Chao Xu - DiVA portalsu.diva-portal.org/smash/get/diva2:857338/FULLTEXT01.pdfstarch for daily life...
Transcript of Chao Xu - DiVA portalsu.diva-portal.org/smash/get/diva2:857338/FULLTEXT01.pdfstarch for daily life...
I m i n e / a z o - l i n k e d m i c r o p o r o u s o r g a n i c p o l y m e r s
- D e s i g n , s y n t h e s i s a n d a p p l i c a t i o n s
Chao Xu
Imine/azo-linked microporous organic polymers -Design, synthesis and applications
Chao Xu
许超
Doctoral thesis 2015
Department of Materials and Environmental Chemistry
Arrhenius Laboratory, Stockholm University
SE-10691 Stockholm, Sweden
Faculty opponent
Prof. Dave Adams
Department of Chemistry
University of Liverpool, United Kingdom
Evaluation committee
Prof. Knut Irgum
Department of Chemistry
Umeå University, Sweden
Docent Michael Malkoch
Department of Fiber and Polymer Technology
KTH Royal Institute of Technology, Sweden
Docent Aji Mathew
Department of Materials and Environmental Chemistry
Stockholm University, Sweden
Substitute
Prof. Mats Jonsson
Department of Applied Physical Chemistry
KTH Royal Institute of Technology, Sweden
©Chao Xu, Stockholm University 2015
ISBN 978-91-7649-274-1
Printed in Sweden by Holmbergs, Malmö 2015
Distributor: Department of Materials and Environmental Chemistry
To my family
Abstract
Microporous organic polymers (MOPs) are porous materials. Owing to their
high surface area, tunable pore sizes and high physicochemical stability, they
are studied for applications including gas capture and separation and hetero-
geneous catalysis. In this thesis, a series of imine/azo-linked MOPs were
synthesized. The MOPs were examined as potential CO2 sorbents and as
supports for heterogeneous catalysis.
The MOPs were synthesized by Schiff base polycondensations and oxida-
tive couplings. The porosities of the imine-linked MOPs were tunable and
affected by a range of factors, such as the synthesis conditions, monomer
lengths, monomer ratios. All the MOPs had ultramicropores and displayed
relatively high CO2 uptakes and CO2-over-N2 selectivities at the CO2 con-
centrations relevant for post-combustion capture of CO2. Moreover, the
ketimine-linked MOPs were moderately hydrophobic, which might increase
their efficiency for CO2 capture and separation.
The diverse synthesis routes and rich functionalities of MOPs allowed fur-
ther post-modification to improve their performance in CO2 capture. A mi-
cro-/mesoporous polymer PP1-2, rich in aldehyde end groups, was post-
synthetically modified by the alkyl amine tris(2-aminoethyl)amine (tren).
The tethered amine moieties induced chemisorption of CO2 on the polymer,
which was confirmed by the study of in situ infrared (IR) and solid-state 13
C
nuclear magnetic resonance (NMR) spectroscopy. As a result, the modified
polymer PP1-2-tren had a large CO2 capacity and very high CO2-over-N2
selectivity at low partial pressures of CO2.
Pd(II) species were incorporated in the selected MOPs by means of com-
plexation or chemical bonding with the imine or azo groups. The Pd(II)-rich
MOPs were tested as heterogeneous catalysts for various organic reactions.
The porous Pd(II)-polyimine (Pd2+
/PP-1) was an excellent co-catalyst in
combination with chiral amine for cooperatively catalyzed and enantioselec-
tive cascade reactions. In addition, the cyclopalladated azo-linked MOP
(Pd(II)/PP-2) catalyzed Suzuki and Heck coupling reactions highly efficient-
ly.
Key words: Microporous organic polymers, CO2 capture and separation,
post-modification, chemisorption, heterogeneous catalysis
List of papers
Paper I:
Microporous adsorbents for CO2 capture – a case for microporous pol-
ymers?
C. Xu and N. Hedin, Materials Today, 2014, 17, 397-403.
Scientific contributions: Lead role in writing
Paper II:
Synthesis of microporous organic polymers with high CO2-over-N2 se-
lectivity and CO2 adsorption
C. Xu and N. Hedin, J. Mater. Chem. A, 2013, 1, 3406-3414.
Scientific contributions: Performed design, synthesis and characterization of
the polymers, lead role in writing
Paper III:
Hydrophobic porous polyketimines for the capture of CO2
C. Xu, E. Dinka, N. Hedin, ChemPlusChem, 2015, DOI:
10.1002/cplu.201500344.
Scientific contributions: Designed the project, performed synthesis and
characterization of the polymers, lead role in writing
Paper IV:
Ultramicroporous CO2 adsorbents with tunable mesopores based on
polyimines synthesized under off-stoichiometric conditions
C. Xu and N. Hedin, Micropor. Mesopor. Mater., accepted.
Scientific contributions: Performed design, synthesis and characterization of
the polymers, lead role in writing
Paper V:
Adsorption of CO2 on a micro-/mesoporous polyimine modified with
tris(2-aminoethyl)amine
C. Xu, Z. Bacsik and N. Hedin, J. Mater. Chem. A, 2015, 3, 16229-16234.
Scientific contributions: Designed the project, performed synthesis and
characterization of the polymers, lead role in writing
Paper VI:
The use of porous palladium (II)-polyimine in cooperatively catalyzed
highly enantioselective cascade transformations
C. Xu, L. Deiana, S. Afewerki, C. Incerti-Pradillos, O. Córdova, P. Guo, A.
Córdova and N. Hedin, Adv. Synth. Catal., 2015, 357, 2150-2156.
Scientific contributions: Performed design, synthesis and characterization of
the polymer and the catalyst, lead role in writing
Paper VII:
A cyclopalladated microporous azo-linked polymer as a heterogeneous
catalyst for eco-friendly Suzuki and Heck coupling reactions
C. Xu, S. Afewerki, A. Córdova and N. Hedin, in manuscript.
Scientific contributions: Designed the project, performed synthesis and
characterization of the polymer and the catalyst, lead role in writing
# Paper III and VI are reprinted with permission by publisher. Copyright
2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim.
Papers not included in the thesis
Paper VIII:
A semiconducting microporous framework of Cd6Ag4(SPh)16 clusters
interlinked using rigid and conjugated bipyridines
C. Xu, N. Hedin, H.-T. Shi and Q.-F. Zhang, Chem. Commun., 2014, 50,
3710-3712.
Paper IX:
Stepwise assembly of a semiconducting coordination polymer
[Cd8S(SPh)14(DMF)(bpy)]n and its photodegradation of organic dyes
C. Xu, N. Hedin, H.-T. Shi, Z.-F. Xin and Q.-F. Zhang, Dalton Trans., 2015,
44, 6400-6405.
Paper X:
UV-vis and plasmonic nano-spectroscopy of the CO2 adsorption ener-
getics in microporous polymers
F. A. A. Nugroho, C. Xu,
N. Hedin and C. Langhammer, Anal. Chem., 2015,
DOI: 10.1021/acs.analchem.5b03108.
Abbreviations
benzimidazole-linked-polymer (BILP)
block co-polymer (BCP)
BarrettJoynerHalenda (BJH)
BrunauerEmmettTeller (BET)
conjugated microporous polymer (CMP)
cross-polarization magic angle spinning (CPMAS)
covalent triazine framework (CTF)
covalent organic framework (COF)
density functional theory (DFT)
dynamic kinetic asymmetric transformation (DYKAT)
dynamic covalent chemistry (DCC)
diastereomeric ratio (dr)
enantiometric ratio (er)
hypercrosslinked polymer (HCP)
inductively coupled plasma optical emission spectrometry (ICP-OES)
infrared (IR)
isosteric heat of adsorption (Qst)
microporous organic polymer (MOP)
metal-organic framework (MOF)
mesocellular foam (MCF)
near-infrared (NIR)
nuclear magnetic resonance (NMR)
one-dimensional (1D)
pore size distribution (PSD)
porous polymer (PP)
porous polymer network (PPN)
porous aromatic framework (PAF)
poly(1-trimethylsilyl-1-propyne) (poly(TMSP)
polymer of intrinsic microporosity (PIM)
scanning electron microscopy (SEM)
temperature-swing adsorption (TSA)
two-dimensional (2D)
three-dimensional (3D)
thermal gravimetric analysis (TGA)
transmission electron microscopy (TEM)
trifluoromethanesulfonic acid (TFMS)
tris(2-aminoethyl)amine (tren)
vacuum-swing adsorption (VSA)
X-ray photoelectron spectroscopy (XPS)
Contents
1. Introduction ................................................................................................. 1
1.1 Porous polymers: a class of porous materials....................................... 1
1.2 Design and synthesis of microporous organic polymers (MOPs) ........ 2
1.3 Application of microporous organic polymers ................................... 13
1.4 Aims of this study .............................................................................. 18
2. Synthesis of imine-linked microporous organic polymers for CO2 capture
and separation ............................................................................................... 19
2.1 Synthesis of imine-linked MOPs ........................................................ 20
2.2 Chemical structural characterizations ................................................ 24
2.3 Porosity analyses ................................................................................ 27
2.4 MOPs for CO2 capture and separation ............................................... 37
3. Adsorption of CO2 on a porous polymer modified with tris(2-
aminoethyl)amine ......................................................................................... 41
3.1 Amine modification of porous polymers ............................................ 41
3.2 Molecular studies by spectroscopy ..................................................... 42
3.3 Porosity analyses ................................................................................ 46
3.4 Studies of CO2 adsorption and separation .......................................... 48
4. Incorporation of Pd(II) species in porous polymers for heterogeneous
catalysis ......................................................................................................... 51
4.1 The use of porous polyimine/Pd(II) in cooperatively catalyzed highly
enantioselective cascade transformations ................................................. 52
4.2 A cyclopalladated microporous azo-linked polymer as a
heterogeneous catalyst for more eco-friendly Suzuki and Heck coupling
reactions ................................................................................................... 58
5. Conclusions ............................................................................................... 71
Populärvetenskaplig sammanfattning ........................................................... 73
Acknowledgements ....................................................................................... 75
References ..................................................................................................... 77
1
1. Introduction
1.1 Porous polymers: a class of porous materials
Polymers are macromolecules or large molecules composed of many repeat-
ed subunits.1 Such polymers may exist in Nature or be synthesized in labora-
tory. Humans have used natural polymers such as silk, wool, cellulose and
starch for daily life since ancient times, and more importantly, polymers
such as DNA and protein constitute most of the structures of life.
Polymer science was first established in the 19th century, and early studies
revolved around modifying natural polymers. Synthetic polymers were
largely developed during World War II, and afterwards, due to the increasing
demand from the industry and military. Nylon and synthetic rubber are typi-
cal examples. Meanwhile, the understanding of polymers at their molecular
level was continuously progressing. First, polymers were understood to be a
collection of small molecules and cyclic structures bound together by un-
known forces, which was represented by the association theory proposed by
Thomas Graham in 1861.2 Hermann Staudinger, the first Noble laureate in
polymer science, developed the first modern macromolecular theory in
1920s that describes polymers as long chains consisting of short repeating
units linked by covalent bonds.3
All polymers have a certain free volume made up of the space between the
polymer chains caused by inefficient packing.4 Polymers with permanent
porosity belong to a class of porous materials exhibiting applications in the
fields of gas storage, separation, catalysis, delivery, etc.5 Such porous poly-
mers merge the properties of both porous materials and polymers. Firstly,
they can be designed to have high surface area and tunable pore size due to
the light elements of the framework and synthetic diversity of polymers.6,7
Secondly, porous polymers are linked by covalent bonds, which give them a
high physicochemical stability.8 Thirdly, they have an advantage of easier
processing than other types of porous materials to produce monoliths or
films, which can enable the porosity to be used in practical applications.9,10
2
1.2 Design and synthesis of microporous organic
polymers (MOPs)
1.2.1 Creation of porosity in polymers
Various methods have been developed to prepare porous polymers, where
creating porosity is a key and challenging step. Direct templating is a simple
and straightforward method.11
Here, the monomers are adsorbed onto a sur-
face or allowed to infiltrate the voids in a template such as mesoporous silica
or anodic aluminum oxide. After in situ polymerization of the monomers,
pores are generated by removing the template. This approach has the ad-
vantage of a precise control of the pore size and structure as the template is
fixed. Self-assembly based on block co-polymers (BCPs) is another way to
prepare porous polymers especially with ordered pore structures.12
BCPs
comprise two or more chemically linked homopolymers. They can serve as
both a template for the pores and a source of the framework. These two tem-
plating methods preferably generate mesopores and macropores; however,
the resulting polymers typically lack micropores.5 Alternatively, direct
polymerization of selected monomers can be used to prepare polymers with
mainly micropores and high specific surface areas. The research on this ap-
proach has expanded during the last decade because of the availability of
diverse synthesis routes and the formation of materials with interesting prop-
erties and applications.13
Examples of these include hypercrosslinked poly-
mers (HCPs),14
conjugated microporous polymers (CMPs),15
polymers of
intrinsic microporosity (PIMs),16
covalent triazine frameworks (CTFs),17
and
covalent organic frameworks (COFs)7. In this thesis, I use the broad term
„microporous organic polymers‟ (MOPs) to refer to all these polymers con-
taining micropores.
1.2.2 Design principles of MOPs
1.2.2.1 Monomer
In order to prepare infinite polymeric structures, the monomers should con-
tain at least two reactive groups. Cross-coupling of building blocks of the
type An + Bn (n is the number of functional groups, n = 2 6) is usually
used, but the homocoupling of building blocks An (n = 2 4) is an alterna-
tively way to synthesize MOPs.18
As the micropores in MOPs experience
high capillary pressures and high surface energies, there is a risk of pore
collapse by the bending or twisting of polymer chains or the stretching of
flexible polymer networks.5,18
Therefore, rigid and conjugated chains or net-
works are necessary for MOPs to demonstrate permanent microporosity.
Monomers with aromatic or aromatic-like structure are essential to build
3
such stable structures. Moreover, the formation of double or triple bonds, or
ring linkages, could further increase the rigidity of the network. Table 1-1
summarizes backbones of selected monomers that have been used to synthe-
size MOPs.
Most of the monomers used for preparing MOPs are relatively expensive.
Hence, MOPs are still mainly synthesized and studied on a laboratory scale.
Developing low-cost MOPs at scale might open for real applications. For
example, HCPs are one example of relatively low cost MOPs that have been
commercialized.13
Micro-/mesoporous poly(melamineformaldehyde) was
prepared by condensation of melamine with formaldehyde.19
Both mono-
mers are inexpensive raw materials available at scale. From a long-term per-
spective, sustainable monomers are preferred for the synthesis of MOPs.
1.2.2.2 Dimension
Most MOPs have two- or three-dimensional (2D or 3D) polymeric struc-
tures (Figure 1-1). Their porosity might be generated by the interpenetration
of 2D polymeric layers with grid networks or come from the cavities formed
within 3D frameworks. Besides the high-dimensional networks, a design of
MOPs with one-dimensional (1D) chains is also possible. For example,
poly(1-trimethylsilyl-1-propyne) (poly(TMSP)) and PIM-1, with polymeric
chain structures, exhibit permanent microporosity and high surface areas.10,20
PIMs are polymers of intrinsic microporosity prepared by dioxane-forming
reactions between monomers that contain ortho-dihydroxy and ortho-
dihalide. Their micropores are generated by a space-inefficient packing of
the rigid and contorted networks.16
Most MOPs have poor solubility even in
polar solvents; however, poly(TMSP) and some low-dimensional PIMs are
soluble in common organic solvents and can be solvent-casted into films.
Such MOP-based films perform well for gas separation, working via mo-
lecular sieving.21
Figure 1-1. Examples of microporous organic polymers displaying different
dimensions: (a) 1D PIM-110
; (b) 2D CMP-515
; (c) 3D PPN-622
. Reproduced
with permissions.
4
Table 1-1. Selected monomer backbones of MOPs, showing different con-
figurations, geometries and sizes.
5
6
1.2.2.3 Porosity
Porosity varies significantly among MOPs. The surface areas range from
hundreds to over 6000 m2/g and the pore size distributions (PSDs) typically
fall in the micropore range (0.5 to 2 nm).23
In some cases, mesopores (250
nm) are also generated in MOPs.24,25
Such micro-/mesoporous polymers with
bimodal porosity are expected to synergetically merge the advantages of
both micro- and mesopores to improve their properties and expand multi-
functional applications.
Porous materials with tunable porosity are desired for many applications.
For instance, small micropores in sorbents are preferred for gas separation,
while large micropores or mesopores increase the mass transport. It is possi-
ble to control surface area and porosity in MOPs, although it is more chal-
lenging to tune these amorphous materials than crystalline materials such as
metal-organic frameworks (MOFs) and COFs.26
The porosity basically de-
pends on the monomers used, the synthesis conditions applied, and the di-
mensionality of the formed MOPs. The size of the monomers has a great
effect on the porosity. For example, Cooper and associates synthesized
CMP-0, CMP-1, CMP-2, CMP-3, CMP-4 and CMP-5 by Sonogashira-
Hagihara coupling using monomers of different lengths, and the porosity
displayed a clear monomer length dependency (Figure 1-2). The micropore
size distribution shifted systematically towards larger pore diameters with
increasing monomer length, while the BET (BrunauerEmmettTeller) sur-
face area (1018515 m2/g) and micropore volume (0.380.16 cm
3/g) de-
creased. The decreasing BET surface area and micropore volume was proba-
bly a consequence of greater conformational freedom and chain interpenetra-
tion with longer struts and chains.6 Their results indicate that short mono-
mers favor microporous structures and that long monomers appear to
generate large pores in the polymers. In addition, the geometry of the mon-
omer also affects the porosity of MOPs. In general, MOPs with 3D struc-
tures usually have high surface area. Such 3D MOPs can be constructed
using monomers with tetrahedral geometry. Monomers with a planar or line-
ar structure tend to build MOPs with low dimensions and low surface areas.
For example, the Yamamoto homocoupling of a tetrahedral shaped mono-
mer, tetrakis(4-bromophenyl)methane resulted in a porous aromatic frame-
work (PAF-1) with an ultra-high BET surface area of 5600 m2/g;
8 however,
the surface area largely decreased to 1665 m2/g when using a similar mono-
mer of tetrakis(4-bromophenyl)ethene but with a planar structure.27
7
Figure 1-2. A series of CMPs constructed by monomers with different
lengths.6 Reproduced with permission.
Furthermore, the synthesis conditions are important and affect the porosi-
ty. For example, metal-catalyzed coupling reactions performed at a low tem-
perature (usually < 100 °C) appear to form MOPs with high surface areas.
The catalyst renders a high reaction rate and a complete polymerization, and
thus produces high-molecular-weight polymers. The formation of an inter-
penetrated or overlapped structure appears to be hindered at such low reac-
tion temperatures. Hence, the pores formed in MOPs under these conditions
are generally less occupied and typically have a large micropore size (> 10
Å), with a correspondingly high internal surface area. For example, Zhou‟s
group used a Yamamoto homocoupling of tetrakis(4-bromophenyl)silane at
room temperature to derive the porous polymer network PPN-4, which has
the highest BET surface area reported among porous materials (6460 m2/g).
