Research Collection

Doctoral Thesis

Synthesis of phosphine-functionalized metal-organic

frameworks and their applications in metal-supported catalysis

Author(s):
Morel, Flavien Lucas

Publication Date:
2016

Permanent Link:
https://doi.org/10.3929/ethz-a-010652144

Rights / License:
In Copyright - Non-Commercial Use Permitted

This page was generated automatically upon download from the ETH Zurich Research Collection. For more
information please consult the Terms of use.

ETH Library
 DISS. ETH NO. 23254

Synthesis of phosphine-functionalized metal-organic

frameworks and their applications
in metal-supported catalysis

A thesis submitted to attain the degree of

DOCTOR OF SCIENCES of ETH ZURICH

(Dr. sc. ETH Zurich)

presented by

FLAVIEN LUCAS MOREL

M. Sc. in Chemical & Biochemical Engineering, EPFL

born on 26.07.1987

citizen of l’Abergement (VD)

accepted on the recommendation of

Prof. Dr. Jeroen A. van Bokhoven, examiner

Prof. Dr. Antonio Togni, co-examiner
Dr. Marco Ranocchiari, co-examiner

2016
“If I have a thousand ideas and only one turns out to be good, I am satisfied.”

 Alfred Nobel
Table of content
Abstract ............................................................................................................................. I
Résumé ............................................................................................................................ III
Chapter 1: Introduction ................................................................................................... 1
1.1 Catalysis .............................................................................................................................. 2
1.2 Metal-organic frameworks: fundamental aspects ................................................................. 4
1.3 Synthesis ............................................................................................................................. 9
1.4 Post-synthetic functionalization of MOFs ............................................................................. 9
1.5 Applications of MOFs ......................................................................................................... 15
1.6 Organophosphine functional groups in MOFs .................................................................... 17
1.7 Rationale of the work ......................................................................................................... 20
Chapter 2: Methods ....................................................................................................... 23
2.1 Introduction ........................................................................................................................ 24
2.2 Synthesis ........................................................................................................................... 24
2.3 Post-synthetic modifications .............................................................................................. 35
2.4 Catalysis ............................................................................................................................ 37
2.5 Characterization................................................................................................................. 40
Chapter 3: Synthesis and characterization of P-MOFs .............................................. 51
3.1 Introduction ........................................................................................................................ 52
3.2 Design of novel P-linkers ................................................................................................... 53
3.3 Synthesis of P-MOFs with MOF-5 topology ....................................................................... 57
3.4 Synthesis of P-MOFs with UMCM-1 topology .................................................................... 60
3.5 Synthesis of P-MOFs with MIL-101 topology ..................................................................... 62
3.6 Conclusion ......................................................................................................................... 67
Chapter 4: Characterization of P-MOFs by XAS and MAS NMR spectroscopy ....... 69
4.1 Introduction ........................................................................................................................ 70
4.2 Determination of the P/P=O ratio ....................................................................................... 71
4.3 Characterization of metal-functionalized P-MOFs .............................................................. 77
4.4 Conclusion ......................................................................................................................... 86
Chapter 5: Decomposition of formic acid over a ruthenium-functionalized P-MOF 87
5.1 Introduction ........................................................................................................................ 88
5.2 Preparation of the catalyst ................................................................................................. 89
5.3 Catalytic decomposition of formic acid ............................................................................... 90
5.4 Conclusion ......................................................................................................................... 97

I
Chapter 6: Supported Grubbs’ catalysts for liquid and gas phase olefin
metathesis ..................................................................................................................... 99
6.1 Introduction ......................................................................................................................100
6.2 Immobilization strategy .....................................................................................................101
6.3 Characterization of Ru-LSK-15b .......................................................................................102
6.4 Self-metathesis of terminal alkenes ..................................................................................105
6.5 Cross-metathesis of functionalized olefins........................................................................107
6.6 Conclusion .......................................................................................................................110
Chapter 7: Rh-catalyzed asymmetric hydrogenation of olefins in the presence of
P-MOFs ........................................................................................................................ 111
7.1 Introduction ......................................................................................................................112
7.2 Hydrogenation of acrolein with Rh/PPh3-LSK-3 ................................................................113
7.3 Zinc P-MOFs and chiral monodentate phosphines as ligands for the asymmetric
hydrogenation of olefins .........................................................................................................113
7.4 Hydrogenation in the presence of amino-functionalized MIL-101 frameworks ..................117
7.5 Conclusion .......................................................................................................................118
Chapter 8: Conclusion and Outlook .......................................................................... 119
8.1 Conclusion .......................................................................................................................120
8.2 Outlook .............................................................................................................................121
Publications................................................................................................................. 127
Curriculum Vitae ................................................................. Error! Bookmark not defined.
Acknowledgments ...................................................................................................... 131
Appendix........................................................................................................................ A-I
Table of abbreviations ............................................................................................................ A-II
References ............................................................................................................................ A-III

II
Abstract
Metal-organic frameworks (MOFs) are porous and crystalline coordination polymers formed
upon self-assembly of organic linkers and inorganic clusters. They are distinguished from
traditional inorganic materials, such as silica and zeolites, by their structural and chemical
versatility—attributes that originate from the large variety of available building units.
Frameworks can thus be designed to accommodate functional groups and organometallic
complexes, forming pockets with immobilized catalytic functions with a high degree of order
and diffusivity.

Organometallic complexes are isolated metal centers surrounded by organic ligands. They
are frequently used in homogeneous catalysis due to their well-defined structures and
appreciable molecular dispersion, providing high activity and selectivity in a wide array of
organic transformations. Notorious rhodium and ruthenium catalysts coordinated by one or
more phosphine ligands use the steric and basic properties of the ligand to fine-tune the
electronic properties and spatial configuration of the metal center during catalysis.
Phosphine-functionalized MOFs (P-MOFs) combine the propitious characteristics of
phosphine groups with the crystalline and porous nature of MOF supports. These materials
possess their own inherent challenges: their development is restricted by the accessibility of
suitable organic linkers, which can require complicated multi-step synthetic procedures, and
P-MOFs are sometimes unstable.

Chapter 1 reviews the recent developments of functionalized MOFs to incorporate molecular

catalysts. The current strategy of choice is based on amino-functionalized frameworks that
often require post-synthetic modification to generate bidentate ligands capable of retaining
organometallic complexes. In contrast, phosphine-MOFs (P-MOFs) benefit from the
increased binding properties of the phosphine groups towards transition metals, allowing their
direct metalation. However, the state-of-the-art in phosphine-functionalized linker synthesis
allows for a relatively restricted number of P-MOFs with limited stability and metal-binding
capabilities.

Chapter 2 describes the experimental procedures for the synthesis and characterization of
the organic linkers, MOF structures and organometallic catalysts used throughout this work.
Characterization was performed using X-ray diffraction (XRD), X-ray absorption spectroscopy
(XAS), nuclear magnetic resonance (NMR) spectroscopy, and IR spectroscopy, as well as
gravimetric techniques, such as nitrogen physisorption, thermogravimetric analysis, and
elemental analysis. A short technical description of the techniques is also presented.

III
In Chapter 3, the synthesis of P-MOFs based on phosphine-functionalized terephthalic acid is
described. A route to a 2-diphenylphosphino-benzene-1,4-dicarboxylic acid (PPh2-bdc) is
reported, using microwave-assisted P-C coupling and silane reduction as the key synthetic
steps. PPh2-bdc was incorporated into several MOFs using a mixed linker approach to
prevent pore crowding. The PPh2-bdc linker was compatible with a variety organic linkers, as
well as metal ions, including zinc(II), chromium(III), and aluminum(III). Several MOF
topologies were targeted, such as MOF-5, MIL-101 and UMCM-1. All materials were
characterized using a combination of X-ray diffraction, nitrogen physisorption and NMR
spectroscopy, to confirm their crystallinity, and permanent porosity. The phosphorus content
was tuned by varying the linker ratio without significant modification of the synthetic
procedures, providing a versatile tool for the future optimization of P-MOFs.

Chapter 4 describes the application of element-specific methods—such as magic-angle

(MAS) NMR spectroscopy, X-ray absorption spectroscopy (XAS), and energy-dispersive
X-ray (EDX) spectroscopy—to characterize phosphine groups in solid samples. These
techniques yield chemical information that is otherwise lost when applying traditional
characterization methods that require digestion of P-MOF samples. A combination of XAS
and NMR spectroscopy represents a powerful methodology for the characterization of the
local coordination shell of organometallic complexes located within P-MOFs. This versatile
approach allows determination of the electronic state of each ligand, thereby validating
impregnation strategies from the perspective of both the support and the metal center. In
parallel, EDX spectroscopy may be used to provide an elemental overview that can be
combined with XAS to generate chemical imaging of phosphine groups within a single crystal.

Chapter 5 describes the synthesis of a ruthenium catalyst immobilized in LSK-15 for the
selective decomposition of formic acid in the gas-phase. Post-synthetic metalation of LSK-15
with [RuCl2(p-cym)]2 provided highly active phosphine-ruthenium molecular centers of the
form [RuCl2(p-cym)(PPh2-MOF)] (Ru-LSK-15a). The support and the immobilized catalyst
were characterized by 31P and 13C MAS NMR spectroscopy, as well as Cl and Ru K-edges
XAS. Ru-LSK-15a was tested in various conditions and consistently displayed high activity
(TOF = 2700 h-1 at 145 °C), stability (TON > 1,290,000 over a period of 23 days), and
exclusive selectivity towards hydrogen and carbon dioxide. LSK-15 effectively stabilizes
highly active molecular complexes and outperforms ruthenium nanoparticle-based systems.
These findings illustrate the suitability of P-MOFs as catalyst support under
industrially-relevant conditions.

IV
Chapter 6 examines the incorporation of a Grubbs-type catalyst in LSK-15. The molecular
complex [RuCl2(IMesH2)(PPh2-MOF)(CHPh)] (Ru-LSK-15b) was formed via a pyridine
intermediate, using a two-step ligand exchange starting from [RuCl2(IMesH2)(PCy3)(CHPh)].
The material was tested for various self- and cross-metatheses of olefins in the liquid phase,
exhibiting a moderate activity compared to the related homogeneous catalyst. Hot filtration
experiments indicated that no active species remained in solution, thus confirming durable
immobilization of the ruthenium centers. Ru-LSK-15b was subsequently tested for gas-phase
self-metathesis of propene, attaining a moderate initial activity (TOF = 152 h-1 at 40 °C) and
stability (TON = 970).

Chapter 7 summarizes our preliminary attempts at rhodium-catalyzed hydrogenation of

olefins in the presence of P-MOFs. Asymmetric hydrogenation of prochiral olefins was
performed using the P-MOFs together with a rhodium precursor and a chiral monodentate
phosphine ligand to form the catalyst in-situ. While enantioselectivity enhancements were
observed in the case of two P-MOFs, rationalization of these results was hindered by
formation of rhodium-phosphine complexes outside the framework. Amino groups were
observed to have a deactivating effect on rhodium centers; therefore an important component
of this chapter evaluates framework effects in amino- and phosphine-functionalized
MIL-101(Al).

Chapter 8 provides an overall conclusion to this work, reviewing the major findings in the
synthesis of P-MOFs, their characterization, and their applications in metal-supported
catalysis. The remaining challenges are presented and possible outlooks for the future design
and optimization of catalytic pockets in P-MOFs are discussed herein.

In summary, the research reported herein describes the rational design of organometallic
centers inside solid porous P-MOF supports. Phosphines are unique functional groups that
provide a strong anchoring point for transition metal-based precursors, producing highly
active and stable heterogeneous molecular catalysts suited to a wide array of key
transformations. The immobilization of metal complexes was confirmed using
element-specific techniques such as MAS NMR spectroscopy and XAS that permit the
observation of the metal ligand sphere, as well as the electronic state of the phosphines. This
work addresses a number of the major challenges involved in the synthesis and
characterization of P-MOFs and illustrates their substantial potential as solid porous ligands
in metal-supported catalysis.

V
Résumé
Les réseaux métallo-organiques (MOFs) sont des polymères de coordination cristallins
poreux, formés à partir de l’assemblage moléculaire entre des lieurs organiques et des ions
métalliques. Ils se distinguent des matériaux inorganiques traditionnels, tel que la silice ou les
zéolites, de par leur versatilité tant chimique que structurelle, attribuée à la grande variété
d’éléments d’assemblage disponibles. Les réseaux peuvent ainsi être élaborés afin de
contenir des fonctions chimiques et organométalliques permettant l’immobilisation de poches
catalytiques possédant un degré d’ordre élevé et une diffusivité accrue.

Les complexes organométalliques sont des centres métalliques entourés de ligands

organiques. Fréquemment utilisés en catalyse homogène grâce à leur structure définie et
leur dispersion moléculaire dans la phase réactionnelle, ces complexes possèdent une très
haute activité catalytique et sélectivité dans une large gamme de réactions chimiques. Les
complexes à base de rhodium et ruthénium sont souvent coordonnés à une ou plusieurs
phosphines. En modulant les propriétés basiques et stériques des phosphines, il est possible
d’influencer de manière contrôlée la configuration spatiale et électronique du centre
métallique durant la catalyse. Les MOFs formés à partir de lieurs organophosphorés
(P-MOFs) possèdent les caractéristiques inhérentes aux phosphines ainsi que la nature
cristalline et poreuse des MOFs. Ces matériaux sont toutefois soumis à de nombreux
enjeux ; leur développement est notamment restreint par la synthèse laborieuse de nouveaux
lieurs organiques qui limitent fortement le nombre de structures disponibles. Les P-MOFs
souffrent aussi souvent d’un manque de stabilité structurelle.

Le Chapitre 1 passe en revue les récentes études sur la fonctionnalisation des MOFs afin
d’incorporer des catalyseurs moléculaires. La stratégie la plus utilisée se base sur la
modification de réseaux contenant des groupes amines, générant des fonctions bidentées
capables de recevoir des complexes organométalliques. Les P-MOFs bénéficient quant à
eux d’une plus grande affinité aux métaux de transition grâce aux phosphines, permettant
ainsi leur métallation directe. L’état de l’art ne permet cependant que la synthèse d’une
quantité limitée de lieurs organophosphorés et par conséquent de P-MOFs.

Le Chapitre 2 décrit les procédures expérimentales pour la synthèse des lieurs organiques,
des MOFs ainsi que des composés organométalliques utilisés au travers cette thèse. Leur
caractérisation a été réalisée à l’aide notamment de la diffraction par rayons X (XRD),
l’absorption par rayons X (XAS), la résonnance magnétique nucléaire (NMR) et la
spectroscopie infrarouge (IR). Des techniques gravimétriques telles que la physisorption
d’azote, l’analyse thermogravimétrique et élémentaire ont aussi été utilisées. Une courte
description technique des principales méthodes est aussi présentée dans ce chapitre.

VI
Le Chapitre 3 résume la synthèse d’une série de P-MOFs basée sur une version substituée
de l’acide téréphtalique. Une voie de synthèse pour l’acide 2-phosphinobenzène-1,4-
dicarboxylique (PPh2-bdc) est décrite, avec comme étapes clefs un couplage P-C assisté par
micro-ondes et une réduction par silane. Le lieur PPh2-bdc est incorporé avec d’autres acides
dicarboxyliques dans différents MOFs, en suivant une stratégie dite de lieurs mixtes, afin de
prévenir un encombrement stérique trop important. Le lieur se démarque par sa bonne
compatibilité avec d’autres acides téréphtaliques, ainsi qu’avec divers ions métalliques dont
le zinc(II), le chrome(III) et l’aluminium(III). Différentes structures ont ainsi pu être
synthétisées, notamment MOF-5, UMCM-1 et MIL-101. Les matériaux générés ont été
caractérisés par diffraction aux rayons X, spectroscopie RMN et adsorption d’azote afin de
confirmer leur nature cristalline et poreuse. La quantité de phosphine a aussi été variée sans
modifications notables des protocoles de synthèse, démontrant ainsi la versatilité de cette
méthode pour l’optimisation future des P-MOFs.

Le Chapitre 4 décrit l’application de méthodes spécifiques à l’observation du phosphore. La

spectroscopie RMN par rotation à l’angle magique (MAS), la spectrométrie d’absorption par
rayons X (XAS) et la spectroscopie par énergie dispersive (EDX) sont utilisées afin de
caractériser les groupes phosphines à l’état solide. Ces techniques permettent d’accéder à
des informations chimiques qui sont généralement perdues lorsqu’on applique à ces
matériaux des techniques basées sur la digestion d’échantillons. De plus, la combinaison de
spectroscopies RMN en solide et de XAS est particulièrement adaptée à la caractérisation de
complexes organométalliques situés dans les P-MOFs, permettant ainsi de valider les
stratégies d’immobilisation proposées. En parallèle, la spectroscopie EDX peut être
combinée avec la spectroscopie XAS afin d’obtenir une cartographie chimique des éléments
présents dans les P-MOFs à l’échelle de cristaux uniques.

Le Chapitre 5 décrit l’immobilisation dans LSK-15 d’un complexe au ruthénium actif dans la
décomposition sélective de l’acide formique en phase gazeuse. La métallation post-
synthétique de LSK-15 avec [RuCl2(p-cym)]2 permet la formation d’un complexe moléculaire
de type phosphine-ruthénium de la forme [RuCl2(p-cym)(PPh2-MOF)] (Ru-LSK-15a). Le
support et le catalyseur ont été caractérisés par MAS RMN (31P et 13
C), ainsi que par
spectrométrie d’absorption aux seuils K de Cl et Ru. En conditions catalytiques, Ru-LSK-15a
démontre une très haute activité (TOF = 2700 h-1 à 145 °C), stabilité (TON > 1'290'000 en
23 jours) et sélectivité avec la formation exclusive d’hydrogène et dioxyde de carbone.
LSK-15 permet de stabiliser un complexe moléculaire dont les performances surpassent
notamment les systèmes à base de nanoparticules supportées. Les découvertes présentées
démontrent l’utilité des P-MOFs comme supports dans des conditions catalytiques
s’approchant des contraintes industrielles.

VII
L’incorporation d’un catalyseur de Grubbs dans les P-MOFs est examinée dans le Chapitre
6. Le complexe [RuCl2(IMesH2)(PPh2-MOF)(CHPh)] (Ru-LSK-15b) est formé à partir de
[RuCl2(IMesH2)(PCy3)(CHPh)] au travers d’un intermédiaire substitué par des ligands
pyridines. Le matériau est utilisé dans diverses réactions de métathèse d’oléfines dans la
phase liquide. Malgré une activité modérée, en comparaison avec les catalyseurs
homogènes correspondants, Ru-LSK-15b a été testé négatif vis-à-vis d’une potentielle
lixiviation de ruthénium dans la phase active, confirmant ainsi le caractère hétérogène de la
réaction. Le complexe a aussi été testé pour la métathèse du propène en phase gazeuse
mais possède une activité initiale (TOF = 152 h-1 à 40 °C) et une stabilité relativement faibles
(TON = 970). Ces exemples démontrent toutefois le potentiel de ces matériaux dans des
conditions catalytiques plus difficiles.

Le Chapitre 7 résume les tests préliminaires dans les hydrogénations d’oléfines catalysées
par des complexes au rhodium en présence de P-MOFs. En particulier, les hydrogénations
asymétriques d’oléfines prochirales ont été réalisées en utilisant une approche dite
combinatoire, où un précurseur au rhodium est introduit dans une solution contenant un
P-MOF et une phosphine chirale monodentée pour former un catalyseur in-situ. Une
amélioration de l’enantiosélectivité a été observée en présence de deux P-MOFs mais
l’interprétation de ces résultats reste difficile à cause de la possible formation de complexes
rhodium-phosphine en solution. Les groupes amine de certains MOFs ont eu une influence
négative sur l’activité du rhodium et cet effet a été étudié en détail pour différentes structures
de type MIL-101(Al) contenant des groups amines et phosphines.

Le Chapitre 8 présente une conclusion générale à ce travail et passe en revue les principales
découvertes dans la synthèse des P-MOFs, leur caractérisation et leur application en
catalyse par métaux supportés. Les enjeux restant sont discutés plus en détails, ainsi que les
futures perspectives dans le design et l’optimisation de nouvelles structures.

En résumé, cette étude décrit la conception rationnelle de composés organométalliques

formés dans le réseau poreux de P-MOFs. Les phosphines sont des groupes fonctionnels
uniques qui permettent l’ancrage de centre moléculaires basés sur des métaux de transition,
formant ainsi des catalyseurs hétérogènes à haute stabilité et actifs dans un large panel de
transformations chimiques. Ce travail adresse certains des enjeux majeurs entourant la
synthèse et caractérisation des P-MOFs et illustre leur énorme potentiel en tant que ligand
solide poreux.

VIII
 Chapter 1

Introduction

Based on: F. L. Morel,* X. Xu,* M. Ranocchiari, J. A. van Bokhoven, in Chem. Organo-

Hybrids Synth. Charact. Funct. Nano-Objects (Eds.: B. Charleux, C. Copéret, E. Lacôte),
John Wiley & Sons, Inc., Hoboken, NJ, USA, 2015, pp. 200-232.

F. L. M. contribution: write-up with X.X.

* Equal contribution from F.L.M. and X.X.

Chapter 1

1.1 Catalysis

Catalysis is one of the most important topics in chemistry. Catalysts have been used by
mankind for thousands of years, and are part of many natural bio-processes. The classic
definition of a catalyst was proposed by Berzelius in 1836 as "a substance which accelerates
the rate of a reaction, while not being consumed during the process". This definition was later
refined by Arrhenius who stated that catalysts “lower the activation energy of a reaction”, and
by Ostwald who described a catalyst as “a substance that increases the rate at which a
chemical system approaches equilibrium”.[1] While all three of these definitions stand true to
date, modern understanding of catalytic processes has shown that catalysts are far from
static species and can be subject to various transformations during their lifecycle, including
activation and deactivation periods. Catalytic performances are usually reported in terms of
three parameters: activity, selectivity, and stability.[2] The activity is best expressed as the
turnover frequency (TOF), the number of molecules converted per unit of time per active site,
and measures how fast a catalyst operates at a given temperature. The selectivity is defined
as the fraction of desired product molecules obtained per substrate molecule converted and
reflects the ability of the catalyst to kinetically discriminate between various reaction
pathways. Finally, the stability of a catalyst is frequently expressed as the turnover number
(TON), the total number of molecules converted per active site before deactivation. Today,
catalysts are used in almost every aspect of industrial chemistry, such as bulk- and fine
chemistry, refining, and waste management. It is estimated that up to 90% of all commercial
chemicals have been in contact with a catalyst during their life cycle.[3]

Catalysts are often separated into three categories: homogeneous, heterogeneous and
biocatalysts. This distinction was created to reflect the physical behavior of the catalyst in
contact with the reactive phase, and provide guidelines for the design of catalytic
experiments. Homogeneous catalysts are active in the same phase as the substrate. They
present the advantage of molecular dispersion within the reaction phase (often liquid) and are
typically well-defined at the atomic level. Representative homogeneous catalysts
include inorganic acids, organic compounds and organometallic complexes. Molecular
organometallic catalysts, first described by Roelen in 1938,[4] are instrumental in many
industrial processes. Ziegler and Natta developed in the 1950’s titanium complexes for the
polymerization of ethylene and propylene[5,6] and Wilkinson reported in the 1960’s a highly
active rhodium catalyst for hydrogenation reactions.[7] These examples were followed by
many large-scale processes, including hydroformylation,[4] metathesis,[8,9] and asymmetric
hydrogenation;[10] the latter leading to the classic production of L-Dopa by the Monsanto
process.[11] The major drawbacks of most homogeneous catalysts are that they are not
volatile and do not possess a high contact surface area, making them unsuitable for

2
 Introduction

gas-phase applications. In addition, they lack easy downstream separation options, which
prevent their efficient recycling and often lead to product contamination.

