
Figure 3: Arrow pushing mechanism between primary amine and CO2 to form a zwitterion
(?), then deprotonate zwitterion to form charged species (?).
Several reported chemisorbents use primary amines to capture CO2 from atmosphere.
Among the most successful of these systems, several have achieved capacities on the order
of 1 to 5 mmol(g?1) [31, 54, 58, 19, 62, 35, 25, 32, 56]. Several reported chemisorption
systems attempted to attach amine-rich monomers to fabric substrates given their high
surface area and ease of processing [58, 32, 56]. Early work by Li et. al. attempted to
functionalize glass fibers using a wet impregnation method, but later work identified the
potential for graft polymerization to form more robust sorbents with higher fractions of
monomer attachment [31, 32].
Reported approaches by Yang et. al. and Wu et. al. to functionalize fabric by graft-
ing amine-rich monomers to fabrics used a two-step initiation procedure; monomers were
initiated with peroxides and fabrics initiated with ionizing radiation [58, 56]. Initiation
10
forms free radicals on which monomers polymerize. Both Yang et. al. and Wu et. al.
reported high CO2 capacities, 6.22 and 4.8 mmol(g
?1) respectively, but synthesized their
sorbent through a two-step process that lends itself to undesired homopolymerization of
monomer and adds additional cost to potential mass production [58, 56, 17]. This work
reports on the fabrication of CO2 sorbents with a novel single step graft polymeriza-
tion procedure to attach amine-rich monomers to consumer-grade polymeric fabrics. We
grafted allylamine and 3-buten-1-amine (butenylamine), shown in figure 4, onto polyethy-
lene terephthalate (PET), high-density polyethylene (HDPE), and Nylon-6 (Nylon) via
radiation-induced graft polymerization (RIGP).
3.2 Methodology
3.2.1 Monomer Selection
A main goal of this study is to develop a single-step synthesis of a commercial-grade
atmospheric CO2 sorbent through the use of radiation-induced graft polymerization. Ad-
ditionally, this study aims to initiate graft polymerization chemically with the use of
peroxides and nitroxides. In order to synthesize fabrics with the capability to capture
CO2, monomers must be attached onto the backbone of the fabric. Thus, the monomer
we select must meet two main criteria: it must be able to attach to our selected fabrics
(PET, HDPE, Nylon) and must be capable of capturing CO2 from ambient air. As a
result, our selected monomers contain a vinyl group which allows the production of a free
radical, providing a site for the monomer to attach onto the substrate. The monomers
also contain primary amine groups, allowing for CO2 sequestration. The two monomers
initially selected were allylamine and butenylamine, with the bond line structure for each
shown in Figure 4 below. Both monomers are toxic as individual molecules. This tox-
icity emerges as the small monomers have vinyl groups and primary amines that could
react with key systems in the body. However, after polymerization and the removal of
unreacted monomers, the compounds can no longer be absorbed into the skin and pose
no risk of reaction in the body. Additionally, the polymers formed are largely unreactive
with the environment and thus unlikely to break off the substrate back into their original
11
form. It is more likely for a non-harmful compound, such as ammonia, to be formed in
the case of a reaction atmospheric ozone or another compound because of the lower bond
energy of C-N bonds compared to C-C bonds.
Figure 4: Bond line structure of (A) allylamine and (B) 3-butenylamine.
In this study, the CO2 sorbency of the selected monomers was verified through im-
pregnation of activated carbon with the selected monomers and characterization of CO2
capacity. According to Gholidoust et al., preparation of activated carbon samples is de-
pendent on wet impregnation, a process in which the monomers physically attach to the
activated carbon. In accordance with this method, each monomer was dissolved in 20
mL of water to form an aqueous solution [20]. This solution was then poured into a
beaker containing 5 g of activated carbon, and stirred for 2 hours at 40 ?. Next, the
samples were dried in a desiccator and massed to determine the degree of attachment
(DoA), calculated using the following equation where m0 is the original mass of the ac-
tivated carbon sample before attachment and m refers to the mass of the sample after
impregnation with monomer.
m?m0
DoA = ? 100 (3)
m0
The activated carbon samples produced were then placed in a chamber filled with 100
% CO2 atmosphere. The CO2 concentration was then measured over time to compare
the CO2 sorbency of the respective monomer solutions. All CO2 sorbency tests were
conducted at atmospheric pressure and 22 ?C for 30 min each. The results shown in
Figure 11 indicate that the impregnation of both allylamine and butenylamine increased
the sorbency of the raw activated carbon by a significant amount.
12
A large portion of the monomer impregnated into the activated carbon was likely
bound into the bulk of the material. These amine-containing monomers bound within
the bulk are likely to contribute less to the CO2 capacity as opposed to surface amines,
as the CO2 can only attach to bulk amines after diffusion into the activated carbon [20].
Due to this phenomenon, the impregnated carbon samples do not effectively model the
expected absorbency of our functionalized fabrics. Instead, CO2 absorbance was used as
a qualitative measure to both confirm the CO2 sorbency of our fabrics and compare the
two monomers. As both impregnated carbon samples showed improved CO2 sorbency as
opposed to raw activated carbon, we concluded that both allylamine and butenylamine
absorb CO2 from ambient air.
As shown by Figure 11, allylamine exhibits a significantly higher CO2 sorbency as
opposed to butenylamine per unit mass of loaded activated carbon. The difference in
mass-normalized CO2 sorbency can likely be attributed to the difference in amine density
between the two monomers we have selected. The bond line structures in Figure 4 show
that allylamine contains one fewer CH2 group than butenylamine, yielding a difference
in molar mass of 14.03 g(mol-1). Despite the additional CH2 group on butenylamine,
both monomers contain the same number of amine groups per molecule. Consequently,
allylamine contains 24.57 % more primary amine groups per gram than butenylamine.
Therefore, 1 g of allylamine would theoretically have a greater CO2 capacity than 1 g of
butenylamine.
The difference in primary amine density may also have an effect on the kinetics of the
CO2 capture reaction depicted in Figure 17. When placed in environments with excess
concentrations of CO2, such as our test chamber, the literature predicts that amine-based
sorbents follow pseudo-first-order kinetics depending on the concentration of unreacted
amines [53]. As allylamine-impregnated carbon is likely to contain a higher concentration
of unreacted amine groups with respect to mass, the CO2 capture reaction for allylamine
is likely to proceed at a faster rate. Thus, the difference in amine density is likely to
create a discrepancy in the kinetics of CO2 capture in addition to overall capacity. As
shown from our experiments, when comparing the two monomers, allylamine exhibited
13
both an improved CO2 capacity and a superior rate of kinetics for the intended reaction.
Unfortunately, after several trials of grafting the two selected monomers, we were
able to determine that the use of butenylamine was no longer cost-effective. Therefore,
we looked into alternative monomers with amine groups for CO2 extraction and vinyl
groups that allow for grafting. The monomer we selected is diallylamine, and the bond
line structure is given in Figure 5 below.
Figure 5: Bond line structure for diallylamine
As opposed to our first two monomer selections, diallylamine contains two vinyl groups
which could help to improve the eventual degree of grafting (DoG) of our monomer onto
the consumer-grade fabrics. We concluded from our previous activated carbon experi-
ments on allylamine and butenylamine that the amine group on the monomer will abstract
CO2 from the atmosphere. Due to this assumption, we concluded that it would not be
necessary to conduct CO2 tests on diallylamine-impregnated activated carbon.
3.2.2 Radiation Grafting
After validating that the selected monomers do chemisorb CO2 from the atmosphere, we
used them to functionalize three commercial-grade fabrics via graft polymerization. This
chapter reports the synthesis of graft copolymers between allylamine as well as buteny-
lamine monomers and PET, HDPE or Nylon backbone chains via radiation-induced graft
polymerization. Figure 6 displays the mechanism of radiation-induced graft polymeriza-
tion for each of the three substrate polymers [15, 18, 23, 38, 17]. Incident electron beam
radiation causes the abstraction of hydrogen?the ejection of a H from an R?RCH2 group
in the backbone polymer?that produces a carbon-centered free radical on the substrate.
This carbon-centered free radical attacks the vinyl group of the monomers, forming a
14
sigma bond and unpaired electron on the newly attached monomer, as indicated in the
first propagation step. Monomers continue to add to the growing grafted polymer chain
through this mechanism, continuing propagation.
Figure 6: Mechanism of radiation-induced graft polymerization (RIGP), including desired
grafting and undesired crosslinking [17]
Before any radiation grafting, the three fabrics were cut into 50-mg square swatches.
To synthesize our novel co-polymers, we needed to remove the protective polylactic acid
(PLA) coating the manufacturer applied to the PET and HDPE. Refluxing both PET
and HDPE in tetrahydrofuran (THF) at a ratio of 20 mL of THF per 50 mg of sample for
4 hours at 70 ?C removed the PLA coating [17]. After refluxing the PET and HDPE, all
three fabrics were washed in methanol and deionized water to remove organic contami-
nants. We expect that atmospheric organic particulates could react with carbon-centered
free radicals to inhibit grafting. After drying, the fabrics were stored and irradiated in
12-mL scintering vials purged with argon.
15
This work reports the first single-step synthesis of fabric CO2 sorbents via radiation-
induced graft copolymerization. The procedure used to irradiate fabrics and grow a graft
copolymer via chain polymerization, in Figure 7, follows work reported by Yang et. al.,
Wu et. al., and Dietz et. al. [58, 56, 13, 17].