PPN-4 also has large pores, with distributions centered at ~15 and ~22 Å.22
In comparison, polycondensation reactions at high reaction temperature
(150200 °C) appear to form MOPs with highly crosslinked or interpene-
trated structures and lower surface area (~1000 m2/g).
28 As a result, (ul-
tra)micropores occur in such polymers.
8
1.2.2.4 Crystallinity
When considering the morphology of polymers that describe the arrange-
ment and microscale ordering of the chains or the networks in space, MOPs
can be divided into amorphous and crystalline cases. Most MOPs are amor-
phous in morphology as they are synthesized by irreversible reactions under
kinetically controlled conditions.23
For example, the Sonagashira–Hagihara
cross-couplings of selected monomers produce amorphous and kinetically
controlled products with high surface areas.6 Compared to the abundant
amorphous MOPs, it has proven very difficult to crystallize the extended
frameworks of porous polymers containing only light elements (C, H, O, N,
etc.). However, in 2005, Yaghi and co-workers successfully crystallized
COFs by applying the concept of reticular chemistry and dynamic covalent
chemistry (DCC).29
DCC is thermodynamically controlled and consists of
the reversible formations of covalent bonds.30,31
The „self-correcting‟ ability
of the reversible covalent bonds has the inherent characteristic of „proofread-
ing‟ and „error-checking‟, reducing the occurrence of structural defects and
assisting the formation of ordered and crystalline structures with high ther-
modynamic stability.32
For instance, the reversible reactions that form borox-
ines, boronate esters and imines have been used to synthesize COFs.7,33,34
The crystal structures of highly crystalline COF can be determined using X-
ray or electron diffraction,35
which could guide the design of new MOPs
with desired properties and applications. In addition, COFs have the ad-
vantages of low densities, high surface areas, tunable pore sizes and tailor-
made structures and properties. However, the formation of the crystalline
products is not generic during the syntheses. The crystallization process is
very sensitive to many factors including solvent, temperature, pressure, and
pH. Hence, a further improvement of the sensitive and delicate synthesis
conditions for COFs is needed. Still, in this thesis, we have mainly studied
various amorphous MOPs and their tentative applications.
1.2.3 Synthesis of MOPs
Due to the rich chemistry of organic polymerization, synthetic methodolo-
gies to prepare highly diverse MOPs have been developed.18,23,36
MOPs have
been synthesized with a range of simple and traditional reactions, either with
or without catalysts. Table 1-2 displays selected reactions that have been
used for MOP synthesis. CC coupling reactions such as Yamamoto cou-
pling,8 Heck reaction,
37 SonogashiraHagihara coupling,
6 Glaser coupling,
38
and Suzuki coupling39
usually catalyzed by noble metal complexes, have
been used to prepare MOPs, typically with high surface areas. The catalysts
could afford MOPs with high degrees of polymerization and high porosity.
For example, the Yamamoto homocoupling reaction of tetrakis(4-
bromophenyl)methane supported by a co-catalyst of Ni(cod)2/2,2-bipyridine
9
resulted in PAF-1, with a high surface area.8 Elemental analysis showed that
there are almost no end-groups in the PAF-1, indicating a complete conver-
sion of the monomer and a correspondingly high degree of polymerization in
the polymer. It is worthy to note that MOPs prepared by coupling reactions
usually have rigid and conjugated structures, and thus belong to the category
of CMPs, initially developed by Cooper‟s group.15
Cyclization reactions can be also used for the preparation of MOPs. For
example, the cyclotrimerization of ethynylbenzene catalyzed by Co2(CO)8
forms new aromatic rings. A series of MOPs with pure aromatic frameworks
and high BET surface areas (> 1000 m2/g) have been prepared by such reac-
tions.40
Alternatively, the cyclotrimerization of nitriles can form rigid, aro-
matic C3N3 heterocycles. Both inorganic and organic catalysts have been
used to efficiently transform nitriles to MOPs.17,41
Antonietti‟s and Thomas‟
groups first conducted the ionothermal cyclotrimerization of nitriles in the
presence of ZnCl2 and produced a number of CTFs.17
X-ray diffraction stud-
ies showed that such CTFs are partly crystalline. A high reaction tempera-
ture and a long reaction time were shown to increase the surface area and
mesoporosity in CTFs.42
However, these polymers lost part of the organic
components and displayed a degree of carbonization under the ionothermal
high temperature conditions applied (up to 700 °C for a long reaction time).
In addition, the ZnCl2 catalyst was also difficult to remove and remained in
the polymers to a certain degree. A suitable organic catalyst can overcome
the drawbacks of ZnCl2. Trifluoromethanesulfonic acid (TFMS), a strong
Brønsted acid catalyst, is a catalyst for the cyclotrimerization of nitriles to
triazines. Recently, Cooper‟s and Dai‟s groups successfully prepared highly
porous CTFs using the TFMS catalyst under otherwise relatively mild condi-
tions.41,43
This new synthetic approach, conducted at low temperature, al-
lowed the formation of a transparent, flexible, and thin triazine-framework-
based porous membrane, TFM-1. The membrane, with functionalized tria-
zine units, exhibits high separation selectivity of CO2 over N2.43
Other MOPs
have been synthesized by cyclization reactions. For example, simple “click”
chemistry between azides and alkynes was used to prepare triazole-based
MOPs.44
With dioxane-forming reactions between monomers that contain
ortho-dihydroxy and ortho-dihalide, soluble PIMs with high porosity were
synthesized.10
The good solubility of these polymers enabled easy pro-
cessing, permitting the preparation of PIM membranes with good permeabil-
ity for gas separations.16
10
Table 1-2. Synthesis routes to microporous organic polymers under (a) cata-
lytic (b) non-catalytic conditions. [Reproduced from Paper I]45
11
The already mentioned HCPs represent an important class of MOPs.46
They are typically prepared by FriedelCrafts alkylation using a Lewis acid
catalyst such as FeCl3. HCPs are usually produced in two different ways: 1)
by inter- and intramolecular crosslinking of preformed polymer chains, or 2)
by a direct step growth polycondensation of suitable monomers.13
For exam-
ple, Davankov and co-workers crosslinked linear polystyrene,
poly(vinylbenzyl chloride), or their crosslinked copolymers by suitable
crosslinkers such as monochlorodimethyl ether and divinylbenzene.47
The
linear polymers and the linkers were first dissolved or swelled in a solvent to
introduce a space between the polymer chains. Then the crosslinking agents
linked the precursors in an expanded form. Porosity was subsequently gener-
ated in the crosslinked polymers by removing the solvent to produce highly
porous HCPs. The crosslinking structure is very important to stabilize the
porosity in HCPs by preventing the pore structures from collapsing to form
12
dense, nonporous polymers.48
In comparison, direct alkylation of small mo-
lecular monomers is an alternative and more straightforward approach to
produce HCPs. To select suitable monomers is essential for the synthesis.
Aromatic bis(chloromethyl) monomers such as dichloroxylene, bis(chloro-
methyl)biphenyl, and bis(chloromethyl)anthracene have been used to pro-
duce such HCPs.49
The catalysts used for above reactions are usually expensive. For example,
noble metal catalysts of Pd(0)/Pd(II) are used for the most common coupling reactions. In addition, some catalysts are less environmentally friendly. For example, Lewis acid catalysts of ZnCl2 and AlCl3 are corrosive and toxic. An alternative approach to prepare MOPs is by catalyst-free polycondensa-tion reactions, which is less expensive, more convenient and environmental-ly friendly. In a condensation reaction, two molecules combine to form a larger molecule together with an elimination of a small molecule, e.g. H2O, HCl, methanol. When both these monomers have two or more reactive func-tional groups, polycondensation reactions occur and polymers form.
Most polycondensation reactions are reversible and thermodynamic fa-vored reactions, as was mentioned in the section 1.2.2.4, and such reactions can be also used for the synthesis of crystalline COFs materials. However, the synthetic methods used for crystalline COFs and amorphous MOPs are quite different. Most of the COFs are obtained by solvothermal synthesis at temperatures of 80120 °C in sealed vessels.
32 Such closed reaction systems
promote reversible reactions, which appear to be needed to form the crystal-line products. In addition, a suitable concentration of monomers and a re-duced pressure are also important for the crystallization. In comparison, sim-ilar polycondensations used for the synthesis of amorphous MOPs usually take place in highly polar solvents at higher temperatures of 150200 °C.
28,50
Many small molecules evaporate at such high temperatures. As the polycon-densations are reversible and the equilibria are tunable by means of external stimuli, removing the condensation products moves the equilibria toward polymerization and produces polymers with higher molecular weight. In addition, the high reaction temperatures assist the formation of cross-linked/interpenetrated structures, which appears to generate micropores or ultramicropores in the polymer.
Various monomers and different types of condensation reactions have been
applied to produce MOPs, as was shown in Table 1-2b. For example, aro-
matic amine monomers with rigid structures, high reactivity and diverse
functionalities have been polymerized into various robust MOPs. Amines
contain a basic nitrogen atom with a lone pair that acts as the active nucleo-
phile. They are divided into four classes: primary amines, secondary amines,
tertiary amines and cyclic amines, among which primary amine is the most
active nucleophile in many reactions. For example, the nucleophilic attack of
a primary amine on aldehydes (ketones), acyl chlorides and anhydrides
forms imines, amides and imides, respectively. Polycondensation between
aromatic amines and aldehydes of the type [An + Bn] (n is the number of
functional group, usually 2 ≤ n ≤ 4) has resulted in a number of imine-, im-
13
idazole-, aminal-linked MOPs.28,51-53
Similar reactions between amine and
dianhydride monomers led to the formation of MOPs connected by imide
bonds.54
In addition, amines also react with halides such as acyl chloride and
cyanuric chloride to form polymers with flexible structures.55
1.3 Application of microporous organic polymers
1.3.1 Gas adsorption
As a result of their high surface areas and tunable pore sizes, MOPs have
been widely explored for the capture and storage of gases such as CO2, CH4
and H2.56
In addition, the availability of various synthesis routes and the rich
functionality of MOPs facilitates post-modification to fabricate on MOPs
functional groups that can further increase the capacity and selectivity for a
specific gas.26,56
As MOPs studied in this thesis have moderate surface areas
(< 1000 m2/g) that are not suitable for gas storage, we mainly focus on their
ability for CO2 capture and separation.
Currently, the combustion of fossil fuels produces a large amount of CO2,
which is generally thought to be a main cause of global warming. Post-
combustion capture of CO2 is a feasible way to control the CO2 emissions in
the short term.57
Amine scrubbing using strongly basic alkyl amines is chem-
ically effective in removing CO2 from the exhaust of coal- and gas-fired
power plants. However, this technology suffers from significant energy pen-
alties and associated corrosion of the devices,58
affecting the cost for the
capture. Hence, alternatives to amine scrubbers have been examined, and
adsorption-based alternatives are actively researched. Physisorption allows
an in-principle low cost for the regeneration of the sorbents by means of
vacuum or temperature swing adsorption (VSA or TSA) techniques.59
MOPs
with high surface area, tunable pore size and modifiable structures are prom-
ising sorbents for CO2 capture. Many MOPs have been studied as sorbents
for post combustion capture of CO2, as shown in Table 1-3.
14
Table 1-3. Examples of microporous organic polymers with CO2 capacity at
0.15 and 1 bar.
MOPs SBET
(m2/g)
CO2 uptake (mmol/g) T (K) Ref.
0.15 bar 1 bar
PPF-1 1740 1.80 6.07 273 [51]
BILP-4 1140 1.99 5.34 273 [53]
POF-1B 917 1.51 4.19 273 [60]
Fe-POP-1 875 1.48 4.30 273 [61]
PPN-6-CH2DETA 555 3.04 4.30 295 [62]
ALP-1 1240 1.73 5.37 273 [63]
As the flue gas exhausted from coal- and gas-fired power plants contains a
low CO2 concentration of 515 vol%, a sorbent for the post-combustion
capture of CO2 requires high capacity at working pressure (0.050.15 bar).64
In order to have high CO2 capacity at the working pressure, physisorbents
require small pores that can favor interactions between the adsorbed CO2 and
the pore walls.26
For example, benzimidazole-linked polymers (BILPs), with
ultramicropores having diameters of ~6 Å, show exceptionally high CO2
uptakes both at 1 bar and at the more relevant 0.15 bar, adsorbing up to 5.34
and 1.99 mmol CO2/g (273 K), respectively.53
In addition, surface area is
important for CO2 capture. In general, sorbents with high surface areas also
have high CO2 capacities. However, there is no linear relationship between
the surface area and CO2 capacity, especially at or below atmospheric pres-
sure.65,66
For example, PAF-1, with an ultra-high surface area of 5600 m2/g,
has a moderate CO2 capacity (1.09 mmol/g at 1 bar and 298 K, 2.05
mmol/g at 1 bar and 273 K).8,67
High surface areas are usually accompa-
nied by large pores in porous materials; hence, the poor CO2 capacity of
PAF-1 can be attributed to the weak interaction between the surface moieties
of the large pores and the CO2 molecules. On the other hand, PAF-1 has a
high CO2 capacity of 29.6 mmol/g at 298 K and 40 bar, making it more rele-
vant to pre- rather than post-combustion CO2 capture.8
The porosity in certain MOPs is difficult to probe by the means of N2 sorp-
tion at 77 K. However, CO2 adsorption measurements at higher temperature
(usually at 273 K) have indicated relatively high CO2 capacities and the ex-
istence of ultramicropores in such polymers. The different responses to N2
and CO2 can be explained in two ways. First, CO2 preferably probes ultrami-
cropores, as CO2 has a smaller kinetic diameter in porous materials than N2
(CO2: 3.4 Å, N2: 3.6 Å).68
It is common that sorbents with ultramicropores
exhibit a high CO2 uptake but a kinetically reduced N2 uptake.69
Second,
some polymers with flexible structures appear to shrink into dense and non-
porous solids at the low working temperature (77 K) used in N2 sorption
porosity characterization.70
The high CO2 but low N2 uptakes usually result
in high CO2-over-N2 selectivities in such MOPs.
15
In addition, CO2 capacity and selectivity can be increased by introducing
specific functional groups or moieties to induce chemisorption of CO2.26
For
example, amine-modified sorbents are efficient for CO2 capture, as has been
studied widely on mesoporous silica,71
MOFs72
and activated carbons73
as
substrates. The strong interaction between basic amine moieties and CO2
molecules generates chemisorption of CO2 with the possible formation of
ammonium carbamates, surface bound carbamates, or carbamic acid.74
In
comparison, MOPs have a synthetic diversity and rich functionality that
make them suitable substrates for post-modification. For example, the poly-
amine-tethered PPN-6 (Figure 1-3) had a high CO2 uptake (3.04 mmol/g at
0.15 bar and 295 K) because of strong interactions between the polymer and
CO2.62
On the other hand, the surface area and N2 uptake of these sorbents
decreased after the modification as the pores were partly occupied by teth-
ered functional groups. Hence, the selectivity of PPN-6 dramatically in-
creased (up to 442 at 295 K) due to the synergetic effect when tethered with
polyamines.
Figure 1-3. Post-synthetic amine modification of PPN-6. (a) CH3COOH/
HCl/H3PO4/HCHO, 90 ºC, 3 days; (b) amine, 90 ºC, 3 days.62
Reproduced
with permission.
The isosteric heat of adsorption (Qst) of CO2 indicates the binding affinity
between an adsorbent surface and CO2. In general, a high Qst value is pre-
ferred for applications in which CO2 concentration is low, for instance, in the
flue gases from coal- and gas-fired power plants. In addition, a high Qst val-
ue indicates a high dependency of the adsorption capacity on temperature,
which is relevant for temperature-swing driven capture. Chemical function-
16
alization of the MOPs is an efficient way to increase the Qst. For example,
polyamine tethered PPN-6-NH2 has very high Qst (up to 60 kJ/mol), whereas
the unmodified substrate has a much lower Qst of 25 kJ/mol.62
1.3.2 Heterogeneous catalysis
The immobilization of catalytic sites onto solid supports to produce recycla-
ble heterogeneous catalysts has been well investigated and proven very use-
ful to the chemical, petrochemical, and pharmaceutical industries.75
Conven-
tional porous materials such as metal oxides, activated carbons and ceramics
have been extensively employed as supports for such immobilization.76
In
contrast, MOPs offer a number of advantages over traditional porous sup-
ports for selected challenges in heterogeneous catalytic processes.77
First,
MOPs, with their defined chemical structure and rich functionality, allow
flexibility in immobilizing catalytic sites. In addition, their high surface area
could deliver, in principle, a high catalytic efficiency. In certain cases, mes-
opores can help to accommodate relatively large catalysts, reactants and
products. Furthermore, their high thermal and physicochemical stability
allow the use of catalytic processes under relatively harsh conditions, for
example, in gas phase transformations at high temperatures.
There are several routes to design MOPs with incorporated or immobi-
lized catalytic sites (Figure 1-4), and post-synthetic modification is the most
common method.78
In this way, functional groups in the MOPs bind catalyti-
cally active metal ions or complexes by coordination or chemical bonds. For
example, nitrogen atoms were used to anchor PtCl2 to a triazine-based
framework (CFT). The obtained Pt-CFT was an efficient catalyst for the
oxidation of methane to methanol at 215 °C and 40 bar. The strongly stable
CFT made the reaction feasible at such a high temperature and pressure.79
Using rigid and well-defined metallo-organic monomers as building
blocks is an alternative way to incorporate catalytic sites in MOPs.80
These
materials can be thought of as metallo-organic polymers, which are amor-
phous analogies to MOFs.23
The “bottom up” approach provides simple and
direct method for the synthesis of heterogeneous catalyst based on MOPs
with well-defined structure and relatively high catalyst loading.75
For exam-
ple, the copolymerization by Sonogashira–Hagihara coupling of 1,3,5-
triethynylbenzene, 1,4-dibromobenzene, and a monomer containing a coor-
dinated iridium complex resulted in a series of CMP-based heterogeneous
catalysts with surface area up to 864 m2/g and iridium loading up to 20.1
wt%. The CMP-CpIr-3 exhibited an activity for reductive amination that was
comparable to those of the homogeneous catalyst.81
17
Figure 1-4. Incorporation of active catalytic sites in MOPs for heterogeneous
catalysis.18
(a) Post-synthetic modification of CFT with Pt(II); (b) Synthesis
of metallo-organic CMP by a bottom-up method; (c) Encapsulation of Pd
nanoparticles in CMPs. Reproduced with permission.