Heterogeneous catalysts are solids that are put in contact with a gas or a liquid containing the
reagents. One of the first reports of heterogeneous catalyst was made by H. Davy in 1817,[12]
who observed that heated palladium and platinum wires catalyzed the oxidation of coal gas,
in contrast to wires made of other metals. E. Davy, Döbereiner and Henry later reported that
palladium, finely-ground or supported on clay, was already active at room temperature,
highlighting the importance of catalyst dispersion.[13] Since then, heterogeneous catalysts
have been at the center of the industrial development in the twentieth century with the
appearance of the contact process (synthesis of sulfuric acid), the Fisher-Tropsch process
(alkane formation from carbon monoxide and hydrogen), catalytic cracking reactions, and the
Haber-Bosch process (nitrogen fixation).[14] The main advantages of supported catalysts over
bulk metals is their improved dispersion of the catalytic phases, which can greatly vary
depending on the nature of the support and its shaping properties. Traditionally used
materials include inorganic supports (SiO2, TiO2, Al2O, clays, ceramics, and zeolites), carbon-
based supports (polymers, fibers), and self-supported metals (filaments, alloys). Virtually
every combination of first- to third-row transition metal has been used in catalytic processes,
often in the presence of alkaline or rare-earth dopants.

The drawback of heterogeneous catalysts is that they operate through a complex mixture of
active centers. This has detrimental repercussions on the catalytic process in terms of
selectivity and hinders the molecular understanding of the catalytic sites. One of the key
challenges in modern catalysis is the controlled design of supported active centers. Molecular
catalysts supported onto porous materials are a promising step towards this goal, as they
provide a narrow distribution of active sites, ideally well-distributed onto the support, and can
be applied in solid-liquid and solid-gas phase reactions.[15] Much research in the past
decades has focused on the immobilization of molecular catalysts supported onto silica and
other inorganic materials.[16] Coordinating directly to the metal center via oxygen bonding, or
via an organic tether, these catalysts were successfully applied in various catalytic reactions,
such as polymerization,[17] hydroformylation,[18] and metathesis.[19] While silica-based supports
are inexpensive and resilient materials, they are rather limited in terms of topological and
chemical versatility. The development of porous polymers and hybrid materials is steadily
increasing to provide a new generation of supports that are far less limited in term of
structural properties and provide a better control for the design of the reactive environment.[20]
Among those, metal-organic frameworks (MOFs) are promising candidates for the
immobilization of molecular catalyst due to their high crystallinity, porosity, and chemical
flexibility.

3
Chapter 1

1.2 Metal-organic frameworks: fundamental aspects

1.2.1 Structural properties, key topologies

Metal-organic frameworks (MOFs) are a porous, crystalline sub-class of coordination

polymers first described by Johnson[21] and Robson,[22] and later popularized by the work of
Yaghi.[23–25] These materials are distinguished from organic polymers and zeolites by the
presence of two structure-determining components: an organic linker and an inorganic
secondary building unit (SBU).[26–28] Linkers coordinate to the SBU, usually via oxygen-metal
or nitrogen-metal bonds to form an interconnected crystalline network (Figure 1.1, a). This
approach results in extensive structure versatility, where the organic linkers can be
varied independently from the SBU to generate a series of isoreticular frameworks
(IRMOFs, Figure 1.1, b), displaying tunable pore size and functionality.[29] Linker mixtures
produce frameworks bearing multiple functionalized linkers (MIXMOFs, Figure 1.1, c), and
the ratio between the linkers can be varied to fine-tune the physical and chemical properties
of the material.[30] The introduction of functional groups within a framework is not restricted by
the synthetic conditions or potential linker incompatibilities. Instead, MOFs are routinely
modified post-synthetically (Figure 1.1, d) using a broad scope of organic and inorganic
transformations, providing a versatile methodology for the incorporation of chemical
functions, including catalytic sites (see Section 1.4).

Early materials were formed from divalent ions such as Zn2+ for [Zn4O(L)3]n (Table 1.1) with
L = 1,4-benzenedicarboxylic acid (bdc), 1,1’-biphenyldicarboxylic acid (bpdc), or longer rigid
di-carboxylated linkers,[25,31] or Cu2+ in [Cu3(H2O)3(H3btc)2]n (HKUST-1, H3btc = benzene-
1,3,5-tricarboxylate).[32] They retain both crystallinity and porosity upon removal of solvent
molecules inside their framework, in contrast to earlier attempts to construct organic-
inorganic frameworks.[22,33,34] This sparked interest in MOFs with high micro- and
mesoporosity, which initiated the use of gradually more complex organic linkers for gas
storage and separation applications.[26] It is difficult to make an exhaustive list of frameworks
designed for gas applications due to the large amount of structures available, and many
excellent reviews cover this particular topic.[28,35–38] Notorious materials with high surface area
include [Zn4O(H3btb)4/3(bdc)]n (UMCM-1, H3btb = 1,3,5-tris(4-carboxyphenyl)benzene),[39]
[Zn4O(H3btb)2]n (MOF-177),[40] and [Zn4O(H3bte)4/3(bpdc)]n (MOF-210, H3bte = 4,4′,4″-
[benzene-1,3,5-triyl-tris(ethyne-2,1-diyl)]tribenzoate),[41] all containing a mixture of di- and tri-
carboxylated linkers. Recent examples have achieved higher surface area (SBET > 6000 m2/g)
and pore volume using longer rigid organic linkers with alternating carbon-carbon triple bonds
and aromatic rings such for [Cu3(H2O)3(L)]n (NU-100, L as a longer analogue of H3bte) and
others.[42–45] Table 1.1 summarizes the compositions and structures of the most discussed
MOFs throughout this chapter.

4
 Introduction

Figure 1.1 Schematics of the four main structural features of MOFs. (a) Prototypical structure
based on coordination of organic linkers with SBUs to form an interconnected tridimensional
network, (b) isoreticular MOFs, (c) Mixed-linker MOFs, and (d) post-synthetic modification.

5
Chapter 1

Frameworks based on zinc-carboxylate coordination bonds present the advantage of

producing highly crystalline materials, often in the form of macro-crystals in the range of
hundreds of micrometers that facilitate their study by X-ray diffraction (XRD). However, their
thermal stability is generally limited up to 150 °C[46] and they are sensitive to moisture, basic
conditions and alcoholic solutions due to the hydrolysis of the ZnO bond.[47] To overcome
this, many frameworks were developed using alternative metal ions. Characteristic examples
are [Zr6O4(OH)4(bdc)6]n (UiO-66), [Al(OH)(bdc)]n (MIL-53(Al)), and [Cr3OF(H2O)(bdc)3]n
(MIL-101(Cr)), which display improved stability due to an increase in metal-O bond strength,
while preserving a large pore space and internal surface area. Strong chemical resilience
was achieved in the nickel-based structure [Ni3(btp)2]n (btp = 1,3,5-tris(1H-pyrazol-4-
yl)benzene), which remained stable in aqueous solutions in a broad pH range (2-14).[48] This
illustrates how these structures do not follow a general trend in terms of chemical and thermal
compatibility, but are unique in the sense that they are defined individually by the combination
of their SBUs and organic linkers.

1.2.2 Isoreticular chemistry and pre-synthetic functionalization

Introduced by Yaghi et al. in 2002,[31] the concept of isoreticular MOF (IRMOF) synthesis is
based on the use of different organic linkers with similar connectivity, while keeping the SBU
constant to generate a series of isostructural frameworks. The authors demonstrated two of
the main features of IRMOFs using a series of framework based on the MOF-5 topology,
where Zn2+ ions were coordinated to various linkers.[31] Different linker lengths create
frameworks with various pore sizes, and non-coordinating moieties can be incorporated onto
the linker to yield similar frameworks with different functional groups (Figure 1.1, b). This
concept was later applied for a variety of topologies beyond zinc-carboxylate coordination.
Férey et al. reported several isoreticular frameworks of [Fe3O(bdc)6]n (MIL-88) obtained by
replacing the bdc linker with 2-aminoterephthalic acid (NH2-bdc), bpdc, fumarate, and
2,6-naphtalene.[49,50] Zirconium-based topologies are among the most notable examples of
isoreticular chemistry. The synthesis of isoreticular frameworks based on the UiO-66 topology
was reported using elongated (UiO-67 and UiO-68)[51] or functionalized bdc linkers
(UiO-66-NH2, -Br, -NO2).[52] UiO-67 were synthesized using the 2,2’-bipyridil-5,5’-
dicarboxylate (bpydc) organic linker to yield a bi-pyridine functionalized framework.[53] Other
isoreticular frameworks were produced,[54] including MIL-53(Al)[55] and MIL-101.[56–58] The
recent advances in linker tailoring make isoreticular chemistry a powerful tool for the rational
design of materials with tuned pore systems, and contribute significantly to broadening the
field of MOF-related applications.[29]

6
 Introduction

Table 1.1 Composition and structure of most-discussed MOFs in this chapter.

7
Chapter 1

Isostructural frameworks are not only achieved by the replacement of organic linkers with
similar connectivity, but can also be generated via metal substitution. It is possible to use
metal ions with the same valence to create SBUs with similar connectivity leading to
isostructural frameworks, although this method is somewhat less predictable than the linker
replacement approach. It is difficult to attribute accurately the first report of rational metal
substitution in MOFs, but early examples include different MIL frameworks.[49,50] Notably,
MIL-53(Al) was reported[59] a few years after the synthesis of MIL-53(Cr)[60] and the method
was later extended to its Fe(III) counterpart.[49] This example was followed to produce
different versions of M(III)-based MOFs such as MIL-100 (Fe and V)[61,62] and MIL-101
(Cr, Fe and Al).[49,57,63] The authors observed that the Fe(III) and Al(III) forms of MIL-101 could
only be synthesized using synthetic conditions non-analogue to the MIL-101(Cr) synthesis,
and only when NH2-bdc was used instead of bdc.[49,57,64] Metal substitution is not limited to
SBUs formed from trivalent metal ions and is also reported for divalent ions. For example, the
synthesis of [M2(dhtp)(H2O)2·8H2O]n (CPO-27/MOF-74, dhtp = 2,5-dihydroxyterephthalic acid)
is reported for several ions,[65,66] such as Zn(II),[67] Ni(II),[68] and Co(II).[69]

1.2.3 MIXMOFs: a route to fine-tuned materials

Following the concept of isoreticular chemistry, Kim et al. reported the systematic study on a
series of frameworks containing different ratios of organic linkers and observed that the
partial and quantitative replacement of the bdc linker by functionalized bdc equally led to the
same topology.[30] This observation was later confirmed for other structures,[70] in particular for
the MOF-5 topology, where partial replacement of the bdc linker by NH2-bdc led to similar
frameworks with alterable sorption properties.[71] Baiker et al. suggested the term
“MIXMOF”,[71] for which the partial replacement of a linker by its functionalized counterpart
produces an isostructural framework (Figure 1.1, c). The concept was later extended to
multi-variable MOFs that contain a mixture of more than two organic linkers with different
chemical functions.[72] Many more examples of MIXMOFs have been reported for well-known
structures, such as UMCM-1[73] and MIL-53(Al).[74] Although MIXMOFs are often assumed to
be produced from a statistical mixture of co-linkers, there are examples where the
crystallization of the framework favors preferential incorporation of linkers in one or more
lattice dimensions. This can be achieved by a substitution of linkers with different donating
groups[75] or of linkers that present a certain degree of steric hindrance,[76] yielding
frameworks of similar topology as their mono-linker counterparts.

8
 Introduction

1.3 Synthesis

The majority of MOFs are synthesized using solvothermal processes,[77] in which organic
linkers and metal sources are dissolved in a solvent or a mixture of solvents, such as
N,N-dimethylformamide (DMF), N,N-diethylformamide (DEF), ethanol and water.[78] The
mixture is subsequently heated, usually between 25 and 250 °C, and left under autogenous
pressure until crystallization is complete. The use of additives, such as organic acids, bases,
and structure-directing agents is common to modulate phase thermodynamics and
crystallization kinetics, and to improve the solubility of the metal ions. Alternative methods to
conventional heating are in focus to improve the controllability of the synthesis or to extend to
continuous flow processes. In particular, microwave- ,[79] mechano-,[78] and sono-assisted[80]
methods were reported for the synthesis of various MOFs, with positive influences on yield,
phase purity, crystal shape and dispersity, among other parameters.

1.4 Post-synthetic functionalization of MOFs

Although MOF structures are often designed “pre-synthesis”, i.e. by using a combination of
ligand design and isoreticular chemistry, many targeted properties or functional groups are
incompatible with regular synthesis conditions. It is possible to modify existing frameworks
post-synthetically to access these functions using the chemical flexibility of MOFs. This
section discusses the methodologies commonly reported in the literature and provides key
examples in the introduction of catalytically active centers.

1.4.1 Post-synthetic modifications and pore tuning

The modification of a MOF pore system post-synthetically was first conceptualized by

Robson et al. in 1990[81] and experimentally verified by Lee et al. a few years later.[82,83]
The notion of post-synthetic modification (PSM) is based on the chemical modification of a
synthesized framework to produce a new isostructural one without relying on dissolution-
recrystallization mechanisms. This was first applied to selected pendant moieties pertaining
to organic linkers and later extended to any form of modification that could produce an intact
framework. Aside from a few early examples,[32,84,85] reports of PSM in the literature remained
scarce until Cohen et al. reported the acetylation of the amino groups in IRMOF-3 by acetic
anhydride in 2007.[86] Since then, the use of PSM has been widely applied to generate
structures with new chemical functions that were featured in many reviews.[87–90] Burrows et
al. have reported the reaction of an aldehyde-functionalized framework isoreticular to
[Zn4O(bpdc)3]n (IRMOF-9) with 2,4-dinitrophenylhydrazine to form the corresponding
hydrazine (Figure 1.2) that could not be formed via direct hydrothermal synthesis.

9
Chapter 1

The authors noticed the lack of reactivity of the methoxy-functionalized bpdc linker towards
hydrazine and subsequently created a MIXMOF containing both linkers to control the extent
of functionalization within the framework thus avoiding excessive pore crowding.[91]

Figure 1.2 Post-synthetic modification of pendant aldehyde groups with hydrazine to produce
the corresponding hydrazone.[91]

Many strategies of framework modifications have involved reactions with pendant

amines[92–94] and azides,[73,95] some of which leading to interesting catalytic applications.
For example, Farrusseng et al. reported a two-step PSM strategy that transformed the amine
groups on [Zn4O(bdc)2(dabco)]n (DMOF-1-NH2, dabco = 1,4-diazabicyclo[2.2.2]octane) into
their azide- and functionalized tryazolyl counterparts.[96] By changing the pKa of the terminal
tryazolyl moieties and the extent of framework functionalization, they were able to
significantly alter or enhance the catalytic properties of the MOF towards the
transesterification of ethyldecanoate in methanol. Additionally, post-synthetic modifications
are not limited to liquid-solid interface reactions. Recently our group has demonstrated the
use of volatile reagents, such as salicyclaldehyde and various anhydrides, to functionalize a
variety of amino-MOFs into their respective condensation products in the vapor phase
(Figure 1.3).[97]

Figure 1.3 Functionalization of amino-functionalized MOFs in vapor-phase.

10
 Introduction

1.4.2 Introduction of metal species

The post-synthetic introduction of metal nuclei within MOFs is a widespread strategy to grant
catalytic properties to frameworks that cannot be included using traditional synthetic routes.
In particular, metal complexes containing vacant coordination sites or weak ligands present
interesting catalytic properties,[98] but are not always compatible with the framework
formation. While the presence of two different metal ions could potentially interfere with the
crystallization process, excess organic linkers in solution often interact with empty
coordination sites, thus preventing the formation of the desired metal complex. Many
strategies have been reported in the literature to overcome these issues and are subject to
discussion.[89,90] The four main strategies to incorporate metal species within MOFs are
depicted in Figure 1.4. They consist of direct metalation of the framework (a), PSM at the
organic linker (b) or SBU (c) prior to metalation, and metal encapsulation within the pore
system (d).

Figure 1.4 Schematic representation of common post-synthetic metalation strategies in

MOFs: (a) direct metalation, (b) PSM at the linker moiety, (c) thermal activation of
the inorganic unit, followed by coordination of an organic ligand and (d) encapsulation.
Red = framework targeted site, blue = secondary coordination site, gray = metal centers
(single atom, complexes or nanoparticles).

Direct metalation (Figure 1.4, a) is one of the shortest routes to access metal-functionalized
frameworks. While notable examples of direct coordination of chromium- and molybdenum
hexacarbonyl complexes onto the aromatic ring of bdc have been reported,[99–103] the majority
of procedures rely on organic linkers with designed metal-binding properties. In an early
example, a homochiral MOF [Cd3Cl6(L)3]n (L = (R)-6,6'-dichloro-2,2'-dihydroxy-1,1'-binaphtyl-
4,4'-bipyridine) was reported to react with Ti(OiPr)4 to yield the Lewis acidic site
Ti(OiPr)2(BINOLate).[85] This strategy was later generalized to a series of isoreticular copper-
and zinc-MOFs (C-MOFs) of generic formula [M2(solvent)2(L)]n, which showed interesting
asymmetric catalytic applications, such as diethylzinc additions to aromatic aldehydes that
processed with high conversion and enantioselectivity values (Figure 1.5).[104,105]

11
Chapter 1

Pyridine-based linkers were also used to generate MOFs containing empty coordination sites,
as it is the case in [Al(OH)(bpydc)]n (MOF-253), where the 2,2’-bipyrdine system was
functionalized with PdCl2 and CuBF4.[106] The same strategy was applied with
[Eu(pdc)1.5(DMF)]n(solv)m (pdc = pyridine-3,5-dicarboxylate), where the coordination site of
the pdc linker incorporated various amount of metal ions (e.g. Cd2+, Zn2+, Mn2+, Co2+, etc.) by
exchanging their corresponding salt in DMF.[107] Several MIXMOFs based on the
2-phenylpyridine-5-,4’-dicarboxylic acid (dcppy) organic linker were functionalized by
cyclometalation of rhodium or iridium complexes to yield catalytic materials active in the
N-alkylation of amines.[108,109] A similar approach was developed independently in several
isoreticular frameworks of UiO-67 containing dcppy linker analogues pre-synthetically
functionalized with various metal salts.[110] This is a good example of how pre- and post-
synthetic approaches can lead to equally satisfactory metal incorporation, granted that the
organometallic complex is stable under synthesis conditions.

Figure 1.5 Direct metalation of a series of homochiral dihydroxy-functionalized MOFs by

titanium isopropoxide and their applications in asymmetric diethylzinc and alkynilzinc
additions.

Yaghi et al. reported two porphyrin-MOFs based on zirconium that can be post-synthetically
metalated with iron chloride.[111] This strategy remains an exception as porphyrin-MOFs are
usually metalated pre-synthesis for stability reasons.[112–115] Re-metalation is also a viable
method in selected examples of salen-MOFs. Schultz et al. reported the de-metalation of
[Zn2(H4tcpb)(dipyridil-Mn(salen))]n (H4tcpb = 1,2,4,5-tetrakis(4-carboxyphenyl)benzene) by
soaking the crystals in a solution of methanol and hydrogen peroxide (Figure 1.6). The
material was subsequently transferred into an aqueous solution of different M(II) metals to
undergo re-metalation.[116] However, formation of defect sites, as well as a certain extent of
metal exchange at the SBUs, were observed by the authors.[114] While catalytic activity of
salen(Mn)-MOFs was demonstrated for enantioselective epoxidation of alkenes,[117,118] their
scope remains limited to date.

12
 Introduction

Figure 1.6 Re-metalation of the H4tcpb linker in [Zn2(H4tcpb)(dipyridil-Mn(salen))]n

In a radically different approach, chloromethylation of MIL-101(Cr) was used to bind a

porphyrin(Sn) ligand onto the bdc organic linker, which was subsequently transformed to
yield a catalyst active in the reaction of carbon dioxide with epoxides.[119] While direct
metalation is largely reported in the literature and is the method of choice to incorporate
catalytic centers within a framework, it is restricted to selected topologies due to the relative
complexity of the organic linkers.

Amino functional groups play a major role in the metalation of MOFs following organic PSM
(Figure 1.4, b). An elegant two-step strategy is the condensation of amine groups with a
variety of N- and O-functionalized substrates, followed by their direct metalation under mild
conditions. Ahn et al. reported that the Mn(acac)2 (acac = acetylacetonato) Schiff’s base can
be condensed in IRMOF-3 and showed the catalytic activity of the manganese complex in
alkene epoxidation.[120] Rosseinsky et al. used IRMOF-3-sal to bind a vanadium complex,
forming IRMOF-3-sal-V(O)(acac).[121] Corma et al. reported the incorporation of a gold
complex within IRMOF-3-sal, which was active in the coupling of N-protected amines with
aldehydes.[122] Similar strategies were reported using a variety of amino-MOFs to incorporate
Fe3+, Cu2+ and In3+ complexes.[123–125] The use of iminopyridine groups was also described in
UMCM-1-NH2[126] and MIL-101(Fe)-NH2,[127] showing that this strategy can generate N-N
chelating groups suitable for palladium and nickel coordination, respectively. The latter
displayed remarkably high activity (TOF > 10,000 h-1 at 30 bar ethylene) in the
oligomerization of ethylene to α-olefins, with a high selectivity towards 1-butene (Figure 1.7).

13
Chapter 1

Figure 1.7 Oligomerization of ethylene catalyzed by a nickel chloride complex immobilized in

MIL-101(Fe)-NH2 via imine condensation.

Reported methods that introduce a secondary metal species within a MOF framework often
assume that the primary metal center forming the SBU cannot be modified without altering
the framework stability. However, many frameworks possess unsaturated or labile
coordination sites occupied with solvent molecules. In an early work, Chui et al. showed that
HKUST-1 can be thermally activated, followed by quantitative replacement of empty
coordination sites by stronger pyridine ligands.[32] Nowadays this concept is regularly applied
for the rational tuning of inorganic nodes. In particular, MIL-101(Cr) is an outstanding
candidate for node modification, due to its excellent thermal and chemical stability. Férey
et al. used different multifunctional amines such as ethylenediamine (ED), diethylenetriamine
(DETA), and 3-aminopropyltrimethoxysilane (APS) to generate MIL-101-ED, MIL-101-DETA,
and MIL-101-APS, respectively.[128] The authors took advantage of the chelating properties of
the grafted ligands to favor encapsulation of noble metal nanoparticles, such as palladium,
platinum, and gold; the former was tested for its catalytic activity in Heck coupling reactions
(Figure 1.8).