Figure 7: Schematic of process used in synthesizing and characterizing novel graft copoly-
meric CO2 sorbents
We accessed the Medical Industrial Radiation Facility (MIRF) at the National Insti-
tute of Standards and Technology (NIST; Gaithersburg, MD) to perform electron beam
irradiation. Dosimetry was performed by correlating the radicals produced in alanine foil
16
to the counts on a Faraday cup behind the sample stage [13, 12, 17]. A linear correlation
between counts on the Faraday cup and dose allowed for the precise tuning of ionizing
radiation incident on the fabrics. Fabrics?consisting of individual fabrics stored in vials
with an inert Argon atmosphere?were irradiated to either 150 kGy or 250 kGy at an
energy of approximately 10.5 MeV to generate approximately 1017 to 1019 free radicals
per gram in PET, HDPE and Nylon [15]. During irradiation, fabrics were kept in a heat
insulated container with dry ice and rotated through the beam path on a turntable, as
depicted in figure 8.
Figure 8: Rendering of turntable used to cycle fabric samples, depicted as white rectangles
in clear vials, through the electron beam path, depicted as the translucent yellow line.
The cryogenic temperatures reduce the rate of radical recombination, increasing the
fraction of radicals available to react with monomer [13, 17]. These conditions con-
tribute to an increase in monomer attachment. After irradiation, approximately 10 mL
of monomer was introduced to each vial containing the substrate fabrics. Upon introduc-
tion of the monomer, free-radical chain graft copolymerization proceeded as depicted in
figure 6.
After grafting, the fabrics were washed to remove any homopolymerization. Washing
consisted of rinsing, vortexing, then sonicating samples in aqueous and organic solvents?
specifically water and methanol. Sample fabrics were then dried in a vacuum desiccator
to remove all solvents while avoiding unintended CO2 sorbency that could interfere with
17
later characterization.
3.2.3 Characterization by Fourier Transformation Infrared Spectroscopy
In a Fourier Transform Infrared (FTIR) Spectrometer, functional groups in a sample
are measured by the wavelength of light they absorb for vibrational transitions [26].
FTIR measures the interference between a split beam of IR light covering a wide range
of wavelengths; one beam passes through the sample and interferes with the one that
does not. This produces an interference pattern that yields a spectra of absorption as a
function of wavenumber when subject to Fourier analysis. This absorption spectra reveals
the functional groups in the sample; each peak corresponds to a characteristic vibrational
transition for each group.
Fourier-Transform infrared attenuated total reflectance (FTIR-ATR) spectrum was
recorded with the Thermo Fisher Scientific (Waltham, MA) Nicolete iS-50 FTIR with a
Smarte iTX diamond window ATR module. In ATR mode, the beam is reflected rather
than transmitted through the sample. Measurements consisted of an average of 120 scans
performed at a resolution of 2 cm?1. Growth in peaks at 764, 1,619, 3,360 and 3,442 cm?1
that correspond to NH wag, bend, symmetric and asymmetric stretch, respectively, along
with 1,261 cm-1 for CN stretch confirm monomer attachment [17, 26].
3.2.4 Characterization by Scanning Electron Microscope and Energy Dis-
persive Spectroscopy
The morphology and elemental composition of the fabric samples after both grafting and
CO2 adsorption testing were determined with scanning electron microscopy (SEM) along
with an energy dispersive spectroscopy (EDX) analyzer. The Hitachi S-3400 Variable
Pressure SEM (Tokyo, Japan) was used to collect the data. SEM images and EDX spectra
were obtained from three unique physical locations on each fabric sample under variable
pressure mode at 100 Pa. Additionally, elemental compositional data obtained from
linescans were analyzed and compared. All measurement processes used an accelerating
voltage of 15 kV, and the magnification used for each image remained in the range of
18
375X to 475X.
3.2.5 CO2 Capacity Characterization
CO2 capacity, typically measured in mmol of CO2 captured normalized by mass or moles
of sorbent, comprises the figure of merit for CO2 sorbent devices [38, 17]. Characterizing
CO2 capacity typically requires exposing the sorbent to an atmosphere with a well con-
trolled CO2 concentration and measuring its decrease. Our work followed the methodol-
ogy established by Ho et. al., where an airtight chamber was filled with a known CO2/Ar
gas mixture and measured the drop in concentration of the former using an IR sensor
[22, 17]. To better resolve drop in CO2 concentration, this study included the design,
fabrication, and use of two chambers?one for fabrics and one for activated carbon. As
indicated in Figure 3, the amount of monomer directly impacts the amount of CO2 ad-
sorbed. Thus, we expected the fabric samples to adsorb far less CO2 given the smaller
total amount of monomer attached per sample as compared to activated carbon [17]. To
properly resolve the smaller expected CO2 adsorption from fabrics, we designed a smaller
chamber. Figure 9 shows a render of both chambers and a schematic of their functional
components.
a. b. c.
Figure 9: A generalized schematic for the (a) test chamber and a (b) detailed rendering
of the test chamber for fabric samples and (c) activated carbon samples [17]
Both the cylinder and lid of the chamber designated for testing activated carbon
samples were manufactured via fused filament fabrication additive manufacturing. We
19
fabricated the box of the chamber designed for fabric samples with laser-cut acrylic joined
with acrylic cement and sealed with silicon rubber sealant.
Both chambers used a CozIR infrared sensor to measure the concentration of CO2 in
atmosphere through the absorbance of characteristic IR frequencies. A metal filament
heated through joule heating generates a range of IR frequencies. A detector at the op-
posite end of the sensor records absorbance of characteristic wavelength bands associated
with CO2: 2.7 ?m, 4.3 ?m, and 15 ?m. To confirm that the chamber and sensor ac-
curately measure CO2 concentration in the chamber, we created a calibration curve to
correlate the measured CO2 fraction with a carefully controlled input CO2/Ar ratio. We
manipulated the input gas mix with the use of two flowmeters. Figure 10 indicates that
the measured CO2 concentration closely matches the input CO2 fraction. This calibration
curve also supports that both fabric and activated carbon tests were performed within
the proportional region of the detector.
Figure 10: Calibration curve of input CO2 concentration and measurements identified by
CozIR sensor [17]
20
Before measuring the capacity of AC or fabric sorbents, we first collected a back-
ground decay in CO2 concentration, which likely stemmed from a leak in the chamber.
Since neither chamber was prepared under clean room conditions, contaminants likely
prevented a perfectly airtight seal [43]. Once ready to measure, the sample was inserted,
and the servo was actuated to lower a lid over the sample. This protected the sample
from unintended CO2 adsorption before measurements were taken. We then flooded the
chamber with a known CO2/Ar ratio. Once the atmosphere in the chamber stabilized at
the desired CO2 concentration the servo was actuated to expose the sample. The CozIR
sensor measured CO2 concentration as a function of time, and equation 4 enabled the
identification of CO2 capacity.
n?CO (1? n)?Ar
capacity(mmol(g?1)) = 10?6?[CO2]nV (
2 + )M?1 (4)
mCO2 mAr
3.3 Results and Discussion
3.3.1 Confirmation of Monomer CO2 Capacity
Following the procedures for wet impregnation of activated carbon from Ho et al. and the
degree of attachment (DoA) measurements determined by equation 3, the attachment of
allylamine and butenylamine on activated carbon was quantified in Table 1.
Table 1: Degree of Attachment for the wet impregnation of both monomers [17]
Monomer Degree of Attachment Standard Error
Allylamine 73.12 7.05
Butenylamine 71.11 11.48
Samples were prepared by mixing the monomer and activated carbon in a 0.8 ratio by
weight, resulting in a theoretical DoA of 80 %. The discrepancy between experimental
and theoretical DoA likely stemmed from the preparation procedure. The attachment
occurs while the carbon is in a slurry; thus, the transfer of the mixture between containers
during sampling leaves behind small amounts of residue.
The CO2 capacities of both unloaded and monomer-impregnated activated carbon can
be found in Figure 6. All CO2 sorbency tests were conducted at atmospheric pressure
21
and 22 ?C for 30 minutes, and the error bars result from standard error calculations
from six samples for each attachment configuration. As evident in figure 11, impregnated
activated carbon outperformed unloaded activated carbon in terms of CO2 capacity by a
significant amount.
Figure 11: Adsorbance capacity of raw and monomer impregnated activated carbon [17]
Much of the monomer impregnated into the activated carbon was likely bound into
the bulk of the material, reducing the overall capacity of the absorbent. Because bulk
amines could only absorb CO2 following diffusion through the activated carbon, they
contributed to CO2 capacity less than surface amines did [20]. As a result, the results
from activated carbon testing can only be used to compare capacity qualitatively and
does not accurately approximate the expected absorbances of functionalized fabrics.
Activated carbon samples impregnated with either monomer exhibited higher capac-
ity than the unprocessed material; therefore allylamine and butenylamine successfully
chemisorb CO2. Due to its significantly higher CO2 capacity on activated carbon, ally-
lamine exhibited better mass-normalized sorbency properties than butenylamine. The
22
disparity between the two monomers can be attributed to the likely higher amine den-
sity with allylamine. Allylamine has a lower molar mass than butenylamine by 14.03
g(mol?1). Because of this difference in molar mass, allylamine contains 24.57 % more
primary amine groups per gram than butenylamine, resulting in a greater CO2 capacity
per gram.