In addition to metal ions or complexes, metal nanoparticles can be also
encapsulated in MOPs by post-synthetic modification.82
The pores in MOPs
play important roles in separating metal nanoparticles, stabilizing them
against aggregation and leaching and thus preventing deactivation. For ex-
ample, CFT with encapsulated Pd nanoparticles has been developed as cata-
lyst for liquid-phase glycerol oxidation; the Pd/CTF had better activity and
in particularly better stability than Pd/activated carbons.83
18
1.3.3 Other applications
With an increasing demand for devices that store energy and power, super-
capacitors, with their high capacity to store power, have been increasingly
studied. Electrode materials such as graphenes,84
carbon nanotubes,85
aero-
gels86
and recently MOPs87
have been developed for supercapacitors. High
specific surface area, optimized pore sizes, and a high conductivity of the
electrode materials are essential factors for the design of supercapacitors.
The conductivity in MOPs can be enhanced by the construction of conjugat-
ed structures and by increasing the polymerization temperature. In addition,
nitrogen doping can enhance supercapacitor performance. For example, aza-
CMPs and CFTs with high conductivity and high nitrogen content obtained
by ionothermal reactions at high temperature displayed relative high capaci-
ties.87,88
Other applications related to photocatalysis,89
light harvesting,27
and sen-
sors90
have been also studied for MOPs. For example, CMPs prepared by
copolymerization are organic photocatalysts for visible-light driven hydro-
gen evolution.91
A polyphenylene-based CMP was found to have light-
harvesting properties and emit blue luminescence.39
1.4 Aims of this study
In this thesis, a series of imine/azo-linked MOPs was synthesized and stud-
ied. The synthesis conditions were optimized to target a high surface area.
Due to the presence of (ultra)micropores and relatively high surface area, the
MOPs were investigated as potential sorbents for CO2. In order to increase
their capacity and selectivity for CO2, chemisorption of CO2 was promoted
in the MOPs by introducing CO2-philic groups. Primary amines were intro-
duced post-synthetically on selected MOPs rich in aldehyde end groups.
As imine and azo moieties can bind metal ions by coordination or chemi-
cal bonds, Pd(II) species were incorporated on selected MOPs, and the re-
sulting materials were studied in heterogeneous catalysis. Their catalytic
activity was examined by cascade transformations, Suzuki coupling and
Heck reactions.
19
2. Synthesis of imine-linked microporous organic polymers for CO2 capture and separation
Imine chemistry was initially developed by Hugo Schiff in the 19th century.
Hence, imines are sometimes denoted as Schiff bases. An imine compound
contains a carbonnitrogen double (C=N) bond, which is prepared by the
condensation of a primary amine and an aldehyde or a ketone. In the synthe-
sis mechanism, the amine nucleophile is added to the carbonyl group to form
a hemiaminal intermediate, from which the elimination of a water molecule
yields the imine. Such reactions are usually supported by acid catalysts, for
example, acetic acid, trifluoroacetic acid and p-toluenesulfonic acid. The
acid catalysts protonate the carbonyl groups, making them more electrophilic,
and thus accelerate the reactions. Scheme 2-1 illustrates the mechanism of
formation of imine compound.
Scheme 2-1. Mechanism of acid-catalyzed imine formation.
20
As discussed in Chapter 1, imine chemistry has been used for the synthe-
ses of both amorphous MOPs and crystalline COFs. Imine-linked amorphous
MOPs were usually prepared by condensing aromatic amines with aromatic
aldehydes in highly polar solvents (DMF, DMSO) at high temperatures
(~150180 °C) without a catalyst. Such high temperatures accelerate the
reaction, obviating the acid catalysts. More importantly, highly crosslinked
and interpenetrated structures with (ultra)micropores are preferably generat-
ed at such high temperatures. The (ultra)micropores of the sorbents are cru-
cial for the CO2 capture at post-combustion conditions.
In papers II, III and IV, we chose an aromatic triamine, 1,3,5-tris(4-
aminophenyl)benzene, as the main building block. Condensations of the
amine and a set of aldehydes or ketones resulted in a series of imine-linked
MOPs with tunable surface areas and pore sizes. The MOPs exhibited large
CO2 adsorption, high CO2-over-N2 selectivity, high thermal stability and
moderate hydrophobicity, which are important properties of potential
sorbents for post-combustion capture of CO2.
2.1 Synthesis of imine-linked MOPs
2.1.1 General methods
In the papers of this thesis, all the imine-linked MOPs were synthesized us-
ing similar methods. In general, the amine monomer (A) and the aldehyde
(or ketone) monomers (B) were dissolved separately in DMSO to form clear
solutions. The aldehyde solution was added dropwise to the amine solution
at 50 °C. Yellow solutions formed. The reactions were heated to 120 °C for
12 h and then 180 °C (or the reflux temperature of DMSO) for 72 h. Yellow
precipitates formed at 120 °C, and the suspension turned to brown during
heating to 180 °C (or the reflux temperature of DMSO). The color change
indicated that polymers with high molecular weight formed at these high
temperatures. The brown precipitates were collected and washed with DMF,
methanol, THF and dichloromethane until the filtrates were colorless, and
then dried at 200 °C in an N2 atmosphere to give the MOP products. All the
MOPs were synthesized in yields of ~9098%.
2.1.2 Synthesis conditions
As the Schiff base formation occurs under a dynamic equilibration with wa-
ter elimination/addition, removing the eliminated water molecules can drive
the reversible reactions towards condensation. Water removal promotes the
formation of polymers with high molecular weight and probably increases
the porosity of the polymers. It can be achieved by evaporation at high reac-
21
tion temperatures and/or by using desiccants. In Paper II, we investigated the
effect of water removal on the porosity of the MOPs by either keeping or
removing the water during the syntheses. Different syntheses of MOP-1
were conducted in (I) a closed system in which the flask was sealed with a
stopper during the reaction; (II) a closed system with the inclusion of molec-
ular sieves (3 Å); (III) an open system in which the flask was left open to the
atmosphere during the reaction; (IV) an open system with added molecular
sieves (3 Å). Based upon N2 adsorption analyses, the MOPs synthesized in
open systems (III, IV) had much higher surface areas and micropore vol-
umes than those synthesized in closed systems (I, IV). Using molecular
sieves as a desiccant slightly further increased the surface areas and mi-
cropore volumes; however, it introduced an extra solidsolid separation pro-
cess to the synthesis. Hence, we used condition (III), the open system, for all
subsequent MOP syntheses.
2.1.3 Synthesis of MOPs with monomers of different types
Scheme 2-2. Synthesis routes of aldimine- and ketimine-linked MOPs. p-
Toluenesulfonic acid was used as a catalyst for the synthesis of MOP-1-CH3
and MOP-1-Ph. [As studied in Papers II and III]66,92
Imines can be synthesized by the condensation of amines and aldehydes but
also with ketones, leading to (ald)imines and (ket)imines, respectively. Al-
dehydes have been used to synthesize various aldimine-linked MOPs.28,51,52
However, ketimine-linked porous polymers had not yet been reported before
Paper III. In comparison with aldehydes, ketones introduce additional func-
tional groups in the polymers, and these can add new properties, for example
hydrophobicity. As shown in Scheme 2-2, condensation of the triamine (A3)
and the dialdehyde (B2) with a molecular ratio of 2:3 forms aldimine-linked
22
MOP-1 (the numbers of amine and aldehyde groups was equal). Likewise,
diketone monomers (B2a and B2b) were used to form ketimine-linked MOPs.
p-Toluenesulfonic acid was used to catalyze the syntheses of polyketimines
(MOP-1-CH3, MOP-1-Ph) and to generate high porosity.
2.1.4 Synthesis of MOPs with monomers of different lengths
Scheme 2-3. Synthesis routes to imine-linked MOPs with monomers of dif-
ferent lengths. [As studied in Paper II]66
23
The porosities of the MOP series described in this thesis were largely de-
pended on the monomer length. In general, long monomers tended to pro-
mote large grids or cavities in the MOPs, as we (Paper II) and others6 have
shown; thus, pores with larger size generated. This tendency provides an
opportunity to control pore size. A set of dialdehyde monomers terephthal-
dehyde (B2), isophthaldehyde (B2c) and 4,4-biphenyldicarboxaldehyde (B2d),
with molecular lengths of 7.14, 6.18 and 11.1 Å, respectively, were used for
the syntheses of MOPs. (Scheme 2-3)
2.1.5 Synthesis of porous polymers with monomers in tunable
ratios
Scheme 2-4. Synthesis routes to imine-linked porous polymers using differ-
ent amine-to-aldehyde ratios. [From paper IV]93
When using cross-coupling or condensation reactions for the synthesis of
MOPs, an equal ratio between different functional groups is usually used to
complete the polymerization and generate high porosity. Recently, Thomas‟
group showed that the surface areas and porosities were affected by the stoi-
chiometric ratio as a result of conformation change or structure defects in the
polymer networks.94
In addition, an unequal stoichiometric ratio allowed the
modulation of the polymer end groups. Further post-modification of the end
groups might introduce new functionalities in the polymers, which we stud-
ied in Paper V. As shown in Scheme 2-4, a series of polymers with tunable
surface areas and pore sizes were produced by tuning the amine-to-aldehyde
ratio in the range of (0.252):1 during the condensation between the triamine
(A3) and 1,3,5-benzenetricarboxaldehyde (B3). The porosity analyses showed
micopores and significant amounts of mesopores to be present in the poly-
24
mers. In order to distinguish „microporous organic polymers‟ referred the
polymers with mainly micropores, we denoted this series polymers with both
micropores and mesopores as „porous polymers‟ (PPs). Specifically, PP2-1,
PP1.5-1, PP1-1, PP1-1.5, PP1-2, PP1-3 and PP1-4 denote the porous poly-
mers, which were synthesized using the amine-to-aldehyde ratio of 2:1, 1.5:1,
1:1, 1:1.5, 1:2, 1:3 and 1:4, respectively.
2.1.6 Post-synthesis heat treatment
DMSO is commonly used as a solvent for certain polycondensations reac-
tions. We chose it as a solvent for the syntheses of MOPs because of its high
polarity and high boiling point. However, it tended to decompose or poly-
merize at high temperatures. The DMSO-based derivatives were difficult to
remove by the method of washing or even by Soxhlet extraction with com-
mon organic solvents; the residues appeared to occupy the pores and thus
reduce the surface area of the synthesized MOPs. Surprisingly, we found that
post-heating of the synthesized MOPs in their solid states was an efficient
way to remove the DMSO derivatives and, consequently, increase the sur-
face area of the MOPs. For example, in Paper IV, we showed that PP1.5-1,
heat treated at 200 °C had a surface area of 256 m2/g and a micropore vol-
ume 0.07 cm3/g; however, the values significantly increased to 547 m
2/g and
0.13 cm3/g after the polymer been heated at 300 °C. Hence, we performed a
heat treatment at 300 °C for all the MOPs to remove most of the DMSO
derivatives but still keep their polymeric structure.
2.2 Chemical structural characterizations
In general, most of MOPs are insoluble and amorphous, which make struc-
tural characterization complicated. Nevertheless, their chemical structures
can be indirectly revealed by the combination of various methods. In this
thesis, we used a large set of tools that include, among others, elemental
analysis, thermal gravimetric analysis (TGA), infrared (IR), Raman spec-
troscopy, and 13
C {1H} cross-polarization (CP) magic angle spinning (MAS)
NMR spectroscopy. These allowed the compositions, chemical and thermal
stabilities, linkages and end groups in such polymers to be determined.18
The imine-linked MOPs described here were characterized using both IR
and solid-state 13
C CPMAS NMR spectroscopy. The IR and 13
C NMR spec-
tra of all these MOPs were quite similar. (Figure 2-1) The IR absorption at a
frequency of ~1670 cm–1
and the 13
C NMR chemical shift at ~163 ppm both
indicated the presence of imine bonds in these MOPs. Signals of unreacted
amine and aldehyde (ketone) groups were detected at ~3300 cm–1
(NH
stretching) and 1792 cm–1
(C=O stretching), respectively, in the IR spectra.
The residual carbonyl groups were clearly detected with bands at ~192 ppm
25
Figure 2-1. IR and 13
C NMR spectra of selected imine-linked MOPs (aster-
isks indicate spinning side bands). MOP-1 and MOP-1-CH3 was synthesized
from triamine A3 and dialdehyde/diketone B2 and B2a, PP1-1 was synthe-
sized from triamine A3 and trialdehyde B3 with an equal amine-aldehyde
ratio. All the samples were heat treated at 200 °C. [Compiled the spectra
from paper II, III and IV]66, 92, 93
in the 13
C NMR spectra. Residual amine and aldehyde groups have been
commonly observed in related networks,28
including the crystalline forms
such as COF-300.34
In particular, the MOPs (also called as PPs) synthesized
with an excess of aldehyde groups had more unreacted aldehyde groups. For
example, the relative intensity of this 13
C NMR spectroscopic band was at
4000 3500 3000 2500 2000 1500 1000 500
C=N
Tra
nsm
itta
nce
Wavenumber (cm-1)
MOP-1
MOP-1a
PP-1
C=O
(a)
PP1-1
MOP-1-CH3
200 150 100 50 0
13C NMR chemical shift (ppm)
PP-1
MOP-1-CH3
MOP-1
(b)
C=O C=N
aromatic
C-S
C-S*
*
*PP1-1
*
26
least double for the aldehyde-rich PP1-4 than for the amine rich PP2-1. All
the MOPs of Papers IIIV showed IR bands at a frequency of ~1030 cm–1
and 13
C NMR bands at chemical shifts of 18~40 ppm. The IR bands can be
attributed to S=O stretching while the NMR bands can be assigned to the
carbons attached to sulfur atoms; together, these are indicative of entrapped
DMSO or its derivatives.
Figure 2-2. (a) In situ IR spectra of MOP-1; (b) solid-state 13
C NMR spectra
of synthesized PP1-1 (asterisks indicate spinning side bands), the samples
were heat-treated at 100, 200, and 300 °C for 12 h before the measurement
(from top to down). MOP-1 was synthesized from triamine A3 and dialde-
hyde B2; PP1-1 was synthesized from triamine A3 and trialdehyde B3 with
equal amounts of amine and aldehyde groups. [From Papers II and IV]66, 93
27
The chemical stability of the MOPs of Papers IIIV was investigated by
in situ IR and solid-state 13
C NMR spectroscopy, as shown in Figure 2-2.
The polymeric structure of the MOPs was stable up to 300 °C, although the
IR and 13
C NMR band intensities for the imine bonds slightly decreased at
that temperature. The band intensities relating to residual amine and alde-
hyde groups significantly reduced after heating the MOPs at 300 °C, indicat-
ing that the polymers further condensed or that the end groups decomposed.
Moreover, the intensities of the signals related to the DMSO derivatives
were reduced, showing that DMSO or its derivatives were largely removed
at this temperature. An even higher temperature completely removed the
DMSO derivatives; however, in situ IR spectroscopy showed that the polyi-
mines started to decompose. In addition, TGA showed that these MOPs had
high thermal stabilities with decomposition onset at temperatures of up to
400 °C.
2.3 Porosity analyses
Gas sorption measurements are very powerful when analyzing the pore char-
acteristics of amorphous porous polymers. They can give information on the
specific surface areas, pore volumes, PSDs and even pore shapes. N2 adsorp-
tion/desorption measurements at its boiling point of 77 K are widely used for
these purposes. However, they usually take long time (up to a few days)
because of the slow diffusion of N2 into the micropores at this low tempera-
ture. In addition, the access of N2 to the ultramicropores is sometimes re-
stricted. As a result, N2 sorption experiments often display significant ad-
sorption/desorption hystereses at low relative pressures. In addition, certain
polymers with ultramicropores show negligible N2 adsorption at 77 K.55
Hence, cryogenic N2 sorption measurements cannot give the correct porosity
characteristics for the ultramicropores.
In comparison, CO2 sorption at 273 K renders easy access to the ultrami-
cropores because of its higher diffusion rate at much higher working temper-
ature and smaller kinetic diameter within porous solids (CO2: 3.4 Å, N2: 3.6
Å).68
The isotherms do usually not show significant adsorption/desorption
hysteresis in the absence of strong chemisorption, which indicates a full
probe to the (ultra)micropores. CO2 adsorption isotherms at 273 K appear to
give more precise information of (ultra)micropores than N2 at 77 K.
Some porous polymers also contain mesopores.19,25,95
Classical ordered
mesoporous materials such as mesostructured silicas and carbons usually
possess cylindrical or spheroidal pores with ordered structures and uni-
formed pore sizes.96,97
Such pores can be easily analyzed by cryogenic N2
sorption measurements and electron microscopy. However, the mesopores in
amorphous porous polymers are non-ordered with irregular pore shapes and
28
broad PSDs, which makes the analyses of such mesopores relatively compli-
cated.
In Papers II, III and IV, we performed cryogenic N2 sorption measure-
ments to characterize the porosity information for all the synthesized imine-
linked MOPs. The effects of synthesis conditions, heat treatment, and mon-
omers on the porosity characteristics were investigated. All of the MOPs
contained ultramicropores as confirmed by CO2 sorption measurements at
273 K. The N2 adsorption/desorption hystereses at intermediate and higher
relative pressures (at 77 K) indicated the existence of mesopores in some of
the MOPs (as studied in detail in Paper IV). Such irregular mesopores were
also visible in the transmission electron microscopy (TEM) images.
2.3.1 Effect of synthesis conditions
As shown in Paper II and in Figure 2-3a, MOP-1III, IV
, synthesized in an open
system, displayed distinct and high adsorption of N2 at low relative pressures
(P/P0 < 0.05), which is a typical characteristic of microporous materials.
These variations of MOP-1 had moderately high surface areas and micropore
volumes (Table 2-1). In comparison, the MOP-1I, synthesized in a closed
system, showed a much lower N2 adsorption at low relative pressure and had
the lowest surface area. Using molecular sieves during the syntheses only
slightly increased the surface areas and micropore volumes. This difference
indicates that using an open reaction system was more efficient than adding
molecular sieves as desiccants to remove the water and construct mi-
croporous configuration in such polymers. Figure 2-3b shows the PSDs of
MOP-1. MOP-1III, IV
displayed similar PSDs, with maxima at ~56 Å and
~13 Å. However, MOP-1I, II
only displayed large micropores with much
smaller micropore volumes. These results indicated that water removal is
crucial for the formation of microporous structure in such polymers.
Table 2-1. Porosity characteristics of MOP-1a synthesized under different
conditions. (N2 adsorption data at 77 K was used for the calculation)
Sample
BET surface area
(m2/g)
Micropore
volume (cm3/g)
Total pore
volume (cm3/g)
MOP-1I 70 0.003 0.15
MOP-1II 142 0.036 0.095
MOP-1III
378 0.11 0.18
MOP-1IV
452 0.12 0.31 a MOP-1 was synthesized from triamine A3 and dialdehyde B2. I, II, III
and IV indicate different synthesis conditions.