Figure 1.8 Example of Heck coupling catalyzed by palladium nanoparticles embed in

MIL-101-ED systems.

Nguyen et al. reported that the secondary amine of dopamine coordinates unsaturated
chromium sites preferentially, leaving pendant hydroxyl moieties that are subsequently
available for VO(acac)2 complexation.[129] Other groups have further expanded this strategy to

14
 Introduction

different frameworks[130,131] and to generate homochiral frameworks.[132] The encapsulation

strategy (Figure 1.4, d) takes advantage of the rigidity of MOFs to trap various guest
molecules within the pore system. Seminal work by Fischer[133] and Kim[134] demonstrated that
vapor-phase deposition of various organometallic complexes can be performed in MOF-5
through weak support-metal interaction. Since then, metal species were incorporated in the
form of nanoparticles,[128,135–143] organometallic complexes,[133,134] polyoxometalates,[63] and
metalloenzymes.[144]

1.4.3 Other post-synthetic methods

PSM remains so far the most prominent framework functionalization method reported in the
literature. Recent advances have, however, led to alternative strategies based on partial
dissolution-recrystallization of the framework. Post-synthetic ligand exchange (PSE), also
called solvent-assisted ligand exchange (SALE), was developed for various topologies,
allowing the replacement of the organic strut with another one, usually in milder thermal
conditions as compared to the synthesis of the original framework.[145–147] Amino- and bromo-
substituted linkers in UiO-66,[148,149] as well as different ZIF frameworks[150–152] were replaced
under near-solvothermal conditions. This strategy was recently used to incorporate a di-iron
complex in UiO-66 as part of a photochemical redox system in water.[153] The SALE
equivalent of SBU exchange is often called transmetalation. For example, Suh et al. reported
the replacement of Zn2+ by Cu2+ ions in [Cu2(bdcppi)(DMF)3·6DMF·4H2O]n (SNU-51,
bdcppi = N,N’-bis(3,5-dicarboxyphenyl)pyromellitic diimide) by immersing the crystals in a
copper nitrate solution in methanol.[154]

1.5 Applications of MOFs

The combination of pre- and post-synthetic design strategies allows for a virtually infinite
number of fine-tuned materials. This is reflected by the variety of MOF-related applications
that drive the search for frameworks with particular properties, and impact the development of
the structures that are discussed throughout this work. Table 1.2 summarizes the most-
reported applications to date, together with some of the materials typically used. Since
Kitagawa et al. reported a MOF series based on the 4,4’-bypiridine (bpy) linker
([M2(bpy)3(NO3)4]n, M = Co, Ni, Zn) and their use in gas sorption, many structures have been
designed with sorption applications in mind. The storage of energy-related gases, such as
methane and hydrogen, has been reported for a large number of structures,[35] and guest-
specific interactions were targeted recently for the capture of greenhouse gases.[37,155] The
chemical flexibility of MOFs has also an enormous potential for the incorporation of
catalytically active centers,[98,156,157] such as homochiral ligands,[158–160] metal
[161–163] [164–167]
nanoparticles, and organometallic complexes.

15
Chapter 1

Table 1.2 Summary of most-reported MOF applications, together with a non-exhaustive list of
materials typically used.

Applications Examples References[a]

Gas technologies
MOF-5
[35,171]
Gas storage (H2, CO2, CH4) MOF-177
ZIF-8
PCN-17 [37,172]
Gas separation
Er(pda)3
MOF-74 [173,174]
Harmful gas sorption
HKUST-1

Catalysis
HKUST-1
[162]
Open-site catalysis MIL-101(Cr)
PCN-124
MIL-100
[157,165–167,175]
Supported metal catalysis MIL-101
MOF-5
Zr(Ru-L)[b]
[159,160,176,177]
Selective catalysis (size, shape, enantiomers, etc.) POST-1
SO-MOF

Biological / Biomedicine
MIL-100(Fe) [168,169]
Drug encapsulation
MIL-53

Storage / Delivery of gasotransmitters CPO-27

[156]
Biomimetic catalysis PCN-222

Electromagnetism
[170]
Luminescence MOF-76
[178]
Magnetism Cu(py)2(bdc)2

Other
MIL-53 [179,180]
Chromatography
ZIF-8
UiO-67 [181]
Water purification
NU-1000
[a]
References include review articles whenever available.
[b]
(Ru-L) = [Ru(binap)(dpen)Cl2]

The use of bio-compatible SBUs and linkers has also allowed the incorporation of drugs
within MOF pores, giving an optimistic view on future medical applications.[156,168,169]
The electromagnetic properties of lanthanides and paramagnetic metal ions can also be
exploited to create luminescent frameworks, used notably in sensing applications.[170]

16
 Introduction

These examples demonstrate the broad scope of application of MOFs and the potential for
rationally-designed structures to answer key of chemical challenges, notably in catalysis.

1.6 Organophosphine functional groups in MOFs

1.6.1 Chemistry of phosphines

Organophosphines are a large class of compounds in which a P(III) atom occupies a central
position and is surrounded by hydrocarbyl substituents or hydrogen, following the general
formula PHnR3-n.[182] Sub-classes of phosphines are usually identified by the number of their
carbon-based substituents, where PH2R, PHR2 and PR3 are referred to as primary-,
secondary- and tertiary-phosphines, respectively. Phosphines are characterized by a lone
electron pair at the phosphorus, which readily coordinates to chalcogens to form P(V)
compounds. The most commonly encountered chemical reaction involving phosphines is
their oxidation to form phosphine oxides, which are the thermodynamically favored
form.[183,184] The reactivity of phosphines is strongly dictated by the carbon substituents.[185]
While small, electron-donating substituents such as methyl or isopropyl usually lead to very
reactive and basic phosphines, bulky electron-withdrawing groups often kinetically limit
oxidation of the P(III) center at room temperature, especially in the solid state. Due to the
strength of the PO bond, phosphines and derivatives were originally used as mild reducing
agents in various deoxygenation reactions, such as the deoxygenation of epoxides to their
corresponding alkenes, or the reaction with hydroperoxide groups, among others.[186] One
notable example of phosphine reactivity is the Wittig reaction, where a phosphonium ylide
reacts with a carbonyl compound to yield a phosphine oxide and an alkene.[187] Phosphines
are also reported to reduce azides to form iminophosphoranes, which can subsequently react
with carbonyl groups to form a C=N bond and a phosphine oxide via a mechanism similar to
the Wittig reaction.[188]

1.6.2 Phosphines in catalysis

Early reports of phosphines include their use as organocatalysts for umpolung and [3+2]
addition reactions, where a tertiary phosphine can perform a nucleophilic attack on an allene
or an alkyne, respectively, formic a zwitterionic intermediate.[189] This leads to an enhanced
reactivity towards nucleophilic or electrophilic substrates, depending on the reaction
conditions, and the subsequent recovery of the phosphine.[188] However, amine and carbene
species have been preferred over the use of phosphines due to their increased resilience
towards oxidation.[190] Phosphines have been increasingly investigated for their role as
ligands in coordination chemistry and organometallic catalysis. In addition to their Lewis basic
properties, phosphines are notable neutral ligands with a strong affinity for late transition
metals of various oxidation states. Generally considered σ-donors and sometimes π-

17
Chapter 1

acceptors, their electronic and steric properties can be modulated by their substituents. They
are usually considered “spectator” ligands and are rarely involved directly with bond formation
or cleavage but influence significantly the properties of organometallic centers. In particular,
tertiary phosphines have been used in key organometallic compounds, such as
[RhCl(PPh3)3], [RuCl2(PCy3)2(CHPh)] and [Pd(PPh3)4], which all had major repercussions in
homogeneous catalysis, both at the preparative and industrial scale. Catalytic uses include
various hydrogenations,[191] hydroformylations[192] and metathesis of olefins,[193] as well as
various coupling reactions.[194,195] In addition to monodentate phosphines, organophosphorus
compounds bearing two phosphine groups have been reported to act as bidentate ligands.
This drove the synthesis of chiral di-phosphine ligands, with notable examples being reported
by Noyori[11] and Knowles[10] for the asymmetric hydrogenation of olefins, ketones and imines,
among others.

While phosphines have been used almost universally in homogeneous catalysis, their
immobilization onto solid supports has also been subject of much attention. Many strategies
have been developed over the years, in particular using silica-based and polymeric supports,
demonstrating the potential of supported phosphines in metal coordination.[196–199]
Triphenylphosphine has often been selected due to its relative resilience towards oxidation
and satisfactory binding affinity towards palladium in cross-coupling reactions.[200–202]
However, these structures often under-perform as compared to their homogeneous
counterparts in term of activity or selectivity. The characterization of their active centers
remains a major challenge, as well as the influence of the support environment due to the
lack of lattice order.

1.6.3 Available linkers and current phosphine-MOFs

Organic linkers containing phosphorus moieties were originally investigated as an alternative

to carboxylic acids to form stable MOFs built around the P-metal coordination.[203–205] Since
then, many forms of phosphorus have been incorporated, including phosphines,[76,206–209]
phosphine oxides,[204,210,211] phosphonates[212] and phosphonium ions.[213,214] In a seminal work
on phosphine-functionalized porous frameworks, Lin et al. proposed the reaction of a 2,2’-bis-
(diphenylphosphino)-1,1’-binaphtyl (binap) organic linker with [Ru(benzene)Cl2]2 and
1,2-diphenylethylenediamine (dpen), followed by the coordination of the complex to a
zirconium SBU to form [Zr[Ru(binap)(dpen)Cl2]]n·4H2O.[215] Although the material displayed
remarkable activity in the asymmetric hydrogenation of ketones, the framework did not
possess long-range order, supposedly due to the complicated nature of the zirconium–
phosphonate coordination. The first highly-ordered MOF containing a “free” organophosphine
linker (i.e. containing a non-coordinated lone electron pair), was reported by Humphrey
et al.[213] 4,4’,4’’-phosphinetriyltribenzoic acid (ptbc) was reacted with Zn(OAc)2 in DMF to

18
 Introduction

yield [H[Zn(ptbc)(HNMe2)]2·1.5DMF]n (PCM-1) that proved to be highly crystalline but

unstable upon solvent removal. Since then, the use of tri(para-carboxylated)
triphenylphosphine and phosphine oxide linkers led to more robust materials; however, their
potential as supports for metal complexes remains relatively unexplored (vide infra).[207,210,211]
Instead, most of the early applications in phosphine- and phosphine oxide-functionalized
frameworks focused on their selective gas sorption[209,214] and luminescent properties.[216]

A common feature for many existing P-MOFs bearing non-coordinated phosphines is their
synthesis from tridentate organic linkers centered on the phosphorus atom. In this case, the
phosphorus atom occupies a so-called “deterministic position”, where the framework topology
revolves around the phosphine pyramidal geometry and strongly depends on the P-R3 bond
angles. This limits the original scope of phosphine groups and hinders the possibility of fine-
tuning their steric and electronic properties. Organophosphine linkers based on linear
dicarboxylic acids should not suffer from these limitations, as the phosphine moiety is
attached to the rigid connection of the framework. With this in mind, our group reported a
series of zinc-based MIXMOFs built from 2-(diphenylphosphino)-[1,1’-biphenyl]-4,4’-
dicarboxylic acid (PPh2-bpdc) and the corresponding oxide, which crystallize in
IRMOF-9-related topologies.[76] These materials have demonstrated interesting applications
in organocatalysis.[217,218] In an alternative approach, Gascon et al. proposed the
post-synthetic chloromethylation of MIL-101(Cr), followed by a nucleophilic substitution of
diphenylphosphine to introduce free phosphine within such frameworks.[219] So far, the
relatively harsh reaction conditions involved prevented the extension of this method to other
frameworks.

1.6.4 Metal-supported applications of P-MOFs

Since the introduction of a ruthenium active center inside a P-MOF by Lin et al.,[215] limited
examples of metal incorporation have been described in the literature. Kaskel et al. reported
the immobilization of palladium and rhodium centers inside EOF-17, the latter being found to
be active in the transfer hydrogenation of cyclohexanone.[220] Although these two materials do
not strictly follow the definition of MOFs due to their low crystallinity, they highlight the
potential of phosphine-functionalized porous coordination polymers in post-synthetic
metalation. Our group recently reported the synthesis of [Zr6O4(OH)4(ptbc)4]n (LSK-1) and its
subsequent functionalization with a gold(I) complex (LSK-1-Au-PA, Figure 1.9).[207] Despite a
relatively low surface area and lack of long-range order, the material displayed permanent
porosity and remarkable activity in the hydration of phenylacetylene and the cyclization of
N-(prop-2-yn-1-yl)benzamide. Recently, Lin et al. revisited the synthesis of their initial binap-
functionalized linker to incorporate terminal carboxylic acids that yielded a crystalline
homochiral zirconium-MOF.[206] They demonstrated that the framework can be functionalized

19
Chapter 1

post-synthetically with ruthenium and rhodium complexes, active in various enantioselective

catalytic applications. While these examples of phosphine-functionalized frameworks all
present remarkable metal-binding properties, they often suffer either from poor long-range
order that prevents the rationalization of their catalytic center, or difficult synthetic procedures
that strongly limit their widespread application to other structures.

Figure 1.9 Structure (left) and catalytic properties (right) of LSK-1-Au-PA in the cyclization of
N-propargyl amides.

1.7 Rationale of the work

The aim of this thesis is to describe the synthesis and characterization of novel phosphine-
MOFs and their applications in metal catalysis. The synthesis of a diphenylphosphine-
functionalized terephthalic acid linker (PPh2-bdc) was targeted, as bdc is used as building
block for the synthesis of common MOF topologies. Throughout this work, the incorporation
of molecular complexes by post-synthetic metalation was demonstrated for ruthenium- and
rhodium complexes, with the help of various characterization techniques. Their catalytic
performances were established in various catalytic applications, such as the dehydrogenation
of formic acid, olefin metathesis, and hydrogenation reactions.

Chapter 2 provides all experimental details about the synthetic procedures, characterization
techniques and catalytic experiments discussed in this work. A short description of the main
characterization methods is also provided.

20
 Introduction

Chapter 3 describes the synthetic approach for phosphine-functionalized linear linkers and
their implications in the synthesis of several MIXMOF structures based on MOF-5, UMCM-1,
and MIL-101 topologies. The novel materials were characterized by XRD, NMR
spectroscopy, nitrogen physisorption and thermogravimetric analysis to confirm their
permanent crystallinity and porosity.

Chapter 4 provides selected examples on how element-specific characterization techniques,

such as magic-angle spinning NMR spectroscopy and XAS provide molecular insights into
P-MOF systems. Two particular topics are discussed throughout this chapter: 1) the accurate
estimation of the phosphine to phosphine oxide ratio in P-MOFs, and 2) the direct and
indirect observation of P-metal coordination in rhodium- and ruthenium-functionalized
frameworks

Chapters 5, 6 and 7 present the developed uses of P-MOFs in metal-supported catalysis.

Chapter 5 describes the incorporation of a [RuCl2(p-cym)]2 complex within the pore system of
LSK-15 (MIXMIL-101(Al)-PPh2-NH2) and its performance in the selective dehydrogenation of
formic acid in vapor-phase. Chapter 6 summarizes the immobilization of a ruthenium-carbene
complex (Grubbs’ catalyst, 2nd generation) through P-Ru coordination and its application in
olefin self- and cross-metathesis in the liquid and gas phase. Chapter 7 discusses the
preliminary application of rhodium-catalyzed hydrogenation in the presence of P-MOFs.

Chapter 8 provides the overall conclusion and discusses outlooks of possible applications for
P-MOFs in metal-supported catalysis and beyond.

21
Chapter 2

Methods
Chapter 2

2.1 Introduction

Metal-organic frameworks (MOFs) are routinely designed by combining new and existing
knowledge from various fields, ranging from organic and coordination chemistry to material
science. Their analysis requires the use of multiple characterization techniques, such as
X-ray diffraction (XRD), infrared (IR) spectroscopy, nitrogen physisorption, electron
microscopy and thermogravimetric analysis. It is capital to understand the electronic and
chemical environment of the metal center and surrounding ligands when considering new
metal catalysts. We used element-specific techniques such as X-ray absorption spectroscopy
(XAS) and nuclear magnetic resonance (NMR) spectroscopy to complement the techniques
mentioned above.

Section 2.2 provides experimental details regarding the synthetic procedures of the organic
linkers, MOFs, and organometallic compounds described in further chapters. Section 2.3
summarizes the methods for post-synthetic modifications, with a particular focus on the
coordination procedures of rhodium and ruthenium compounds in phosphine-functionalized
MOFs. Typical experimental conditions and setups used for catalytic applications are
summarized in Section 2.4. Section 2.5 provides a short theoretical background of the main
methods used to characterize the different compounds throughout this work, including
a technical description.

2.2 Synthesis

2.2.1 Organic precursors and linkers

All chemicals were purchased from commercially available sources and used without further
purification, unless otherwise stated. Microwave reactions were performed in a CEM MARS
system with EasyPrep (100 mL) or Xpress (20 mL) PTFE autoclaves under autogenous
pressure. Phosphine reduction procedures were carried out under argon atmosphere and
solvents used during synthesis and work-up were dried over activated alumina
(Brockmann I, basic), degassed using standard Schlenk techniques and stored over
molecular sieves (4 Å).

24
 Methods

Synthesis of 3-(diphenylphosphoryl)-4-methylbenzoic acid

In a procedure adapted from Rummelt et al.,[221,222] 3-iodo-4-methylbenzoic acid (1.440 g,

5.50 mmol), Pd/C (0.06 g, 0.056 mmol unreduced Pd), K2CO3 (3.035 g, 22.0 mmol),
diphenylphosphine oxide (1.440 g, 7.12 mmol) and water (50 mL) were loaded into a 100 mL
PTFE vial. The mixture was heated for 1 h at 180 °C under microwave irradiation (isothermal
mode, maximum of 400 W) and vigorous stirring. The crude mixture was filtered over Celite
and neutralized to pH = 5 using HCl (3 M). The precipitate was recovered by filtration,
washed with water and dried under vacuum. Recrystallization of the solid in EtOH/H2O (1:1)
yielded pure 3-(diphenylphosphoryl)-4-methylbenzoic acid as white needles. Yield: 1.679 g
(91%). 1H NMR (500 MHz, DMSO-d6) δ (ppm) 13.06 (s, 1H), 8.05 (d, J = 7.7 Hz, 1H),
7.69 – 7.56 (m, 11H), 7.51 (dd, J = 7.6, 3.4 Hz, 1H), 2.37 (s, 3H).31P NMR (202 MHz, DMSO-
d6) δ (ppm) 28.20 (s).

Synthesis of 2-(diphenylphosphoryl)-terephthalic acid (POPh2-bdc)

3-(diphenylphosphoryl)-4-methylbenzoic acid (2.200 g, 6.54 mmol) was mixed with KMnO4

(3.100 g, 19.62 mmol) in an aqueous solution of KOH (5% w/w, 10 mL). The slurry was
heated at 120 °C overnight until the color changed from purple to red-brown. After cooling
down to room temperature, the crude mixture was filtered over a glass filter. The filtrate was
acidified using HCl (conc.) until pH = 1 and 2-(diphenylphosphoryl)-terephthalic acid
(POPh2-bdc) was recovered as a white precipitate. Recrystallization in EtOH/H2O led to
colorless needles suitable for single-crystal X-ray diffraction analysis. Yield: 1.981 g (83%).

25
Chapter 2

1
H NMR (500 MHz, DMSO-d6) δ (ppm) 13.43 (s, 2H), 8.23 (dt, J = 8.1, 1.4 Hz, 1H), 8.11
13
(dd, J = 13.4, 1.7 Hz, 1H), 7.96 (dd, J = 8.0, 3.5 Hz, 1H), 7.62 – 7.49 (m, 10H). C NMR
(126 MHz, DMSO-d6) δ (ppm) 167.57 (d, J = 2.6 Hz), 165.93 (s), 140.56 (d, J = 6.3 Hz),
134.64 (d, J = 11.0 Hz), 133.18 (d, J = 107.5 Hz), 132.86 (d, J = 2.4 Hz), 132.43 (d, J = 11.3
Hz), 132.25 (d, J = 97.3 Hz), 131.71 (d, J = 2.7 Hz), 131.33 (d, J = 9.8 Hz), 130.27 (d, J = 8.3
Hz), 128.51 (d, J = 12.1 Hz). 31P NMR (202 MHz, DMSO-d6) δ (ppm): 28.62 (s). HR-ESI-MS:
[M-H]- m/z calcd. for C20H14O5P-: 365.0584, found: 365.0592.

Synthesis of 2-(diphenylphosphino)-terephthalic acid (PPh2-bdc)

2-(diphenylphosphoryl)-terephthalic acid (1.000 g, 2.73 mmol) was added to toluene (22 mL)
and stirred for 10 minutes under argon. Methyldiethoxysilane (1.75 mL, 10.92 mmol) and
diphenylphosphate (0.103 g, 0.41 mmol) were added under constant stirring and the mixture
was heated to 130 °C. After 72 h, the solution was filtered and reduced under vacuum. The
remaining solution was extracted with a saturated solution of sodium bicarbonate (3 x 10 mL).
The aqueous phase was washed with dichloromethane (DCM, 3 x 10 mL) and acidified using
HCl (3 M) until pH = 1. The solid was filtered, washed with water (3 x 10 mL) and dried under
vacuum for 24 h. 2-(diphenylphosphino)-terephthalic acid (PPh2-bdc) was recovered as a
yellow solid. Yield: 0.583 g (61%). 1H NMR (500 MHz, DMSO-d6) δ (ppm) 13.27 (s, 2H), 8.09
(dd, J = 8.1, 3.7 Hz, 1H), 8.01 (dd, J = 8.0, 1.7 Hz, 1H), 7.51 (dd, J = 4.0, 1.8 Hz, 1H),
13
7.41 – 7.34 (m, 6H), 7.22 – 7.16 (m, 4H). C NMR (126 MHz, DMSO-d6) δ (ppm) 167.34
(d, J = 2.2 Hz), 166.39 (s), 140.15 (d, J = 28.9 Hz), 138.63 (d, J = 19.2 Hz), 137.64
(d, J = 12.0 Hz), 134.37 (s), 133.44 (d, J = 20.8 Hz), 133.29 (s), 130.53 (d, J = 2.4 Hz),
129.39 (s), 128.86 (s), 128.71 (d, J = 7.2 Hz). 31P NMR (202 MHz, DMSO-d6) δ (ppm) -5.62
(s). HR-ESI-MS: [M-H]- m/z calcd. for C20H14O4P-: 349.0635, found: 349.0626.