The higher primary amine density in allylamine may also result in faster kinetics
of the CO2 capture reaction. In excess concentrations of CO2, literature dictates that
amine-based sorbents follow pseudo-first order or second order kinetics depending on the
unreacted amine concentration [53]. Within the thirty minutes allotted for capacity mea-
surements, more monomers of allylamine reacted to absorb CO2. It can be concluded that
allylamine exhibited greater CO2 capacity than butenylamine and a non-loaded standard
because of its greater amine density and faster kinetics of absorption.
3.3.2 Monomer Attachment
After grafting allylamine and butenylamine onto the selected fabrics, we used three meth-
ods to confirm the formation of a graft copolymer composed of clean fibers and amine-rich
monomers: 1. gravimetric analysis; 2. Fourier-transform infrared (FTIR) spectroscopy;
and 3. energy dispersive spectroscopy (EDX). This study also employs the use of scanning
electron microscopy (SEM) in order to assess morphological changes to the fibers. This
study implements gravimetric analysis in order to confirm that our clean fabrics exhibit
an increase in weight upon grafting. Assuming that no mass is gained or lost due to
irradiation, any mass change can be attributed to the addition of nitrogenous monomers
onto the fabric. This mass change is reported via degree of grafting (DoG), calculated
using equation 5.
m?m0
DoG = ? 100 (5)
m0
It is essential to thoroughly clean the fabrics to ensure that any increase of mass is due to
the grafting of monomer onto the backbone, rather than the formation of homopolymer
physically attached to the substrate. After three cycles of washing and desiccation, the
23
DoG values for all six polymers were calculated and are shown in Figure 12 below.
Figure 12: Degree of grafting of allylamine (red) and butenylamine (blue) as a function
of dose for all three fabrics [17].
As shown in the figure, fabrics grafted with butenylamine exhibited a greater monomer
attachment than those grafted with allylamine. This disparity in degree of grafting can
likely be attributed to the different structures of our two monomers. Butenylamine con-
tains one extra CH2 group on the chain. Therefore, the butenylamine has an additional
carbon on the backbone separating the amine and the vinyl group. As indicated by
the aforementioned free-radical mechanisms, the graft copolymerization step requires re-
actions between free radicals generated on the vinyl group and the ?-bond of another
monomer. Thus, we can conclude that the rate of copolymerization is determined by
the reactivity of the free radical generated by radiation. Unfortunately, the presence of
amine groups have been shown to stabilize and subsequently eliminate free radicals [55].
This phenomenon may be responsible for the difference in DoG values, as the amine
group on allylamine is one carbon closer to the vinyl group when compared to buteny-
lamine. Additionally, as expressed previously, allylamine exhibits a greater amine density
when compared to butenylamine. When combining these effects, the allylamine-centered
radicals are less reactive, which may explain the difference in grafting between the two
monomers.
In addition to gravimetric analysis, this study also employs the use of FTIR to con-
firm the presence of specific chemical bonds following the grafting procedure. Specifi-
24
cally, we aim to identify the growth of certain peaks associated with bonds present on
our monomers. Each of our two monomers will only possess one functional group once
grafted to the substrate: primary amines bonded to an sp3 hybridized carbon. On an
FTIR spectra, primary amines exhibit peaks at 764, 1619, 3360, and 3442 cm?1, corre-
sponding to the vibrational modes of NH wag, bend, symmetric and asymmetric stretch,
respectively [8]. Additionally, successful monomer attachment is indicated by the pres-
ence of a peak at 1261 cm?1, which corresponds to the CN stretch [27]. Upon comparison
between un-grafted and grafted fabrics, we can confirm attachment of the nitrogenous
monomer with the identification of the peaks given above. Figure 13 below shows exam-
ples of three sample spectra for each fabric.
25
Figure 13: FTIR spectra of HDPE, PET, an2d6 Nylon highlighting the growth in the peaks
associated with primary amines [17]
As shown by the figure, both PET and HDPE exhibit small growth in peaks for the
NH stretching mode when grafted with allylamine and butenylamine. These small peaks
at 3442 cm?1 and 3360 cm?1 combine to form one broad peak centered between the two
values above. Unlike PET and HDPE, however, Nylon does contain an amine group.
Because the grafted monomers make up a very small fraction of the final Nylon-monomer
copolymer, the growth of the NH stretch and wag peaks due to monomer attachment is
concealed by the characteristic peaks of Nylon. The signal from Nylon?s secondary amine
also conceals the growth of the CN stretch peak. The CN stretch corresponds to the
location of a CO stretch in PET [28, 34]. The signal from the CO group in PET may
overwhelm the peak growth at 1261 cm?1 due to monomer attachment, explaining the
lack of increased intensity shown by PET samples. Given the growth of both NH stretch
peaks at 3442 cm?1 and 3360 cm?1 in all samples and NH bend and wag at 1619 cm?1 and
764 cm?1 in PET and HDPE, we can qualitatively confirm monomer attachment. Table
2 below highlights the DoG values of the fabrics used to produce the spectra highlighted
in Figure 13.
Table 2: Degree of Grafting for the fabrics with FITR spectra displayed in figure 13
Fabric DoG Butenylamine grafted DoG Allylamine grafted
PET 17.36 3.76
HDPE 7.05 4.85
Nylon 11.85 2.30
Finally, EDX was implemented to confirm the success of grafting through elemental
analysis. Specifically, we look to use EDX to confirm the increase in the nitrogen content
present in our fabrics, due to the addition of nitrogen-rich monomers. Figure 14 below
provides a comparison between the nitrogen content present in the clean fibers and the
grafted fabrics.
27
Figure 14: Nitrogen to carbon ratios obtained from EDS for both PET and Nylon fabrics
generated with allylamine and butenylamine compared to ungrafted samples [17]
As shown in the figure above, each grafted sample exhibited a greater nitrogen content
with respect to each ungrafted fabric. The error bars shown represent the standard
deviation between the nitrogen concentration recorded at three different points across
the fabrics. Although EDX provides us with raw values, the comparison of elemental
concentration between grafted samples is primarily used as a qualitative measure of the
success of the grafting process. Thus, we can use Figure 14 to qualitatively confirm the
attachment of our nitrogenous monomers onto the various fabrics. Qualitatively, the
growth in nitrogen content of HDPE was much lower than the growth shown in PET and
Nylon samples. This is likely due to the fact that the volume and texture of the HDPE
fibers was such that only a small percentage of the interaction volume of the electrons
used in EDX overlapped with the bulk fiber. This produced a spectra with a very low
signal-to-noise ratio, thus the EDX results for HDPE samples were not reported.
Additionally, the EDX mapping feature allows for a direct view of the dispersion of
the amine-rich monomer; an example is provided in Figure 15 below. The figure clearly
shows the presence of nitrogen in the fabric. As raw PET does not contain any nitrogen,
we can conclude that the presence of nitrogen confirms the attachment of our monomers.
28
Figure 15: EDS map of PET grafted with butenylamine, DoG of 14.48 %. Observed
nitrogen is colored red and the secondary electron image is colored red [17]
Finally, SEM was used to evaluate the morphological changes to the fibers due to
radiation and the introduction of our monomer. An image of PET fibers before and after
grafting is given in Figure 16 below.
29
Figure 16: PET fibers before (A) and after (B) grafting of allylamine monomer at 150 kGy.
No significant structural damage or homopolymer buildup was observed after radiation
[17]
The figure shows that the fibers composing the fabrics maintained similar physi-
cal characteristics following the grafting procedure. Additionally, there was little to no
buildup of homopolymerized allylamine or butenylamine on the surface of the fibers, en-
suring that any monomer present is indeed covalently bonded to the backbone of the
substrate. The figure also indicates that the grafting was not localized, but instead grew
uniformly along the polymer backbone. The absence of visible localized polymerization
can indicate a low probability of propagation along grafted chains. Rather, the reaction
favors the grafting of small chains attaching to the backbone. Figure 16 also shows the
minimal amount of degradation due to radiation, as the grafted fabric looks morpholog-
ically similar to the raw fabric. The polymers that are used as a substrate for grafting
30
primarily undergo crosslinking as a result of ionizing radiation; however, it is expected
that limited aging via chain degradation may have occurred [15, 3].
3.3.3 CO2 Capacity, Kinetics, and Mechanisms
Figure 17 depicts a characteristic decay curve of CO2 concentration as a function of time
when exposed to a PET-allylamine copolymer. As indicated in figure 3, the reaction
between CO2 and primary amines reported in this study depends on both the concentra-
tion of primary amines and atmospheric CO2. Thus, the reaction theoretically matches
second-order kinetics. Given that the method of characterizing CO2 adsorption produces
time-dependent decay curves, this study includes analysis of the reaction kinetics be-
tween our novel sorbents and CO2. The CO2 concentration decay over time was used to
confirm second order kinetic behavior and identify the rate constants or adsorption for
each copolymer.
Figure 17: A characteristic decay in CO2 concentration from adsorption on allylamine-
grafted PET as a function of time. The insert displays a second order kinetics fit with
an effective rate constant of 6.54 x 10-5 (M(s))?1 [17]
31
The decay curve of CO2 in atmosphere in the presence of fabric depicted in figure 17
was used to identify the rate constants for the reaction between the sorbents and CO2. As
indicated by a comparison of R2 values between first and second-order kinetic models, the
latter better describes the CO2 sorbent reaction [17]. This contradicts previous findings
of pseudo-first order kinetic behavior [53]. Table 3 gives the second order kinetic rate
constants for each copolymer. Given the similar rate constants between each copolymer,
the kinetic behavior of the material system described in this study does not depend
significantly on the fabric or monomer.