29
Figure 2-3. (a) N2 adsorption (filled symbols) and desorption (open symbols)
isotherms of MOP-1 recorded at 77 K; (b) PSDs of MOP-1 determined from
N2 adsorption isotherms at 77 K using the density functional theory (DFT)
method. [From Paper II]66
MOP-1I and MOP-1
II displayed significant adsorption/desorption hystere-
sis at low relative pressures, which was consistent with a certain amount of
small pores. In order to probe such small pores, we performed CO2 adsorp-
tion measurement on the polymers. As shown in Figure 2-4a, MOP-1I dis-
played a considerable CO2 adsorption at 273 K and 0–1 bar while MOP-1II
had a much higher CO2 adsorption, which is consistent with the results from
N2 adsorption data at 77 K. PSD analyses based on CO2 adsorption iso-
therms (CO2-DFT model) in Figure 2-4b indicated that both MOP-1I and
MOP-1II had ultramicropores in the size range of 4–8 Å. Such ultrami-
cropores were probably not fully accessible for N2 at 77 K but they can be
easily probed by CO2 at 273 K. As a result, despite their quite different N2
0.0 0.2 0.4 0.6 0.8 1.0
0
100
200
1 2
0.0
0.1
0.2
0.3
0.4
(b)
MOP-1IV
MOP-1II
MOP-1I
MOP-1III
Quantity
adsorb
ed (
cm
3/g
)
P/P0
(a)
Pore
volu
me (
cm
3/g
)
Pore size (nm)
MOP-1I
MOP-1II
MOP-1III
MOP-1IV
30
adsorption isotherms, MOP-1II had similar CO2 adsorption isotherms and
PSDs to MOP-1III
and MOP-1IV
.
Figure 2-4. (a) CO2 adsorption isotherms of MOP-1 at 273 K. (b) PSDs of
MOP-1 obtained by analyzing the CO2 adsorption isotherms at 273 K. MOP-
1 was synthesized from triamine A3 and dialdehyde B2, I, II, III and IV indi-
cate different synthesis conditions. [From Paper II]66
2.3.2 Effect of monomer type
We used dialdehyde and diketone as monomers for the synthesis of polyal-
dimine and polyketimine, respectively. The two types of polyimine had simi-
lar molecular structures; however, they showed significantly different po-
rosity. Terephthaldehyde (B2), 1,4-diacetylbenzene (B2a) and 1,4-
dibenzoylbenzene (B2b), all based upon the same backbone structure, were
used as monomers for the synthesis of MOP-1, MOP-1-CH3 and MOP-1-Ph,
0.0 0.2 0.4 0.6 0.8 1.0
0
1
2
3
0.5 1.0
0.0
0.2
0.4
0.6
0.8
MOP-1III
MOP-1I
MOP-1II
MOP-1IV
Qu
an
tity
ad
so
rbe
d (
mm
ol/g
)
Pressure (bar)
(a)
(b)
Po
re v
olu
me
(cm
3/g
)
Pore size (nm)
MOP-1I
MOP-1II
MOP-1III
MOP-1IV
31
respectively. The resulted polymers had similar molecular structures. In
comparison with the H atom on MOP-1, the methyl and phenyl groups in
MOP-1-CH3 and MOP-1-Ph were likely to occupy the micropores and cause
a slightly decrease of surface area and micropore volume. However, their N2
adsorption isotherms at 77 K were very different to that of MOP-1. MOP-1
was a typical microporous material with a type I adsorption isotherm, as
shown in Figure 2-3a. However, the ketimine-linked MOP-1-CH3 and MOP-
1-Ph appeared to be mainly nonporous, adsorbing negligible amounts of N2
at 77 K, and had very low surface areas (6 and 35 m2/g, respectively).
Figure 2-5. (a) CO2 adsorption isotherms of MOP-1, MOP-1-CH3, and
MOP-1-Ph at 273 K. (b) PSDs of MOP-1, MOP-1-CH3, MOP-1-Ph derived
from the CO2 adsorption isotherms at 273 K using a DFT model. MOP-1,
MOP-1-CH3 and MOP-1-Ph were synthesized from triamine A3 and dialde-
hydes/diketones B2, B2a, B2b, respectively. [Compiled the figures from Paper
II and III]66,92
0.0 0.2 0.4 0.6 0.8 1.0
0
1
2
3
0.4 0.5 0.6 0.7 0.8 0.9 1.0
0.0
0.2
0.4
0.6
0.8
MOP-1-Ph
MOP-1
MOP-1-CH3
Qu
an
tity
ad
so
rbe
d (
mm
ol/g
)
Pressure (bar)
(b)
Po
re v
olu
me
(cm
3/g
)
Pore size (nm)
MOP-1
MOP-1-CH3
MOP-1-Ph
(a)
32
Surprisingly, both MOP-1-CH3 and MOP-1-Ph showed relatively high CO2 adsorption with high corresponding BET surface areas and micropore volumes (CO2, 273 K); in fact, they were comparable with those of MOP-1, as shown in Figure 2-5 and Table 2-2. They demonstrated ultramicropores with size of 49 Å according to the PSD analyses (CO2-DFT model) calcu-lated based upon the CO2 adsorption isotherms recorded at 273 K. The low N2 adsorption at 77 K can be attributed to a diffusion limitation, kinetic hin-drance and/or a shrinking of the polymers at the low temperature used.
Table 2-2. Porosity characteristics of MOP-1, MOP-1-CH3 and MOP-1-Ph
analyzed by N2 and CO2 adsorption.
MOPs
BET surface area
(m2/g)
Micropore volume
(cm3/g)
N2, 77 K CO2, 273 K N2, 77 K CO2, 273 K
MOP-1 378 245 0.11 0.11
MOP-1-CH3 6 229 - 0.10
MOP-1-Ph 35 267 - 0.12
Note: (-) means negligible value; MOP-1, MOP-1-CH3 and MOP-1-Ph were
synthesized from triamine A3 and dialdehydes/diketones B2, B2a and B2b.
2.3.3 Effect of monomer length
As previous studies had revealed that the monomer length could affect the
porosity characteristics of MOPs,6 we chose aldehydes with different lengths
(B2, B2c and B2d; refer to Scheme 2-3) for the syntheses of MOP-1, MOP-2
and MOP-3. All of the MOPs were typical microporous materials and dis-
played distinct steep N2 adsorption at low relative pressures (Figure 2-6a).
MOP-2, synthesized from the shortest aldehyde, B2c, displayed the highest
surface area of 614 m2/g (N2, 77 K, BET). MOP-1 and MOP-3, built from
longer monomers, displayed lower surface areas of 378 and 589 m2/g. The
decreased surface areas may be attributed to an increased flexibility and
degree of interpenetration in the polymers. The ultramicropore size and vol-
ume displayed a clear dependence on monomer length. MOP-2, MOP-1 and
MOP-3 demonstrated ultramicropore sizes of ~4, 5, and 8 Å, respectively,
following the same order as monomer length (B2c: 6.18 Å, B2: 7.14 Å, B2d:
11.1 Å) (Figure 2-6b). Therefore, shorter monomers assisted the formation
of microporous structures, whereas longer monomers appear to construct
large pores with thicker “walls”.
33
Figure 2-6. (a) N2 adsorption (filled symbols)/desorption (open symbols)
isotherms of MOP-1, MOP-2 and MOP-3, recorded at 77 K; (b) PSDs of
MOP-1, MOP-2 and MOP-3 calculated from the N2 adsorption isotherms.
MOP-1, MOP-2 and MOP-3 were synthesized from triamine A3 and dialde-
hydes B2, B2c and B2d. [From Paper II]66
2.3.4 Effect of monomer ratio
In paper IV, a series of porous polymers (PPs) with tunable porosities were
synthesized by tuning the monomer ratio of the triamine (A3) and the trialde-
hyde (B3) during the polycondensation. The porosity characteristics of the
PPs were evaluated by N2 (77 K) and CO2 (273 K) adsorption measurements.
Figure 2-7 shows the N2 adsorption/desorption isotherms of the PPs. All the
PPs except PP2-1 had distinct N2 adsorption at low relative pressures (P/P0 <
0.0 0.2 0.4 0.6 0.8 1.0
0
50
100
150
200
250
1 2
0.0
0.2
0.4
0.6
(b)
MOP-1
MOP-2
MOP-3
Qua
ntity
ad
so
rbe
d (
cm
3/g
)
P/P0
(a)
Pore
vo
lum
e (
cm
3/g
)
Pore size (nm)
MOP-1
MOP-2
MOP-3
34
0.05), which was indicative of microporous structures. PSD analyses of the
N2 adsorption isotherms using a DFT model (N2-DFT) indicated that the PPs
Figure 2-7. N2 adsorption (filled symbols) and desorption (open symbols)
isotherms of porous polymers PP2-1, PP1.5-1, PP1-1, PP1-1.5, PP1-2, PP1-3
and PP1-4. PPX-Y were synthesized from triamine A3 and trialdehyde B3
using amine-to-aldehyde ratios of X:Y. [From paper IV]93
0.0 0.2 0.4 0.6 0.8 1.00
100
200
300
400
0.0 0.2 0.4 0.6 0.8 1.00
100
200
300
400
0.0 0.2 0.4 0.6 0.8 1.00
100
200
300
400
0.0 0.2 0.4 0.6 0.8 1.00
100
200
300
400
0.0 0.2 0.4 0.6 0.8 1.00
100
200
300
400
0.0 0.2 0.4 0.6 0.8 1.00
100
200
300
400
0.0 0.2 0.4 0.6 0.8 1.00
100
200
300
400
Quantity
adsorb
ed (
cm
3/g
)
P/P0
Quantity
adsorb
ed (
cm
3/g
)
P/P0
Quantity
adsorb
ed (
cm
3/g
)
P/P0
PP1-4
PP1-1.5
PP1-3
PP1.5-1
PP1-1
PP2-1
PP1-2
Quantity
adsorb
ed (
cm
3/g
)
P/P0
Quantity
adsorb
ed (
cm
3/g
)
P/P0
Quantity
adsorb
ed (
cm
3/g
)
P/P0
Quantity
adsorb
ed (
cm
3/g
)
P/P0
35
had micropores with diameters of ~12 Å (Figure 2-8b). They had moderately
large BET surface areas (445694 m2/g, N2, 77 K) and micropore volumes
(0.100.20 cm3/g). PP2-1, synthesized with the greatest excess of amine
functionalities, had a low BET surface area of 35 m2/g (N2, 77 K). In com-
parison to the significant dissimilarities of the N2 adsorption between PP2-1
and other PPs, all the PPs had less varied CO2 adsorption characteristics
(Table 2-3). For example, PP2-1 had a relatively high CO2 capacity of 2.20
mmol/g at 1 bar and 273 K with ultramicropores with diameters of 49 Å
(CO2-DFT model) (Figure 2-8a) and ultramicropore volume of 0.09 cm3/g.
These values are comparable with other PPs having much higher BET sur-
face areas (N2, 77 K). N2 adsorption and desorption isotherms of porous
polymers PP2-1, PP1.5-1, PP1-1, PP1-1.5, PP1-2, PP1-3 and PP1-4 at 77 K
are shown in Figure 2-7. It appears that the ultramicropores in the PPs were
not fully accessed by N2 at 77 K, and using CO2 adsorption data at 273 K
better probed the ultramicropores due to the smaller kinetic diameter of CO2
(in porous materials) and a faster diffusion rate at 273 K.98
Table 2-3. Porosity characteristics of porous polymers, analyzed by N2 and
CO2 adsorption.
Porous polymersa
SBET
(m2/g)
b
Vmicro
(cm3/g)
b
Vultramicro
(cm3/g)
c
Vtotal
(cm3/g)
b
PP2-1 17 - 0.09 0.016
PP1.5-1 547 0.13 0.10 0.34
PP1-1 639 0.18 0.13 0.54
PP1-1.5 536 0.13 0.11 0.59
PP1-2 648 0.17 0.12 0.67
PP1-3 694 0.20 0.13 0.45
PP1-4 449 0.13 0.11 0.24
a Porous polymers PPX-Y were synthesized from triamine A3 and trialde-
hyde B3 using amine-to-aldehyde ratios of X:Y; b Calculated from N2 ad-
sorption isotherms at 77 K; c Calculated from CO2 adsorption isotherms at
273 K. [From paper IV]93
The N2 adsorption/desorption isotherms of the PPs, published in Paper IV,
displayed dissimilarities at high relative pressures. The high-pressure hyste-
resis loops of PP1.5-1, PP1-1, PP1-1.5, PP1-2 and PP1-3 were consistent
36
with mesopores with broad PSDs (calculated using the Barrett-Joyner-
Halenda (BJH) model). The mesopore size grew when the relative fraction
of aldehyde was increased up to a certain point, see Figure 2-8c. PP1-3
showed a smaller mesopore size and mesopore volumes than PP1-2, PP1-1.5
and PP1-1. PP1-4, which was synthesized with the largest excess of alde-
hyde, displayed almost no mesopores. PP1.5-1, PP1-1 and PP1-3 displayed
significant desorption steps at a specific relative pressures (P/P0 = 0.49),
which indicated the mesopores were partly blocked.
Furthermore, we used TEM to image the mesopores (Figure 2-9a). The
contrast variations in the TEM image of PP1-1.5 indicated mesopores. Un-
like classical mesoporous materials with ordered mesopore structures and
uniform pore size,96,97
the porous polymer displayed irregular and non-
ordered mesopores. The scanning electron microscopy (SEM) image of PP1-
1.5 also showed the foam-like meso- and macropores (Figure 2-9b).
Figure 2-8. PSDs at different porosity scales: (a) ultramicropores (using CO2
adsorption isotherms at 273 K and the DFT model); (b) micropores (using N2
adsorption isotherms at 77 K and the DFT model); (c) mesopores (using N2
adsorption isotherms at 77 K and the BJH model). PPX-Y were synthesized
from triamine A3 and trialdehyde B3 using amine to aldehyde ratios of X:Y =
1:4, 1:3, 1:2, 1:1.5, 1:1, 1:1.5 and 2:1. [From Paper IV]93
37
Figure 2-9. TEM (left) and SEM (right) images of PP1-1.5. PP1-1.5 was
synthesized from triamine A3 and trialdehyde B3 using an amine-to-aldehyde
ratio of 1:1.5. [From Paper IV]93
2.4 MOPs for CO2 capture and separation
The MOP and PP series described in this thesis had ultramicropores and
displayed relatively high CO2 adsorption. Their low pressure (0.15 bar) ad-
sorption surpassed that of many related polymers and compared well with
those of top-performers, such as BILPs (1.251.99 mmol/g at 0 °C).53
In
particular, some of the MOPs and PPs had high CO2 adsorption but negligi-
ble N2 adsorption. These observations promoted us to study their ability for
potential separation of CO2 from N2. The CO2 and N2 adsorption isotherms
were recorded at 273 K and 01 bar from which the CO2-over-N2 selectivity
was estimated from a simple selectivity estimation using the single compo-
nent gas uptakes of CO2 and N2 and the relative pressures of those gases the
single component: S = (qA/qB)/(pA/pB) (S: selectivity; q: uptake; p: partial
pressure; A, B: gas components).62,99
All the MOPs and PPs displayed rela-
tively high CO2-over-N2 selectivities, as shown in Table 2-4. PP2-1, which
was synthesized with the highest amine excess, had the highest estimated
selectivity of 90. Such a high selectivity can be attributed to additional ef-
fects of molecular sieving or kinetic hindrance of N2 adsorption by the ul-
tramicropores. In addition, MOP-1-CH3 and MOP-1-Ph showed relatively
high CO2-over-N2 selectivities of 26 and 25, respectively, at 40 ºC. This
temperature is relevant for realistic CO2 capture after an effective heat ex-
change of a flue gas.
In order to investigate the binding affinity of the MOPs and PPs toward
CO2, we calculated the Qst of CO2 using adsorption data at 273 K and 293 K
from the Clausius–Clapeyron equation (equation I) and the isosteres.
38
Sorbents with high Qst values are preferred to capture CO2 at low partial
pressures and are relevant for temperature-swing adsorption.100
A too-high
Qst value would induce high energy penalties for the regeneration of the
sorbents.64,101,102
Hence, the adsorption performance and the Qst value should
be balanced.103
All the MOPs with high CO2 adsorption capacities had Qst of
2535 kJ/mol, typical for the physisorption of CO2. The physisorption pro-
cess is beneficial to the regeneration of the sorbents with low energy penal-
ties. Table 2-4 summarizes the CO2 and N2 capacities, selectivities and Qst
for CO2 adsorption for the MOPs and PPs studied in this thesis.
Table 2-4. Summary of gas adsorption, selectivity and heat of CO2 adsorp-
tion on microporous organic polymers (MOPs) and porous polymers (PPs).
MOP
/PP
CO2 adsorption at
273 K (mmol/g)
N2 adsorption at
273 K (mmol/g) SCO2/N2
a
Qstb
(kJ/mol) 0.15 bar 1 bar 0.85 bar 1 bar
MOP-1 1.20 2.66 0.10 0.12 68 30
MOP-1-
CH3 0.93 2.41 0.14 0.17 37 25
MOP-1-
Ph 1.26 2.99 0.21 0.25 34 34
MOP-2 1.02 2.71 0.10 0.12 56 34
MOP-3 0.93 2.25 0.08 0.10 65 28
PP2-1 0.93 2.20 0.06 0.07 90 33
PP1.5-1 1.09 2.60 0.20 0.23 31 31
PP1-1 1.58 3.14 0.26 0.3 34 33
PP1-1.5 1.35 2.74 0.23 0.26 34 31
PP1-2 1.47 3.04 0.23 0.25 39 32
PP1-3 1.58 3.28 0.29 0.33 31 35
PP1-4 1.42 2.76 0.18 0.21 44 35
a SCO2/N2 = [nCO2(0.15bar)/nN2(0.85bar)]/(0.15/0.85);
b Heat of adsorption of CO2 at low
coverage.
39
In general, sorption kinetics is quite important for adsorption-driven sepa-
ration processes. A rapid adsorption might shorten the time and reduce the
cost of a swing sorption processes. A kinetic study of the adsorption of CO2
showed a quite rapid adsorption of CO2 for PP1-3. Figure 2-10 displays the
adsorption kinetics of CO2 at pressures of 0.15 bar and 0.008 bar and at a
temperature of 273 K. A more rapid adsorption was observed for the higher
pressure condition than was expected from Darken‟s relation.104
Figure 2-10. Kinetic CO2 adsorption curves of PP1-3 at CO2 pressures of
0.008 and 0.15 bar, 273 K. PP1-3 was synthesized from triamine A3 and
trialdehyde B3 using an amine-to-aldehyde ratio of 1:3. [From Paper IV]93
As flue gases emitted from power plants are humid, the competition be-
tween adsorption of water and CO2 typically depresses the efficiency of cap-
turing CO2. Drying of the flue gases is important when using hydrophilic
sorbents, such as zeolites; however, the drying process costs an extra ~$10
per ton of captured CO2.105
Hence, sorbents with a hydrophobicity feature
are expected to increase the separation efficiency and reduce the cost. Porous
polymers, with their diverse synthesis routes and straightforward post-
modification processes, can be potentially designed as hydrophobic sorbents.