26
 Methods

Synthesis of 2-(diphenylphosphoryl)-[1,1':4',1''-terphenyl]-4,4''-dicarboxylic acid

(POPh2-tpdc)

1,4-benzene diboronic acid (0.250 g, 1.52 mmol), 4-iodobenzoic acid (0.374 g, 1.52 mmol),
K2CO3 (1.20 g, 9.1 mmol), Pd/C (0.024 g, 1.5 mol% Pd ) and water (10 mL) were loaded into
a 20 mL PTFE vial and the mixture was stirred at room temperature for 24 h.
3-bromo-4-iodobenzoic acid (0.518 g, 1.59 mmol) was added and the mixture was stirred for
24 h. Diphenylphosphine oxide (0.366 g, 1.82 mmol) was added thereto and the mixture was
heated to 160 °C under microwave irradiation for 2 h (isothermal mode, maximum of 400 W).
The mixture was filtered over Celite, neutralized to pH = 1 using HCl (37% w/w) and the
precipitate was filtered, washed with water (3 x 10 mL), MeOH (3 x 10 mL) and dried
overnight under vacuum to yield 2-(diphenylphosphoryl)-[1,1':4',1''-terphenyl]-4,4''-
dicarboxylic acid. ESI-MS: [M+H] m/z calc. for C32H24O5P+: 519.1356, found 519.1367.
+

2.2.2 Metal-organic frameworks

Unless otherwise stated, all chemicals were purchased from commercially available sources
and used without further purification. Prior to synthesis, N,N-dimethylformamide (DMF), DCM,
acetone and toluene were dried over activated alumina (Brockmann I, basic) and stored over
molecular sieves (4 Å) under inert atmosphere. For all procedures involving organophosphine
linkers, synthesis, handling and storage were carried under inert atmosphere and solvents
were degassed following standard Schlenk techniques. Supercritical CO2 drying was
performed in a Tousimis Samdri 931 critical point dryer.

27
Chapter 2

Synthesis of MIXIRMOF-9-PPh2 (LSK-3)

Biphenyl-4,4’-dicarboxylic acid (0.039 g, 0.10 mmol), PPh2-bpdc (0.173 mg, 0.08 mmol) and
Zn(NO3)2·4H2O (0.191 g, 1.01 mmol) were mixed together with DMF (10 mL) in a 20 mL
scintillation sealed vial and the mixture was heated to 90 °C in an oven continuously flushed
with nitrogen. After 72 h, the vials were cooled down to room temperature and the yellow
crystals of LSK-3 were suspended in fresh DMF (3 x 10 mL). Samples were stored in DMF
for 3 days, followed by a solvent exchange with DCM and toluene (3 days each, solvent
refreshed every day). Prior to nitrogen adsorption the sample was exchanged with acetone
(3  10 mL), dried using supercritical CO2 and activated at 90 °C for 24 h under reduced
pressure.

Synthesis of IRMOF-3

2-aminoterephthalic acid (NH2-bdc, 0.150 g, 0.83 mmol) and Zn(NO3)2·4H2O (0.600 g,

2.30 mmol) were combined in a round-bottom flask together with DMF (20 mL). The mixture
was stirred for 5 min at room temperature. The solution was divided into two 20 mL
scintillation sealed vials and placed in an oven at 105 °C for 24 h. After cooling down to room
temperature, the brown crystals were suspended in DMF (3  10 mL) over a period of 3 days.
The crystals were subsequently exchanged with DCM and stored in toluene under inert
atmosphere.

28
 Methods

Synthesis of MIXMOF-5

NH2-bdc (0.091 g, 0.5 mmol), terephthalic acid (bdc, 0.249 g, 1.5 mmol) and Zn(NO3)2·4H2O
(1.568 g, 6 mmol) were dissolved in DMF (100 mL). The mixture was stirred for 10 min at
50 °C. The solution was divided into ten 20 mL scintillation sealed vials and placed in an oven
at 95 °C for 24 h. After cooling to room temperature, the brown crystals were suspended in
DMF (3  10 mL) over a period of 3 days. The crystals were subsequently exchanged with
DCM and stored in toluene under inert atmosphere.

Synthesis of MIXMOF-5-PPh2 (LSK-11)

In a typical synthesis, PPh2-bdc (0.264 g, 0.75 mmol), bdc (0.250 g, 1.5 mmol) and
Zn(NO3)2·4H2O (1.773 g, 6.77 mmol) were dissolved in DMF (80 mL). The mixture was stirred
for 10 min at 50 °C. The solution was divided into ten 20 mL scintillation sealed vials and
placed in a nitrogen-flushed oven at 95 °C for 48 h. After cooling down to room temperature,
the yellow crystals of MIXMOF-5-PPh2 (LSK-11) were transferred to a single vial and
suspended in DMF (3  10 mL) over a period of 3 days. Prior to nitrogen adsorption the
sample was exchanged with acetone (3  10 mL), dried using supercritical CO2 and activated
at 150 °C for 24 h under reduced pressure. The recovered mass was 0.230 g. In subsequent
experiments, the molar ratio between PPh2-bdc and bdc was varied from 2:1 to 1:5.

29
Chapter 2

Synthesis of MIXMOF-5-PPh2-NH2 (LSK-13)

The synthesis of LSK-13 was adapted from a procedure reported by Xu.[223] PPh2-bdc
(0.015 g, 0.05 mmol), NH2-bdc (0.032 g, 0.18 mmol) and Zn(NO3)2·4H2O (0.178 g,
0.68 mmol) were dissolved in DMF (5 mL) at 50 °C in a sealed vial under argon atmosphere.
The mixture was placed in a nitrogen-flushed oven at 105 °C for 24 h. The vials were allowed
to cool down to room temperature and the crystals were washed with DMF (3  10 mL) over a
period of 3 days and suspended in acetone (10 mL) overnight. After washing, the crystals
were dried using supercritical CO2 and stored under argon atmosphere before further
analysis.

Synthesis of MIL-101(Cr)

In a procedure adapted from Hatton et al.,[224,225] bdc (0.332 g, 2.00 mmol) was added to an
aqueous solution of tetramethylammonium hydroxide (TMAOH, 10 mL, 0.05 M) and stirred
for 10 min. Cr(NO3)3·9H2O (0.800 g, 2.00 mmol) was added and the mixture was stirred for
an additional 20 min. The mixture was placed in a PTFE-lined autoclave and heated at
210 °C for 24 h under nitrogen atmosphere. After cooling down to room temperature, the
green powder was collected using centrifugation and washed with water (3  20 mL), DMF
(3  10 mL) and placed in a Soxhlet extractor for 24 h (ethanol, 200 mL) to yield MIL-101(Cr).
Prior to analysis, the sample was activated at 150 °C for 24 h under reduced pressure. The
recovered mass was 0.120 g.

30
 Methods

Synthesis of MIXMIL-101(Cr)-POPh2 (LSK-12)

PPh2-bdc (0.100 g, 0.29 mmol) and bdc (0.285 g, 1.71 mmol) were added to an aqueous
solution of TMAOH (10 mL, 0.05 M) and stirred for 10 min. Cr(NO3)3·9H2O (0.800 g,
2.00 mmol) was added and the mixture was stirred for an additional 20 min before being
heated in a PTFE-lined autoclave at 210 °C for 24 h under nitrogen atmosphere. After cooling
down to room temperature, the green powder was centrifuged and washed with water
(3  20 mL), DMF (3  10 mL) and placed in a Soxhlet extractor for 24 h (ethanol, 200 mL) to
yield MIXMIL-101(Cr)-POPh2 (LSK-12). Prior to analysis, the sample was activated at 150 °C
for 24 h under reduced pressure. The recovered mass was 0.235 g. The molar ratio between
PPh2-bdc and bdc was varied in subsequent experiments from 1:6 to 1:1.

Synthesis of MIL-101(Al)-NH2

In a procedure adapted from Hartmann and Fischer,[64] NH2-bdc (0.272 g, 1.5 mmol) was
dissolved in DMF (40 mL) at 110 °C. A solution of AlCl3·6H2O in DMF (0.724 g, 3.0 mmol in
20 mL) was added dropwise over a period of 2 h under constant stirring. The mixture was
stirred for 3 h and then allowed to stand still at 110 °C for 18 h. The solid was isolated by
centrifugation and washed with DMF (3  20 mL) and ethanol (3  20 mL) to yield
MIL-101(Al)-NH2. Prior to analysis, the sample was further washed by Soxhlet extraction for
24 h (ethanol, 200 mL) and dried at 120 °C for 24 h under vacuum. The recovered mass was
0.142 g.

31
Chapter 2

Synthesis of LSK-14

PPh2-bdc (0.263 g, 0.75 mmol) and bdc (0.125 g, 0.75 mmol) were dissolved in DMF (40 mL)
under argon atmosphere at 110 °C. Various amounts of water (0, 10 and 500 equivalents)
were added to the mixture in subsequent experiments. A solution of AlCl3·6H2O in DMF
(0.724 g, 3.0 mmol in 20 mL) was added drop-wise over a period of 1 h under stirring. The
mixture was stirred for 3 h and then permitted to stand still at 110 °C for 18 h. The solid was
isolated by centrifugation and washed with DMF (3 x 20 mL) and ethanol (3 x 20 mL) to give
a pale yellow solid (LSK-14). Prior to analysis, the sample was further washed by Soxhlet
extraction for 24 h (ethanol, 200 mL) and dried at 150 °C for 24 h under vacuum.

Synthesis of MIXMIL-101(Al)-PPh2-NH2 (LSK-15)

In a typical synthesis, PPh2-bdc (0.175 g, 0.5 mmol) and NH2-bdc (0.181 g, 1.0 mmol) were
dissolved in DMF (40 mL) at 110 °C. A solution of AlCl3·6H2O in DMF (0.724 g, 3.0 mmol in
20 mL) was added dropwise over a period of 2 h under constant stirring. The mixture was
stirred for 3 h and then permitted to stand still at 110 °C for 18 h. The solid was isolated by
centrifugation and washed with DMF (3  20 mL) and ethanol (3  20 mL) to give
MIL-101(Al)-NH2-PPh2 (LSK-15). Prior to analysis, the sample was further washed by Soxhlet
extraction for 24 h (ethanol, 200 mL) and dried at 120 °C for 24 h under vacuum. The
recovered mass was 0.156 g.

32
 Methods

Synthesis of MIXUMCM-1-PPh2 (LSK-17)

PPh2-bdc (0.016 g, 0.045 mmol), terephthalic acid (0.015 g, 0.09 mmol), 1,3,5-tris(4-
carboxyphenyl)benzene (0.021 g, 0.048 mmol) and Zn(NO3)2·4H2O (0.140 g, 0.54 mmol)
were added in a 20 mL scintillation vial according to a procedure reported by Xu.[223] DMF
(5 mL) was added and the mixture was placed in a nitrogen-flushed oven at 85 °C for 48 h.
The vials were cooled down to room temperature and the crystals were washed with fresh
DMF (3  10 mL) and DCM (3  10 mL) over a period of 3 days to yield MIXUMCM-1
(LSK-17), subsequently stored in toluene.

2.2.3 Organometallic complexes

All reactions involving organometallic complexes were carried out under inert atmosphere
using standard Schlenk techniques. All solvents were dried and degassed using the
procedures described above, with the exception of DCM and pyridine that were distilled over
CaH2 under argon. [Ru(IMesH2)(PCy3)Cl2(CHPh)] was purchased from commercially
available sources and used without further purification.

Synthesis of [RhCl(COD)]2

RhCl3·nH2O (40.4% Rh w/w, 2.0 g, 7.85 mmol) was placed in a round-bottom flask together
with EtOH (17.7 mL), H2O (3.3 mL) and 1,5-cyclooctadiene (COD, 3 mL, 24.5 mmol).[226,227]
The mixture was heated to reflux under argon for 24 h, and subsequently filtered.
[RhCl(COD)]2 was recovered as a yellow powder and was washed with H2O/MeOH
(1:5, 60 mL) and hexane (20 mL), and dried under vacuum overnight. Yield: 1.546 g (80%).
1
H NMR (500 MHz, CDCl3) δ (ppm) 4.23 (s, 4H), 2.52 (m, 4H), 1.75 (m, 4H).

33
Chapter 2

Synthesis of [Rh(COD)2]BF4

In a procedure adapted from Bosnich et al.[228] [RhCl(COD)]2 (0.5 g, 1.0 mmol) was mixed
with DCM (10 mL) and COD (0.374 mL, 3.05 mmol) was subsequently added to the solution.
A solution of AgBF4 (0.452 g, 2.32 mmol) in acetone (3.4 mL) was added and the mixture was
stirred for 20 min at room temperature. The precipitate was removed by filtration and THF
(6.8 mL) was added to the filtrate. The mixture was subsequently reduced under vacuum.
The brown crystals were washed with hexane (3 x 5 mL), diethyl ether (3 x 10 mL) and dried
under vacuum to yield [Rh(COD)2]BF4. Yield: 0.65 g (79%). 1H NMR (500 MHz, CDCl3)
δ (ppm) 5.37 (s, 4H), 2.70 - 2.43 (m, 8H).

Synthesis of [Rh(COD)(PPh3)2]BF4

The synthesis was adapted from a reported procedure.[229] [Rh(COD)2]BF4 (0.100 g,

0.25 mmol) was dissolved in toluene (75 mL). Triphenylphosphine (0.197 g, 0.75 mmol) was
added thereto and the mixture was stirred for 48 h at room temperature.
[Rh(COD)(PPh3)2]BF4 was recovered as a yellow solid, which was washed with pentane
(10 mL) and dried under vacuum overnight. Yield: 0.124 g (67%). 1H NMR (500 MHz, CDCl3)
δ (ppm) 7.37 (dt, J = 14.7, 7.1 Hz, 30H), 4.59 (s, 4H), 2.60 (d, J = 9.4 Hz, 4H), 2.25
(d, J = 8.8 Hz, 4H). 31P NMR (202 MHz, CDCl3) δ (ppm) 26.22 (d, J = 145.0 Hz).

Synthesis of [RuCl2(p-cym)]2

RuCl3·3H2O (1.20 g, 4.71 mmol) and α-phellandrene (6 mL, 37.5 mmol) were dissolved in
EtOH (60 mL) in a round-bottom flask.[230] The mixture was heated to reflux for 4 h in a
nitrogen atmosphere. After cooling down to room temperature, the solution was filtered off
and the filtrate was reduced to half-volume under reduced pressure. The solution was placed
in a refrigerator overnight and the resulting orange crystals were separated by filtration,
washed with cold hexane and dried under vacuum to yield [RuCl2(p-cym)]2.
Yield: 1.04 g (74%). 1H NMR (500 MHz, CDCl3) δ (ppm) 5.40 (dd, J = 67.9, 5.9 Hz, 4H), 2.92
(hept, J = 7.0 Hz, 1H), 2.15 (s, 3H), 1.27 (d, J = 6.9 Hz, 6H).

Synthesis of [RuCl2(p-cym)(PPh3)]

PPh3 (0.336 g, 1.28 mmol) was added to [RuCl2(p-cym)]2 (0.400 g, 0.640 mmol) in DCM (10
mL). The mixture was stirred at room temperature for 15 min according to the literature.[231]
The solvent was removed under vacuum and the solid was repeatedly washed with hexane.
After drying under vacuum, [RuCl2(p-cym)(PPh3)] was recovered as a dark-orange solid.
Yield: 0.508 g (70%). 1H NMR (500 MHz, CDCl3) δ (ppm) 7.93 – 7.70 (m, 6H), 7.45 – 7.28 (m,
9H), 5.19 (d, J = 6.0 Hz, 2H), 4.99 (d, J = 6.2, 1.4 Hz, 2H), 2.96 – 2.73 (m, 1H), 1.87 (s, 3H),
1.10 (d, J = 6.9 Hz, 6H). 31P NMR (202 MHz, CDCl3) δ (ppm) 24.19.

34
 Methods

Synthesis of [RuCl2(IMesH2)(C5H5N)2(CHPh)]

The synthesis of [RuCl2(IMesH2)(C5H5N)2(CHPh)] (IMesH2 = 1,3-dimesityl-4,5-

dihydroimidazol-2-ylidene) was adapted from a procedure reported by Grubbs et al.[232] In a
sealed vial, [RuCl2(IMesH2)(PCy3)(CHPh)] (PCy3 = tricyclohexylphosphine, 0.200 g, 0.236
mmol) was dissolved in toluene (0.5 mL) and pyridine (1.5 mL, 37 mmol) was added. The
solution was stirred at room temperature for 10 min after which the solid changed from a red
to a bright green color. The solution was precipitated in pentane (3 mL) at -10 °C and the
resulting bright green solid was washed with pentane (4  1 mL) and dried overnight under
reduced pressure. Yield: 0.100 g (60%). 1H NMR (500 MHz, benzene-d6) δ (ppm) 19.64
(s, 1H), 8.70 (s, 2H), 8.36 (s, 2H), 8.08 – 8.01 (m, 2H), 7.14 (s, 1H), 6.83 – 6.03 (m, 9H), 3.39
(d, J = 43.0 Hz, 4H), 2.76 (s, 6H), 2.43 (s, 6H), 2.00 (s, 6H).

2.3 Post-synthetic modifications

This section summarizes the post-synthetic procedures used throughout this work. The
description of the immobilization of rhodium and ruthenium compounds is limited to
impregnation procedures that yielded isolated materials. The addition procedures of transition
metal complexes and P-MOFs during catalytic experiments are described in Section 2.4.

2.3.1 Organic procedures

Reaction of LSK-15 with phenyl isothiocyanate

LSK-15 (0.030 g, approx. 0.11 mmol amino groups) and phenyl isothiocyanate
(160 μL, 1.1 mmol) were added to DCM (1 mL) in a glass vial and stirred for 48 h at room
temperature. The material was washed with DCM (3  0.5 mL) and dried under reduced
pressure before use to yield the thiourea-functionalized LSK-15 (LSK-15-TU).

Reaction LSK-15 with acetic anhydride

Adapting a procedure reported by Cohen et al.,[86] LSK-15 (0.030 g, approx. 0.11 mmol amino
groups) and acetic anhydride (18.9 μL, 0.2 mmol) were added to DCM (1 mL) in a glass vial
and stirred for 5 days at room temperature. The material was washed with DCM (3  0.5 mL)
and dried under reduced pressure before use to yield the amide-functionalized LSK-15
(LSK-15-AM).

35
Chapter 2

2.3.2 Immobilization of rhodium complexes

Synthesis of Rh-LSK-3

LSK-3 (0.060 g dry crystals) was mixed with [Rh(COD)2]BF4 (2.5  10-3 g, 6.0  10-3 mmol) in
DCM (0.5 mL) at room temperature for 24 h. The orange crystals were subsequently washed
with DCM (3  1 mL) and stored in DCM to yield Rh-LSK-3.

Synthesis of Rh/PPh3-LSK-3

Rh-LSK-3 (0.030 g dry crystals, 3.0  10-3 mmol Rh) was suspended in a solution of PPh3
(3.0  10-3 mmol in 0.5 mL DCM) at room temperature for 24 h. The yellow crystals were
subsequently washed with DCM (3  1 mL) and stored in DCM to yield Rh/PPh3-LSK-3.

Synthesis of Rh-LSK-11

LSK-11 (0.010 g dry crystals) and [Rh(COD)2]BF4 (3.0  10-3 g, 7.0  10-3 mmol) were added
to DCM (0.3 mL). The mixture was kept still for 48 h at room temperature, after which the
crystals were washed in DCM (3  0.5 mL) to yield Rh-LSK-11. The solvent was
subsequently exchanged with toluene prior to analysis.

Synthesis of [Rh(acac)(COD)]@LSK-15

LSK-15 (0.01 g) was suspended in toluene (1 mL) together with [Rh(acac)(COD)]

(acac = acetylacetonato, 6.7  10-3 g, 0.020 mmol) for 1 h at room temperature. The powder
was isolated by centrifugation, washed with toluene (3  1 mL) and dried under vacuum
overnight to yield a light orange solid [Rh(acac)(COD)]@LSK-15.

2.3.3 Immobilization of ruthenium complexes

Synthesis of Ru-LSK-15a

LSK-15 (0.150 g) was suspended under argon in a solution of [RuCl2(p-cym)]2

(0.025 g, 0.04 mmol) in DCM (1.5 mL). The suspension was stirred for 24 h at room
temperature. The solid was separated by decantation, washed with DCM (4 x 2 mL) and
dried under reduced pressure at room temperature to yield Ru-LSK-15-b. The ruthenium
loading was measured by atomic absorption spectroscopy (AAS).

Synthesis of Ru-LSK-15b

In a typical procedure, LSK-15 (0.05 g) and [RuCl2(IMesH2)(pyridine)2(CHPh)] (0.018 g,

0.025 mmol) were suspended in DCM (0.65 mL) in a glass vial. The mixture was stirred for
48 h at room temperature. The solid was recovered by decantation and washed with DCM
(5  1 mL) and dried under reduced pressure to yield green-brown material. The ruthenium

36
 Methods

loading was measured by AAS. Similar procedures (minus the drying step) were applied to
the impregnation of LSK-11 and LSK-17 to form Ru-LSK-11b and Ru-LSK-17b, respectively.

Synthesis Ru@MIL-101(Al)-NH2

The synthesis was carried out by a wetness-impregnation procedure where [RuCl2(p-cym)]2

(0.015 g, 0.002 mmol) in DCM (0.5 mL) was added dropwise to MIL-101(Al)-NH2 (0.100 g) at
room temperature and stirred for 24 h. The solid was dried under reduced pressure at room
temperature to yield Ru@MIL-101(Al)-NH2. The ruthenium loading was measured by AAS.

2.4 Catalysis

All reactions were carried out under inert gas atmosphere, using dried and degassed solvents
using standard Schlenk techniques. Analytical methods for the determination and
quantification of the products are described in Section 2.4.5.

2.4.1 Synthesis of organic substrates

2-benzoloxy-3-butene

In a procedure adapted from the literature,[233] benzoyl chloride (1.653 mL, 14 mmol) was
added to a solution of 3-butene-2-ol (1.214 mL, 14 mmol) in pyridine (5 mL) at 0 °C. The
mixture was allowed to warm up to room temperature and was stirred for 2 h. Saturated brine
(30 mL) and ether (30 mL) were added to the mixture. The organic phase was separated by
decantation and sequentially washed with HCl (5 wt. % in water), brine, saturated aqueous
NaHCO3 and brine. The organic phase was dried over MgSO4 and the solvent was removed
under vacuum. The resulting oil was further purified by silica gel chromatography (120 g
Silica 60) using hexane and ethyl acetate (10:1) as eluent to yield 2-benzoloxy-3-butene as a
transparent oil. The purity of the product was measured at 97% by gas chromatography –
mass spectrometry (GC-MS). Yield: 1.88 g (76%).

2.4.2 Formic acid dehydrogenation catalyzed by supported ruthenium catalysts

The gas-phase dehydrogenation of formic acid and the detection of products were carried out
in a setup described by Beloqui Redondo et al.[234] Formic acid vapor was introduced through
heated lines into a continuous flow reactor containing [RuCl2(p-cym)(LSK-15)] (typically
0.02 g) via a saturator containing formic acid with nitrogen as carrier gas. Products and non-
converted formic acid were analyzed by gas chromatography.