Table 3: Rate constants for second order kinetic behavior for CO2 sorbency [17]
keff Standard deviation n
(mole(s))?1 (mole(s))?1
Allylamine 4.61?10?5 1.96?10?5 8
PET
Butenylamine 3.04?10?5 1.42?10?5 10
Allylamine 3.83?10?5 2.13?10?5 10
HDPE
Butenylamine 4.51?10?5 3.21?10?5 9
Allylamine 5.11?10?5 4.62?10?5 9
Nylon
Butenylamine 4.27?10?5 2.09?10?5 5
Figure 18 displays the CO2 capacities of each novel copolymer synthesized via RIGP:
PET, HDPE, and Nylon grafted with allylamine and butenylamine. The error bars in the
figure represent a 2.5 % uncertainty per recommendation by CozIR?e.g. error bars of
?1.25 % for a raw concentration of 50 %. These error bars are treated as shot noise and
give the noise floor of the CO2 concentration measurements. Any fabric where the error
bars do not capture the 0 mmol(g?1) value exhibited statistically significant absorbance.
All CO2 sorbency tests were performed in a well controlled atmosphere of 50 % CO2 and
50 % Ar at 22 ?C, per the procedure described in 3.2.5.
Figure 18.a. presents CO2 sorbency in units of mmol of CO2 absorbed normalized
by the mass of each sorbent, including both fabric and grafted monomer. This is the
most common figure of merit for CO2 sorbency and enables comparison between sor-
bents reported in this study and prior work reported in literature [48, 60, 61]. Figure
18.b. presents CO2 sorbency in units of mmol of CO2 absorbed per mmol of monomer
attached. This measure allows for an evaluation of the extent of reaction between pri-
32
mary amines in the grafted monomer and CO2. Given that the CO2 sorbency reaction
presented in Figure 3 predicts a mmol(mmol)?1 capacity of exactly 0.5, these values allow
for an evaluation of the fraction of monomers that react to absorb CO2.
All copolymers with a DoG of higher than 4 % achieved statistically significant CO2
capacity, indicating that they do behave as CO2 sorbents. This study is the first to
report the successful fabrication of CO2 sorbents through single-step radiation-induced
graft polymerization. Table 4 further confirms that the single step RIGP produced CO2
sorbents; all untreated fabrics did not adsorb CO2 with statistical significance, but the
grafted fabrics did.
Table 4: CO2 capacities of untreated fabrics and fabrics functionalized via RIGP; the
former did not exhibit statistically significant capacity while the latter does [17]
DoG (%) CO2 capacity mmol(g
?1) Instrumental Error mmol(g?1)
Untreated 5.1?10?5 .00025
PET Allylamine 3.516 .00044 .00033
Butenylamine 14.487 .0012 .00031
Untreated .00014 .00061
HDPE Allylamine 3.920 .0013 .00085
Butenylamine 7.058 .0014 .00085
Untreated .00014 .00031
Nylon Allylamine 4.848 .00053 .00032
Butenylamine 12.390 .00076 .00028
33
a. b.
Figure 18: CO2 sorbent capacity reported in mmol(g
?1) (a) and mmol(mmol?1) (b) [17]
As indicated in figure 18, the copolymers synthesized via the novel single-step RIGP
method investigated in this study achieved statistically significant CO2 sorbency. This
result quantitatively validates the single-step RIGP synthesis as a successful means of
fabricating CO2 sorbents. As indicated in figure 18, CO2 capacity increases as a function
of degree of attachment, similar to CO2 sorbents reported in literature [58, 56]. Thus,
enhancing monomer attachment would increase CO2 capacity, potentially to a value com-
parable to those reported in literature.
Figure 18.b. reports capacity in mmol(mmol?1), a measure useful for identifying the
extent of amine-CO ?12 sorbent reaction. Interestingly, the capacity in mmol(mmol ) sug-
gests the presence of alternate mechanisms of sorbency as some of the highest grafted
fabrics achieve a capacity higher than 0.5 mmol(mmol?1), the theoretical maximum for
amine-based chemisorption. One potential mechanism for CO2 extraction may stem
from the formation of charged groups from chemisorption, as indicated in figure 3. These
charged groups may contribute to physisorption of CO2 given their propensity for strong
intermolecular interactions with CO2 molecules with charge separation between oxygen
34
and carbon atoms. The potential for physisorption is purely speculative based on our
understanding of the dominant chemisorption reaction. Though further work is required
to confirm physisorption, figure 18.b clearly evidences mechanisms other than posited in
figure 3.
4 Comparison of Novel CO2 Graft Copolymer Syn-
thesis by Radiation and Chemical Initiated Graft
Polymerization
4.1 Introduction to Chemical-Induced Graft Polymerization (CIGP)
Prior research by both Yang et. al. and Wu et. al. reported high CO2 capacities?6.22
and 4.8 mmol(g?1) respectively?through a two-step synthesis process involving both
chemical and radiation-induced graft polymerization [58, 56, 17]. We published the first
successful single-step synthesis of copolymeric CO2 sorbents through RIGP, but our low
monomer attachment precluded high CO2 capacity [17]. In this chapter, we report the
fabrication of allylamine and diallylamine grafted onto PET and Nylon using peroxide
and nitroxide initiators. Chemically initiated graft polymerization (CIGP) offers several
advantages over RIGP, namely in total environmental cost of energy for radicalization
and relative ease of grafting procedure. We also discuss characterization of monomer
attachment through gravimetric analysis, FTIR, and EDX.
4.2 Methodology of Peroxide and Nitroxide-Induced Grafting
While radiation-induced grafting provides better control in terms of radiation dose and
procedure, it causes relatively low radical yield as discussed in previously published liter-
ature [47, 46]. We explored chemical-induced grafting using peroxides and nitroxides as a
potential alternative, so as to increase CO2 capacity while retaining a digestible, effective
grafting procedure.
35
Benzoyl peroxide and (2,2,6,6-Tetramethylpiperidin-1-yl)oxyl (TEMPO) were explored
as initiators for chemical-induced graft polymerization. Dissociation of such initiators
creates the primary free radical species of C6H5COO
? and/or the secondary free radical
species of C6H
?
5 during polymerization. Through hydrogen abstraction, these primary
and secondary free radical species form carbon-centered PET macroradicals, like those
obtained through radiation. The free radical on the PET attacks the vinyl group of the
monomers, creating a sigma bond and unpaired electrons. After this first propagation
step, more monomers continue to bind onto the growing grafted polymer chain.
Following the same preprocessing procedure as in section 3.2.3, 50-mg square swatches
of PET and Nylon-6 fabric were refluxed in THF to remove the protective polylactic
acid coating. To begin, we first dissolved the initiators: benzoyl peroxide in toluene and
TEMPO in ethanol, at 10-20 wt. % and 5-10 % respectively. These ratios followed closely
with those of literature [46]. Approximately 5 mL of each solution were prepared in the
10-mL vials used, filling half the vial. Solutions were thoroughly purged with argon to
remove dissolved oxygen, which if present can form unwanted peroxides by reacting with
radicals on the polymer backbone. These radicals must remain available for our intended
monomer to polymerize onto. The PET and Nylon fabrics were added into the solutions
for 4 hours before removing the solutions from the vials. Afterwards, approximately 10
mL of 100 % monomer solutions of allylamine and diallylamine were introduced to the
fabric samples in the vials and the monomers were left to polymerize overnight.
After grafting, potential homopolymers were washed off with several cycles of rinsing,
vortexing, and sonication. Rinses were done alternating between water and methanol.
Fabrics were then dried in vacuum desiccators and inert atmosphere for storage before
CO2 capacity testing.
4.3 Results and Discussion
In addition to radiation-induced graft polymerization (RIGP), chemically initiated graft
polymerization (CIGP) was also investigated. Benzoyl peroxide as well as nitroxide were
each used as radical initiators. The attachment of nitrogenous monomers was charac-
36
terized gravimetrically via degree of grafting calculation, FTIR spectroscopy, and EDX
elemental analysis.
Gravimetrically, CIGP-treated fabrics showed attachment of nitrogen-containing monomer.
Figure 19 shows that while many fabrics achieved positive degrees of grafting, there were
multiple samples that exhibited negative DoG values. These negative values can be
attributed to ?-scission of the fibers? polymer chains, which is a known result of the
radicalization of benzoyl peroxide as well as other peroxy derivatives [14]. This scission
results in the loss of small fragments of fiber due to post-grafting processing. In addition,
the organic and aqueous solvents used swell both nylon and PET, enhancing the proba-
bility of fibers entering the solution during post-grafting washes. This loss of fiber mass
may have artificially decreased the measured degree of grafting values to levels below 0.
Figure 19: Degree of grafting values for allylamine-grafted (orange) and diallylamine-
grafted (blue) PET and Nylon fabrics that have undergone chemically initiated graft
polymerization (CIGP) treatment.
Observed using FTIR, characteristic peaks at 764, 1,619, 3,360 and 3,442 cm?1 indi-
cate the presence of primary amine functional groups; additionally, the peak at 1,261 cm?1
represents the CN stretch that would form between the grafted monomer and the fabric
37
substrate [8]. Figure 20 also shows the collected FTIR spectra for ungrafted, allylamine-
grafted, and diallylamine-grafted fabrics. The Nylon spectra do not show great increase
in intensity at the aforementioned peak locations; however, this lack of growth can be
attributed to nylon?s structure, which already contains amines in great quantity. These
amines likely conceal the growth of the characteristic peaks. Although the PET spectra
do not show a dramatic increase in any of the characteristic peak intensities, the peak
corresponding to the CN stretch did noticeably increase. This increase qualitatively con-
firms monomer attachment in PET. However, monomer attachment on nylon cannot be
confirmed with FTIR alone.