Among all synthesized MOPs, MOP-1-CH3 and MOP-1-Ph, with me-
thyl/phenyl groups, are expected to be the most hydrophobic. We performed
water vapor adsorption measurements for MOP-1-CH3 and MOP-1-Ph. As
shown in Figure 2-11, the adsorption of water vapor on the MOPs was small
at low pressures, which is indicative of hydrophobicity. At a high relative
humidity, the water adsorption was high probably due to large (extra parti-
cle) pores or possibly the swelling of the polymers. When compared with
zeolites or activated carbons,106
the MOPs had much lower adsorption of
water at low pressures and were, hence, comparably hydrophobic.
0 100 200 300 400
1.35
1.40
1.45
Time (s)
Qu
an
tity
ad
so
rbe
d a
t 0
.15
ba
r (m
mo
l/g
)
0.00
0.05
0.10
0.15
0.20
Qu
an
tity a
dso
rbe
d a
t 0.0
08
ba
r (mm
ol/g
)
0.008 bar
0.15 bar
40
Figure 2-11. Water vapor adsorption (filled symbols)/desorption (open sym-
bols) isotherms of MOP-1-CH3 and MOP-1-Ph recorded at 293 K. (●: MOP-
1-CH3, ▲: MOP-1-Ph) MOP-1-CH3 and MOP-1-Ph were synthesized from
triamine A3 and diketones B2a and B2b, respectively. [Reproduced from Paper
III]92
In summary, imine-linked MOPs and PPs with tunable porosities were
synthesized. All these MOPs and PPs had high CO2 adsorption and CO2-
over-N2 selectivities under the conditions of post-combustion CO2 capture
because of their ultramicropores. They displayed moderate Qst values, rapid
CO2 adsorption, and hydrophobicity, which might further increase the effi-
ciency of separating CO2 from flue gas and reduce the cost of their use. With
these collected advantages, we expect that such MOPs and PPs can be fur-
ther optimized for potential use as sorbents for post-combustion capture of
CO2 from flue gases.
0.01 0.1 1
0
2
4
6
8
10
Qua
ntity
adsorb
ed (
mm
ol/g)
Pressure (bar)
41
3. Adsorption of CO2 on a porous polymer modified with tris(2-aminoethyl)amine
3.1 Amine modification of porous polymers
As described in chapter 2 and in Paper IV, porous polymers were synthe-
sized from reactions between 1,3,5-tris(4-aminophenyl)benzene (A3) and
1,3,5-benzenetricarboxaldehyde (B3) with tunable amine-to-aldehyde ratios.
All the porous polymers had ultramicropores and displayed high CO2 capaci-
ties and high CO2-over-N2 selectivities. Mesopores were generated under
most amine-to-aldehyde ratios. The PPs synthesized with an excess of alde-
hyde also had rich aldehyde end groups, as confirmed by solid-state IR and 13
C CPMAS NMR spectroscopy. Such aldehyde groups can be further modi-
fied by multifunctional alkyl amines to introduce the potential for CO2
chemisorption and hence increase the capacity and selectivity for CO2.
Moreover, the mesopores in the porous polymers provide space for the post-
modification and also favor mass transport during the adsorption and desorp-
tion processes.
The aldehyde-rich PP1-2 displayed both ultramicropores and mesopores
and was judged to be a good substrate for the amine modification to target
chemisorption of CO2. Tris(2-aminoethyl)amine (tren), a strongly basic
alkyl amine with pKb ~ 4, was tethered to the substrate (Scheme 3-1). Specif-
ically, PP1-2 (100 mg, the as-synthesized sample was heat-treated at 200 ºC
under a N2 atmosphere for 12 h) and tren (10 mL) were added to a 25-mL
round-bottom flask under an atmosphere of N2. The flask was sealed and
heated to 100 ºC for 1 day. The solid was collected and washed several times
with THF and methanol, then dried at 100 ºC under N2.
42
Scheme 3-1. Synthesis and amine modification of the porous polymer PP1-2.
[From Paper V]107
3.2 Molecular studies by spectroscopy
PP1-2 and tren reacted, as confirmed by the change in the IR and solid-state 13
C CPMAS NMR spectra. The 13
C NMR spectrum of PP1-2 shows bands
for the carbonyl carbons of the aldehyde end groups at a chemical shift of
193 ppm and for the carbons of imine groups at 161 ppm. In comparison, the
NMR spectrum of PP1-2-tren shows no carbonyl carbons and a slight in-
crease in the relative intensity of the carbons of the imine groups. (Figure 3-
1a) Hence, the aldehyde end groups in the porous polymer were consumed
by the reaction with tren, and new imine bonds were formed. The new band
at 1630 cm–1
in the IR spectrum of PP1-2-tren also indicates the formation of
new imine bonds. (Figure 3-1b) The primary amine groups of tethered tren
were detected by near infrared (NIR) spectroscopy, which is an important
spectroscopic method that uses the NIR region of electromagnetic spectrum
(wavelength: ~8002500 nm) to examine molecular overtones and combina-
tion vibrations. (Figure 3-1c) The NIR spectrum of PP1-2-tren clearly shows
a combination band (at 4930 cm–1
) and an overtone band (at 6512 cm–1
) be-
longing to primary amine groups. NIR spectra of primary amines show such
bands.108
In addition, IR and NMR spectroscopy detected the alkyl groups of
tethered tren. In the solid-state 13
C NMR spectrum in Figure 3-1a, the chem-
ical shifts at 40 and 54 ppm are assigned to the CH2 groups in tren (the sig-
nal for the alkyl groups at 40 ppm partly overlapped with the side bands). In
the IR spectrum in Figure 3-1b the bands at 2923 cm–1
and 2846 cm–1
are
assigned to the C–H stretching modes in the CH2 groups of the tren moie-
ties.
43
Figure 3-1. (a) Solid-state 13
C NMR (asterisks indicate spinning side bands)
and (b) IR spectra of PP1-2 and PP1-2-tren. (c) NIR spectrum of tris(2-
aminoethyl)amine and PP1-2-tren. [From paper V]107
(a)
4000 3500 3000 2500 2000 1500 1000 500
PP1-2-tren
C=N
Wavenuber (cm-1)
N-H C-H
C=N
PP1-2
Tra
nsm
itta
nce
(b)
6500 6000 5500 5000 4500
Wavenumber (cm-1)
Ab
sorb
an
ce
[ν(N-H) + δ(N-H2)
2νa(N-H)
PP1-2-tren
tris(2-aminoethyl)amine6512
4930
2v (C-H)
(c)
aromatic
aliphatic
* *
*
300 250 200 150 100 50 0 -50 -100
13C NMR chemical shift (ppm)
PP1-2
PP1-2-tren
44
The molecular details of the interaction between CO2 and PP1-2-tren were
revealed by in situ IR and solid-state 13
C NMR spectroscopy. Previous study
of amine-CO2 chemistry in amine-modified silica indicated a possible for-
mation of ammonium carbamates, surface bound carbamates or carbamic
acid, depending on the nature and concentration of the amines, the distance
between amine groups, the humidity, and other factors.71,74,109-111
As tren has
pKb ~ 4, chemisorption of CO2 in PP1-2-tren was hypothesized. The IR and
NMR spectra showed that CO2 molecules were indeed chemisorbed onto
PP1-2-tren (Figure 3-2). Chemisorbed CO2 was detected by the bands in the
range of 11001715 cm–1
and 3425 cm–1
. The band with the highest intensity
at 1715 cm–1
was assigned to carbonyl groups with a frequency suggestive of
a carbamic acid (or a similar compound with a carbonyl group). The band at
3425 cm–1
was assigned to the N–H stretching mode in carbamic acid. The
broad band between 3600 and 2500 cm–1
was assigned to the O–H stretching
modes typical for hydrogen bonded acids. However, no strong characteristic
bands of NH3+ and (NH)COO
– moieties were observed, which indicated the
absence of ammonium carbamate ion pairs. Such characteristic bands would
have been observed at ~1484 and 1633 cm–1
for NH3+ and 1580 cm
–1 for
(NH)COO–.74
The long distances between the amine groups within the tren molecules
appeared to have prevented the formation of ammonium carbamates. The
change in the solid-state 13
C NMR spectrum of PP1-2-tren upon treatment
with CO2 gas also suggested chemisorption of CO2. PP1-2-tren subjected to
CO2 showed further increase in the relative intensity of the signal at a 13
C
NMR chemical shift of ~160165 ppm, in comparison with that of fresh
sample. This was consistent with the formation of the carbonyl group in
PP1-2-tren upon chemisorption of CO2 (the signals of carbonyl and imine
carbons are partially overlapped). In addition, physisorbed CO2 was visible
in the IR spectrum with the bands at frequencies of 2269 (antisymmetric
stretching band for 13
CO2), 3580 and 3688 cm–1
(combination bands), corre-
sponding to physisorbed CO2 molecules. In conjunction with the IR and 13
C
NMR spectra, we conclude that CO2 was both physisorbed and chemisorbed
onto PP1-2-tren. Carbamic acid formed during the chemisorption of CO2 in
PP1-2-tren (Scheme 3-2).
45
Figure 3-2. (a) Comparison of 13
C NMR spectrum of fresh PP1-2-tren (top)
and CO2 adsorbed PP1-2-tren (middle) and their differences (bottom); (b) In
situ difference IR spectrum of adsorbed CO2 on PP1-2-tren. (A degassed
fresh sample was used to measure the IR background spectrum and the
bands in the difference spectrum relate to the changes upon CO2 adsorption)
[From paper V]107
0
0.01
0.02
0.03
0.04
4000 3500 3000 2500 2000 1500 1000
Wavenumber (cm-1)
Ab
sorb
an
ce
Physisorbed CO2
ν(13CO2)
368
8*
3580
* 3425
2269
*
1715
1539
1225
1640
1403
1470
N-H
(ν3+ν1)(CO2)
(ν3+2ν2)(CO2) C=O
(b)
(a)
300 250 200 150 100 50 0 -50 -100
13C NMR chemical shift (ppm)
C=NC-S
aliphatic
46
Scheme 3-2. Illustration of the formation of carbamic acid in PP1-2-tren
upon chemisorpion of CO2. [From paper V]107
3.3 Porosity analyses
Amine modification on porous sorbents usually results in reduced surface
area, pore size and pore volume because of pore blocking by the bulky
amine groups.62,112,113
As expected, the porosity characteristics of the sub-
strate PP1-2 were greatly affected by tethered amines. Figure 3-3a shows the
N2 adsorption and desorption isotherms of PP1-2 and PP1-2-tren recorded at
77 K. The steep N2 uptake at low relative pressures (P/P0 < 0.05) that was
observed for PP1-2 is a typical feature of microprous materials, and a PSD
analysis (Figure 3-3b) revealed micropores centered at 0.7 and 1.3 nm. In
comparison, PP1-2-tren had much lower N2 uptake at low relative pressures
and displayed no regular micropores. The surface area was reduced from 405
to 72 m2/g and the total pore volume was reduced from 0.54 to 0.27 cm
3/g
upon tren tethering. The reduction of surface area, pore volume and pore
size revealed the successful modification of the substrate from another point
of view. The high-pressure regime (P/P0 > 0.9) showed a steep uptake of N2
and a corresponding hysteresis for both samples, which is typical for a sam-
ple with large mesopores or interparticle voids. Such steep N2 uptakes at
high relative pressures are commonly observed in related porous
polymers.19,28,51,95
47
Figure 3-3. (a) N2 adsorption and desorption isotherms of PP1-2 (▲) and
PP1-2-tren (●). (b) PSDs of PP1-2 (▲) and PP1-2-tren (●) calculated from
the N2 adsorption isotherms measured at 77 K using the DFT model. PP1-2
and PP1-2-tren were degassed at 200 and 100 °C, respectively, for 12 h be-
fore the measurement. [From paper V]107
Note that an inefficient pore blocking by the amines in the substrate PP1-2
could leave some ultramicropores unaffected or generate a certain amount of
ultramicropores in the amine modified polymer PP1-2-tren. It is relatively
common that such ultramicropores in porous polymers cannot be detected by
N2 sorption at 77 K.55,69
Using CO2 sorption at 273 K is a feasible way to
probe the ultramicropores;98
however, porosity analysis by CO2 adsorption
isotherms is not appropriate in this case because of the existence of strong
chemisorption.
0.0 0.2 0.4 0.6 0.8 1.0
0
100
200
300
400
0.5 1.0 1.5 2.0 2.5 3.0 3.5
0.0
0.1
0.2
0.3
0.4
0.5
Quantity
adsorb
ed (
cm
3/g
)
P/P0
(a)
Pore
volu
me (
cm
3/g
)
Pore size (nm)
(b)
48
3.4 Studies of CO2 adsorption and separation
Generally, amine-modified sorbents display significantly higher CO2 capaci-
ty, especially at low relative pressures, than the corresponding unmodified
sorbents because of the strong CO2-amine interactions.62,72,112
In paper V,
analyses of the IR and NMR spectra revealed that CO2 can be both phy-
sisorbed and chemisorbed on the PP1-2-tren. Hence, the sorbent was ex-
pected to show high CO2 capacity at the conditions relevant to a post-
combustion capture of CO2. As shown in Figure 3-4, PP1-2-tren shows dis-
tinct uptake of CO2 at low relative pressures due to the chemisorption on the
basic tren moieties. The uptake of CO2 at pressures of 0.15 and 0.05 bar for
the amine modified polymer (1.45 and 1.13 mmol/g, 273 K) were 61 and
144% higher, respectively, than those of the unmodified porous polymer
(0.90 and 0.46 mmol/g, 273 K).
Figure 3-4. CO2 and N2 adsorption isotherms of PP1-2 (▲: CO2; Δ: N2) and
PP1-2-tren (●: CO2; ○: N2) at 273 K. [From paper V]107
The binding affinities of CO2 to the sorbents were quantified by determin-
ing the Qst. These values are important for an adsorption-driven capture of
CO2. Sorbents with a high Qst value are preferred for capturing CO2 from
diluted flue gases. However, a too-high Qst value could result in a large en-
ergy penalty for the regeneration of the sorbents. The Qst of PP1-2 and PP1-
2-tren was calculated from the temperature-dependent CO2 adsorption iso-
therms (273 and 293 K) using a dual-site Langmuir model that was initially
developed by Mason et al.100
The use of a dual-site (or triple-site, multi-site)
Langmuir model has advantages in that it allows the separation and identifi-
cation of different adsorption sites with significantly different energetics in
sorbents that have inherently heterogeneous pore surfaces and high binding
affinity to sorbates.114
PP1-2 had a Qst value of ~28 kJ/mol at low coverage,
0.0 0.2 0.4 0.6 0.8 1.0
0.0
0.5
1.0
1.5
2.0
2.5
Qu
an
tity
ad
so
rbe
d (
mm
ol/g
)
Pressure (bar)
P=0.05 bar
49
which is typical for the physisorption of CO2. In comparison, the PP1-2-tren
had a very high Qst of ~80 kJ/mol (site A) at a low coverage of CO2, and
remained at a relatively high level of ~40 kJ/mol (site B) for a higher cover-
age of CO2 (Figure 3-5). To the best of our knowledge, ~80 kJ/mol is the
highest Qst for CO2 among all reported for porous polymers but it is still
lower than that of aqueous amines (usually > 100 kJ/mol).115
Site A, with
high Qst value, describes the chemisorption of CO2 and concomitant for-
mation of carbamic acid. Following the consumption of accessible primary
amines, the Qst was significantly reduced at higher CO2 loadings. The reduc-
tion in Qst was consistent with the CO2 isotherm showing lower gradients at
high relative pressures. Site B, with the lower Qst value, can be assigned to
somewhat weaker CO2-NH2 interactions such as hydrogen bonding of
O(CO2)…N(NH2) and/or physisorption of CO2 in the ultramicropores in
PP1-2-tren.66,116
Figure 3-5. Heat of adsorption of CO2 for PP1-2 (▲) and PP1-2-tren (●),
calculated using the dual-site Langmuir model. [From paper V]107
As mentioned earlier, the CO2-over-N2 selectivity is very important to a
potential adsorption-driven capture of CO2. Amine modification of the po-
rous polymer PP1-2 significantly increased its CO2 capacity and on the other
hand reduced its N2 uptake; hence, PP1-2-tren was expected to display a
high CO2-over-N2 selectivity. The selectivity was estimated from the single-
component adsorption data using the equation S = (qA/qB)/(pA/pB) (S: selec-
tivity; q: uptake; p: partial pressure; A, B: gas components).62,99
PP1-2-tren
had very high selectivities (456 and 1040) for CO2 in gas mixtures of CO2/N2
with compositions of 15/85 vol% and 5/95 vol%, respectively, at 273 K (Ta-
ble 3-1). The high selectivity of PP1-2-tren is consistent with the formation
of carbamic acids and may include an added kinetic contribution from sepa-
0.0 0.5 1.0 1.5 2.0
20
30
40
50
60
70
80
He
at
of a
dso
rptio
n (
kJ/m
ol)
CO2 loading (mmol/g)
50
ration in ultramicropores. The unmodified polymer PP1-2 had moderate
selectivities of 31 and 48, respectively, under the same conditions.
Table 3-1. Estimated CO2-over-N2 selectivities of PP1-2 and PP1-2-tren for
two CO2/N2 mixtures. [From paper V]107
Mate-
rial
CO2/N2 mixture (15/85 vol%) CO2/N2 mixture (5/95 vol%)
CO2 uptake
0.15 bar
(mmol/g)
N2 uptake
0.85 bar
(mmol/g)
S
CO2 uptake
0.05 bar
(mmol/g)
N2 uptake
0.95 bar
(mmol/g)
S
PP1-2 0.901 0.163 31 0.463 0.182 48
PP1-
2-tren 1.446 0.018 456 1.132 0.021 1035
To conclude, porous polymers with multimodal porosity are promising
substrates for amine-modification to target high efficiency in CO2 capture
and separation. The amine-modified polymer studied in Paper V demon-
strates high CO2 capacity and CO2-over-N2 selectivity at conditions relevant
to post-combustion CO2 capture. The high Qst value is beneficial for captur-
ing CO2 from flue gases with low CO2 concentrations and relevant for a
temperature-swing adsorption process. However, the high Qst value might
cause large energy penalty for regeneration of the sorbents. Hence, quantita-
tive modification of the polymers to optimize and balance the surface area,
pore size, heat of adsorption, uptakes and selectivities could be important
aspects for future studies.
51
4. Incorporation of Pd(II) species in porous polymers for heterogeneous catalysis
Palladium-catalyzed reactions play an important role in organic chemistry
and have led to the synthesis of a large amount of new organic molecules.