2.4.3 Olefin metathesis experiments

Cross-metathesis of terminal olefins catalyzed by homogeneous ruthenium catalysts

In a typical experiment, [RuCl2(IMesH2)(C5H5N)2(CHPh)] (1.6  10-3 g, 2.2  10-3 mmol) was

added to DCM (1 mL) together with dodecane (5.9  10-3 mL, 2.6  10-3 mmol) as internal

37
Chapter 2

standard in a 25 mL two-neck round-bottom flask fitted with a condenser. 5-hexenylacetale

(0.031 mL, 0.22 mmol) and 1-hexene (0.028 mL, 0.22 mmol) were added and the mixture
was stirred at 45 °C for 24 h. Aliquots were taken at various reaction times, filtered through a
glass filter and quickly analyzed by GC-MS. Similar experiments were performed with
[RuCl2(IMesH2)(PCy3)(CHPh)], as well as with other olefins (see Chapter 6).

Cross-metathesis of terminal olefins catalyzed by supported ruthenium catalysts

In a typical experiment, [RuCl2(IMesH2)(LSK-15)(CHPh)] (0.02 g, 1.2  10-3 mmol Ru) was

suspended in DCM (2 mL) together with dodecane (5.9  10-3 mL, 2.6  10-3 mmol) in a 25 mL
two-neck round-bottom flask fitted with a condenser. 1-hexene (0.028 mL, 0.22 mmol) and
5-hexenylacetale (0.031 mL, 0.22 mmol) were added and the mixture was stirred at 45 °C for
72 h. Aliquots were taken at various reaction time, filtered through a glass filter and quickly
analyzed by GC-MS. Similar experiments were performed with various olefins and P-MOFs
(see Chapter 6).

Self-metathesis of propylene in continuous flow

In a typical experiment, [RuCl2(IMesH2)(LSK-15)(CHPh)] (0.02 g, 1.2  10-3 mmol Ru) was

loaded in a continuous flow reactor (4.1 mm internal diameter). The setup consists of a
mixing unit, where propylene is introduced together with helium, a reactor, and a downstream
line connected to an online mass spectrometer (Prisma 80 equipped with a QMA 200,
Pfeiffer) and a sampling port for gas chromatography (Figure 2.1). A needle valve was used
upstream of the reactor to regulate the flowrate at low values (below 20 ml/min). Samples
were exposed at temperatures between 30 to 120 °C at various concentrations of propylene.

Figure 2.1 Setup schematic for the metathesis of propylene in continuous flow. MFC = mass
flow controller, FI = flow indicator, TI = temperature indicator, TC = temperature controller.

38
 Methods

2.4.4 Hydrogenation experiments

Hydrogenation of acrolein in the presence of Rh/PPh3-LSK-3

Rh/PPh3-LSK-3 (0.03 g, 3  10-3 mmol Rh) was suspended in DCM (0.5 mL) in a scintillation
glass vessel and acrolein (0.01 mL, 0.15 mmol) was added before the mixture was placed
into a steel autoclave. The mixture was stirred under 3 bar H2 at 25 °C for 24 h. Conversion
and selectivity were recorded by GC-MS.

Hydrogenation of acrolein in the presence of [Rh(COD)(PPh3)2]BF4

A solution of [Rh(COD)(PPh3)2]BF4 in DCM (0.5 mL, 3  10-3 mmol Rh) and acrolein (100 mL,
0.15 mmol) were mixed in an open scintillation glass vessel that was placed into a steel
autoclave. The mixture was stirred under 3 bar H2 at 25 °C for 24 h. Conversion and
selectivity were recorded by GC-MS.

Hydrogenation of olefins with [Rh(COD)2]BF4, P-MOFs and monodentate chiral

phosphines

[Rh(COD)2]BF4 (3  10-3 g, 7.5 × 10-3 mmol) was added together with DCM (2 mL) in a
Schlenk tube under argon atmosphere. Various P-MOFs (see Table 7.1, Chapter 7) and
(S)-(+)-neomenthyldiphenylphosphine ((S)-NMDPP) were added to the solution so that the
total phosphorus-to-rhodium ratio equal 2.2. The solution was stirred for 1 h and the olefin
(typically dimethyl itaconate, 105.5 μL, 7.5 × 10-1 mmol) was then added. The mixture was
saturated with hydrogen by rapid alternation of vacuum and hydrogen atmosphere (4 cycles)
and kept under hydrogen (1 bar). Aliquots were taken after 2 and 24 h and analyzed
by GC-MS. Subsequent hydrogenations of methyl 2-acetamidoacrylate, (z)-methyl
2-acetylamino-3-phenylacrylate, and (z)-methyl-2-acetamido-3-(4-nitrophenyl)-acrylate were
performed in the presence of LSK-11.

Hydrogenation of olefins with [Rh(COD)2]BF4, PPh3 and monodentate chiral phosphines

[Rh(COD)2]BF4 (3  10-3 g, 7.5 × 10-3 mmol) was added together with DCM (2 mL) in a
Schlenk tube under argon atmosphere. Various amounts (see Table 7.2, Chapter 7) of PPh3
and (S)-NMDPP were added and the solution was stirred for 1 h before the olefin (typically
dimethyl itaconate, 105.5 μL, 7.5 × 10-1 mmol) was then added. The mixture was saturated
with hydrogen by rapid alternation of vacuum and hydrogen atmosphere (4 cycles) and kept
under hydrogen (1 bar). Aliquots were taken after 2 and 24 h and analyzed by GC-MS.
Subsequent hydrogenations of methyl 2-acetamidoacrylate, (z)-methyl 2-acetylamino-3-
phenylacrylate, and (z)-methyl-2-acetamido-3-(4-nitrophenyl)-acrylate were performed with in
the presence of [Rh(COD)2]BF4 and (S)-NMDPP.

39
Chapter 2

2.4.5 Analytical methods

GC-FID

Gas chromatography – flame ionization detector (GC-FID) analyses were carried out in an
Agilent 6890A GC (HP-5 crosslink 5% Ph Me Silicone column, 50 m  0.32 mm  1.05 μm)
coupled to a FID detector.

GC-MS

Gas chromatography – mass spectroscopy (GC-MS) analyses were carried out in an Agilent
789A GC (J&W Cyclosil-B column, 30 m  0.25 mm  0.25 μm or HP-5 MS 5% Ph Me
Silox column, 30 m  0.25 mm  0.25 μm) coupled to an Agilent 5975C mass spectrometer
(triple axis MSD detector, EI mode).

Determination of conversion and selectivity of formic acid dehydrogenation reactions

Conversion and selectivity were monitored with an Agilent 7890A lab GC-FID and an Inficon
3000A micro GC equipped with a thermal conductivity detector. The carbon monoxide
detection limit was 20 ppm. Conversion of formic acid was calculated by the peak area
decrease in the chromatograms. The selectivity to hydrogen (SH2) was calculated from the
hydrogen yield divided by the conversion of formic acid:

/
∙ (2.1)
∙

2.5 Characterization

2.5.1 X-ray diffraction

MOFs are commonly characterized by a high degree of lattice order. X-ray diffraction (XRD)
is often used to observe the structure of their crystal lattice, as short wavelengths, in the
order of 1 Å, can interact with matter in a non-destructive fashion. When a monochromatic
beam of X-rays passes through a lattice plane, photons interact with the electron density of
the atoms, which elastically scatters the X-ray radiation in all directions. In the case of
amorphous materials, i.e. that do not possess an ordered lattice, X-rays form little to no
interferences, which results in almost no detected energy for a given geometry. In the case of
a crystalline material, the order between planes of the lattice creates a constructive
interference in specific directions resulting in the strengthening of the wave front, also called
the diffraction phenomenon. The condition for which diffraction occurs in a three-dimensional
periodic structure is called Bragg’s law:[235]

2 sin (2.2)

40
 Methods

Where n is the order of diffraction, λ the wavelength, d the spacing distance between lattice
planes and θ the angle between the incident and diffracted beam. The recording of the
energy of the diffracted beam at various θ-angle is called a diffraction pattern.

Single-crystal X-ray diffraction

In a single-crystal XRD experiment,[236] a series of two-dimensional diffraction patterns are

recorded at various spatial orientations of the crystal, which are then merged into a three-
dimensional representation (also called reciprocal space) of the electron densities. From the
reciprocal space, the structure factors, F(hkl), can be extracted:

(2.3)

where h, k and l are Miller indices of the diffraction planes, N the number of atoms in the cell
and xj, yj, and zj the fractional coordination of the j-atom in the cell. The structure factors are
vectors in the reciprocal space that are equivalent to the electron densities in the physical
space. In practice, the amplitude of |F(hkl)| are extracted from the measurements and the
relative phase, φ(hkl), needs to be approximated and refined using specialized software to
yield a crystal structure in the physical space.

Suitable single-crystals of MOFs or organic linkers were selected under an optical

microscope (Leica DFC425 C digital microscope camera) using polarized light. Samples were
mounted on the tip of a Mark-tube borosilicate glass and covered with perfluorinated
polyether oil to protect the crystal from excessive exposure to air. Tubes were placed on a
cold nitrogen beam and measured at various temperatures. The data collection was
performed on a Brucker SMART diffractometer equipped with a charge-coupled device
(CCD) detector. Data treatment and space group assignments were performed using
WinGX.[237] The structures were solved using the SHELXL-97 (and higher versions) direct
method[238,239] and further refinement was performed following the least-square method in
Olex2.[240] If applicable, the structures were refined using the Platon SQUEEZE method for
the removal of solvent molecules within MOFs pore.[241]

Powder X-ray diffraction

It is possible to measure a diffraction pattern from a powder sample, in which the size of the
individual crystals is too small for single-crystal X-ray analysis.[242] The resulting random
orientation of each crystallite leads to a diffraction cone for each individual intensities, as
opposed to spots in a single-crystal. In practice the measurement of the intensities is often

41
Chapter 2

limited to a one-dimensional diffraction pattern that focuses the analysis towards phase
determination and tracking of changes in the unit cell.

In this work, powder XRD patterns of MOFs were recorded at room temperature on a
PANanalytical Empyrean diffractometer at 45 kV, 40 mA with CuKα (λ = 1.541 Å) radiation.
Samples were ground using a mortar and pestle and deposited evenly on a low-background,
single-crystal Si(111) sample holder. Measurements were carried out at diffraction angles 2θ
between 2 and 70° with a step of 0.033° and a total counting time of 100 s. The sample
holder was constantly rotated to prevent preferential orientation of the sample.

2.5.2 X-ray absorption spectroscopy

For a material of thickness x, the intensity, I, of an incident X-ray beam correlates to the
incoming photon intensity, I0, and the linear absorption coefficient, μ(E), of the material
following Lambert-Beer’s law:

  (2.4)

where the energy of the photon, E, is related to its frequency, ν, or its wavelength, λ, following
Eq. (2.5) (Planck’s constant h = 6.626  10-34 Js, c = 2.997  108 ms-1 the speed of light in
vacuum):

⁄ (2.5)

At distinct energies, the absorption coefficient increases drastically to give rise to an

absorption edge, where the incoming photon has the required energy to eject a core electron
of the absorber atom above the Fermi level. The energy at which the incoming photon is
equal to the binding energy is called the edge energy, E0. Each edge is located at an energy
that is characteristic of the binding energy of the core electron, the atomic weight of the
absorber, as well as the quantum number of the shell (K, L, M, etc.) (Figure 2.2, a,b). If more
energy is transferred to the absorber, the surplus is carried as kinetic energy by the
photoelectron in the form of a spherical wave to the neighboring atoms. The wave is
subsequently back-scattered forming constructive and destructive interferences patterns
responsible for the oscillating behavior of the spectra above the edge jump (Figure 2.2, d).

X-ray absorption spectroscopy (XAS) is a technique that measures the cross-section

absorbance in the vicinity (usually 50 eV before and several 100 eV after) of one or more of
these edges.[243,244] The region around the absorption edge is often separated into three
distinct regions (Figure 2.2 c): the pre-edge region, the X-ray absorption near-edge structure
(XANES), and the extended X-ray absorption fine structure (EXAFS).

42
 Methods

XAS allows the extraction of electronic and structural information in an element-specific

fashion, which is of particular interest for the characterization of materials with complex
structures, such as MOFs.

Figure 2.2 a): behavior of the X-ray absorption coefficient μ/ρ for P (Z = 15) and Ru (Z = 44).
Data represented from the NIST database.[245] b): Schematic representation of K and L shell
transitions where electrons are extracted above the Fermi level. c): normalized X-ray
absorption spectrum at the Ru K-edge with XANES and EXAFS region. d): representation of
radiating and back-scattering processes upon X-ray absorption (color: absorber atom in red,
neighbors in black).

Transmission XAS

Transmission X-ray absorption spectra were collected at the SuperXAS beamline[246] of the
Swiss Light Source, Villigen PSI, using a QEXAFS monochromator mounted with a Si(311)
channel-cut crystal and an approximate 2   2 mm2 beam size. Ru K-edge spectra were
measured from 22.0 to 25.3 keV and the energy was calibrated using a ruthenium metal foil
standard (E0 = 22.117 keV). For each sample, 597 spectra were recorded using the QEXAFS
mode[247] and processed using the JAQ software.[248]

Fluorescence XAS

Fluorescence XAS measures the photon emission due to the secondary electron filling the
core hole created in the absorption process. This method is usually preferred when the
absorber is highly diluted within the sample, which results in transmission spectra with poor
signal-to-noise ratio. The absorption coefficient is then measured as the ratio between the
fluorescence yield, IF, and I0:

/ (2.6)

Fluorescence X-ray absorption spectra were collected at the PHOENIX beamline at the
Swiss Light Source, Villigen PSI, using a detector with respectively one and four silicon drift
diodes (Vortex, USA). The light source of the beamline is an undulator and monochromatic

43
Chapter 2

light is generated by a double crystal monochromator using Si(111) crystals. To remove

higher harmonics contributions from the light, the beam is reflected by a set of two planar
mirrors, where the reflection angle is set to reflect only photons with energies below 6 keV.
The detector was mounted at 90° with respect to the incoming beam. The detector distance
was selected so that the electronic dead time remained below 20% during the experiment to
avoid spectral distortion. A dead time correction was applied to the raw data during post-
processing.

MOFs and reference compounds were measured without dilution at the P K-edge. Reference
compounds for Ru L-edge spectra were diluted with cellulose to 1% Ru w/w. Samples were
ground using a mortar and pestle and pressed to pellets. Unless otherwise specified, the
samples were irradiated using an unfocused beam with an approximate 1  2 mm2 beam
size, moderated so as to reduce the flux density on the sample and prevent radiation
damage. I0  was determined using the total electron yield signal taken from a 0.5 µm thin
polyester foil coated with a 50 nm nickel thin film located upstream in a vacuum chamber,
which is held at a pressure of about 10-7 mbar.

Analysis of XANES and EXAFS

Over-absorption corrections were performed for P K-edge XANES using the self-absorption
correction algorithm reported by Haskel,[249] as implemented into the ATHENA software
package.[250] Unless otherwise stated, spectra were processed by subtracting the smooth pre-
edge background fitted with a straight line from the measured data. Each spectrum was then
normalized to unity above the absorption edge, where the EXAFS oscillations were no longer
visible (circa 200-400 eV above the edge). When required, the XANES spectra at the
P K-edge were processed using the Linear Combination Fitting (LCF) procedure
implemented in ATHENA,[250] using reference spectra.

2.5.3 Nuclear magnetic resonance

NMR in solution

Nuclear magnetic resonance (NMR) spectroscopy is a technique that provides local

structures of nuclei with non-zero overall nuclear spin number, I, in the presence of a
magnetic field.[251,252] Measured nuclei are called respectively dipolar if I = 1/2 (such as 1H,
C, 15N and 31P) and quadrupolar in the case of I > 1/2 (14N, 27Al, etc.). In the presence of an
13

external magnetic field of constant field strength, B0, a nuclei of spin I = 1/2 will align with the
magnetic field in a parallel (+1/2) or antiparallel fashion (-1/2), separated by a transition
energy ΔE (also called the Zeeman splitting effect).

44
 Methods

This energy difference is proportional to the strength of the applied magnetic field:

∆ (2.7)

where the nuclear magnetic moment, μ, is proportional to the spin number and the
gyromagnetic ratio, γ:

(2.8)

The transition energy can be converted to its respective frequency, ν, using Eq. (2.5). The
operative frequency of the nuclei then becomes:

⁄2 (2.9)

This is only valid for an isolated nucleus. In practice, the surrounding electrons produce an
induced local magnetic field that partially shields the nucleus from the external field. Electron
density from neighboring atoms produces a similar effect that results in a shift in the operative
frequency. The sum of these effects is regrouped in one mathematical term called σ. In
practice, this shift is often referenced to a standard compound frequency νref at the same field
strength and expressed as the chemical shift, δ:

⁄2 1 (2.10)

(2.11)

Due to the organic nature of MOFs, NMR spectroscopy in solution is routinely applied to
digested samples. In a typical digestion experiment, a MOF sample (5 mg) is dissolved using
a mixture of DCl in D2O (100 μL, 20 w% DCl) and d6-DMSO (500 μL). 1H, 13C and 31P NMR
spectra in solution were recorded at room temperature on a Bruker AVANCE 1 500
spectrometer. 1H and 13
C NMR chemical shifts are reported in parts per million (ppm) to a
31
tetramethylsilane external reference. P NMR spectra in ppm are referenced to an 85%
H3PO4 standard. Positive numbers indicate downfield shift.

Magic-angle spinning NMR

When NMR spectroscopy is applied to a solid sample, it is necessary to use a more

sophisticated description of the interactions between two nuclei. For this, a Hamiltonian can
be expressed as the sum of the different spin interactions:

2 ⋯ (2.12)

45
Chapter 2

Where the first two terms of the equation are equivalent to Eq. (2.10) and the two last terms
describe the scalar, through-bond or indirect, and dipolar, through-space or direct, coupling
between two spins. In the case of high field approximation, the chemical shift term HCS can be
written in a vector form that takes into account the orientation of the nuclei in respect to the
magnetic field:

(2.13)

1
3 cos 1 (2.14)
3
, , , ,

The shielding tensor, σzz, can be divided into isotropic and anisotropic terms, where θ is the
orientation angle to B0. In solution, only isotropic chemical shifts are observed due to
molecular tumbling; however, orientation-dependent effects are strongly dominating in a fixed
lattice.
To suppress, or at least reduce, these effects, the sample is oriented at the “magic angle”
(θ = 54,74°) such as 3cos2(θ)-1 = 0 and spun rapidly. Similar effects are achieved in the other
orientation-dependent terms of Eq. (2.12), thus allowing the measurement of dipolar nuclei in
a quasi-isotropic fashion.[253]

In this work, 1H, 27

Al and 31
P MAS NMR spectra were recorded in a Brucker AVANCE
AMX-400 spectrometer or a Brucker 700 spectrometer using a 4 or 2.5 mm zirconium oxide
rotor at 10 or 20 kHz spinning rate, respectively. Prior to measurement, spectra were
calibrated using NH4H2PO4, NH4Al(SO4)2 and adamantane external references. Chemical shift
conventions were applied as described above for NMR spectroscopy in solution.

Cross-polarization MAS-NMR

Cross-polarization (CP) is a pulse technique that is frequently used for the measurement of
low-sensitivity nuclei. It is particularly useful for solid materials containing carbon and
nitrogen, for which the recording of a spectrum with satisfying signal-to-noise ratio may take
days to weeks. A first π/2 pulse is applied to a high sensitivity nucleus, I (usually 1H),
followed by a double irradiation period (Figure 2.3). During the irradiation time, spin-lock
fields are applied to both nuclei, so as to match the Hartmann-Hahn condition:[254]

, , (2.15)

In this condition, magnetization transfer from I to S happens through heteronuclear dipolar

coupling, where I = 1H and S = 13C in the present work. This has the advantage of enhancing
the signal from the low-gamma nucleus S proportionally to γI /γS. Another advantage is that

46
 Methods

the recycling of the magnetization is only dependent on T1,I, which is in the ms to s range in
the case of 1H and allows for rapid spectra measurements. A possible drawback from this
method is that magnetization transfer will occur at different rates depending on the nature
and proximity of I and S. In the case of a carbon-based material such as MOFs, hydrogen-
rich functional groups such as -CH2 and -CH3 will have enhanced signals compared to
aromatic carbons. This method should therefore not be applied when quantitative data are
required.

Figure 2.3 General principle of the cross-polarization pulse sequence for the high- (I) and
low- (S) sensitivity nuclei. After a π/2 pulse, a spin-lock field is applied to both nuclei to
achieve Hartmann-Hahn condition. The nucleus S is then measured while decoupled from I.

2.5.4 Nitrogen adsorption experiments

Adsorption experiments measure the uptake of gas physisorbed on a solid material at the
equilibrium pressure for a given temperature. The collection of uptake values across
a pressure range, expressed in relation to the saturation pressure, p°, represents the
adsorption isotherm of a material. Nitrogen isotherms are often used to describe the pore
system of solid supports such as zeolites, silicates, and MOFs. The IUPAC has classified the
different isotherm types in six groups that reflect the different layout of pore systems inside a
material (Figure 2.4).[255]

Figure 2.4 Physisorption isotherms as defined by the IUPAC classification. Adapted from
ref.[255]

47
Chapter 2

In the case of MOFs, Type I, II and IV are commonly encountered. Type I is characterized by
a steep gas uptake a low partial pressure that is attributed to the filling of micropores. Type II
represents the case of non-microporous and macroporous materials for which adsorption is
not limited to a single layer of gas molecules. Type IV best describes mesoporous materials
for which the filling of the pores leads to the condensation of gas molecules at high partial
pressure, characterized by the presence of hysteresis between the adsorption and the
desorption isotherms.

In a typical nitrogen sorption experiment, 60 mg of material was placed in a glass tube and
thermally activated under reduced pressure for 24 h. The sample was quickly transferred to a
Micromeritics Tristar II 3020 and adsorption and desorption isotherms were recorded at
77 K.

Brunauer-Emmett-Teller surface area

The surface area was calculated from the nitrogen sorption experiments using the Brunauer-
Emmett-Teller (BET) adsorption model.[256] The BET model describes the physisorption
process onto a homogeneous surface, for which the first layer of molecules can be used as
substrate for subsequent adsorption. The two main assumptions in this model are 1) the
adsorption enthalpy is constant for each layer, with the exception of the first layer adsorbed
on the surface, and 2) an infinite amount of layers are formed when p→p°. Although this
model represents a simplification of the adsorption process onto a heterogeneous support, it
is widely used in its linearized form to calculate a specific surface area, SBET.

1 1
∙ (2.16)
° ° °

(2.17)

Where Q is the measured specific gas uptake at the relative partial pressure, p/p°. The BET
monolayer capacity, nm, and the BET constant, C, are then used to calculate the surface area
using Avogadro’s number, N, and the molecular cross-sectional area, σ (0.162 nm2 for
nitrogen). Due to the required linear relationship between p/Q(p°‐p) and p/p°, the model is
applied to a relative pressure range between 0.05 and 0.3.