38
Figure 20: FTIR spectra of ungrafted, allylamine-grafted, and diallylamine-grafted Nylon
(top) and PET (bottom).
EDX was also used to confirm monomer attachment. Linescans obtained of both PET
and Nylon that underwent CIGP showed that these materials contained higher amounts
of nitrogen, which is indicative of successful grafting of nitrogen-containing allylamine and
diallylamine. Figure 21 shows the average ratio between nitrogen-attributed and carbon-
39
attributed X-ray counts for ungrafted, allylamine-grafted, and diallylamine-grafted PET
and Nylon samples. The rather high variance in the obtained data can be attributed to the
uneven geometry present in fibrous materials, which results in a proportionately uneven
spatial emission of X-rays from the sample to the EDX detector. The low signal-to-noise
ratio caused by this unevenness likely resulted in such high variance. Nevertheless, the
data show an increase in nitrogen content in the analyzed grafted samples.
40
Figure 21: EDS-obtained ratios between nitrogen-attributed and carbon-attributed X-ray
counts on ungrafted, allylamine-grafted, and diallylamine-grafted PET and Nylon.
The degrees of grafting for the samples used in the EDX and FTIR analyses are
shown in Table 5 below. Specifically, samples with low DoG values were analyzed with
EDX to demonstrate that even fabrics with gravimetrically low attachment of monomer
still exhibit higher nitrogen content than ungrafted samples, thus indicating successful
41
grafting.
DoG (%)
Diallylamine Allylamine
PET 2.08 0.66
Nylon 0.21 8.14
Table 5: Degree of grafting values for samples analyzed via EDS and FTIR
4.4 Conclusions on Comparing Radiation Induced and Chemical-
Initiated Graft Polymerization
Chemically initiated graft polymerization (CIGP) offers several advantages as compared
to RIGP, specifically with respect to throughput of sorbents and environmental impact of
the synthesis process. Our successful development of a single-step RIGP process required
the use of an electron beam capable of outputting 10 MeV electrons and delivering 250
kGy of total dose [17]. This equipment is expensive, difficult to obtain, and requires high
energy costs. Thus, CIGP-driven synthesis lowers the barrier to commercial or research
production of the novel CO2 sorbents reported here. In addition, the nitroxide initiator
used is soluble in water and ethanol, two environmentally friendly solvent. Thus, the
CIGP procedure described in this chapter enables environmentally-friendly synthesis of
CO2 sorbents.
Both RIGP and CIGP reactions are highly vulnerable to peroxide formation; an un-
desired reaction in which molecular oxygen in the atmosphere or solution reacts with
a radical on the polymer backbone to form peroxide. Peroxide formation eliminates
backbone-centered radicals necessary for grafting. We performed CIGP with ethanol as
the solvent for nitroxides; given the high solubility of molecular oxygen in ethanol, which
has an Oswald distribution value of 0.233, preventing peroxide formation from inhibit-
ing graft polymerization posed a considerable challenge [2]. Despite using a glovebox to
perform reactions in an inert atmosphere and sparging of solvents with argon, we still
achieved far less grafting with CIGP when compared to that of RIGP. Though we achieved
approximately 17 % DoG with RIGP, we could only achieve 9 % DoG with CIGP. Effec-
tive application of CIGP to fabricating CO2 sorbents therefore requires extreme caution
42
to prevent molecular oxygen from suppressing grafting. CIGP also poses specific chal-
lenges with respect to characterization. Gravimetric analysis, which produced the degree
of grafting measure commonly used to assess monomer attachment in graft polymeriza-
tion, may lose accuracy in chemically initiated copolymers [13, 17]. Gravimetric analysis
relies on the assumption that any change in mass stems from added monomer. However,
the initiating solvent used in CIGP typically swells the polymer, potentially removing
fibers from the fabrics. This would decrease the mass of the fabrics, artificially lowering
the measured DoG. Since Nylon and PET swell to different degrees in ethanol, this effect
differs between fabric samples [57].
5 Economic and Environmental Analysis
5.1 Economic Analysis
In the economic analysis of the synthesis of carbon capturing fabrics, four main aspects
were considered: cost of initiation, cost of monomers, cost of fabric, and cost of CO2
removal. The cost of initiating the polymerization was considered using the electricity cost
associated with the electron beam or the cost of chemical initiators. The capital cost of
the electron beam is $300,000-$500,000; however, this cost was neglected for this analysis
because, in the long run, a net positive operating cost will overcome the capital cost and
the electron beam can be used for other tasks. The bulk cost of amine-based monomers,
chemical initiators, and fabrics were obtained from sources online including Sigma-Aldrich
and ICIS; however, these costs are likely higher than the minimum purchasing cost [69,
70, 71, 72, 73, 74]. The steam was assumed to be the more expensive high pressure steam,
and the cost was determined from Warren Seider?s chemical process design textbook [49].
The masses of monomer and chemical initiator needed to synthesize the fabric were
assumed to be equal to 5 times the mass of the fabric, similar to the ratios used in
our chemical grafting procedure. This assumption is reasonable because the amount of
monomer grafted to each sample is certainly less than the initial mass of the fabric, and,
43
even when optimized, not all of the substrate will initiate. The amount of steam required
to remove a mmol of CO2 from the fabric was determined by equation (6).
MSteamRemoval = (?CO2(SatWater))(mmolFactor) (6)
Equation (7) gives the total operating cost normalized per gram of fabric.
Costoperating = CostInitiation+CostMonomer+CostFabric+n(CostSteam)(SC)(CapCO2) (7)
Furthermore, as shown in Table 6-8, the mass of fabric required to capture one ton of
CO2 is calculated by equation (8).
22730000
MassTon,CO2 = (8)n(Capacity)
The operating cost can be leveraged along with the capacity to determine the cost of
capturing one ton of CO2. The monetary cost to capture one ton of CO2 is calculated
with equation (9)
CostTon,CO2 = CostOperating(MassTon,CO2) (9)
Ideally, the cost per ton of CO2 captured is between $100 and $350 in order to maintain
economic viability based upon the market price of captured CO2 [44].
5.2 Environmental Analysis
In preliminary assessments of the functionalized fabrics, both radiation and chemically
initiated grafting yielded minor, yet promising, results in terms of environmental impact.
Evaluating the net environmental effect of the functionalized fabrics requires assessment
of the energy used to radicalize the fabric and the CO2 captured by the fabric over time.
To quantify the environmental impact of energy used to radicalize the fabrics, this
study calculated the equivalent CO2 emissions necessary to produce the energy required.
Because the peroxide-induced and nitroxide-induced grafting procedure only requires
thorough mixing of peroxides, nitroxides, fabric, solvents, and monomers, there were
44
no direct sources of CO2 emissions. Radiation-induced grafting requires the use of an e-
beam, however. The energy to operate the beam at the required dose of 5.56 x 10?5 kWh
per gram fabric translates to an approximate 0.566 mmol CO2 emitted per gram fabric.
The synthesized fabrics can only be effective at addressing climate change if the CO2
emitted in production is offset by the CO2 captured during the lifetime of the fabric. Net
CO2 sequestration is calculated with equation (10).
NetCO2Sequestration = n(Capacity)? Emissions (10)
Table 6: Economic and environmental analyses of one-time use of the functionalized fabric
at experimentally-derived CO2 capacities. The best monomer-substrate combinations
are evaluated at each radicalization method, with radiation being on allylamine-PET,
peroxide on diallylamine-Nylon, and nitroxide also being on diallylamine-Nylon.
Radiation Peroxide Nitroxide
Net CO2 Sequestration (mmol CO2/g fabric) -0.566 2.48? 10?5 5.19? 10?4
Total Operating Cost ($/g fabric) 0.497 2.09 32.40
Functionalized Fabric Required to Capture 5.17? 1010 9.18? 1011 4.38? 1010
One Ton of CO2 (g fabric)
Cost per Ton of CO2 captured 2.57? 1010 1.92? 1012 1.42? 1012
Table 7: Economic and environmental analyses of extended use (50 cycles) of the function-
alized fabric at experimentally-derived CO2 capacities. The best monomer-substrate com-
binations are evaluated at each radicalization method, with radiation being on allylamine-
PET, peroxide on diallylamine-Nylon, and nitroxide also being on diallylamine-Nylon
Radiation Peroxide Nitroxide
Net CO2 Sequestration (mmol CO2/g fabric) -0.544 1.24? 10?3 0.0259
Total Operating Cost ($/g fabric) 0.497 2.09 32.40
Functionalized Fabric Required to Capture 1.03? 109 1.84? 1010 8.76? 108
One Ton of CO2 (g fabric)
Cost per Ton of CO2 captured 5.14? 108 3.84? 1010 2.48? 1010
Table 8: Economic and environmental analyses of one-time use of the functionalized fabric
at a theoretical CO2 capacity of 2 mmol/g. The best monomer-substrate combination of
allylamine-PET is shown.