For example, the carboncarbon bond formation reactions catalyzed by pal-
ladium that were developed by Heck, Negishi and Suzuki (2010 Nobel Prize
winners in chemistry) made great contributions to synthetic organic chemis-
try.117
This chemistry has found important applications in the fine chemical
and pharmaceutical industries. Today, homogeneous catalysts including
Pd(II) salts and organopalladium are widely used in many synthetic process-
es because of their remarkable catalytic activity and selectivity.118
However,
such homogeneous catalysts usually need further separation and purification
steps, and thus produce extra waste and increase costs. In this manner, they
are not environmentally and eco-friendly. Heterogeneous catalysts, on the
other hand, allow easy separation steps but suffer from drawbacks of metal
leaching and comparably low catalytic activity and selectivity.119
Porous materials with large internal surface area and appropriate pore siz-
es are shown as ideal supports to load homogenous catalysts for the use of
heterogeneous catalysis.120
For example, activated carbons, silicas and MOFs
have been studied as supports for catalysts effecting various transfor-
mations.121-123
Their porosities allow for the immobilization of homogeneous
catalysts, efficient conversions and rapid mass transport. Homogeneous cata-
lysts can be immobilized by physisorption, complexation or by tethering
using chemical bonding. Physisorbed catalysts have a higher risk of leaching
out from the interconnected pore channels in the porous supports during the
reactions. For example, the important hydrogenation catalyst palladium on
carbon (Pd/C), suffers from weak interactions between the Pd particles and
the porous carbon, and consequently results in a leaching of Pd and poor
recyclability.124
In comparison, chemically tethered homogeneous catalysts
display better stability than physisorbed one. By means of chemical tethering
or complexation (coordination bonding), the catalysts can be incorporated
and stabilized by using the rich chemistry of several porous supports such as
MOFs, amine-functionalized siliceous mesocellular foam (MCF), and porous
polymers.75,121,125
Here, porous polymers offer unique advantages to tether
catalysts. The synthetic diversity of porous polymers is quite large. Specific
functional groups can be introduced in various manners that can be used to
52
bind metal catalysts.23
In the papers relating to this chapter, we developed
imine- and azo-linked porous polymers as supports to incorporate Pd(II)
species taking advantages of the coordination behaviors of imine and azo
groups. The porous polymer-Pd(II) compositions were good heterogeneous
catalysts for various organic reactions, as shown in Papers VI-VII.
4.1 The use of porous polyimine/Pd(II) in cooperatively catalyzed highly enantioselective cascade transformatio-ns
4.1.1 Synthesis of porous Pd(II)-polyimine (Pd2+
/PP-1)
Scheme 4-1, from Paper VI, shows the synthesis routes for PP-1 and
Pd2+
/PP-1. PP-1 was synthesized via a procedure for porous polymers as
described in section 2.1.5. Monomers of 1,3,5-tris(4-aminophenyl)benzene
(A3) and 1,3,5-benzenetricarboxaldehyde (B3) in a molar ratio of 1.5:1 were
used for the synthesis. The collected solid was dried at 200 °C under a N2
atmosphere to obtain PP-1.
To prepare Pd2+
/PP-1, a fine powder of PP-1 (100 mg) and Pd(OAc)2 (100
mg) was mixed in CH2Cl2 (12 ml) and stirred for 24 h under N2. The solid
was collected by centrifugation and further purified by Soxhlet extraction
with CH2Cl2 for 24 h. The obtained solid was subsequently dried under dy-
namic vacuum.
Scheme 4-1. Synthesis route to the porous polyimine PP-1, and the catalyst
Pd2+
/PP-1 (an idealized structure of PP-1 only). [From paper VI]126
53
4.1.2 Characterizations of PP-1 and Pd2+
/PP-1
PP-1 had a similar molecular structure as the related porous polymers pre-
sented in chapter 2. The imine bonds of PP-1 were detected by IR spectros-
copy (C=N stretching, 1669 cm–1
) and solid-state 13
C NMR spectroscopy
(13
C NMR chemical shift, 193 ppm) (Figure 4-1a-b), and the amount of Pd
loaded on Pd2+
/PP-1 was determined by inductively coupled plasma-optical
emission spectrometry (ICP-OES) to be ~2.3 wt%. A high-resolution XPS
(X-ray photoelectron spectroscopy) spectrum of Pd2+
/PP-1 shows the charac-
teristic doublet in the Pd 3d region. The peak at 337.9 eV corresponded to
Pd3d5/2 of Pd(II) (Figure 4-1c). Similar Pd3d5/2 XPS bands of Pd(II) were ob-
served in Pd/COF-LZU1 (337.7 eV) and a model complex of Pd/Phen (337.8
eV).127
In contrast, for free Pd(OAc)2 that band was detected at 338.4 eV.
Figure 4-1. (a) Solid-state 13
C{1H} NMR spectrum of the porous polyimine
(PP-1); spinning side bands denoted by (*); (b) IR spectrum of PP-1; (c)
high-resolution XPS spectrum (Pd 3d) of the catalyst Pd2+
/PP-1. [From paper
VI]126
54
As for most polymers in this thesis, the porosities of PP-1 and Pd2+
/PP-1
were probed by N2 sorption at 77 K. As shown in Figure 4-2, the steep up-
take at low relative pressures (P/P0 < 0.05) indicated that both materials were
(ultra)microporous. The large hysteresis between the adsorption and desorp-
tion isotherms corresponded to mesopores and/or swelling. The post-
synthetic modification of porous materials usually reduces their surface area
and pore volume because of the increased density and the pore blocking
effect. As expected, the surface area was reduced from 256 for PP-1 to 92
m2/g for Pd
2+/PP-1. According to the analyses of the PSDs, the (ul-
tra)micropore volume was significantly reduced upon Pd(OAc)2 loading.
Figure 4-2. Nitrogen adsorption (filled symbols) and desorption (open sym-
bols) isotherms of the porous polyimine PP-1 (■) and the catalyst Pd2+
/PP-1
(●) at 77 K. Inset shows PSDs of PP-1 (■) and Pd2+
/PP-1 (●) analyzed on the
N2 adsorption isotherms with DFT method. [From paper VI]126
4.1.3 Enantioselective cascade transformations catalyzed by
Pd2+
/PP-1/chiral amine
Asymmetric or enantioselective synthesis is a very important method to ob-
tain enantiopure compounds in the field of pharmaceutical industry and life
science.128
In this context, asymmetric catalysis is a subfield of asymmetric
synthesis. Here, only a catalytic amount of the catalyst is employed to trans-
form achiral materials into enantiopure compounds. Hence, the process is
more efficient and cost-effective than non-catalyzed processes. Furthermore,
sophisticated chemical reactions can be designed using so-called cascade
reactions, which allow the one-pot synthesis of complex molecules. In addi-
tion, cascade reactions allow the formation of carboncarbon and car-
bonheteroatom bonds, polycyclic structures and all-carbon quaternary cen-
55
ters, which are all important for the synthesis of fine chemicals and pharma-
ceuticals.129
Most cascade reactions are catalyzed by transition metals, but
organocatalysts offer an alternative way for cascade reactions to generate
multiple bonds in one-pot processes.130
Recently, the concept of synergistic
catalysis, which combines transition metal catalysts and organocatalysts, has
emerged as a powerful strategy in organic synthesis.131-133
Here, we show
that Pd2+
/PP-1 works in synergy with a chiral amine to give an excellent
catalyst for cooperatively catalyzed and enantioselective cascade reactions.
The catalytic performance of the synergetic catalysts of Pd2+
/PP-1/chiral
amine for the dynamic kinetic asymmetric transformations (DYKAT) of the
enals 1 and propargylic nucleophiles 2 was investigated in paper VI. Previ-
ous studies have shown that this reaction could give the corresponding car-
bocycles 4 an all-carbon stereocenters via Michael intermediates 3 when
supported by Pd2+
/chiral amine catalytic systems, as shown in Scheme 4-2.129
An initial screening of the conditions revealed that highly enantioselective
cascade reactions were achieved when performing the synthesis in toluene
with the co-catalysts of Pd2+
/PP-1/chiral amine. Cinnamic aldehyde 1a was
converted to 4a in yield (85%) with a diastereomeric ratio (dr) of 95:5 and
an enantiometric ratio (er) of 97.5:2.5. Both Pd2+
/PP-1 and the chiral amine 5
catalyst had to be present to form the corresponding product 4, demonstrat-
ing the cooperation of the two catalysts for the cascade sequence.
Scheme 4-2. Possible dynamic kinetic asymmetric transformations
(DYKAT) between the enals 1 and propargylic nucleophiles 2 by merging
Pd2+
/PP-1 and chiral amine catalysts. [From paper VI]126
The co-catalytic system of Pd2+
/PP-1/chiral amine was further investigated
for the dynamic cascade transformations of a set of enals 1 with propargylic
nucleophile into various carbocycles. These Michael–carbocyclization reac-
56
tions were highly enantioselective and the corresponding cyclopentenes 4
were isolated in high yields with up to 97:3 dr and 99.5:0.5 er, as shown in
Table 4-1. Remarkably, the electron-donating or electron-withdrawing
groups on the β-aryl-substituted of enals 1 did not affect the efficiency or
stereoselectivity of the reactions. In addition, the catalytic system Pd2+
/PP-
1/chiral amine catalyzed the conjugate/carbocyclization cascade transfor-
mation between a set of enals 1 and the 3-substituted oxindole 6 into the
spirocyclic oxindoles 7 with high enantioselectivity up to 99.5:0.5 er.
(Scheme 4-3) Note that the spirocyclic oxindoles have an all-carbon quater-
nary stereocenter, which is a valuable structural motif in several natural
products and pharmaceuticals. In comparison with Pd loaded on other sup-
ports such as amine-functionalized mesocellular foam and amine modified
silica,125,134
Pd2+
/PP-1 showed faster transformations and similar (or better)
stereoselectivity for same cascade reactions.
Scheme 4-3. Stereoselective cascade synthesis of spirocyclic oxindoles 7.
[From paper VI]126
Another important aspect in heterogeneous catalysis is the recovery of the
solid catalyst. The Pd2+
/PP-1 catalyst showed excellent recyclability with
respect to the yield of product 4a, and the enantioselectivity of the cascade
sequence was not significantly affected. (Table 4-2) Moreover, elemental
analysis of the filtrate obtained after hot filtration and completion of the re-
action showed very low Pd contents of 7.5 and 15.1 ppm, respectively,
whereas the background Pd content was 9 ppm. These results indicated no
significant Pd leaching was observed during the reactions.
57
Table 4-1. Scope of the reaction between enals and progargylic nucleophiles
promoted by the heterogeneous Pd2+
/PP-1/chiral amine catalyst. [From paper
VI]126
58
Table 4-2. Recycling of the heterogeneous Pd2+
/PP-1 in the DYKAT process
[From paper VI]126
4.2 A cyclopalladated microporous azo-linked polymer
as a heterogeneous catalyst for more eco-friendly Suzuki and Heck coupling reactions
Polymers containing azo groups (N=N) have been widely studied for their
interesting properties such as photo- and thermochemical induced cis-trans
isomerization, optical nonlinearities, and photo-controlled reversible proper-
ties.135,136
Various azo polymers were synthesized during the last decades,
including side-chain polymers, main-chain polymers, cross-linked polymers
and block copolymers.137
Recently, a series of microporous aromatic azo
polymers were synthesized. So far, azo-linked porous polymers have been
synthesized either via the homocoupling of aromatic amine (nitro) mono-
mers or the heterocoupling of aromatic amine and aromatic nitro monomers.
These polymers are promising CO2 sorbents because of the Lewis basic fea-
ture of azo group and their electron rich pore surfaces.63,138,139
On the other hand, aromatic azo compounds are active ligands to coordi-
nate or chemically bond many transition metals. For example, azobenzene can bind Pd(II) species in different coordination modes, as shown in Scheme 4-4. In general, azobenzene usually reacts with Pd(II) by ortho-palladation, forming a chloro-bridged dimer of cyclopalladated azobenzenes (Type I).
140
59
In the presence of neutral ligands such as DMF, DMSO and pyridine, a monocyclopalladated azobenzene is preferably formed (Type II).
141,142 The
cyclopalladated complexes (type I and II) are quite chemically stable even upon treatment with strong Lewis bases such as phosphine and amine be-cause of the strong PdC bonds.
140 The formation of dichloro-
bis(azobenzene)palladium with solely PdN coordination bonds (Type 3) is not common.
143 This complex is probably a kinetically stable product and
has been suggested as an intermediate during cyclopalladation of azoben-zene.
144 The interesting coordination chemistry of azobenzene/Pd(II) in-
spired us to develop aromatic azo-linked porous polymers loaded with Pd(II) for heterogeneous catalysis. In Paper VII, an aromatic azo-linked porous polymer was synthesized via the oxidative homocoupling of a tetramine monomer. The porous polymer was studied as a functional support able to form cyclopalladated polymer, which performed well for heterogeneous Suzuki and Heck coupling reactions.
Scheme 4-4. Coordination modes between azobenzene and Pd(II).
4.2.1 Synthesis
Aromatic azo compounds are conventionally synthesized by two steps from
anilines via diazonium salts or nitrobenzene intermediates using nitrite salts
or other oxidants.145,146
These oxidants are usually toxic and not environmen-
tally friendly. Zhang et al. reported a convenient method to synthesize azo
compounds under mild conditions using a copper(I) catalyst and air as an
oxidant.147
With a slightly modified procedure, Arab et al. prepared a series
of azo-linked porous polymers from various aromatic amines.63
In Paper VII,
we followed a similar procedure and synthesized a new azo-linked porous
polymer (PP-2), as shown in Scheme 4-4. An aromatic tetramine monomer,
of N,N,N',N'-tetrakis(4-aminophenyl)-1,4-benzenediamine (A4) was used as
the building block. CuBr and pyridine were used as catalysts and air was
used as an oxidant. Although toluene was the best solvent for the synthesis
60
of aromatic azo compounds with high yields in Zhang‟s study,147
the solvent
mixture of THF/toluene assured high monomer solubility and high
porosity.63
Specifically, 100 mg of A4 was dissolved in a THF/toluene (20
mL, vol:vol = 1:1), and CuBr (20 mg) and pyridine (90 mg) were subse-
quently added. The mixture was stirred under an air atmosphere at room
temperature for 24 h, and then at 60 °C for 24 h and finally at 80 °C for 24 h.
The obtained solid was soaked in HCl (1 M, 100 mL) for 24 h to remove the
CuBr, then filtered, and washed with NaOH (1 M, 200 mL) and a large
amount of water. After drying the solid in an oven at 100 °C for 24 h, PP-2
was isolated as fine brown powder (94 mg, yield: 96%).
Scheme 4-5. Synthesis route of azo-linked porous polymer PP-2 and the
catalyst Pd(II)/PP-2. [From paper VII]
The cyclopalladation of PP-2 was followed by a typical procedure for
post-modification of such porous polymers with metal salts. In detail, PP-2
(100 mg) and PdCl2 (100 mg) were mixed in CH2Cl2 (10 mL) and stirred for
24 h at room temperature under a N2 atmosphere. The solid in the suspension
was collected by filtration and washed by water several times to remove free
PdCl2. After the sample was dried under dynamic vacuum, the catalyst
Pd(II)/PP-2 was isolated as a black powder. Both PP-2 and Pd(II)/PP-2 were
insoluble in common solvents and insensitive to air and moisture. PP-2 was
even stable in aqueous acid and base solutions. They had high thermal stabil-
ity with decomposition temperatures of > 400 °C.
61
4.2.2 Characterizations
Figure 4-3. (a) IR and (b) Raman spectra of azo-linked porous polymer PP-2.
[From Paper VII]
The molecular details of PP-2 were investigated by IR, Raman, and 13
C
NMR spectroscopies and XPS. In the IR spectrum of PP-2 (Figure 4-3a), the
two main peaks at 1496 and 1591 cm–1
were assigned to the stretching vibra-
tions of the aromatic structure, and the strong peak at 1259 cm–1
was as-
signed to the CN stretching of the tertiary amine moieties. A small band
observed at 3353 cm–1
was assigned to the asymmetric NH stretching of
4000 3500 3000 2500 2000 1500 1000 500
400 600 800 1000 1200 1400 1600 1800
Tra
nsm
itta
nce
Wavenumber (cm-1)
(b)
Inte
nsity
Raman shift (cm-1)
(a)
62
residual amine groups of the A4 monomer. The weak shoulder band at fre-
quencies of ~14001450 cm–1
can be probably assigned to the stretching of
the trans-N=N linkage. In principle, the aromatic azo compounds display
stretching vibrations at ~14651380 cm–1
(weak) and ~1510 cm–1
(strong)
for N=N linkage in trans and cis configurations, respectively.143
However,
the bands of azo (N=N) linkages were not strong in the IR spectrum. The
richness of the IR spectra of the aromatic azo compounds or polymers in the
range of ~16001300 cm–1
makes it difficult to detect and assign the N=N
stretching vibrations. In comparison, Raman spectroscopy afforded more
precise information on the N=N linkages. As shown in Figure 4-3b, the
strong bands at frequencies of 1448 and 1402 cm–1
in the Raman spectrum
directly confirmed trans N=N linkages in PP-2.
PP-2 and Pd(II)/PP-2 had similar IR and Raman spectra; however, they
displayed quite different solid-state 13
C CPMAS NMR spectra, which in-
formed on chemical changes of PP-2 upon loading with Pd(II). In the 13
C
NMR spectrum of PP-2 as shown in Figure 4-4a, the band at ~150 ppm was
assigned the carbons linked to azo group (a) and the carbons attached to N
atoms from the monomer (d,e). The shoulder at ~120 ppm could be assigned
to the ortho carbons (b) of azobenzene moieties. The band at a chemical shift
of ~130 ppm was assigned to other aromatic carbons of PP-2. In comparison,
the shoulder ~120 ppm (b) in PP-2 was significantly shifted to a chemical
shift of ~158 ppm (b’) in Pd(II)/PP-2 (Figure 4-4b), which is a strong evi-
dence of formation of cyclopalladated azobenzene; this was probably a chlo-
ro-bridged dimer structure (type I mode) because no neutral ligands was
used during the synthesis.142,148
Elemental analysis of Pd(II)/PP-2 (Table 4-3)
indicated that the molecular ratio of Pd:Cl was approximately 1:1.3, whereas
a type I cyclopalladated azobenzene should have a 1:1 ratio of Pd:Cl. The
extra chloride content suggested that a minor amount of Pd(II) was coordi-
nated to the azo N atoms with a type III mode.
The support PP-2 could (mainly chemically) bind up to 20.2 wt% Pd. The
high loading of Pd can be attributed to the microporous structure of the sup-
port PP-2 and the formation of strong organometallic PdC bond and coor-
dinative PdN bond in Pd(II)/PP-2. Thus, when designing a heterogeneous
catalyst with a given Pd content, the high Pd loading capacity of PP-2 could
minimize the amount of the support required, which could limit costs (as
porous polymers are usually expensive.)
63
Figure 4-4. Solid-state 13
C NMR spectra with assignment of carbons in sup-
posed molecular structure of (a) PP-2 and (b) cyclopalladated Pd(II)/PP-2; (c)
the spectral differences between (a) and (b). [From Paper VII]
Table 4-3. Elemental composition of Pd(II)/PP-2 revealed by elemental
analysis using ICP-OES.