2.5.5 Thermogravimetric analysis

MOF samples (typically 5-10 mg) were placed inside an alumina holder and the
thermogravimetric analysis was performed under nitrogen atmosphere (50 mL/min) using a
Mettler Toledo TGA/SDTA 851 balance equipped with a TSO800GC1 gas control, a
TSO801RO sample robot, a platinum pan and heated at a rate of 10 °C/min from 30 °C
to 800 °C.

48
 Methods

2.5.6 Liquid chromatography

In a typical digestion experiment, a MOF sample (1 mg) was dissolved in a NaOH solution
(400 μL 1 M) at 80 °C. Ethanol (400 μL) and phosphoric acid (60 μL, 85% w/w) were added
to the mixture, which was then filtered and analyzed in a Waters Acquity UPLC
H-Class system equipped with a photodiode array detector and a Waters BEH C18
(1.7 μm  1.0 mm  150 mm) column.

2.5.7 Infrared spectroscopy

Prior to measurement, samples (typically 2-3 mg) were mixed with KBr (30 mg) and pressed
into pellets. Infrared spectra were recorded in the range of 4000-400 cm-1 on a PerkinElmer
2000 FTIR spectrometer.

2.5.8 Electron microscopy

Transmission electron micrographs and electron-dispersive X-ray spectra were obtained

from a Hitachi aberration-corrected dedicated scanning transmission electron microscope
(HD-2700CS, cold FEG, 200 kV).

2.5.9 Elemental analysis

In a typical preparation, the sample (~20 mg) was dissolved in aqua regia (nitric and
hydrochloric acids, 1:3 v/v) for 24 h. Then the solution was placed in an ultrasonic bath for 1 h
before analysis. Atomic emission spectroscopy was performed on a Varian SprectrAA 220 FS
spectrometer. A commercial ruthenium standard solution (1000 ± 5 ppm) was used for
calibration.

49
 Chapter 3

Synthesis and characterization of P-MOFs

Based on:

F. L. Morel, M. Ranocchiari, J. A. van Bokhoven, Ind. Eng. Chem. Res. 2014, 53, 9120–9127

F. L. M. contribution: synthesis, characterization, write-up

Chapter 3

3.1 Introduction

Phosphine ligands are commonly encountered in homogeneous catalysis due to their strong
affinity for transition metals.[257–261] In addition to their ligating properties, organophosphines
are among the functional groups with the highest chemical versatility, allowing fine-tuning
of their steric and electronic properties to impact the catalytic behavior of the metal
center.[262–267] Metal-organic frameworks containing phosphine moieties (P-MOFs) were
originally investigated as an alternative to carboxylic acids, using their metal-binding
properties to form stable frameworks built around the phosphine-metal
[203–205]
coordination (Figure 3.1, a). Following this trend, Lin et al. proposed to use a
2,2’-bis-(diphenylphosphino)-1,1’-binaphtyl organic linker together with a ruthenium complex
and a zirconium cluster to form a framework with phosphine-ruthenium precursors. They
showed that the material was available for metal-catalysis, despite a lack of long-range
order.[215] The first highly-ordered MOF pertaining a free organophosphine linker, PCM-1, was
reported by Humphrey et al.[213] The carboxylate groups on 4,4’,4’’-phosphinetriyltribenzoic
acid (ptbc) were coordinated to zinc to form octahedral zinc-oxygen-carbon clusters, yielding
a framework with free phosphorus-containing groups that proved unstable upon solvent
removal. Since then, the use of tri(para-carboxylated) triphenylphosphine and phosphine
oxide linkers has led to more robust materials allowing for catalytic applications;[207,210,211]
however, their potential as supports for metal complexes remains largely unexplored.

A common feature for many existing P-MOFs is that they are synthesized from tridentate
organic linkers centered on the phosphorus atom (Figure 3.1, b). In this pyramidal geometry,
the phosphorus atom occupies a position that strongly determines the framework structure
and stability depending on the relative angle of the phosphorus atom with its substituent. This
limits the scope of phosphine groups and hampers the possibility of fine-tuning their steric
and electronic properties. Organophosphine linkers based on linear dicarboxylic acids do not
suffer from these limitations, as the phosphine moiety is only pendant to the rigid connection
of the framework (Figure 3.1, c). With this in mind, we recently reported a series of
zinc-based MIXMOFs built on 2-(diphenylphosphino)-[1,1’-biphenyl]-4,4’-dicarboxylic acid
(PPh2-bpdc) and its related oxide, which crystallize in topologies related to IRMOF-9.[76,217,218]
Although these materials presented interesting structural and organo-catalytic properties,
metal-supported applications remained scarce due to their relative chemical sensitivity, in
particular towards protic solvents, and to the restricted pore volume induced by framework
interpenetration. This chapter describes the synthesis of alternative organophosphine linkers
based on linear dicarboxylates, with a particular focus on 2-(diphenylphosphino)terephthalic
acid (PPh2-bdc), and presents the phosphine-functionalized frameworks based on commonly
reported topologies known for their high porosity and chemical resilience.

52
 Synthesis and characterization of P-MOFs

Figure 3.1 Selected modes of coordination for phosphine-functionalized linkers: (a)

phosphine coordinated to the inorganic unit and free phosphine with (b) tridentate and (c)
bidentate modes.

3.2 Design of novel P-linkers

Organic linkers based on linear dicarboxylic acids are commonly reported for a variety of
framework topologies.[29] The use of functionalized bdc, and to some extent bpdc, have been
reported for a variety of pendant moieties, including amines, halogens, alkyls, and
alcohols.[58,72,268] The synthesis of linear phosphine-functionalized linkers remains however an
exception with only few reported examples.[76,206] In this section, we describe a synthetic route
to PPh2-bdc and discuss its implications for the synthesis of phosphine-functionalized linkers
based on longer chains.

3.2.1 2-(Diphenylphosphino)-terephthalic acid

2-(Diphenylphosphoryl)terephthalic acid (POPh2-bdc) represents the first step towards the

synthesis of a phosphine-functionalized terephthalic acid. Several direct P-C coupling
reactions were initially investigated, considering iodo- or bromo-substituted terephthalic acid
as starting substrates. Palladium-catalyzed couplings of diphenylphosphine onto halogenated
arenes containing carboxylic acid were reported by Stelzer et al.[269] Sun and co-workers
described the coupling of 3-iodo-4-(methoxycarbonyl)benzoic acid with diphenylphosphine
and Pd(OAc)2 in the presence of triethylamine.[270] Despite our efforts, no conversion was
observed when halogenated terephthalic acids were used in the presence of
diphenylphosphine and Pd(OAc)2. Rummelt et al. reported a P-C coupling strategy of
diphenylphosphine oxide with halogeno-arenes using microwave irradiation.[221]

53
Chapter 3

Coupling of diphenylphosphine oxide with the iodo- or bromo-substituted terephthalic acid

was attempted; however the reaction exclusively yielded terephthalic acid as dehalogenation
product. Presumably, the carboxylate group in the ortho-position sterically hindered
palladium-mediated reactions.

Since the direct synthesis of PPh2-bdc and POPh2-bdc from halogenated terephthalate
precursors was not successful, we considered a multi-step approach instead. A P-C coupling
catalyzed by Pd/C in water under microwave irradiation of 3-iodo-4-methylbenzoic acid with
diphenylphosphine oxide afforded the desired 3-(diphenylphosphoryl)-4-methylbenzoic acid
in 91% yield (Figure 3.2). Oxidation of 3-iodo-4-methylbenzoic acid with KMnO4 in water
produced the desired diphenylphosphoryl terephthalic acid, adapting a procedure reported by
Meek et al. for the oxidation of 2-fluoroterephthalic acid.[271] Advantageously, this preparative
procedure for POPh2-bdc was carried out in water, did not require anoxic conditions, had no
major by-products, and involved simple purification steps. This makes it highly
environmentally friendly, as well as only route to POPh2-bdc so far.

Figure 3.2 Two-step synthesis of POPh2-bdc involving a microwave-assisted P-C coupling

with diphenylphosphine oxide.

Combined high resolution mass spectrometry (HRMS) and nuclear magnetic resonance
(NMR) spectroscopy analyses were used to confirm the structure of the desired product
(Chapter 2). Crystallization of POPh2-bdc with a mixture of ethanol and water led to the
formation of colorless crystals suitable for single-crystal X-ray diffraction (XRD).
The structural analysis (Figure 3.3) revealed that the ligand co-crystallizes with two molecules
of water (refined as O(6) and O(7)). The unit cell was triclinic with the following dimensions:
a = 9.2111(7) Å, b = 9.9850(8) Å, c = 10.5450(8) Å, α = 95.976(1)°, β = 90.822(1)°,
γ = 105.171(1)° (R = 3.85%). The refined structure indicated that the POPh2-bdc
ligand adopted a tetrahedral geometry around the phosphorus atom, as expected for
triarylphosphines.[272]

54
 Synthesis and characterization of P-MOFs

Figure 3.3 Representation of POPh2-bdc co-crystallized with water molecules after

refinement of the XRD crystal structure (296 K, thermal ellipsoids at 30%, hydrogen atoms
attached to carbons are omitted for clarity). C atoms are displayed in gray, O in red and P in
orange. R-factor = 3.85%.

An additional step was needed to reduce the phosphine oxide group on POPh2-bdc into its
phosphine derivative PPh2-bdc to ultimately access the desired phosphine-functionalized
MOFs. Our experience in reduction of arylphosphines bearing carboxylic functions led us to
initially use trichlorosilane, a well-known reducing agent reported for its compatibility with
carboxylic acids.[76] Experimental conditions with a three- to fifteen-fold excess of
trichlorosilane were tested with systematic degradation of POPh2-bdc into oligomeric
structures. We hypothesize that this incompatibility may again arise from the presence of the
carboxylic acid in ortho-position. This type of reaction occurs via a Si···O···P···H
intermediate[273] and the relatively close distance between the two oxygen atoms
(O(1)-O(2) = 2.919 Å, Figure 3.3) may interfere with the standard reaction pathway. Residual
water molecules may also play a disruptive role despite that extensive drying of the starting
material was performed prior to reduction.

Alternative reduction procedures were subsequently screened. Lithium aluminum hydride[274]

and diisobutylaluminum hydride[275] were discarded due to their known incompatibilities with
carboxylic acids. Less conventional methods, including the use of copper catalysts[276] and
the mild reducing agent diphenylsilane,[277] yielded no success with our substrate. Recently,
Beller et al. reported a procedure to reduce functionalized phosphine oxides that we
successfully adapted for POPh2-bdc.[278] Reduction in the presence of methyldiethoxysilane
and diphenylphosphate as catalyst in toluene at 110 °C for 72 h produced the desired
phosphine PPh2-bdc in 61% isolated yield (Figure 3.4).

55
Chapter 3

Slow oxidation of the phosphine was observed in solution and the final product usually
contained a low (< 5%) amount of phosphine oxide. The structure was confirmed by HRMS
and NMR spectroscopy. Although this reduction was slow compared to other procedures, the
high selectivity and yield towards the product open new perspectives in the synthesis of
PPh2-bdc in large scale. A metal-free reduction also prevents the presence of metal
contaminants during the MOF synthesis.

Figure 3.4 Reduction of POPh2-bdc to the corresponding organophosphine PPh2-bdc.

3.2.2 2-(Diphenylphosphoryl)-[1,1’:4’,1’’-terphenyl]-4,4’’-dicarboxylic acid

Following the P-C coupling strategy used in POPh2-bpdc,[76] the synthesis

of 2-(diphenylphosphoryl)-[1,1’:4’,1’’-terphenyl]-4,4’’-dicarboxylic acid (POPh2-tpdc) was
attempted by successive Suzuki coupling reactions of 1,4-benzene diboronic acid with
4-iodobenzoic acid and 3-bromo-4-iodobenzoic acid, followed by a palladium-mediated P-C
coupling with diphenylphosphine oxide under microwave irradiation (Figure 3.5). Product
formation was confirmed by HRMS, although the difficult separation from by-products did not
allow for the accurate characterization of the product by NMR spectroscopy. Further
optimization of the reaction conditions was not attempted. Nevertheless, we demonstrated
that the scope of the P-C coupling strategy using Pd/C under microwave irradiation can be
extended to the synthesis of other phosphine-functionalized linkers.

Figure 3.5 Synthesis of POPh2-tpdc with two successive Suzuki couplings and a microwave-
assisted P-C coupling with diphenylphosphine oxide catalyzed by Pd/C.

56
 Synthesis and characterization of P-MOFs

3.3 Synthesis of P-MOFs with MOF-5 topology

Developed by Yaghi et al., the MOF-5 framework series is often reported for its high
tolerance towards functional group incorporation and as a textbook example of isoreticular
chemistry.[31,72] PPh2-bdc was initially used herein together with bdc to produce a framework
with the MOF-5 topology. We diluted the bulky diphenylphosphino functional groups to
prevent an excessive framework strain and pore crowding, thereby leaving enough free
space to accommodate metal complexes. The initial synthesis of LSK-11 was performed with
PPh2-bdc and bdc linkers, in a molar ratio of 1:2 in DMF under solvothermal conditions, at
95 °C for 24 h. The light-yellow cubic crystals of LSK-11 were washed repeatedly with fresh
solvent before characterization. The single-crystal X-ray diffraction pattern at 230 K
(Figure 3.6) revealed that the material possesses a cubic unit cell (a = 25.65 Å) refined within
a Fm(-3)m space group (R = 7.88%), comparable to the MOF-5 topology.[31] In contrast to
LSK-3 (IRMOF-9-PPh2), the phosphine groups were not preferentially localized in any unit
cell dimension.[76] Instead, they were randomly distributed across all possible linker positions,
which, in addition to dilution effects, is responsible for the localized disorder that forbids the
refinement of the phosphine group inside the crystal structure. Nevertheless the recorded
powder XRD pattern confirmed the single-crystal structure of the material (Figure 3.7, left)
and its strong similarity with other representative of the IRMOF-1 series.[31,72]

Figure 3.6 View along the b-axis of the crystal structure of LSK-11 based on single-crystal
XRD measurements at 230 K. Gray = carbon, red = oxygen, blue = zinc. Thermal ellipsoid set
at 30%. R-factor = 7.88%. Gray box represents the unit cell boundaries.

57
Chapter 3

Figure 3.7 Characterization of LSK-11: (left) powder XRD patterns; (right) nitrogen
adsorption isotherm at 77 K.

Prior to nitrogen sorption experiments, LSK-11 was dried using supercritical CO2,[279] followed
by an activation period in vacuum at 120 °C. Nitrogen adsorption experiments indicated that
LSK-11 had a BET surface area of 2320 m2/g, (Figure 3.7, right) comparable to functionalized
materials of MOF-5 topology.[72] We digested the sample in a mixture of DCl / d6-DMSO to
quantify the amount of phosphorus groups present inside the framework by 1H NMR
spectroscopy. The spectrum revealed a PPh2-bdc to bdc ratio of 2:7 (Figure 3.8, left). We
also observed that the organophosphine linker oxidized under digestion conditions.
31
A P magic-angle spinning (MAS) NMR spectrum of LSK-11 was recorded to confirm the
nature of the phosphorus group in the solid state (Figure 3.8, right). Two distinct signals
associated with phosphine (A) and phosphine oxide (B) groups were observed at -6 ppm and
+37 ppm, respectively. These signals are consistent with reported values in P-MOFs and
other phosphine-containing materials.[213,280]

Figure 3.8 Characterization of LSK-11 by NMR spectroscopy: (left) 1H NMR spectrum after
digestion with DCl / d6-DMSO; (right) solid-state 31P MAS NMR spectrum recorded at 20 kHz.

58
 Synthesis and characterization of P-MOFs

Differences between solid- and liquid-state NMR spectroscopy, as well as alternative

methods for the quantification of the phosphine to phosphine oxide ratio (P/P=O), will be
discussed in detail in Chapter 4. Oxidation of PPh2-bdc to the corresponding phosphine oxide
is believed to occur in contact with residual oxygen in solution during the solvothermal
synthesis of LSK-11, as is often encountered for phosphine ligands in solution.
Triarylphosphines are known for their resistance towards oxidation, particularly in their solid
state forms.[281] LSK-11 resisted oxidation when stored under an inert atmosphere for a
prolonged period of time. Slow oxidation occurred upon exposure to air, the more so if
suspended in aerated solvents. Phosphine oxidation reduces the number of coordination
sites available for metal nuclei, but does not negatively impact the framework crystallinity.
1
H-13C cross-polarization (CP) MAS NMR spectrum revealed three main carbon signals
(Figure 3.9, left). Signals at +129 and +136 ppm correspond to the presence of the
terephthalic acid linker. The signal at +175 ppm, related to the carbon on the carboxylate
group, is approximately 5 ppm upfield from the free linker in solution, which confirms the
formation of the zinc-oxygen bond. Specific signals from the phosphine linker are difficult to
discriminate from the terephthalic acid due to the intrinsic broadening of the signals in
solid-state NMR spectroscopy,[282] as well as the relative dilution of the linker within the
material. The thermal stability of LSK-11 was determined using thermogravimetric analysis
after the same activation procedure as described for nitrogen physisorption experiments. The
analysis indicated that virtually no residual solvent remained within the pores of the material
and no mass loss was observed below 480 °C (Figure 3.9, right). The latter was assigned to
the carbonization of the organic linkers and the collapse of the framework, which is
comparable with reported values for zinc-MOFs.[31]

Figure 3.9 Analysis of LSK-11: (left) 1H-13C CP MAS NMR recorded at 10 kHz, spinning
sidebands are denoted with (*); (right) thermogravimetric analysis, 10 °C/min, 50 mL/min N2.

59
Chapter 3

In a second step, the ratio between PPh2-bdc and bdc linkers was varied to evaluate how the
amount of phosphine incorporated inside LSK-11 was affecting the physical properties of the
material (Table 3.1, 1a-e). We observed that a PPh2-bdc to bdc ratio up to 2:1 did not reduce
the crystallinity of the material (1a), although a reduced BET surface area and pore volume
were noted due to the increased presence of the bulky phosphine ligand inside the pore
system. PPh2-bdc to bdc ratios lower than the original conditions (1b) were difficult to
synthesize, with a new amorphous phase starting to appear in the crystallization liquor with
ratios below 1:3 (1d). PPh2-bdc to bdc ratios below 1:5 (1e) did not yield any crystalline
material.

The synthesis of MIXMOF-5-NH2-PPh2 (LSK-13) was also attempted using a procedure

adapted from the synthesis of IRMOF-3.[223] PPh2-bdc and NH2-bdc were mixed in a 1:12
molar ratio in DMF at 95 °C for 24 h to yield yellow cubic crystals. LSK-13 exhibited a similar
nitrogen uptake as compared to LSK-11, with a BET surface area of 1606 m2/g and a pore
volume of 0.8 cm3/g. The main difference between the two P-functionalized frameworks is
their relative sensitivity to air exposure. The crystallinity of the dried LSK-13 decreased
quickly upon air exposure, limiting its characterization by X-ray diffraction.

Table 3.1 Summary of physical parameters of attempted syntheses of P-MOFs with MOF-5
topology.

P/R-bdc Surface area pore volume

Name Phase
ratio (m2/g) (cm3/g)
1a 2:1 crystalline 1490 0.66
1b 1:1 crystalline 2060 0.91
[a]
1c LSK-11 1:2 crystalline 2320 0.98
1d 1:3 mixed n/a n/a
1e 1:5 amorphous n/a n/a
1f LSK-13[b] 1:12 crystalline[c] 1606 0.80
[a]
R = H, [b]R = NH2, [c]dry single-crystals decomposed under exposure to air.

3.4 Synthesis of P-MOFs with UMCM-1 topology

UMCM-1 is a well-studied alternative topology developed by Matzger et al. based on zinc-

carboxylate bonds.[39] The framework structure [Zn4O(H3btb)4/3(bdc)]n has been featured in
seminal works in post-synthetic modifications[73,125,283] due to its large pores composed of
cages surrounding mesoporous-like channels of 32 Å. It is also a good candidate for
quantitative or partial linker substitution at the bdc linker position.[284] A phosphine-
functionalized UMCM-1 was synthesized in DMF by replacing partly the bdc linker with
PPh2-bdc, while maintaining the ratio of linear linkers to H3btb equal to the original synthesis
procedure. The mixture was heated at 85 °C for 48 h and produced pale-yellow

60
 Synthesis and characterization of P-MOFs

needle-shaped crystals (LSK-17). Comparison of the LSK-17 powder X-ray diffraction pattern
with the pattern of UMCM-1 simulated from the single-crystal XRD pattern confirmed that
LSK-17 possesses with the desired parent topology (Figure 3.10, left) despite notable
differences in Bragg peak intensities. Disparities with the simulated pattern were also
observed in the original material,[39] as well as in functionalized framework variants,[284] for
which LSK-17 resembles the most. Nitrogen physisorption isotherm (Figure 3.10, right)
confirmed the permanent porosity of the framework, with a BET surface area recorded at
2330 m2/g and a pore volume of 1.22 cm3/g. Despite a lower nitrogen uptake for LSK-17, as
compared to the parent material,[39] the shape of the isotherm displayed characteristic traits of
UMCM-1 with a sharp increase at p/p° = 0.2 corresponding to the filling of the mesoporous
channels. However, the relative uptake at a relative pressure p/p° superior to 0.2 is very
similar in both cases. This effect could be attributed to the presence of phosphine groups
partially blocking the microporous pores of the cages, while leaving the larger channels
relatively unhindered.

Figure 3.10 Characterization of LSK-17: (left) powder XRD patterns and (right) nitrogen
physisorption at 77 K.
31
P MAS NMR spectroscopy confirmed the presence of free phosphine groups and the sum
of A1 and A2 signals, at -2 and -4 ppm respectively, accounted for 88% of the total signal
(Figure 3.11). Two minor impurities were found at +1 and +4 ppm, as well as phosphine oxide
at +40 ppm (Figure 3.11, B). Similarly to LSK-11, the relatively sharp phosphorus signals are
a consequence of the high order in this material. The linker composition in LSK-17 was
determined by 1H NMR of a digested sample (Figure 3.11, left). The PPh2-bdc to bdc ratio
was measured at 1:7.3 and the total linear linker to H3btb ratio at 1:1.1. The latter was below
the literature value of 1:1.3, but still fell within the optimal range for UMCM-1 synthesis.[39]
A higher amount of linear linker within the structure may indicate residual impurities inside the
pore system of LSK-17, as observed by 31P MAS NMR spectroscopy.

61
Chapter 3

Figure 3.11 NMR spectroscopy of LSK-17: (left) 1H NMR of digested sample and (right) 31P
MAS NMR spectrum recorded at 20 kHz on a 700 MHz magnet.