Radiation
CO2 Capacity Required to Sell One Ton of CO2 for $100 (mmol/g fabric) 2.30? 104
Total Operating Cost ($/g fabric) 0.502
Functionalized Fabric Required to Capture One Ton of CO2 (g fabric) 1.14? 107
Cost per Ton of CO2 captured 5.71? 106
45
Table 9: Economic and environmental analyses of extended use (50 cycles) of the func-
tionalized fabric at a theoretical CO2 capacity of 2 mmol/g. The best monomer-substrate
combinations of allylamine-PET is shown
Radiation
CO2 Capacity Required to Sell One Ton of CO2 for $100 (mmol/g fabric) 3.36? 103
Total Operating Cost ($/g fabric) 0.739
Functionalized Fabric Required to Capture One Ton of CO2 (g fabric) 2.27? 105
Cost per Ton of CO2 captured 1.68? 105
Radiation-induced polymerization requires much more energy than chemically initi-
ated does due to the electron beam usage. In this analysis, the energy used for perform-
ing the chemical synthesis was neglected due to its relatively low magnitude. Due to the
higher energy usage, radiation-induced grafting produces more CO2 per gram of fabric,
which must be overcome by a larger CO2 capacity. To offset the emissions specifically, a
gram of fabric must be able to capture more than 0.566 mmol of CO2 over its lifetime.
Fabrics created through CIGP can afford to have a slightly lower capacity; however,
the production cost per gram of fabric of CIGP is much greater than that of RIGP. This
higher cost makes it much more difficult to achieve the economic viability threshold of
$100 per ton of CO2 captured. As indicated in Table 9, at the specified operating cost
and assuming a capacity of 2 mmol/g and 50 stripping cycles, the removal of a ton of
CO2 from the atmosphere would require $168,000 using radiation and $7.42 million using
nitroxides. This result indicates that radiation-induced grafting has greater economic
potential so long as it can achieve the capacity required.
Based upon the gathered data, the maximum capacity of peroxides is much lower than
that achieved by nitroxides and therefore the peroxide method is less cost-efficient despite
its lower cost per gram. With $100, 135 g of fabric can be generated through RIGP when
the fabric undergoes 50 cycles of steam stripping. With the same cost limitations, only 3
g can be generated when put through 50 steam stripping cycles with nitroxide grafting.
As a result, to capture one ton of CO2 at a cost of $100 with the calculated operating
cost, the capacity must be 3,358 mmol/g for radiation-grafting and 148,406 mmol/g us-
ing nitroxide grafting. These capacities are much greater than those achieved by other
carbon capture methods; therefore, the $100 per ton of CO2 goal can only be achieved
46
through a combination of improving CO2 capacity, extending the life cycle of each fabric,
and reducing operating costs.
6 Conclusion
6.1 Comparison Between CIGP and RIGP
Both chemically initiated and radiation-induced graft polymerization offer distinct ad-
vantages when looking to synthesize a graft copolymer for CO2 extraction. The methods
for chemical initiation discussed in this study only requires the mixing of chemicals, so no
CO2 is produced in order to initiate radicals on our fabrics. Conversely, the operation of
an electron beam requires the emission of CO2. In order to become a truly carbon nega-
tive option for sequestration, the fabrics must feature a high CO2 capacity to overcome
this barrier from electron beam operation.
Due to this, chemically initiated graft polymerization can afford to have a slightly
lower capacity due to its lack of CO2 emissions. Unfortunately, production costs for the
synthesis of a fabric produced through CIGP are significantly higher than those produced
through RIGP. Specifically, as shown in Table 9, the cost per ton of CO2 removed from
the atmosphere is much lower for RIGP fabrics than CIGP fabrics. As a result, we can
conclude that radiation grafting has a greater economic potential so long as it can achieve
the CO2 capacities required to overcome the barrier produced by the electron beam op-
eration.
Each of these methods also offer benefits and drawbacks in terms of their accessibility
and ease of operation. RIGP requires the use of an electron beam or alternative radiation
source, which can require costly reservations. However, CIGP only requires the mixing of
chemicals, so the synthesis can be performed in any available fume hood. Additionally,
all of the chemicals used in CIGP to initiate the fabric are water/ethanol soluble, allowing
for an environmentally friendly synthesis of a CO2 sorbent. Unfortunately, our ability
to characterize our CIGP fabrics was limited by the degradation of the fabrics during
47
initiation. As our fabrics swelled with our peroxide/nitroxide initiator, individual fibers
were detached from the fabric, leading to mass loss. This limited the effectiveness of our
gravimetric analysis, which served as one of the main quantitative measures in this study.
Conversely, as depicted in Figure 12, there is little to no degradation shown in the fabrics
produced through RIGP, allowing for the use of gravimetric analysis as a quantitative
measure of monomer attachment to the backbone.
6.2 Uses for Captured CO2
While outside the scope of our research, investigating the use of captured carbon pro-
vides understanding of the financial in addition to environmental benefits. In section 5, a
competitive price of $100 per gallon of sequestered CO2 is used for calculations, but more
detailed insight on the usage leads us to believe that the market for captured carbon adds
incentive for those researching methods of CO2 capture.
Captured aqueous carbon has risen in commercial value as society begins to develop
methods of using it. Enhanced oil recovery (EOR) serves as the most prominent use of
captured CO2. Injecting CO2 into oil fields decreases the viscosity and increases pressure
of the crude oil, allowing for augmented flow [41]. Additionally, 90-95 % of CO2 used
remains stored within the land using the closed-loop systems of the EOR facilities. In
1972, when this technique was first implemented, natural gas separation produced CO2,
which would then be used for the process. With more recent, effective methods of carbon
capture, EOR will remain the dominant use of sequestered CO2 [9, 10]. While the oil
industry clearly adds to the climate change problem, the positive implications of CO2
storage within the crude oil fields as well as the boosting of fracking efficiency helps to
minimize the negative environmental impact.
Long term storage in deep saline aquifers offers another option for captured carbon.
Through injection into very deep water-permeable rock formations, CO2 dissolves within
brine and remains geo-chemically trapped for thousands to millions of years [50]. Al-
though researched for as little as a few decades, geo-sequestration provides a permanent
solution for the excess CO2 that will be captured during the coming years.
48
6.3 Future Work
The main issues faced with the current grafting methodology were the qualitative mea-
suring of grafting and the relatively low CO2 capacity of the functionalized fabrics. While
gravimetric, FTIR, and EDX measurements were qualitatively helpful, they were unable
to provide sound quantitative confirmation of attachment. If provided time and ac-
cess to more surface analysis instruments capable of techniques like X-ray photoelectron
spectroscopy (XPS) and Auger electron spectroscopy (AES), we may be able to better
quantify attachment of monomers.
Given more time, we would also quantify mechanical properties of the bare and func-
tionalized fabrics. Tensile and degradation testing before and after functionalization will
provide insight on long-term viability in a commercial setting. Chemical specificity of
the CO2 capture mechanism may also be explored more by conducting tests with Ar/N2
and Ar/O2 gas mixtures and comparing with Ar/CO2 capacity data.
Additionally, the novel single-step synthesis method using RIGP could be further ex-
plored given extended use of the NIST electron beam. With more time, we could test
higher doses of radiation and potentially improve overall CO2 capacities of the sorbents
as a result. To further affirm radicalization, we could also characterize blank irradiations
of the substrates with gel permeation chromatography (GPC) analysis.
Another method aside from RIGP and CIGP that exhibits potential is with copoly-
mers. Through introduction of already-formed copolymer chains of primary amines to
the commercial-grade fabric with further catalysis, we may form a sorbent with higher
CO2 capacity.
49
References
[1] Akingbogun, O.; Muhr, L.; Vallieres, C. Capture of Carbon Diox-
ide by an Ion-Exchange Resin: Influence of Gas Humidity on Capture
Mechanisms. Separation Science and Technology 2017, 52 (10), 1761?1767.
https://doi.org/10.1080/01496395.2017.1293094.
[2] Battino, R. Oxygen and Ozone: Solubility Data Series; Elsevier, 2015.
[3] Bell, V. L.; Pezdirtz, G. F. Effects of Ionizing Radiation on Linear Aromatic
Polyesters. Journal of Polymer Science: Polymer Chemistry Edition 1983, 21 (11),
3083?3092. https://doi.org/10.1002/pol.1983.170211106.
[4] Blauwhoff, P. M. M.; Versteeg, G. F.; Van Swaaij, W. P. M. A Study on the Reac-
tion between CO2 and Alkanolamines in Aqueous Solutions. Chemical Engineering
Science 1984, 39 (2), 207?225. https://doi.org/10.1016/0009-2509(84)80021-4.
[5] Cavenati, S.; Grande, C. A.; Rodrigues, A. E. Adsorption Equilibrium of Methane,
Carbon Dioxide, and Nitrogen on Zeolite 13X at High Pressures. Journal of Chemical
& Engineering Data 2004, 49 (4), 1095?1101. https://doi.org/10.1021/je0498917.
[6] Chiang, Y.-C.; Chen, Y.-J.; Wu, C.-Y. Effect of Relative Humidity on Adsorption
Breakthrough of CO2 on Activated Carbon Fibers. Materials (Basel) 2017, 10 (11).
https://doi.org/10.3390/ma10111296.
[7] Cho, Y.; Lee, J.-Y.; Bokare, A.; Kwon, S.-B.; Park, D.-S.; Jung, W.-S.; Choi, J.-
S.; Yang, Y.-M.; Lee, J.-Y.; Choi, W. LiOH-Embedded Zeolite for Carbon Dioxide
Capture under Ambient Conditions. Journal of Industrial and Engineering Chem-
istry 2015, 22, 350?356.
[8] Craver, C. D. The Coblentz Society Desk Book of Infrared Spectra,; NATIONAL
STANDARD REFERENCE DATA SYSTEM, NATIONAL STANDARD REFER-
ENCE DATA SYSTEM, 1977.