Atom C H N Cl Pd
Weight percentage (%) 45.04 3.05 8.88 9.21 20.2
250 200 150 100 50 0
N N
N
N N
N NN
N N
a'
b'
Pd(II)/PP-2
c'd'e'
f'
PdCl
ClPd
NN
N N
N
N N
N NN
N N
ab
cde
f
PP-2
13C NMR chemical shift (ppm)
a d e c f
b
b'
d' e'
a' c' f'(a)
(b)
(c)
64
The oxidation state of Pd was confirmed by X-ray photoelectron spectros-
copy (Figure 4-5). The XPS spectrum of Pd(II)/PP-2 exhibited two main
peaks at 343.4 and 338.2 eV, which corresponded to Pd3d3/2 and Pd3d5/2 of
Pd(II). In comparison, the binding energy of free PdCl2 was detected at
343.7 and 338.4 eV in the Pd 3d region.149
The negative shifted in the bind-
ing energy indicated strong coordination and chemical bonding of Pd(II)
with azobenzene groups in Pd(II)/PP-2.
Figure 4-5. High-resolution XPS spectrum (Pd 3d) of cyclopalladated
Pd(II)/PP-2. [From Paper VII]
The porosities of PP-2 and Pd(II)/PP-2 were probed by N2 sorption meas-
urement at 77 K. Both compositions showed typical sorption isotherms for
microporous materials with their steep adsorption of N2 at low relative pres-
sures (P/P0 < 0.05), as displayed in Figure 4-6a. The BET surface area of PP-
2 was reduced from 623 to 306 m2/g upon cyclopalladation. PSD analyses, in
Figure 4-6b, using the DFT model based upon the N2 adsorption isotherms
indicated that the ultramicropores in PP-2 (with size ~0.6 nm) vanished and
that the micropore volume was significantly reduced on cyclopalladation.
The reduced surface area, micropore volume and pore size can be explained
by increased density and pore blocking effect upon cyclopalladation of the
polymers.
350 345 340 335 330
Binding energy (ev)
65
Figure 4-6. (a) N2 sorption isotherms of PP-2 (▲) and cyclopalladated
Pd(II)/PP-2 (●), recorded at 77 K. (b) PSDs of PP-2 and Pd(II)/PP-2 calcu-
lated from the N2 adsorption isotherms using a DFT model. [From Paper VII]
0.5 1.0 1.5 2.0
0.0
0.5
1.0
0.0 0.2 0.4 0.6 0.8 1.0
0
100
200
300
400
(b)
Po
re v
olu
me
(cm
3/g
)
Pore size (nm)
(a)
Qu
an
tity
ad
so
rbe
d (
cm
3/g
)
P/P0
66
4.2.3 Application for heterogeneous Suzuki and Heck coupling
reactions
The high Pd loading and the microporous structure of the cyclopalladated
Pd(II)/PP-2 allowed us to study it as a heterogeneous catalyst. Its catalytic
activity was examined in representative Suzuki and Heck coupling reactions.
Suzuki coupling reactions have the advantage that they can be conducted in
both aqueous solution and organic solvents. Water is both a green and an
economical solvent. In addition, the high solubility of the required base and
boronic acids in water favors the reactions. However, many supports and
catalysts are sensitive to water. The cyclopalladated Pd(II)/PP-2, with its
high chemical stability, efficiently catalyzed various Suzuki couplings in
aqueous solution. The Suzuki coupling reactions between aryl halides 8 and
arylboronic acids 9 with either an electron-donating or an electron-
withdrawing group afforded the products 10 in excellent yields (90–98%).
0.5 mol% of the cyclopalladated catalyst (Pd(II)/PP-2) was used together
with the base K2CO3. Moreover, the use of bromobenzenes also gave the
corresponding products with excellent yields (90–95%, Table 4-4, entries 2,
4, 8). In addition, the chemical transformation also tolerated a hetero-
aromatic functionality, which is known to be challenging for Suzuki cou-
plings. (Table 4-4, entries 6 and 11) Such reactions can be used to introduce
heteroatoms and Cheteroatom bonds into compounds for specific applica-
tions.
The palladium-catalyzed Heck coupling reaction, which forms CC bonds,
is an important reaction in organic syntheses.150
The cyclopalladated
Pd(II)/PP-2 was also tested as a catalyst for the reaction between aryl halides
and styrenes. Investigations of the effect of solvent on the catalysis showed
that the reaction afforded higher yields in DMF than in other organic sol-
vents or water. Using conditions similar to those used for the Suzuki cou-
pling reactions (but with a longer reaction time of 16 h), a set of aryl halides
8 and styrenes 11 were coupled to the corresponding products 12 and 13
with total yields of 87–95%, as shown in Table 4-5. The more active iodo-
benzene gave slightly higher yields (95%, Table 4-5, entry 1) than bromo-
benzene (91%, Table 4-5, entry 2). The presence of electron-withdrawing
groups in the aryl halides appeared to decrease the yield (87%, Table 4-5,
entry 3). Correspondingly, the Heck reaction also tolerated hetero–aromatic
functionality giving high yielded product (Table 4-5, entry 5).
67
Table 4-4. Scope of the Suzuki coupling reaction catalyzed by cyclopalladat-
ed Pd(II)/PP-2. [From Paper VII]
68
Table 4-5. Scope of the Heck coupling reaction catalyzed by cyclopalladated
Pd(II)/PP-2. [From Paper VII]
The recyclability of Pd(II)/PP-2 in heterogeneous catalysis was examined
for the Suzuki coupling reaction of iodobenzene with phenylboronic acid.
The catalyst could be reused at least seven times without any significant loss
of catalytic activity, as shown in Table 4-6. Elemental analysis on the filtrate
indicated that no Pd content (~0.1 ppm) had leached into the aqueous solu-
tion. The excellent recyclability and absence of Pd leaching in aqueous solu-
tion further confirmed that the catalytic processes were essentially supported
by the cyclopalladated Pd(II)/PP-2 rather than free PdCl2.
69
Table 4-6. Recycle test of cyclopalladated Pd(II)/PP-2 in the Suzuki cou-
pling reaction between iodobenzene and phenylboronic acid. [From Paper
VII]
To conclude, porous polymers with high surface area, high physicochemi-
cal stability and rich coordination sites are ideal supports to load transition
metal species for heterogeneous catalysis. Pd(II) species were successfully
incorporated by coordination to the imine-linked (PP-1) and by cyclopallada-
tion to the azo-linked (PP-2) porous polymer. A set of polysubstituted cyclo-
pentenes and spirocyclic oxindoles, including the all-carbon quaternary ste-
reocenters, were synthesized in high yields with support of the catalytic sys-
tem of Pd2+
/PP-1/chiral amine. High diastereo- and enantioselectivities were
achieved for these DYKAT of enals with propargylic nucleophiles. In particular, the azo-linked porous polymer (PP-2) can accommodate up
to 20.2 wt% Pd, primarily by cyclopalladation to yield organometallic PdC bonds. Due to its microporous structure, high Pd loading, and strong chemi-cal interactions between Pd(II) species and the support, the cyclopalladated Pd(II)/PP-2 showed high efficiency and excellent recyclability in the cataly-sis of Suzuki and Heck coupling reactions. The use of water as a solvent for the Suzuki coupling reaction is economically and environmentally ad-vantages.
70
71
5. Conclusions
Imine- and azo-linked MOPs were synthesized and studied. Due to their moderately high surface area, tunable pore size, rich functionality, and high physicochemical stability, these MOPs were investigated as potential sorbents for CO2 capture and supports for heterogeneous catalysis.
A set of imine-linked MOPs were synthesized by polycondensations of various amine and aldehyde monomers in refluxing DMSO. Removing the eliminated water molecules during the synthesis was important to form the microporous structures. DMSO and its derivatives appeared to template the formation of (ultra)micropores. Using ketone monomers with pendant me-thyl and phenyl groups for the synthesis introduced hydrophobicity to the MOPs. In addition, performing off-stoichiometric synthesis conditions pro-duced polymers with ultramicropores, tunable mesopores and rich end groups. Separately, azo-linked MOPs were synthesized by oxidative cou-pling reaction of aromatic amines.
All these MOPs had relatively high CO2 adsorption capacities and CO2-over-N2 selectivities (up to 77 for mixture of CO2/N2 in 15/85 vol%) at con-ditions relevant to post-combustion CO2 capture because of their ultrami-cropores. The moderate Qst values (2540 kJ/mol at low coverages) indicat-ed that CO2 molecules were kinetically separated and captured in the MOPs mainly by means of physisorption. Tethering alkyl amine on the polymer by post-modification dramatically enhanced its capacity, selectivity (456 and 1040, respectively, for CO2/N2 mixtures with 15/85 vol% and 5/95 vol%) and Qst (~80 kJ/mol). The study of in situ IR and solid-state
13C NMR spec-
troscopy revealed chemisorption of CO2 in the amine-modified polymer with the formation of carbamic acid species. The high separation efficiency and high Qst value, which was dominated by chemisorption, of CO2 is beneficial for capturing CO2 from flue gases with low CO2 concentrations; however, it might incur a large energy penalty upon regenerating the sorbents. We ex-pect future studies of such porous polymers with quantitative modification to balance the separation efficiency of CO2 and energy combustion needed for reactivation of the sorbents.
Homogeneous catalysts of Pd(OAc)2 and PdCl2 were immobilized on the imine-linked porous polymer (PP-1) and chemically bonded by cylopallada-tion on the azo-linked porous polymer (PP-2), respectively. The Pd(II) was coordinated to the imine groups and cyclopalladated with the azobenzene species. In cooperation with a chiral amine, Pd
2+/PP-1 catalyzed DYKAT to
synthesize polysubstituted cyclopentenes and spirocyclic oxindoles with high diastereo- and enantioselectivities. Interestingly, the azo-linked MOP could be loaded with up to 20.2 wt% Pd due to the formation of stable cy-
72
clopalladated azobenzene species. The cyclopalladated Pd(II)/PP-2 efficient-ly catalyzed Suzuki and Heck coupling reactions, and their easy separation and reuse were both more economical and more environmentally friendly than the use of homogeneous Pd(II) catalysts. We expect that such metal-bound porous polymers could be widely used as catalysts or co-catalysts for various asymmetric transformations and coupling reactions involving CC and Cheteroatom bond formation.
73
Populärvetenskaplig sammanfattning
Microporösa organiska polymerer (MOPs) är en ny klass av porösa material
som består av kovalent sammanlänkade organiska byggstenar. Typiskt har
sådana polymerer stor specifik ytarea, reglerbar porstorlek och god
fysikalkemisk stabilitet. Deras egenskaper och kemi studeras generellt men
också med ett fokus på tillämpningar inom lagring och separation av
gasmolekyler samt inom heterogen katalys.
I den här avhandlingen behandlades specifikt syntes och tillämpningar av
olika imin- och azolänkade MOPs. De iminlänkade polymererna hade hög
porvolym med specifikt mycket små porer. Ultramikroporerna gav dessa
MOPs en stor förmåga till att ta upp koldioxid med en hög selektivitet från
kvävgasblandingar. När polymererna modifierades med amingrupper ökade
koldioxidaffiniteten. Hos dessa MOPs kunde koldioxid reagera kemiskt med
amingrupper genom en så kallad kemisorption. Båda typerna av MOPs är
relevanta för processer för ett borttag av koldioxid från rökgaser. Ett
storskaligt uppfång av koldioxid vid punktkällor är nödvändigt för
avskiljning och lagring av koldioxid (CCS). CCS anses av många forskare
och beslutsfattare som nödvändig för att effektivt minska de globala
utsläppen av klimatstörande gaser.
Genom katalys snabbas kemiska reaktioner upp genom en katalysator som
inte själv förbrukas. Sådana katalysatorer kan byggas in i MOPs och detta är
ett nytt sätt att utveckla heterogena katalysatorer. Heterogena katalysatorer
är typiskt fasta ämnen som snabbar på kemiska reaktioner i gas- eller
vätskefas. I den här avhandlingen byggdes palladium(II) joner in i MOPs
genom komplexbildning och/eller kovalenta bindningar. Dessa
palladiumrika MOPs studerades som heterogena katalysatorer för reaktioner
i vätskor. Specifikt studerades olika enantioselektiva reaktioner samt olika
kopplingsreaktioner. Katalysatorerna var generellt både effektiva och
specifika.
74
75
Acknowledgements
First of all, I would like thank my supervisor Prof. Niklas Hedin for all his
support during the years. Without your broad knowledge, timely supervision
and continuous encouragement, I could not imagine finishing my PhD. You
created a free atmosphere for me that made me really enjoy the research and
appreciate the time working with you. Also, I would like to thank my co-
supervisor, Prof. Lennart Bergström, for all the help and advice.
I want to thank Prof. Armando Córdova and his group at Mid Sweden
University for the nice collaboration in catalysis. Your suggestions and in-
puts to the projects were very useful and important. Also thanks Prof. Chris-
toph Langhammer and his PhD student Ferry at Chalmers University of
Technology for the collaboration on studying adsorption with new method. I
really appreciate to have collaboration with all of you. I hope we can keep
the connection in the future.
To other members in Niklas‟ group: Zoltan, thank you for your help with
adsorption and spectroscopy experiments, I have learned so much from you.
Samson, thank you for the nice collaboration and your hard working, you
helped me a lot with organic chemistry and catalysis studies. Wenming, you
have been so helpful when I started my life and study in Stockholm. Ocean,
thanks for the discussion and advice of adsorption experiments. Jordon, Li-
hong, Ermiase, you are great project students, thank you for your contribu-
tion to the projects and papers. Fredrik, Przemek, Kian, Tamara, Rouzbeh,
Eva, Panagiotis, Mikaela, Baroz, and Orlando, I have enjoyed working to-
gether with you. Special thanks to Tamara, Samson and Wenming for read-
ing and improving my thesis.
I would also like to thank many colleagues in MMK. Peng, you had very
nice TEM images for my polymers. Thank you Xiaodong for your useful
discussion and input to the paper. German, Valentina, Andreas, thank you
for your help to the SAXS experiment. Hong, Jie, Leifeng, you are very
good crystallographers, thanks for all the help and discussion in crystallog-
raphy. Mats, Jekabs, Lars, Feifei, Aatto, Alexander, James, Junliang, Wei,
Cheuk-wai, and Kjell, I have gained a lot of knowledge from your interest-
ing lectures.
I would like to give thanks to my former supervisor, Prof. Qian-Feng
Zhang at Anhui University and Technology. You introduced me to the area
of scientific research and helped me getting interested with it. Thank you for
76
all the help and encouragement. Also thanks to my colleagues there, Huatian,
Zhifeng, Aiquan, thanks for your nice collaborations.
During my PhD period, I have spent one month in Prof. Yuping Wu‟s
group at Fudan University, China, to study electrochemistry. Thank you for
your warm welcome and supports during my stay. It was a really nice expe-
rience to work with your group.
I am so grateful for all of the administrative staffs, Gunnar, Camilla, Dan-
iel, Tatiana, Helmi, Pia, Ann, Hanna, Ann-Britt, and Anita, you made my
study here very convenient.
To football players in MMK, Zoltan, Wenming, German, Arnaud,
Haoquan, Junzhong, Dickson, Yanhang, Leifeng, Guido, Changjiu, Ge,
Yunxiang, Julien, Farid, Ken, Qingxia, and other players on the pitch, we
had good time to play football together.
Also to other friends in Sweden, Guang, Yuan (Zhong), Yifeng, Duan, Jie,
Qingpeng, Fei, Yang, Ji, Yuan (Ning), Justin, Yuncheng, Yonglei, Shichao
Bin, Jianfeng, Yongsheng, Junzhan, Xiaohui, and Bo, you made my life in
Sweden so happy and interesting. To my friends in China, Qing, Zheng,
Biwen, and Qian, thank you for your overseas encouragement during the
years.
There are so many people I would like to express my thanks, but I cannot
list all of your names here. Thank you all for your kind help.
Finally, I would like to thank my parents, my wife Hongmei for your con-
tinuous love, supporting and understanding along the way, to my angel
daughter Ruixue for making our life so wonderful.
77
References
1. H. R. Allcock, F. W. Lampe, J. E. Mark and H. Allcock, Contemporary Polymer Chemistry, Pearson Education Upper Saddle River, NJ, 2003.
2. T. Graham, Philosophical Transactions of the Royal Society of London, 1861, 183-224.
3. R. Muelhaupt, Angewandte Chemie International Edition, 2004, 43, 1054-1063.
4. M. S. Silverstein, N. R. Cameron and M. A. Hillmyer, Porous Polymers, John Wiley & Sons, 2011.
5. D. Wu, F. Xu, B. Sun, R. Fu, H. He and K. Matyjaszewski, Chemical Reviews, 2012, 112, 3959-4015.
6. J.-X. Jiang, F. Su, A. Trewin, C. D. Wood, H. Niu, J. T. Jones, Y. Z. Khimyak and A. I. Cooper, Journal of the American Chemical Society, 2008, 130, 7710-7720.
7. H. M. El-Kaderi, J. R. Hunt, J. L. Mendoza-Cortés, A. P. Côté, R. E. Taylor, M. O'Keeffe and O. M. Yaghi, Science, 2007, 316, 268-272.
8. T. Ben, H. Ren, S. Ma, D. Cao, J. Lan, X. Jing, W. Wang, J. Xu, F. Deng, J. M. Simmons, S. Qiu and G. Zhu, Angewandte Chemie, 2009, 121, 9621-9624.
9. M. Rose, N. Klein, I. Senkovska, C. Schrage, P. Wollmann, W. Böhlmann, B. Böhringer, S. Fichtner and S. Kaskel, Journal of Materials Chemistry, 2011, 21, 711-716.
10. P. M. Budd, B. S. Ghanem, S. Makhseed, N. B. McKeown, K. J. Msayib and C. E. Tattershall, Chemical Communications, 2004, 230-231.
11. Y. Meng, D. Gu, F. Zhang, Y. Shi, H. Yang, Z. Li, C. Yu, B. Tu and D. Zhao, Angewandte Chemie, 2005, 117, 7215-7221.
12. D. A. Olson, L. Chen and M. A. Hillmyer, Chemistry of Materials, 2007, 20, 869-890.
13. J.-X. Jiang and A. I. Cooper, in Functional Metal-Organic Frameworks: Gas Storage, Separation and Catalysis, Springer, 2010, pp. 1-33.
14. Y. Luo, B. Li, W. Wang, K. Wu and B. Tan, Advanced Materials, 2012, 24, 5703-5707.
15. A. I. Cooper, Advanced Materials, 2009, 21, 1291-1295. 16. N. B. McKeown and P. M. Budd, Chemical Society Reviews, 2006,
35, 675-683. 17. P. Kuhn, M. Antonietti and A. Thomas, Angewandte Chemie
International Edition, 2008, 47, 3450-3453.
78
18. Y. Xu, S. Jin, H. Xu, A. Nagai and D. Jiang, Chemical Society Reviews, 2013, 42, 8012-8031.
19. M. X. Tan, Y. Zhang and J. Y. Ying, ChemSusChem, 2013, 6, 1186-1190.
20. T. Masuda, E. Isobe, T. Higashimura and K. Takada, Journal of the American Chemical Society, 1983, 105, 7473-7474.
21. M. Carta, R. Malpass-Evans, M. Croad, Y. Rogan, J. C. Jansen, P. Bernardo, F. Bazzarelli and N. B. McKeown, Science, 2013, 339, 303-307.