3.5 Synthesis of P-MOFs with MIL-101 topology

In contrast to the MOF-5 and UMCM-1 topologies, MIL-101 can be formed from different
trivalent metal ions such as chromium,[63,285] aluminum[57] and iron.[49] The MIL-101 structure is
well-known for its chemical resilience, as well for its large pore system and high surface area,
suitable for a large range of applications,[63,268,286,287] including catalysis.[64,127,137,138,288–290]
We incorporated phosphine functional groups into MIL-101(Cr) and MIL-101(Al)-NH2.
The synthesis of MIXMIL-101(Cr)-PPh2 (LSK-12) was adapted from the HF-free synthesis of
MIL-101(Cr) reported by Hatton et al.[224,225] LSK-12 was synthesized by combining PPh2-bdc
and bdc organic linkers (1:6 molar ratio) in an aqueous solution of tetramethylammonium
hydroxide (TMAOH) at 210 °C for 24 h. The powder X-ray diffraction pattern of LSK-12
(Figure 3.12, left) showed that the resulting material was highly crystalline and possessed a
pattern similar to that of MIL-101(Cr).[224] LSK-12 displayed the same thermal and chemical
stability as MIL-101(Cr) and was washed with ethanol and activated at 150 °C without
apparent loss of crystallinity. Nitrogen physisorption confirmed the high porosity of the
material with an isotherm shape characteristic of the MIL-101 topology (Figure 3.12, right).
The BET surface area (3020 m2/g) and pore volume (1.47 cm3/g) were lower than that of the
phosphine-free material, an observation that was repeatedly confirmed for all the P-MOFs
presented in this work.

62
 Synthesis and characterization of P-MOFs

Figure 3.12 Characterization of LSK-12: (left) diffraction pattern and (right) nitrogen
physisorption isotherm at 77 K.

A sample of LSK-12 was digested in a solution of sodium hydroxide in ethanol, titrated with
phosphoric acid, and analyzed by ultrahigh pressure liquid chromatography (Chapter 2). This
revealed that PPh2-bdc occupied approximately 5% of the total organic linker positions
available. Moreover, NMR spectroscopy of a digested sample in a mixture of DCl/d6-DMSO
did not yield any quantitative linker ratio due to the paramagnetism of Cr(III). As in LSK-11,
we investigated the effect of the PPh2-bdc to bdc ratio on the material crystallinity and
porosity. Materials synthesized with PPh2-bdc to bdc molar ratios of 1:2 and 1:1 were
compared with the original material (molar ratio of 1:6, vide supra). Although the powder
X-ray diffraction patterns showed that all materials were crystalline (Figure 3.13, left), the
BET surface area decreased significantly from 3020 m2/g for the 1:6 ratio to 1860 m2/g and
835 m2/g for 1:2 and 1:1, respectively (Figure 3.13, right). All isotherms presented the
characteristics of MIL-101 pore system. We did not observe the formation of a secondary
amorphous phase, in contrast to LSK-11. This suggests that the large cages of MIL-101 can
accommodate multiple phosphine groups without framework strains due to the large free
space available.

63
Chapter 3

Figure 3.13 Comparison of the physical properties of LSK-12 synthesized with different
PPh2-bdc to bdc ratios. (left) Powder XRD patterns and (right) nitrogen physisorption
isotherm of 1:6 (●), 1:2 (□) and 1:1 ( ) ratios.

To further extend the scope of the PPh2-bdc linker towards different SBUs, we attempted to
add phosphine groups to the MIL-101(Al)-NH2 framework. MIL-101(Al)-NH2 represents an
interesting alternative to MIL-101(Cr), as it contains amino functional groups that could
potentially be modified post-synthetically.[89] The synthetic route reported by Hartmann and
Fischer[64] proceeds under mild conditions that help preventing the formation of phosphine
oxide during the synthesis. The framework contains Al(III) centers that are not paramagnetic,
thus allowing characterization by NMR spectroscopy.[57] Unlike LSK-12, the synthesis was
performed in an open vessel under a stream of argon. A solution of aluminum nitrate
hexahydrate in DMF was added progressively during 2 h into a solution of PPh2-bdc and
NH2-bdc in DMF at 110 °C (Chapter 2). After 18 h, we recovered the pale yellow crystalline
material MIXMIL-101(Al)-NH2-PPh2 (LSK-15). The analysis of the X-ray diffraction patterns
confirmed that LSK-15 has the same structure as MIL-101(Al)-NH2 (Figure 3.14, left).
Nitrogen uptake and BET surface area (2690 m2/g) of LSK-15 were lower than LSK-12, but
comparable to MIL-101(Al)-NH2.[64] 1H NMR spectroscopy of the digested sample showed
that the PPh2-bdc linker occupied approximately 12% of the organic positions in the material.

The synthesis of MIXMIL-101(Al)-PPh2 (LSK-14) was attempted by using PPh2-bdc and bdc
organic linkers in similar synthesis conditions as for LSK-15. The resulting material displayed
only partial crystallinity with broad reflections in the diffraction pattern (Figure 3.14, left) that
are more related to MIL-53(Al).[291] The nitrogen uptake of LSK-14 also contrasted strongly
with LSK-15 and displayed a shape related to a type I isotherm, without the cage filling
behavior in the range of p/p° = 0.1-0.2, characteristic of the MIL-101 topology.[63] BET surface
area (907 m2/g) and pore volume (0.47 cm3/g) also felt within the reported range for
functionalized MIL-53(Al).[55] Despite our optimization efforts, no crystalline phase was
isolated.

64
 Synthesis and characterization of P-MOFs

During the synthesis, we noticed a strong dependency of the water content in the synthesis
mixture on the crystallinity of the resulting material. These findings were correlated to
MIL-101(Al)-NH2 and LSK-15, where a small addition of water (from 0.5 to 1 equivalent to
aluminum) was strongly detrimental to the framework crystallinity. Higher amounts did not
yield any crystalline materials. These observations are in agreement with the recent work
from Goesten et al. on MIL-101(Al)-NH2, which demonstrated that framework formation
proceeds through a dissolution-recrystallization of the MIL-235(Al)-NH2 phase into the desired
topology, promoted by an H-Cl-DMF adduct.[292] Thus it is likely that water, even in
stoichiometric amount, plays an interfering role at this stage of the synthesis.

Figure 3.14 (left) Powder XRD patterns of LSK-15 and related phases and (right) nitrogen
physisorption isotherms at 77 K of LSK-15 (●), MIL-101(Al)-NH2 (□), and LSK-14 ( ).

The 31P MAS NMR spectrum of LSK-15 revealed three distinct signals at +35, -6 and -10 ppm
(Figure 3.15, left). Signals A1 at -6 ppm and A2 at -10 ppm, respectively accounting for
approximatively 85% of the total phosphorus sites, were attributed to the free phosphine sites
and signal B to the phosphine oxide. The 1H-13C CP MAS NMR spectrum confirmed the
presence of amino groups at +150 and +116 ppm and carboxylate-aluminum linkage at
+172 ppm (Figure 3.15, right). It is however difficult to directly observe the signals arising
from the phosphine groups due to the relative dilution of the linker and the presence of
multiple overlapping carbon signals in the aromatic region (i.e. from +140 to +120 ppm).
27
The framework integrity of LSK-15 was confirmed by Al MAS NMR spectroscopy
(Figure 3.15, middle), which showed octahedral alumina appearing as a lone signal at -0.46
ppm, thereby ruling out instances of P-O-Al and tetrahedral aluminum coordinations.[293]

65
Chapter 3

Figure 3.15 Solid-state NMR spectra of LSK-15: (left) 31P with P(III) (A1 and A2) and P(V) (B)
environments; (middle) 27Al and (right) 1H-13C CP MAS NMR. Spinning sidebands are
signaled by (*) and residual solvent contributions by (s).

Thermogravimetric analysis of the two P-functionalized MIL-101 frameworks indicated the

presence of residual solvent molecules within the pore. The recorded mass loss of LSK-12
and LSK-15 (Figure 3.16) followed a more complex profile than LSK-11, with a slower
transition above 300 °C. In the case of LSK-15, the transition temperature is more difficult to
assess precisely due to a continuously decreasing thermogram. This behavior has been
already reported in the literature for MIL-101(Al)-NH2 and was attributed to the desorption of
water molecules, followed by DMF.[57]

Figure 3.16 Thermogravimetric analysis of selected P-MOFs with MIL-101 topology: (left)
LSK-12 and (right) LSK-15. Heating ramp 10 °C/min, 50 mL/min N2.

66
 Synthesis and characterization of P-MOFs

3.6 Conclusion

In this chapter, we synthesized a series of P-MOFs based on a rigid terephthalic acid

derivative bearing a diphenylphosphine moiety. Our multi-step approach produced the
PPh2-bdc linker in high yield, without the use of expensive catalysts and tedious separation
procedures. This modular method, involving sequential oxidation of 3-(diphenylphosphoryl)-4-
methylbenzoic acid and reduction of the phosphine oxide group, creates future opportunities
to generate finely-tuned phosphine sites on a broader range of terephthalic acid derivatives.
PPh2-bdc was successfully used to prepare new structures based on MOF-5, UMCM-1 and
MIL-101 topologies; thus bridging the gap between these well-studied structures and MOFs
containing phosphine moieties. All presented materials were synthesized using a mixture of
ligands (bdc or NH2-bdc) to avoid excessive pore hindrance by the bulky diphenylphosphino
moiety and to preserve high crystallinity and phase purity. The P-MOFs all displayed high and
permanent porosity. We also demonstrated that these materials can incorporate different
amounts of phosphine without modification of the synthetic conditions. In pursuit of solid
porous ligands, with high chemical stability (particularly towards protic solvents), we have
investigated organophosphine-containing MOFs with chromium- and aluminum-based
inorganic units. These materials demonstrate potential in single-site catalysis due to their
easy synthesis and the ability of phosphine functional groups to coordinate discrete transition
metal nuclei. Their applications in metal-supported catalysis will be further discussed in
Chapters 5 to 7.

67
 Chapter 4

Characterization of P-MOFs by XAS and MAS NMR

spectroscopy

Based on:

F. L. Morel, S. Pin, T. Huthwelker, M. Ranocchiari, J. A. van Bokhoven, Phys. Chem. Chem.

Phys. 2015, 17, 3326–3331.

F. L. M. contributions: synthesis, characterization, write-up

Chapter 4

4.1 Introduction

Organophosphines are a class of P(III) ligands that features an easily accessible electron
lone pair, which binds to chalcogens and transition metals. Phosphine oxidation however
yields to a P(V) center with a poor ligand ability. Although phosphine oxides are the most
thermodynamically stable form of phosphines, their oxidation is often kinetically limited at
room temperature.[184,294] When considering phosphine-functionalized MOFs, it is therefore
critical to identify and quantify the different phosphorus species present within the framework.
Accessing the oxidation state of phosphine groups in the solid state using non-specific
spectroscopy techniques remains however a challenge for many materials. Powder X-ray
diffraction (XRD) cannot distinguish between phosphine and phosphine oxide, because
phosphine oxidation has often a negligible effect on the framework structure. Infrared (IR)
spectroscopy can be used to identify the P=O stretching bands in phosphine oxide with
relative ease, but there is no characteristic and isolated bond stretching in organophosphines
that allows for their identification.

Element-specific techniques, such as X-ray absorption spectroscopy (XAS) and nuclear

magnetic resonance (NMR) spectroscopy, can be used to overcome these limitations. NMR
spectroscopy is a powerful characterization method,[251] often used for the naturally abundant
31
P isotope.[295–297] However, it may not be suitable when paramagnetic centers such as
Fe(III) and Cr(III) are present in the material, which is the case for MOFs, such as
MIL-101(Cr)[63] and MIL-100(Fe).[62] XAS analysis methods, such as X-ray absorption
near-edge structure (XANES) and extended X-ray absorption fine structure (EXAFS), are
widely used techniques the for characterization of solid catalysts (see Chapter 2).[244,298] XAS
applied to MOFs has already provided useful structural insights and can be used to
complement other techniques such as XRD, IR spectroscopy, NMR spectroscopy, and
neutron diffraction.[299] Many studies were conducted to specifically elucidate the local
environment of the MOF inorganic unit. Prestipino et al. performed a comparative XANES
and EXAFS study at the Cu K-edge on the solvated and dehydrated forms of HKUST-1 to
determine the change in local Cu2+ coordination.[300] Valenzano et al. detected local
deformations in the desolvated UiO-66 structure using EXAFS at the Zr K-edge,
complemented by ab initio calculations.[301]

XAS can be used to probe the interaction of unsaturated inorganic units with guest
molecules, for instance Cr3(1,3,5-benzenetricarboxylate)2[302] and Ni-CPO-27[303] presented
unique spectroscopic features upon sorption with different gases. Finally, XAS is used to
characterize catalytically active metallic species located inside a MOF.

70
 Characterization of P-MOFs by XAS and NMR spectroscopy

Fischer et al. observed Cu and ZnO nanoparticles loaded into MOF-5[304] and our group
previously used XAS at the Au L3-edge to characterize gold complexes immobilized in
the phosphine-functionalized LSK-1.[207] Huang et al. recently reported the EXAFS
characterization of Pt ions incorporated within a series of functionalized UiO-66.[305,306]

The structure of MOF inorganic units has been generally probed by XAS in the hard X-ray
regime (i.e. > 10 keV), while the characterization of the low Z elements present in the organic
linker is usually carried out by 1H liquid NMR spectroscopy and elemental analysis, which
often require the digestion of the samples. Phosphorus represents an interesting case of a
light element featuring a K-edge in the soft X-ray regime (~2145 eV), easily accessible by
synchrotron radiation, as well as a naturally pure isotope (31P) active in NMR spectroscopy.
Throughout this chapter, we demonstrate how the parallel characterization by NMR
spectroscopy and XAS can provide information about the oxidation state and local
coordination of the phosphorus atom in P-MOFs presented in Chapter 3. Synchrotron
radiation was also used to probe electronic properties of transition metal complexes
coordinated into selected P-MOFs.

4.2 Determination of the P/P=O ratio

The synthesis of P-MOFs is commonly performed under solvothermal conditions that require
the dissolution of organic precursors at a temperature between 90 and 210 °C over a period
of 1 to 3 days. This is often detrimental to the stability of the phosphine moieties if trace
amounts of oxygen are present in the system. The accurate characterization and
quantification of the phosphorus functional groups inside the framework are therefore
important to establish the material’s properties, particularly towards metal complexation. Two
P-MOFs based on the MIL-101 topology, namely MIXMIL-101(Cr)-POPh2 (LSK-12) and
MIXMIL-101(Al)-NH2-PPh2 (LSK-15) were first selected to provide a comparison between
non-specific techniques and element-specific spectroscopy for the quantification of phosphine
and phosphine oxide.[307,308]

4.2.1 Infrared spectroscopy

The immediate drawback of IR spectroscopy is the lack of strong characteristic bands for
tertiary arylphosphines. P-C stretching bands are often very weak and present in a frequency
range mostly populated by other C-C aromatic bands, which limits their direct observation.[309]
On the other hand, tertiary arylphosphine oxides typically display a strong P=O stretching
band in the 1320-1140 cm-1 region.[310,311] In the case of the POPh2-bdc linker, a broad band
at 1173 cm-1 was observed alongside C=O and C-O stretching bands at 1702 and 1257 cm-1,
respectively (Figure 4.1). A similar contribution from the POPh2-bpdc linker was observed in
[Zn4O(bpdc)1(PPh2-bpdc)2]n (LSK-3) at 1165 cm-1.[223]

71
Chapter 4

LSK-12 and LSK-15 were also characterized by IR spectroscopy in the solid-state

(Figure 4.1); however their respective framework contribution with multiple vibration bands
arising from C=O, C-O, and C=C stretching makes the identification of the phosphine oxide
groups difficult.

Figure 4.1 IR spectra of POPh2-bdc (top), LSK-12 (middle), and LSK-15 (bottom) at high
(left) and low (right) wavenumber.
31
4.2.2 P NMR spectroscopy

NMR spectroscopy is a routinely-used method to characterize functional groups within MOFs,

and is often performed in the liquid phase after sample digestion in a mixture of deuterated
solvents and acids.[88] In particular, 1H NMR spectroscopy benefits from the high sensitivity
and short relaxation time of protons to provide a quantitative analysis of the organic linkers
present in a MOF with mixed functional groups (MIXMOFs).[88,312] The 1H NMR spectrum of
LSK-15 after digestion revealed that the material has a phosphorus- to nitrogen-containing
31
linker molar ratio of 1:7.3 (Figure 4.2, left). The P NMR spectrum could not be used to
estimate the P/P=O ratio due to phosphine oxidation under digestion conditions, confirmed by
a signal at +31 ppm in the 31P NMR spectroscopy (Figure 4.2, middle). A small triplet signal
(signal i in Figure 4.2, middle) was also observed in the spectrum and is possibly linked to
further decomposition of the linker under digestion conditions. The 1H NMR spectrum further
confirmed complete phosphine oxidation, in which only POPh2-bdc signals were observed.
31
A P MAS NMR spectrum was recorded (Figure 4.2, right) to identify the phosphorus
oxidation state of the phosphine moieties in a non-digested sample. Signals observed
at -10 (A1) and -6 (A2) ppm were attributed to two non-equivalent phosphine positions of the
phosphine inside the framework.[307] 31P MAS NMR spectroscopy at 100 K did not highlight
any significant changes in the ratio of A1 to A2, ruling out rotational effects. A signal was
observed at +35 ppm (Figure 4.2, R3P=O) and was assigned to the oxidized phosphine using
reported literature value.[313]

72
 Characterization of P-MOFs by XAS and NMR spectroscopy

Figure 4.2 Comparison of the LSK-15 characterization by NMR spectroscopy. (left) 1H

(middle) and 31P NMR spectra of a digested sample, (right) 31P MAS NMR spectrum recorded
at 20 kHz on a 700 MHz magnet. Impurities are signaled with (i).

A phosphine fraction of 85 ± 10% was calculated relative to the peak areas of the combined
phosphine sites against that of phosphine oxide. Conditions of sample preparation for MAS
NMR spectroscopy induced virtually no oxidation of LSK-15, as opposed those for the
31
digestion. However, this value should be considered cautiously since the P nucleus is
characterized by a long spin-lattice relaxation time, T1, in the order of seconds to minutes
(typical values for 1H are in the order of hundreds of milliseconds). In order to improve their
signal-to-noise ratio, 31P NMR spectra are commonly recorded with a time delay below the
5T1 criteria for which a nucleus is usually considered fully relaxed.[314] Notable differences in
T1 between phosphorus species may lead in this case to systematic errors, albeit of lower
magnitude than the one caused by sample digestion. The main restricting factor of NMR
spectroscopy remains its poor characterization performance in the presence of paramagnetic
species, such as the Cr(III) centers in LSK-12. For the quantification of the phosphine content
in this type of material, we employed other characterization techniques such as XAS.

4.2.3 X-ray absorption spectroscopy at the P K-edge

XAS measurements of PPh2-bdc and POPh2-bdc linkers at the P K-edge were performed to
provide a spectral reference for the characterization of LSK-12 and LSK-15. A shift in the
white line from 2147.3 to 2150.6 eV is observed in the XANES spectra upon oxidation of the
phosphorus nuclei (Figure 4.3, A and B). Pre-edge resonances (Figure 4.3, AI and BI) are
attributed to the effect of phenyl groups in the higher coordination shells of the
phosphorus.[315] A more pronounced pre-edge feature relative to its white line is observed in
the spectrum of PPh2-bdc, compared that of POPh2-bdc.

73
Chapter 4

Spectral features are not affected equally by the presence of an oxygen atom, with a 2.2 eV
shift for AI to BI features compared to one of 3.3 eV for the A to B white lines. PPh2-bdc
displayed an additional spectral feature, AII, at 2150.3 eV, which coincides with the white line
31
of POPh2-bdc and indicates partial oxidation of the sample in agreement with P NMR
spectroscopy data (< 5%, see Chapter 3).

Figure 4.3 P K-edge XANES spectra of the organic linkers PPh2-bdc and POPh2-bdc (left)
and P-MOFs (right).

Figure 4.3 shows the P K-edge XANES spectra of LSK-15 and LSK-12 and the main spectral
features are summarized in Table 4.1. The spectrum belonging to LSK-15 is characterized by
four different absorption features (Figure 4.3, features C to CIV), which identify the presence
of multiple phosphorus oxidation states. LSK-15 displayed spectral features closely related to
the white lines of the PPh2-bdc (C) and POPh2-bdc (CII) spectra at 2145.7 and 2150.2 eV,
respectively. The relative intensity of CII to C ratio is higher in LSK-15 than the corresponding
AII to A ratio in PPh2-bdc, which evidences of phosphine oxidation within the material. The
CI pre-edge feature is attributed to the characteristic resonance of the phosphine
environment in PPh2-bdc. Fewer features were recorded in the XANES spectrum of LSK-12
(Figure 4.3, D to DIII), as compared to LSK-15. Comparison based solely on the absorption
edge positions of the two organic linkers (Figure 4.3, A and B) is difficult due to a perceived
overlap between A and BI, which could both be contributors to the DI feature. However, the
absence of the corresponding AI pre-edge feature in LSK-12, and to a lesser extent the small
edge shift between C and DI, are indicative of phosphine oxide as the primary phosphorus
species in LSK-12. The relative intensities of the white line D and the pre-edge feature DI also
compare well with POPh2-bdc. A shift in post-edge features of the two MOFs is consistent
with the observations concerning PPh2-bdc and POPh2-bdc.

74
 Characterization of P-MOFs by XAS and NMR spectroscopy

Table 4.1 Summary of the main spectral features of organic linkers and P-MOFs.

Compound Feature Energy (eV)

A 2147.3
AI 2145.7
PPh2-bdc
AII 2150.3
AIII 2155.4
B 2150.6
POPh2-bdc BI 2147.9
BII 2159.6
C 2147.6
CI 2145.7
LSK-15 CII 2150.2
CIII 2156.0
CIV 2158.6
D 2150.6
I
D 2147.9
LSK-12
DII 2156.2
DIII 2159.3

The spectrum of LSK-15 provides the opportunity to evaluate the accuracy of phosphine
content quantification inside the material using XAS at the P K-edge. The mild thermal
conditions employed during synthesis promote the framework formation, leaving the majority
of phosphine functional groups unoxidized. The XANES region is particularly sensitive to the
oxidation state of the phosphorus, as shown by the spectra of PPh2-bdc and POPh2-bdc
(Figure 4.3, left). We used the linear combination fitting (LCF) procedure available in
ATHENA (see Chapter 2) to compute the percentage of phosphine groups based on the
deconvoluted spectrum (Figure 4.4, left). The obtained scaling factors revealed that
phosphine groups account for 72 ± 4% of the total phosphorus sites, the rest being
phosphine oxide.

75
Chapter 4

Figure 4.4 P K-edge XANES spectrum of LSK-15 with the linear fit (dashed line) and the two
weighted contributions of PPh2-bdc and POPh2-bdc.