50
[9] Dai, Z.; Middleton, R.; Viswanathan, H.; Fessenden-Rahn, J.; Bauman, J.; Pawar,
R.; Lee, S.-Y.; McPherson, B. An Integrated Framework for Optimizing CO 2 Se-
questration and Enhanced Oil Recovery. Environmental Science & Technology Let-
ters 2014, 1 (1), 49?54. https://doi.org/10.1021/ez4001033.
[10] Dai, Z.; Stauffer, P. H.; Carey, J. W.; Middleton, R. S.; Lu, Z.; Jacobs, J. F.;
Hnottavange-Telleen, K.; Spangler, L. H. Pre-Site Characterization Risk Analysis
for Commercial-Scale Carbon Sequestration. Environ. Sci. Technol. 2014, 48 (7),
3908?3915. https://doi.org/10.1021/es405468p.
[11] Danckwerts, P. V. The Reaction of CO2 with Ethanolamines. Chemical Engineering
Science 1979, 34 (4), 443?446. https://doi.org/10.1016/0009-2509(79)85087-3.
[12] Dietz, T. (ORCID:0000000187333227); Fastow, E.; Tsoi, M.; Tsinas, Z.; Pazos, I.;
Al-Sheikhly, M. Enhancement of the Extraction of Uranium from Seawater; DOE-
UMD-0008437-1 15-8663; Univ. of Maryland, College Park, MD (United States),
2018. https://doi.org/10.2172/1489218.
[13] Dietz, T. C.; Tomaszewski, C. E.; Tsinas, Z.; Poster, D.; Barkatt, A.; Adel-Hadadi,
M.; Bateman, F. B.; Cumberland, L. T.; Schneider, E.; Gaskell, K.; LaVerne, J.;
Al-Sheikhly, M. Uranium Removal from Seawater by Means of Polyamide 6 Fibers
Directly Grafted with Diallyl Oxalate through a Single-Step, Solvent-Free Irradiation
Process. Industrial & Engineering Chemistry Research 2016, 55 (15), 4179?4186.
https://doi.org/10.1021/acs.iecr.5b03401.
[14] Diop, M. F.; Torkelson, J. M. Ester Functionalization of Polypropylene via Con-
trolled Decomposition of Benzoyl Peroxide during Solid-State Shear Pulverization.
Macromolecules 2013, 46 (19), 7834?7844. https://doi.org/10.1021/ma401628u.
[15] Dole, M. The Radiation Chemistry of Macromolecules; Elsevier, 2013.
[16] Dutcher, B.; Fan, M.; Russell, A. G. Amine-Based CO 2 Capture Technology Devel-
opment from the Beginning of 2013?A Review. ACS Applied Materials & Interfaces
2015, 7 (4), 2137?2148. https://doi.org/10.1021/am507465f.
51
[17] Fastow, E.; Cook, S.; Dean, P.; Ott, P.; Wilson, J.; Yoon, H.; Dietz, T.; Bateman,
F.; Al-Sheikhly, M. Single-Step Synthesis of Atmospheric CO2 Sorbents through
Radiation-Induced Graft Polymerization on Commercial-Grade Fabrics. Radiation
Research 2019, 192 (2), 219?230. https://doi.org/10.1667/RR15362.1.
[18] Faterpeker, S. A.; Potnis, S. P. Radiation Grafting of Poly(Ethylene Terephtha-
late)(PET) Fibre, I. Grafting with Vinyl Acetate. Angew. Makromol. Chem. 1980,
90 (1), 69?81. https://doi.org/10.1002/apmc.1980.050900107.
[19] Fisher, J. C.; Tanthana, J.; Chuang, S. S. C. Oxide-Supported Tetraethylenepen-
tamine for CO2 Capture. Environ. Prog. Sustainable Energy 2009, 28 (4), 589?598.
https://doi.org/10.1002/ep.10363.
[20] Gholidoust, A.; Atkinson, J. D.; Hashisho, Z. Enhancing CO2 Adsorption via Amine-
Impregnated Activated Carbon from Oil Sands Coke. Energy & Fuels 2017, 31 (2),
1756?1763. https://doi.org/10.1021/acs.energyfuels.6b02800.
[21] Hauchhum, L. Carbon Dioxide Adsorption on Zeolites and Activated Carbon by
Pressure Swing Adsorption in a Fixed Bed. International Journal of Energy and
Environmental Engineering 2014, 5 (4), 349?356. https://doi.org/10.1007/s40095-
014-0131-3.
[22] Ho, T. M.; Tuyen, T.; Howes, T.; Bhandari, B. R. Method of
Measurement of CO2 Adsorbed into ?-Cyclodextrin by Infra-Red CO2
Probe. International Journal of Food Properties 2016, 19 (8), 1696?1707.
https://doi.org/10.1080/10942912.2015.1084005.
[23] Holland, B. J.; Hay, J. N. The Thermal Degradation of PET and Analogous
Polyesters Measured by Thermal Analysis?Fourier Transform Infrared Spectroscopy.
Polymer 2002, 43 (6), 1835?1847. https://doi.org/10.1016/S0032-3861(01)00775-3.
[24] House, K. Z.; Baclig, A. C.; Ranjan, M.; Nierop, E. A. van; Wilcox, J.; Herzog, H. J.
Economic and Energetic Analysis of Capturing CO2 from Ambient Air. Proceedings
52
of the National Academy of Sciences of the United States of America 2011, 108 (51),
20428?20433. https://doi.org/10.1073/pnas.1012253108.
[25] Islamoglu, T.; Rabbani, M. G.; M. El-Kaderi, H. Impact of Post-Synthesis Mod-
ification of Nanoporous Organic Frameworks on Small Gas Uptake and Selective
CO 2 Capture. Journal of Materials Chemistry A 2013, 1 (35), 10259?10266.
https://doi.org/10.1039/C3TA12305G.
[26] Jackson, P.; Robinson, K.; Puxty, G.; Attalla, M. In Situ Fourier Transform-Infrared
(FT-IR) Analysis of Carbon Dioxide Absorption and Desorption in Amine Solutions.
Energy Procedia 2009, 1 (1), 985?994. https://doi.org/10.1016/j.egypro.2009.01.131.
[27] Jasso, M.; Krump, H.; Hudec, I.; St?ahel?, P.; Kova?c?ik, D.; S???ra, M. Coating of
PET Cords at Atmospheric Pressure Plasma Discharge in the Presence of Butadi-
ene/Nitrogen Gas Mixtures. Surface and Coatings Technology 2006, 201 (1), 57?62.
https://doi.org/10.1016/j.surfcoat.2005.10.024.
[28] Krimm, S.; Liang, C. Y.; Sutherland, G. B. B. M. Infrared Spectra of
High Polymers. II. Polyethylene. J. Chem. Phys. 1956, 25 (3), 549?562.
https://doi.org/10.1063/1.1742963.
[29] Lee, S.; Filburn, T. P.; Gray, M.; Park, J.-W.; Song, H.-J. Screening Test of Solid
Amine Sorbents for CO2 Capture. Ind. Eng. Chem. Res. 2008, 47 (19), 7419?7423.
https://doi.org/10.1021/ie8006984.
[30] Li, W., Choi, S., Drese, J. H., Hornbostel, M., Krishnan, G., Eisen-
berger, P. M., & Jones, C. W. (2010, June 23). Steam-Stripping for
Regeneration of Supported Amine-Based CO2 Adsorbents. Retrieved from
https://onlinelibrary.wiley.com/doi/abs/10.1002/cssc.201000131
[31] Li, K.-M.; Jiang, J.-G.; Tian, S.-C.; Chen, X.-J.; Yan, F. Influence of Silica Types on
Synthesis and Performance of Amine?Silica Hybrid Materials Used for CO2 Capture.
J. Phys. Chem. C 2014, 118 (5), 2454?2462. https://doi.org/10.1021/jp408354r.
53
[32] Li, P.; Zhang, S.; Chen, S.; Zhang, Q.; Pan, J.; Ge, B. Preparation and
Adsorption Properties of Polyethylenimine Containing Fibrous Adsorbent for
Carbon Dioxide Capture. J. Appl. Polym. Sci. 2008, 108 (6), 3851?3858.
https://doi.org/10.1002/app.27937.
[33] Li, W.; Choi, S.; Drese, J. H.; Hornbostel, M.; Krishnan, G.; Eisen-
berger, P. M.; Jones, C. W. Steam-Stripping for Regeneration of Sup-
ported Amine-Based CO2 Adsorbents. ChemSusChem 2010, 3 (8), 899?903.
https://doi.org/10.1002/cssc.201000131.
[34] Liang, C. Y.; Krimm, S. Infrared Spectra of High Polymers. III. Polytetrafluo-
roethylene and Polychlorotrifluoroethylene. J. Chem. Phys. 1956, 25 (3), 563?571.
https://doi.org/10.1063/1.1742964.
[35] Lu, W.; Sculley, J. P.; Yuan, D.; Krishna, R.; Zhou, H.-C. Carbon Dioxide Capture
from Air Using Amine-Grafted Porous Polymer Networks. J. Phys. Chem. C 2013,
117 (8), 4057?4061. https://doi.org/10.1021/jp311512q.
[36] Mahajani, V. V.; Joshi, J. B. Kinetics of Reactions between Carbon Diox-
ide and Alkanolamines. Gas Separation & Purification 1988, 2 (2), 50?64.
https://doi.org/10.1016/0950-4214(88)80013-6.