22. D. Yuan, W. Lu, D. Zhao and H. C. Zhou, Advanced Materials, 2011, 23, 3723-3725.
23. R. Dawson, A. I. Cooper and D. J. Adams, Progress in Polymer Science, 2012, 37, 530-563.
24. K. V. Rao, R. Haldar, C. Kulkarni, T. K. Maji and S. J. George, Chemistry of Materials, 2012, 24, 969-971.
25. R. Dawson, F. Su, H. Niu, C. D. Wood, J. T. Jones, Y. Z. Khimyak and A. I. Cooper, Macromolecules, 2008, 41, 1591-1593.
26. R. Dawson, A. I. Cooper and D. J. Adams, Polymer International, 2013, 62, 345-352.
27. Y. Xu, L. Chen, Z. Guo, A. Nagai and D. Jiang, Journal of the American Chemical Society, 2011, 133, 17622-17625.
28. P. Pandey, A. P. Katsoulidis, I. Eryazici, Y. Wu, M. G. Kanatzidis and S. T. Nguyen, Chemistry of Materials, 2010, 22, 4974-4979.
29. A. P. Cote, A. I. Benin, N. W. Ockwig, M. O'Keeffe, A. J. Matzger and O. M. Yaghi, Science, 2005, 310, 1166-1170.
30. S. J. Rowan, S. J. Cantrill, G. R. Cousins, J. K. Sanders and J. F. Stoddart, Angewandte Chemie International Edition, 2002, 41, 898-952.
31. M. E. Belowich and J. F. Stoddart, Chemical Society Reviews, 2012, 41, 2003-2024.
32. X. Feng, X. Ding and D. Jiang, Chemical Society Reviews, 2012, 41, 6010-6022.
33. A. P. Cote, H. M. El-Kaderi, H. Furukawa, J. R. Hunt and O. M. Yaghi, Journal of the American Chemical Society, 2007, 129, 12914-12915.
34. F. J. Uribe-Romo, J. R. Hunt, H. Furukawa, C. Kl ck, M. O‟Keeffe and O. M. Yaghi, Journal of the American Chemical Society, 2009, 131, 4570-4571.
35. Y.- . Zhang, J. Su, H. Furukawa, Y. Yun, F. G ndara, . Duong, X. Zou and O. M. Yaghi, Journal of the American Chemical Society, 2013, 135, 16336-16339.
36. F. Vilela, K. Zhang and M. Antonietti, Energy & Environmental Science, 2012, 5, 7819-7832.
37. L. Sun, Z. Liang, J. Yu and R. Xu, Polymer Chemistry, 2013, 4, 1932-1938.
38. J.-X. Jiang, F. Su, H. Niu, C. D. Wood, N. L. Campbell, Y. Z. Khimyak and A. I. Cooper, Chem. Commun., 2008, 486-488.
79
39. L. Chen, Y. Honsho, S. Seki and D. Jiang, Journal of the American Chemical Society, 2010, 132, 6742-6748.
40. S. Yuan, B. Dorney, D. White, S. Kirklin, P. Zapol, L. Yu and D.-J. Liu, Chemical Communications, 2010, 46, 4547-4549.
41. S. Ren, M. J. Bojdys, R. Dawson, A. Laybourn, Y. Z. Khimyak, D. J. Adams and A. I. Cooper, Advanced Materials, 2012, 24, 2357-2361.
42. P. Kuhn, A. Forget, D. Su, A. Thomas and M. Antonietti, Journal of the American Chemical Society, 2008, 130, 13333-13337.
43. X. Zhu, C. Tian, S. M. Mahurin, S.-H. Chai, C. Wang, S. Brown, G. M. Veith, H. Luo, H. Liu and S. Dai, Journal of the American Chemical Society, 2012, 134, 10478-10484.
44. J. R. Holst, E. St ckel, D. J. dams and . I. Cooper, Macromolecules, 2010, 43, 8531-8538.
45. C. Xu and N. Hedin, Materials Today, 2014, 17, 397-403. 46. V. Davankov and M. Tsyurupa, Comprehensive Analytical
Chemistry, 2011, 56, 315-358. 47. M. Tsyurupa and V. Davankov, Reactive and Functional Polymers,
2006, 66, 768-779. 48. S. Xu, Y. Luo and B. Tan, Macromolecular Rapid Communications,
2013, 34, 471-484. 49. C. D. Wood, B. Tan, A. Trewin, H. Niu, D. Bradshaw, M. J.
Rosseinsky, Y. Z. Khimyak, N. L. Campbell, R. Kirk and E. Stöckel, Chemistry of Materials, 2007, 19, 2034-2048.
50. G. Li and Z. Wang, Macromolecules, 2013, 46, 3058-3066. 51. Y. Zhu, H. Long and W. Zhang, Chemistry of Materials, 2013, 25,
1630-1635. 52. M. G. Schwab, B. Fassbender, H. W. Spiess, A. Thomas, X. Feng
and K. Mullen, Journal of the American Chemical Society, 2009, 131, 7216-7217.
53. M. G. Rabbani and H. M. El-Kaderi, Chemistry of Materials, 2012, 24, 1511-1517.
54. J. Weber, Q. Su, M. Antonietti and A. Thomas, Macromolecular Rapid Communications, 2007, 28, 1871-1876.
55. H. Zhao, Z. Jin, H. Su, J. Zhang, X. Yao, H. Zhao and G. Zhu, Chemical Communications, 2013, 49, 2780-2782.
56. Z. Chang, D.-S. Zhang, Q. Chen and X.-H. Bu, Physical Chemistry Chemical Physics, 2013, 15, 5430-5442.
57. S. Chu and A. Majumdar, Nature, 2012, 488, 294-303. 58. A. A. Olajire, Energy, 2010, 35, 2610-2628. 59. N. Hedin, L. Andersson, L. Bergström and J. Yan, Applied Energy,
2013, 104, 418-433. 60. A. P. Katsoulidis and M. G. Kanatzidis, Chemistry of Materials,
2011, 23, 1818-1824. 61. A. Modak, M. Nandi, J. Mondal and A. Bhaumik, Chemical
Communications, 2012, 48, 248-250.
80
62. W. Lu, J. P. Sculley, D. Yuan, R. Krishna, Z. Wei and H. C. Zhou, Angewandte Chemie International Edition, 2012, 51, 7480-7484.
63. P. rab, M. G. Rabbani, . K. Sekizkardes, T. slamo lu and H. M. El-Kaderi, Chemistry of Materials, 2014, 26, 1385-1392.
64. D. M. D'Alessandro, B. Smit and J. R. Long, Angewandte Chemie International Edition, 2010, 49, 6058-6082.
65. R. Dawson, E. Stöckel, J. R. Holst, D. J. Adams and A. I. Cooper, Energy & Environmental Science, 2011, 4, 4239-4245.
66. C. Xu and N. Hedin, Journal of Materials Chemistry A, 2013, 1, 3406-3414.
67. T. Ben, C. Pei, D. Zhang, J. Xu, F. Deng, X. Jing and S. Qiu, Energy & Environmental Science, 2011, 4, 3991-3999.
68. S. Sircar, Industrial & Engineering Chemistry Research, 2006, 45, 5435-5448.
69. D. Ko, H. A. Patel and C. T. Yavuz, Chemical Communications, 2015, 51, 2915-2917.
70. N.-R. Chiou, L. J. Lee and A. J. Epstein, Journal of Materials Chemistry, 2008, 18, 2085-2089.
71. N. Hiyoshi, K. Yogo and T. Yashima, Microporous and Mesoporous Materials, 2005, 84, 357-365.
72. . Demessence, D. M. D‟ lessandro, M. L. Foo and J. R. Long, Journal of the American Chemical Society, 2009, 131, 8784-8786.
73. J. Wang, M. Wang, B. Zhao, W. Qiao, D. Long and L. Ling, Industrial & Engineering Chemistry Research, 2013, 52, 5437-5444.
74. Z. Bacsik, N. Ahlsten, A. Ziadi, G. Zhao, A. E. Garcia-Bennett, B. Martín-Matute and N. Hedin, Langmuir, 2011, 27, 11118-11128.
75. Y. Zhang and S. N. Riduan, Chemical Society Reviews, 2012, 41, 2083-2094.
76. M. Rose, ChemCatChem, 2014, 6, 1166-1182. 77. P. Kaur, J. T. Hupp and S. T. Nguyen, ACS Catalysis, 2011, 1, 819-
835. 78. E. Verde-Sesto, E. M. Maya, Á. E. Lozano, G. Jose, F. Sánchez and
M. Iglesias, Journal of Materials Chemistry, 2012, 22, 24637-24643.
79. R. Palkovits, M. Antonietti, P. Kuhn, A. Thomas and F. Schüth, Angewandte Chemie International Edition, 2009, 48, 6909-6912.
80. L. Chen, Y. Yang and D. Jiang, Journal of the American Chemical Society, 2010, 132, 9138-9143.
81. J. X. Jiang, C. Wang, A. Laybourn, T. Hasell, R. Clowes, Y. Z. Khimyak, J. Xiao, S. J. Higgins, D. J. Adams and A. I. Cooper, Angewandte Chemie International Edition, 2011, 50, 1072-1075.
82. P. Zhang, Z. Weng, J. Guo and C. Wang, Chemistry of Materials, 2011, 23, 5243-5249.
83. C. E. Chan-Thaw, A. Villa, P. Katekomol, D. Su, A. Thomas and L. Prati, Nano Letters, 2010, 10, 537-541.
84. J. Yan, Z. Fan, T. Wei, W. Qian, M. Zhang and F. Wei, Carbon, 2010, 48, 3825-3833.
81
85. E. Frackowiak and F. Beguin, Carbon, 2002, 40, 1775-1787. 86. J. Li, X. Wang, Q. Huang, S. Gamboa and P. Sebastian, Journal of
Power Sources, 2006, 158, 784-788. 87. Y. Kou, Y. Xu, Z. Guo and D. Jiang, Angewandte Chemie, 2011,
123, 8912-8916. 88. L. Hao, J. Ning, B. Luo, B. Wang, Y. Zhang, Z. Tang, J. Yang, A.
Thomas and L. Zhi, Journal of the American Chemical Society, 2015, 137, 219-225.
89. B. Kiskan, M. Antonietti and J. Weber, Macromolecules, 2012, 45, 1356-1361.
90. Z. Xiang and D. Cao, Macromolecular Rapid Communications, 2012, 33, 1184-1190.
91. R. S. Sprick, J.-X. Jiang, B. Bonillo, S. Ren, T. Ratvijitvech, P. Guiglion, M. A. Zwijnenburg, D. J. Adams and A. I. Cooper, Journal of the American Chemical Society, 2015, 137, 3265-3270.
92. C. Xu, E. Dinka and N. Hedin, ChemPlusChem, 2015, DOI: 10.1002/cplu.201500344.
93. C. Xu and N. Hedin, Microporous and Mesoporous Materials, accepted.
94. J. Schmidt, D. S. Kundu, S. Blechert and A. Thomas, Chemical Communications, 2014, 50, 3347-3349.
95. G. Li, B. Zhang, J. Yan and Z. Wang, Journal of Materials Chemistry A, 2014, 2, 18881-18888.
96. R. Ryoo, S. H. Joo, M. Kruk and M. Jaroniec, Advanced Materials, 2001, 13, 677-681.
97. D. Zhao, J. Feng, Q. Huo, N. Melosh, G. H. Fredrickson, B. F. Chmelka and G. D. Stucky, Science, 1998, 279, 548-552.
98. J. Jagiello and M. Thommes, Carbon, 2004, 42, 1227-1232. 99. A. Myers and J. M. Prausnitz, AIChE Journal, 1965, 11, 121-127. 100. J. A. Mason, K. Sumida, Z. R. Herm, R. Krishna and J. R. Long,
Energy & Environmental Science, 2011, 4, 3030-3040. 101. R. S. Haszeldine, Science, 2009, 325, 1647-1652. 102. G. T. Rochelle, Science, 2009, 325, 1652-1654. 103. L.-C. Lin, A. H. Berger, R. L. Martin, J. Kim, J. A. Swisher, K.
Jariwala, C. H. Rycroft, A. S. Bhown, M. W. Deem and M. Haranczyk, Nature Materials, 2012, 11, 633-641.
104. D. M. Ruthven, Chemical Engineering Science, 1992, 47, 4305-4308.
105. M. F. Hasan, E. L. First and C. A. Floudas, Physical Chemistry Chemical Physics, 2013, 15, 17601-17618.
106. J.-H. Kim, C.-H. Lee, W.-S. Kim, J.-S. Lee, J.-T. Kim, J.-K. Suh and J.-M. Lee, Journal of Chemical & Engineering Data, 2003, 48, 137-141.
107. C. Xu, Z. Bacsik and N. Hedin, Journal of Materials Chemistry A, 2015, 3, 16229-16234.
108. J. Workman Jr and L. Weyer, Practical Guide to Interpretive Near-Infrared Spectroscopy, CRC Press, 2007.
82
109. A. Danon, P. C. Stair and E. Weitz, The Journal of Physical Chemistry C, 2011, 115, 11540-11549.
110. C. Kn fel, C. Martin, V. Hornebecq and P. L. Llewellyn, The Journal of Physical Chemistry C, 2009, 113, 21726-21734.
111. J. Tanthana and S. S. Chuang, ChemSusChem, 2010, 3, 957-964. 112. V. Guillerm, Ł. J. Weseliński, M. lkordi, M. I. H. Mohideen, Y.
Belmabkhout, A. J. Cairns and M. Eddaoudi, Chemical Communications, 2014, 50, 1937-1940.
113. Z. Wang and S. M. Cohen, Chemical Society Reviews, 2009, 38, 1315-1329.
114. D.-X. Xue, A. J. Cairns, Y. Belmabkhout, L. Wojtas, Y. Liu, M. H. Alkordi and M. Eddaoudi, Journal of the American Chemical Society, 2013, 135, 7660-7667.
115. Y. Belmabkhout and A. Sayari, Adsorption, 2009, 15, 318-328. 116. R. Vaidhyanathan, S. S. Iremonger, G. K. Shimizu, P. G. Boyd, S.
Alavi and T. K. Woo, Science, 2010, 330, 650-653. 117. J. Bäckvall, The Royal Swedish Academy of Sciences, 2010. 118. A. Zapf and M. Beller, Topics in Catalysis, 2002, 19, 101-109. 119. N. T. Phan, M. Van Der Sluys and C. W. Jones, Advanced Synthesis
& Catalysis, 2006, 348, 609-679. 120. X. Zhao, X. Y. Bao, W. Guo and F. Y. Lee, Materials Today, 2006,
9, 32-39. 121. J. Lee, O. K. Farha, J. Roberts, K. A. Scheidt, S. T. Nguyen and J. T.
Hupp, Chemical Society Reviews, 2009, 38, 1450-1459. 122. F. Rodríguez-Reinoso, Carbon, 1998, 36, 159-175. 123. V. Polshettiwar, C. Len and A. Fihri, Coordination Chemistry
Reviews, 2009, 253, 2599-2626. 124. N. Erathodiyil, S. Ooi, A. M. Seayad, Y. Han, S. S. Lee and J. Y.
Ying, Chemistry-A European Journal, 2008, 14, 3118-3125. 125. L. Deiana, L. Ghisu, S. Afewerki, O. Verho, E. V. Johnston, N.
Hedin, Z. Bacsik and A. Córdova, Advanced Synthesis & Catalysis, 2014, 356, 2485-2492.
126. C. Xu, L. Deiana, S. Afewerki, C. Incerti‐Pradillos, O. Córdova, P. Guo, A. Córdova and N. Hedin, Advanced Synthesis & Catalysis, 2015, 357, 2150-2156.
127. S.-Y. Ding, J. Gao, Q. Wang, Y. Zhang, W.-G. Song, C.-Y. Su and W. Wang, Journal of the American Chemical Society, 2011, 133, 19816-19822.
128. S. Afewerki, G. Ma, I. Ibrahem, L. Liu, J. Sun and A. Córdova, ACS Catalysis, 2015, 5, 1266-1272.
129. S. Santoro, L. Deiana, G.-L. Zhao, S. Lin, F. Himo and A. Córdova, ACS Catalysis, 2014, 4, 4474-4484.
130. D. Enders, C. Grondal and M. R. Huettl, Angewandte Chemie International Edition, 2007, 46, 1570-1581.
131. Z. Du and Z. Shao, Chemical Society Reviews, 2013, 42, 1337-1378. 132. Z. Shao and H. Zhang, Chemical Society Reviews, 2009, 38, 2745-
2755.
83
133. A. E. Allen and D. W. MacMillan, Chemical Science, 2012, 3, 633-658.
134. L. Deiana, S. Afewerki, C. Palo-Nieto, O. Verho, E. V. Johnston and A. Córdova, Scientific Reports, 2012, 2, 851.
135. A. Natansohn and P. Rochon, Chemical Reviews, 2002, 102, 4139-4176.
136. S. Xie, A. Natansohn and P. Rochon, Chemistry of Materials, 1993, 5, 403-411.
137. D. Wang and X. Wang, Progress in Polymer Science, 2013, 38, 271-301.
138. H. A. Patel, S. H. Je, J. Park, D. P. Chen, Y. Jung, C. T. Yavuz and A. Coskun, Nature Communications, 2013, 4, 1357.
139. J. Lu and J. Zhang, Journal of Materials Chemistry A, 2014, 2, 13831-13834.
140. A. C. Cope and R. W. Siekman, Journal of the American Chemical Society, 1965, 87, 3272-3273.
141. D. abic, M. Curic, K. i. Mol anov, G. Ilc and J. Plavec, Inorganic Chemistry, 2008, 47, 10446-10454.
142. M. Curic, D. abic, . Višnjevac and K. Molcanov, Inorganic Chemistry, 2005, 44, 5975-5977.
143. A. L. Balch and D. Petridis, Inorganic Chemistry, 1969, 8, 2247-2252.
144. G. Khare, R. Little, J. T. Veal and R. J. Doedens, Inorganic Chemistry, 1975, 14, 2475-2479.
145. H. A. Dabbagh, A. Teimouri and A. N. Chermahini, Dyes and Pigments, 2007, 73, 239-244.
146. K. Krageloh, G. H. Anderson and P. J. Stang, Journal of the American Chemical Society, 1984, 106, 6015-6021.
147. C. Zhang and N. Jiao, Angewandte Chemie, 2010, 122, 6310-6313. 148. M. Ćurić, D. abić, Ž. Marinić, L. Paša-Tolić, V. utković, J.
Plavec and L. Tušek- ožić, Journal of Organometallic Chemistry, 2003, 687, 85-99.
149. M. C. Militello and S. J. Simko, Surface Science Spectra, 1994, 3, 402-409.
150. X. F. Wu, P. Anbarasan, H. Neumann and M. Beller, Angewandte Chemie International Edition, 2010, 49, 9047-9050.