[37] Maschmeyer, T.; van de Water, L. An Overview of Zeolite, Zeotype and Mesoporous
Solids Chemistry: Design, Synthesis and Catalytic Properties. Catalysts for Fine
Chemical Synthesis: Microporous and Mesoporous Solid Catalysts; Vol. 4, pp 1?38.
[38] Matyjaszewski, K.; Davis, T. P. Handbook of Radical Polymerization; John Wiley
& Sons, 2003.
[39] McCann, N.; Maeder, M.; Attalla, M. Simulation of Enthalpy and Capacity of
CO2 Absorption by Aqueous Amine Systems. Ind. Eng. Chem. Res. 2008, 47 (6),
2002?2009. https://doi.org/10.1021/ie070619a.
54
[40] IPCC Special Report on Carbon Dioxide Capture and Storage; Metz, B., Inter-
governmental Panel on Climate Change, Eds.; Cambridge University Press, for the
Intergovernmental Panel on Climate Change: Cambridge, 2005.
[41] Melzer, L. S. Carbon Dioxide Enhanced Oil Recovery (CO2 EOR): Factors Involved
in Adding Carbon Capture, Utilization and Storage (CCUS) to Enhanced Oil Re-
covery. Center for Climate and Energy Solutions 2012.
[42] IPCC Special Report on Carbon Dioxide Capture and Storage; Metz, B., Inter-
governmental Panel on Climate Change, Eds.; Cambridge University Press, for the
Intergovernmental Panel on Climate Change: Cambridge, 2005.
[43] Nahas, K. A.; Cama, J.; Schaich, M.; Hammond, K.; Deshpande, S.; Dekker,
C.; Ryadnov, M. G.; Keyser, U. F. A Microfluidic Platform for the Charac-
terisation of Membrane Active Antimicrobials. Lab Chip 2019, 19 (5), 837?844.
https://doi.org/10.1039/C8LC00932E.
[44] Peters, A. (2019, June 17). We have the tech to suck CO2 from the
air?but can it suck enough to make a difference? Retrieved from
https://www.fastcompany.com/90356326/we-have-the-tech-to-suck-CO2-from-
the-air-but-can-it-suck-enough-to-make-a-difference
[45] Pinkas, J. CHEMISTRY OF SILICATES AND ALUMINOSILICATES. Ceramics -
Silika?ty 2005, 49, 287?298.
[46] Sacak, M.; Celik, M. Hydrogen Peroxide Initiated Grafting of Acry-
lamide onto Poly(Ethylene Terephthalate) Fibers in Benzyl Alcohol. J. Appl.
Polym. Sci. 1996, 59 (7), 1191?1194. https://doi.org/10.1002/(SICI)1097-
4628(19960214)59:7?1191::AID-APP17?3.0.CO;2-8.
[47] Sacak, M.; Oflaz, F. Benzoyl-Peroxide-Initiated Graft Copolymerization of
Poly(Ethylene Terephthalate) Fibers with Acrylic Acid. J. Appl. Polym. Sci. 1993,
50 (11), 1909?1916. https://doi.org/10.1002/app.1993.070501107.
55
[48] Samanta, A.; Zhao, A.; Shimizu, G. K. H.; Sarkar, P.; Gupta, R. Post-Combustion
CO2 Capture Using Solid Sorbents: A Review. Ind. Eng. Chem. Res. 2012, 51 (4),
1438?1463. https://doi.org/10.1021/ie200686q.
[49] Seider, W. Product and Process Design Principles: Synthesis, Analysis and Evalua-
tion, 4th Edition,Wiley
[50] Silva, G. D., Ranjith, P., & Perera, M. Geochemical aspects of CO2 se-
questration in deep saline aquifers: A review. Fuel, 2015 155, 128?143. doi:
10.1016/j.fuel.2015.03.045
[51] Thiruvenkatachari, R.; Su, S.; An, H.; Yu, X. X. Post Combustion CO2 Capture by
Carbon Fibre Monolithic Adsorbents. Progress in Energy and Combustion Science
2009, 35 (5), 438?455. https://doi.org/10.1016/j.pecs.2009.05.003.
[52] Vaidya, P. D.; Kenig, E. Y. CO2-Alkanolamine Reaction Kinetics: A Re-
view of Recent Studies. Chem. Eng. Technol. 2007, 30 (11), 1467?1474.
https://doi.org/10.1002/ceat.200700268.
[53] Vaidya, P. D.; Kenig, E. Y. A Study on CO2 Absorption Kinetics by Aqueous Solu-
tions of N,N-Diethylethanolamine and N-Ethylethanolamine. Chem. Eng. Technol.
2009, 32 (4), 556?563. https://doi.org/10.1002/ceat.200800573.
[54] Wang, J.; Wang, M.; Li, W.; Qiao, W.; Long, D.; Ling, L. Ap-
plication of Polyethylenimine-Impregnated Solid Adsorbents for Direct
Capture of Low-Concentration CO2. AIChE J. 2015, 61 (3), 972?980.
https://doi.org/10.1002/aic.14679.
[55] Wayner, D. D. M.; Clark, K. B.; Rauk, A.; Yu, D.; Armstrong, D. A. C-H Bond Dis-
sociation Energies of Alkyl Amines: Radical Structures and Stabilization Energies.
J. Am. Chem. Soc. 1997, 119 (38), 8925?8932. https://doi.org/10.1021/ja971365v.
[56] Wu, Q.; Chen, S.; Liu, H. Effect of Surface Chemistry of Polyethyleneimine-
Grafted Polypropylene Fiber on Its CO 2 Adsorption. RSC Advances 2014, 4 (52),
27176?27183. https://doi.org/10.1039/C4RA01232A.
56
[57] Wypych, G. Handbook of Polymers; Elsevier, 2016.
[58] Yang, Y.; Li, H.; Chen, S.; Zhao, Y.; Li, Q. Preparation and Characterization of a
Solid Amine Adsorbent for Capturing CO2 by Grafting Allylamine onto PAN Fiber.
Langmuir 2010, 26 (17), 13897?13902. https://doi.org/10.1021/la101281v.
[59] Yoshida, K.; Toyoura, K.; Matsunaga, K.; Nakahira, A.; Kurata, H.; Ikuhara, Y.;
Sasaki, Y. Atomic Sites and Stability of Cs+ Captured within Zeolitic Nanocavities.
Scientific Reports 2013, 3. https://doi.org/10.1038/srep02457.
[60] Yu, C.-H. A Review of CO2 Capture by Absorption and Adsorption. Aerosol and
Air Quality Research 2012. https://doi.org/10.4209/aaqr.2012.05.0132.
[61] Yu, Q.; Brilman, D. W. F. Design Strategy for CO2 Adsorption from
Ambient Air Using a Supported Amine Based Sorbent in a Fixed
Bed Reactor. Energy Procedia 2017, 114 (Supplement C), 6102?6114.
https://doi.org/10.1016/j.egypro.2017.03.1747.
[62] Zelenak, V.; Halamova, D.; Gaberova, L.; Bloch, E.; Llewellyn, P. Amine-Modified
SBA-12 Mesoporous Silica for Carbon Dioxide Capture: Effect of Amine Basicity on
Sorption Properties. Microporous and Mesoporous Materials 2008, 116 (1), 358?364.
https://doi.org/10.1016/j.micromeso.2008.04.023.
[63] Zeman, F. Energy and Material Balance of CO2 Capture from Ambient Air. Environ.
Sci. Technol. 2007, 41 (21), 7558?7563. https://doi.org/10.1021/es070874m.
[64] Zukal, A.; Kub?u, M.; Pastva, J. Two-Dimensional Zeolites: Adsorption of Carbon
Dioxide on Pristine Materials and on Materials Modified by Magnesium Oxide. Jour-
nal of CO2 Utilization 2017, 21, 9?16. https://doi.org/10.1016/j.jcou.2017.06.013.
[65] Climate Change Overview. International Debates 2009, 7 (9), 4?7.
[66] The Social Cost of Carbon. https://19january2017snapshot.epa.gov/climatechange/social-
cost-carbon-.html (accessed Apr 26, 2020).
57
[67] 2020 Electricity Rates By State (Updated Feb 2020). (2020, February 4). Retrieved
from https://paylesspower.com/blog/electric-rates-by-state/
[68] Carbon and Usage Calculation Methods. (n.d.). Retrieved from
https://carbonfund.org/calculation-methods/
[69] Linear Accelerator Pricing Guide. (n.d.). Retrieved from
https://www.oncologysystems.com/resources/linear-accelerator-guides/used-linac-
price
[70] Chemical Profile - NYLON-6 and NYLON-6/6.
https://www.icis.com/explore/resources/news/2005/12/02/549810/chemical-
profile-nylon-6-and-nylon-6-6/ (accessed Mar 30, 2020).
[71] TEMPO 214000. https://www.sigmaaldrich.com/catalog/product/aldrich/214000?lang=en?ion
(accessed Mar 30, 2020).
[72] Luperox?A98, Benzoyl peroxide 179981. https://www.sigmaaldrich.com/catalog/product/sial/179981?lang=en?ion
(accessed Mar 30, 2020).
[73] https://www.alibaba.com/product-detail/POLY-ALLYLAMINE-
HYDROCHLORIDE-71550-12- 4-62008872588.html?spm=a2700.7724857.normalList.18.462ed964Nq1WUW
(accessed Mar 30, 2020).
[74] https://www.alibaba.com/product-detail/Diallylamine-with-best-price-CAS-124-
60705764425.ht ml?spm=a2700.7724857.normalList.25.43e92acbhLupU0 (accessed
Mar 30, 2020).
58