The 31P MAS NMR spectrum of LSK-15 was recorded to provide a comparison to the XANES
spectra (Figure 4.2). We encountered a small difference in the sensitivity of the two
characterization methods towards the phosphine environment, with phosphine content
31
measured at 85 ± 10% by NMR spectroscopy. Additionally, the P MAS NMR spectrum
revealed that two non-equivalent phosphine sites are present in the framework
(Figure 4.2, A1 and A2). These two sites could not be discriminated in the XANES spectrum
using PPh2-bdc and POPh2-bdc, which strongly supports topological differences rather than a
different phosphorus oxidation state or coordination of the phosphorus with an aluminum
atom. This illustrates how a combination of MAS NMR spectroscopy and XAS can be
complementary for the detection of phosphorus sites inside a MOF without inducing
phosphine oxidation.
31
In the case of LSK-12, for which P MAS NMR spectroscopy was inconclusive due to
paramagnetic Cr(III) centers, XANES was solely used to characterize phosphorus functional
groups in the solid state. Quantification of the occupancy of different phosphorus sites inside
LKS-12 was more straightforward than in LSK-15, because the absence of pre-edge features
related to phosphine groups implies that total oxidation of phosphine sites inside the material
has occurred. A modification of the phosphorus environment by beam damage can be ruled
out as we did not observe any alteration in the spectra of the organic linkers, nor of the
P-MOFs after repeated irradiation of the samples. Equivalent characterization methods allow
meaningful comparisons to be made between the impact of the respective synthesis
conditions for LSK-12 and LSK-15. The level of oxidation in the two materials can be
rationalized by the different thermal conditions used during synthesis (Chapter 3).

76
 Characterization of P-MOFs by XAS and NMR spectroscopy

The EXAFS spectra of the two linkers (Figure 4.5, top) in the k2-weighted space displayed
relatively mild oscillatory behavior, as anticipated by the absence of strong scattering nuclei
in the phosphorus coordination shells. The two spectra were reasonably similar despite the
inclusion of an oxygen atom in POPh2-bdc. The region 3.5 < k < 5 Å-1 is characterized by a
broad peak in the case of POPh2-bdc and sharper features in the case of PPh2-bdc. The
same features were observed in the two MOFs (Figure 4.5, bottom), with a more distinct
resemblance of LSK-15 to PPh2-bdc and LSK-12 to POPh2-bdc, in accordance with our
observations in the XANES region. However, it is difficult to go beyond a qualitative
comparison of the EXAFS spectra due to the short k-range available, aggravated by the weak
oscillations generated by carbon and oxygen in the phosphorus coordination shells.

Figure 4.5 k2-weighted chi(k) functions of POPh2-bdc and PPh2-bdc (top), and LSK-12 and
LSK-15 (bottom).

4.3 Characterization of metal-functionalized P-MOFs

The next step in the characterization of P-MOFs is to provide information on their phosphine
groups when they are coordinated to transition metals. It is capital to discriminate
physisorbed complexes within the MOF pore system from immobilized species, coordinated
to the phosphine functional groups, as phosphines are well-known to impact the stability and
activity of transition metal catalysts.[280] The extent of functionalization, as well as spatial
distribution of the organometallic complexes within a crystal are significant parameters when
considering potential catalytic applications. Examples are presented in this section to
illustrate the potential of XAS and MAS NMR spectroscopy applied to different P-MOFs
functionalized with transition metals.

77
Chapter 4

4.3.1 Ru-functionalized LSK-15

The MIL-101 topology is often reported to accommodate metallic species within its pore
system such as nanoparticles[268] and organometallic complexes.[127] However, the
immobilization of active catalysts is mainly achieved via physisorption,[135,137,138] which limits
the scope of possible support effect and may result in leaching of the active species.
In our approach, the phosphine functional groups in LSK-15 are used to coordinate the
[RuCl2(p-cym)]2 (p-cym = para-cymene) complex via a procedure similar to the preparation of
[RuCl2(p-cym)(PPh3)].[316,317] LSK-15 was suspended for 24 h in a solution of [RuCl2(p-cym)]2
in DCM at room temperature to yield [RuCl2(p-cym)(LSK-15)] (Ru-LSK-15a, Figure 4.6) after
washing. Complexes of the form [RuX2(η6-arene)(L)] are reported precursors for
homogeneously-catalyzed reactions such as transfer hydrogenation[318,319] and decomposition
of formic acid,[320–325] for which phosphine ligands play a key role in the stability and activity of
the catalyst. Thus, it is critical to ensure that the ruthenium complex is immobilized into the
P-MOF through a P-Ru coordination bond. Catalytic performances of Ru-LSK-15a are further
discussed in Chapter 5.

Figure 4.6 Functionalization of LSK-15 with [RuCl2(p-cym)]2

31
P MAS NMR spectroscopy was first used to characterize the phosphine groups inside
Ru-LSK-15a (Figure 4.7). Two additional spectral features were observed at +19 and
+33 ppm that may be attributed to multiple modes of phosphine-ruthenium
coordination.[316,317] In the case of the +33 ppm signal, it is however difficult to separate it from
the phosphine oxide signal (+35 ppm). Indeed, the relative broadness of the peak likely arises
from a mixture of multiple phosphorus species. The two phosphine signals from LSK-15
(-6 and -10 ppm) remain clearly visible, which suggests that the ruthenium complex does not
saturate all phosphine sites. The 27Al MAS NMR spectrum confirmed that the structure is not
affected by the presence of ruthenium. The 1H-13C CP MAS NMR spectrum displayed
features from the coordinated p-cymene ligand from +17 to +35 ppm and at +80 ppm,
which confirms that the ruthenium sites are in a chemical environment analogous to
[RuCl2(p-cym)(PPh3)].[316,317] High intensity of the p-cymene signals compared to the ones
from LSK-15 is due to the increased sensitivity of methyl groups in cross-polarization
conditions (Chapter 2).

78
 Characterization of P-MOFs by XAS and NMR spectroscopy

Figure 4.7 31P (left), 27Al (middle) and 1H-13C CP (right) MAS NMR spectra of Ru-LSK-15a
recorded at 20 kHz on a 700 MHz magnet. Spinning sidebands are marked with (*). Impurity
from sample contamination is marked with (i).

The ruthenium coordination environment was examined using XAS at the Cl and Ru
K-edges. In this analysis, reference spectra of [RuCl2(p-cym)]2 (Figure 4.8, A) and
[RuCl2(p-cym)(PPh3)] (B) were compared to a Ru-LSK-15a sample (C). Table 4.2
summarizes the main feature energies for each sample and their relative energy difference to
A. We observed a common feature in all Cl K-edge XANES spectra at ~2821 eV, previously
attributed to the 1s Cl → 4d Ru transition in similar examples.[326,327] Pre-edge features at the
Cl K-edge have been reported to be very sensitive to the electronic state of the ruthenium
d-orbital, as well as for other transition metals.[326–329] Reaction of [RuCl2(p-cym)]2 with PPh3
resulted in a -1.1 eV relative energy difference to [RuCl2(p-cym)]2 for the terminal chlorines,
similar to what was observed with Ru-LSK-15a (-0.6 eV). In addition to a noticeable energy
shift, the presence of bridging chlorine atoms (µ-Cl) was only observed in the
[RuCl2(p-cym)]2, which is consistent with the formation of mono-ruthenium complexes upon
reaction with phosphine groups. Other features at the Cl K-edge cannot be compared easily
in this system due to the presence of chlorine ions inside the framework of LSK-15.

79
Chapter 4

Figure 4.8 Normalized fluorescence XANES at Cl K-edge (left), normalized fluorescence

XANES at the Ru L2-edge, and transmission XANES at the Ru K-edge (right) for
[RuCl2(p-cym)]2 (A), [RuCl2(p-cym)(PPh3)] (B), and Ru-LSK-15a (C).

Spectra at the Ru L2-edge (Figure 4.8, middle) did not reveal any significant shifts in the white
line for the three samples. While small spectral feature differences in the overall XANES may
be interpreted using ab initio calculations to extract structural information in ruthenium
complexes,[330,331] the coordination of PPh3 to [RuCl2(p-cym)]2 did not cause any significant
structural or electronic changes that can be used in qualitative analysis. The Ru K-edge
XANES spectra displayed significant white line shifts for [RuCl2(p-cym)(PPh3)] (-6.2 eV) and
Ru-LSK-15a (-5.7 eV) as compared to [RuCl2(p-cym)]2, giving further confirmation of the
analogous electronic and structural properties of Ru-LSK-15a to its organometallic
counterpart (Figure 4.8, right). The energy difference between the [RuCl2(p-cym)(PPh3)]
complex and Ru-LSK-15a may be attributed to the slight alteration of the electronic properties
of the phosphorus in PPh3 and PPh2-bdc. No shoulder pertaining to [RuCl2(p-cym)]2 was
observed in the XANES of Ru-LSK-15a, which suggest a virtually complete coordination of
the ruthenium onto the LSK-15 phosphine groups.

Table 4.2 Summary of selected feature positions for [RuCl2(p-cym)]2 (A) and its relative
energy differences with [RuCl2(p-cym)(PPh3)] (B) and Ru-LSK-15a (C). Energy positions E0
and differences ∆E0 are given in eV.

Cl K[a] Ru L2 Ru K
A 2821.6 2968.7 22129.8
∆(B-A) -1.1 0 -6.2
∆(C-A) -0.6 0 -5.7
[a]
1s Cl → 4d Ru transition

80
 Characterization of P-MOFs by XAS and NMR spectroscopy

4.3.2 Rh-functionalized LSK-3

Attempts at recording a rhodium complex within a P-MOF were first conducted on Rh-LSK-3.
LSK-3 was mixed together with [Rh(COD)]2BF4 in DCM at room temperature for 24 h to yield
Rh-LSK-3 as bright orange crystals. After washing, the sample was dried under argon and
31
characterized by P MAS NMR spectroscopy (Figure 4.9, c), along with a fresh (a) and
oxidized (b) sample of LSK-3. Comparison of Rh-LSK-3 with fresh LSK-3 (+28 ppm)
highlighted a new signal at +24 ppm, consistent with a P-Rh coordination environment.[332]
However, the oxidized sample revealed that multiple phosphine oxide environments are
present in the system (+23 and +27 ppm), overlapping with the observed feature of
Rh-LSK-3. Unfortunately, the signal broadening in solid-state NMR spectroscopy limited the
resolution of the spectra and prevented the observation of the 1J(P,Rh) coupling in Rh-LSK-3.
Striking differences in the spinning sidebands intensities between (b) and (c) suggest that the
phosphorus species responsible for the signals at +23 and +24 ppm may be chemically
different. This example represents a typical case where XAS at the P K-edge would be
determinant for the discrimination of phosphine oxide from phosphine-metal environment;
however, the fragility of LSK-3 in dry conditions hindered any XAS measurements.

Figure 4.9 (a) 31P MAS NMR spectra of a fresh sample of LSK-3 at 20 kHz, (b) oxidized
sample of LSK-3, (c) and Rh-LSK-3 at 10 kHz. Spinning sidebands are marked with (*).
All spectra were recorded on a 400 MHz magnet.

81
Chapter 4

4.3.3 Rh-functionalized LSK-11

To understand the electronic and structural properties of phosphine groups in the case of
rhodium-functionalized P-MOFs, LSK-11 was impregnated with a solution of [Rh(COD)2]BF4
in DCM at room temperature for 24 h. LSK-11 was selected for its structural resemblance to
LSK-3 and its higher stability in dry conditions. The crystals, originally light-yellow, turned
orange showing the incorporation of the rhodium complex in the crystal. P K-edge XAS
spectra were measured at the surface of Rh-LSK-11 using a micro-focused beam (Figure
4.10). A LCF strategy similar to the one used for LSK-15 (vide supra) was
employed, using PPh2-bdc and POPh2-bdc as external references. The spectrum of
[Rh(COD)(PPh3)2]BF4 was used as the third fitting component (Figure 4.10, left).
A satisfactory fit (R-factor = 0.014) was obtained when the reference data were corrected for
self-absorption using the FLUO algorithm available in ATHENA.[249] Poor data fitting was
observed when no self-absorption correction was applied (R-factor = 0.120) and when the
rhodium reference was not included in the fit (R-factor = 0.130). In the latter, the residual
spectrum had a pronounced feature at 2148.7 eV, which corresponds to the dominating
contribution of the spectrum of the rhodium complex. This is a strong indication of a P-Rh
coordination in Rh-LSK-11. Using the best fit values, we estimated that phosphine oxide
accounted for ~35% of the phosphorus and that the rhodium coordinated ~60% of the
remaining available phosphine groups.

Figure 4.10 (left) Normalized P K-edge XANES spectra of Rh-LSK-11 (black) and LCF fitting
(dashed line) using weighted components. [Ph3P-Rh] = [Rh(COD)(PPh3)2]BF4 (red), PPh2-bdc
(green), and POPh2-bdc (blue). (right) Normalized Rh L2,3-edge XAS spectra of Rh-LSK-11.

A Rh L2,3-edge XAS spectrum of Rh-LSK-11 was measured at the same crystal position
(Figure 4.10, right). The spectrum displayed white lines at 3007 and 3149 eV for the
L2,3-edge, respectively. Comparison with literature values rules out the presence of
rhodium(0) that features an L3-edge white line at 3003 eV.[333]

82
 Characterization of P-MOFs by XAS and NMR spectroscopy

However, a finer assignment of the absorption edge position is challenging without further
measuring additional reference compounds, or using ab initio calculations, as reported
L-edge spectra for rhodium(I) and rhodium(III) species are restricted to complexes that do not
contain phosphine ligands.[334]

4.3.4 Chemical mapping in Rh-LSK-11

While P K-edge XAS provided insights into the LSK-11 phosphorus coordination in the
presence of rhodium, single-point XAS measurements at the external surface of the crystal
may not be representative of the spatial distribution of rhodium and phosphine species within
the crystal. The control of the metal dispersion within porous materials is a key engineering
challenge,[16] the more so in large crystals, for which a significant portion may be unused.
Elemental imaging studies in MOFs were reported only recently using fluorescence
microscopy,[335] PT-IR,[336] and X-ray spectro-tomography.[337] However, these techniques do
not allow for chemical speciation, and chemical mapping in MOFs remains a major challenge.
To address this topic in P-MOFs, we performed fluorescence mapping of a hand-cut crystal
of Rh-LSK-11 using a micro-focused X-ray beam at 2150 eV (Figure 4.11).

Figure 4.11 Optical image of a hand-cut crystal of Rh-LSK-11. (a) top view, (b) 45° right-hand
side view emulating the fluorescence detector viewpoint.

The phosphorus fluorescence map displayed strong intensity differences across various
crystal regions (Figure 4.13, b). This effect was attributed to the irregular exposed surface of
the crystal, which was confirmed by a similar pattern in the zinc fluorescence map (Figure
4.13, a) where maxima in the fluorescence signal are recorded for out-of-plane surfaces. X-
ray shadows (dark-blue regions in Figure 4.13, a-c) are also observed at the top-right and
bottom-left region of the maps and likely originate from a combination of high planes and the
45° measurement geometry. This provided relative information about the crystal elemental
distribution, but prevented to know the exact overlap between the optical picture and the
map.

83
Chapter 4

Normalization of the phosphorus map by its zinc counterpart (Figure 4.13, d) did not reveal
any significant intensity differences within the crystal (y-positions between 100 and 290 µm).
However, a variation of the normalized rhodium signal was observed at one edge of the
crystal (Figure 4.13, d, y-positions between 150 and 250 µm). The sharp features in the
normalized maps observed at the crystal edges (at y-positions of 70 µm and 300 µm) are
attributed to the difference in energy of the fluorescence lines resulting in a slightly shifted
measured edge due to penetration depth effects.

Figure 4.13 Fluorescence mapping of Zn, P, and Rh at 2150 eV. (a) Temperature map:
blue = minimum, red = maximum fluorescence signal, (b) normalized fluorescence along
x-distance = 110 μm.

In parallel to the fluorescence maps, individual XAS spectra at the P K-edge were recorded at
selected regions of the crystal (Figure 4.14, left). The spectra were normalized by the zinc
fluorescence signal to minimize topological effects in the post-edge features. Dominant
spectral features were attributed to the phosphine oxide (P=O, 2150.5 eV) and the
phosphine-rhodium (P-Rh, 2148.5 eV) environments. The high extent of phosphine oxidation
is explained by the long storage period (>1 month) of the crystals in solvent that ensued
between the initial measurements at the surface (Figure 4.10) and the acquisition of the
fluorescence maps (Figure 4.13). A LCF analysis of the XAS spectra including the PPh2-bdc
spectrum did not result in a significant improvement of the fit due to the small contribution of
the phosphine coordination. The P-Rh/P=O ratio was calculated instead from the spectral
features intensities, and compared to the ratio obtained by fluorescence at the same position
(Figure 4.14, right and Table 4.3). The differences between the two techniques may indicate
the presence of non-coordinated rhodium species within the framework, possibly due to the
sample prolonged storage period or an incomplete washing procedure.

84
 Characterization of P-MOFs by XAS and NMR spectroscopy

Figure 4.14 XAS using micro-focused beam at selected region of the crystal.
(left) Normalized P K-edge XANES spectra (right) normalized fluorescence profile at
x-distance = 110 μm.

Table 4.3 P-Rh/Ptot ratio calculated from XANES features and fluorescence signal

P-Rh/Ptot
y-position (µm) XANES[a] Fluorescence[b]
140 1.1 1.2
170 0.9 1.5
200 1.0 1.7
210 1.2 1.7
[a]
calculated from P-Rh/P=O, [b]calculated from P/Rh.

The large contribution of the phosphine-rhodium spectral feature (Figure 4.14, left) to the total
signal across all measured points is an indication of the high dispersity of the rhodium within
the crystal cross-section. Equal distribution of the phosphorus and rhodium across the crystal
may be attributed to the binding ability of the phosphine, together with the mixed-linker nature
of LSK-11 (Chapter 3) that provides an enhanced diffusivity within the pore system, in a
manner reminiscent to a previous study on MIXMOF-5-NH2.[166] To confirm our XAS
observations, scanning electron microscopy (SEM) coupled to energy-dispersive x-ray
spectroscopy (EDX) was performed on the same crystal cut (Figure 4.15). EDX spectroscopy
at 20 keV at the P K-alpha and Rh L-alpha emission lines confirmed our observations that
both elements are well-dispersed for the majority of the crystal cut, albeit several smaller
regions that were mainly populated by phosphorus only (Figure 4.15, c). While EDX
spectroscopy provides a high spatial resolution of the key elements present within the crystal,
the combination of fluorescence mapping and micro-focused XAS allows for space-resolved
chemical speciation of the phosphorus, providing a distinction between phosphine-ruthenium
complexes and inactive phosphine oxide. Thus, the combination of these techniques offers a
unique picture of this class of materials and their potential in transition metal immobilization.

85
Chapter 4

Figure 4.15 (a) SEM image of the cross-section of the Rh-LSK-11 crystal, (b) EDX
spectroscopy mapping at the P K-alpha line (yellow), and (c) overlay of the P K-alpha line
(yellow) and Rh L-alpha line (pink) maps.

4.4 Conclusion

In this chapter, we successfully demonstrated the applications of NMR spectroscopy and

XAS techniques for the chemical observation of phosphine functional groups within various
P-MOFs. Where non-specific spectroscopic techniques or characterization of digested
31
samples only provided limited information on the phosphorus center, P MAS NMR
spectroscopy and XAS were able to discriminate and quantify various phosphorus nuclei
such as phosphines, phosphine oxides, and phosphine-metal complexes. In the case of
rhodium- and ruthenium-substituted frameworks, these two techniques were used in parallel
to provide meaningful insights both from the MOF and the metal complex perspective,
confirming the metal-binding potential of the P-MOFs presented in Chapter 3.

86
 Chapter 5

Decomposition of formic acid over a ruthenium

functionalized P-MOF

Based on: A. Beloqui Redondo,* F. L. Morel,* M. Ranocchiari, J. A. van Bokhoven, ACS

Catalysis, 2015, 5, 7099-7103.

F. L. M. contributions: material synthesis, characterization with A.B.R., write-up with A.B.R.

*Equal contribution from A.B.R and F.L.M.

Chapter 5

5.1 Introduction

The use of hydrogen as energy feedstock represents an important field due to the growing
demand in alternative energy technologies.[338–340] Due to the difficulty of hydrogen storage,
research has focused on alternative energy carriers, such as alcohols, amines, and formic
acid that can be catalytically decomposed to produce hydrogen.[341,342] Formic acid is a
promising candidate due to its safe handling and low toxicity.[343] Decomposition of formic acid
can proceed via dehydrogenation (Eq. 5.1) to form hydrogen and carbon dioxide, and
dehydration (Eq. 5.2), producing water and carbon monoxide. The latter is a well-known
poison for catalysts and represents a great challenge for downstream applications, notably
fuel cells.[344] This has triggered the search for catalysts with high selectivity towards
dehydrogenation. Liquid-phase formic acid decomposition has been performed at room
temperature with palladium nanoparticles embedded within MIL-125, achieving a turnover
frequency (TOF) of 214 h-1.[345] A similar system was prepared with Au-Pd nanoparticles
immobilized in MIL-101.[346] However, these systems and other supported catalysts are often
outperformed by homogeneous catalysts in terms of activity, selectivity, and stability.[322,347]
In recent years, notable advances in the design of homogeneous catalysts for formic acid
decomposition have been reported with ruthenium,[324,325,348–350] rhodium,[351] and iron[352–354]
complexes, sometimes used in conjunction with amine additives. Despite the considerable
number and variety of ruthenium catalysts used for this reaction, few heterogeneous
ruthenium catalysts have been developed.[355–357]

HCOOH → H2 + CO2 (5.1)

HCOOH → CO + H2O (5.2)

In Chapter 3, we described the preparation of a bifunctional MOF called LSK-15 that exhibits
the MIL-101(Al) topology (Figure 5.1, a) and which contains phosphino- and amino functional
groups.[307,358] LSK-15 is an ideal material for formic acid decomposition because (a) the
MIL-101 topology shows good chemical and thermal stability and its large pores can easily
accommodate transition metal complexes and substrates for reaction,[63,127,268] (b) phosphine
moieties are strong anchoring sites for transition metal complexes, and (c) amine moieties
provide a local basic environment that is thought to facilitate the activation of formic acid, in a
manner reminiscent of homogeneous systems.[325] In this chapter, we report a very stable and
selective MOF catalyst based on LSK-15 for the vapor-phase dehydrogenation of formic acid.

88
 Decomposition of formic acid over a ruthenium-functionalized P-MOF

Figure 5.1 (a) General representation of the MIL-101 pore system (green = aluminum, red =
oxygen, gray = carbon). (b) Molecular representation of the local environment in Ru-LSK-15a
with phosphine and amine linkers (pink = phosphorus, blue = nitrogen, orange = ruthenium,
light green = chloride). Hydrogen atoms are omitted for clarity. (c) Schematic representation
of the ruthenium, phosphine and amine functionalities present in the MOF cages. Reprinted
with permission from ref.[359] Copyright 2015 American Chemical Society.

5.2 Preparation of the catalyst

Inspired by the activity of ruthenium molecular catalysts in liquid-phase formic acid

decomposition, we introduced a ruthenium precursor into LSK-15 by post-synthetic
metalation. Mild conditions were selected to ensure the formation of a ruthenium-phosphine
molecular complex within the pores of LSK-15, adapting existing procedures for related
organometallic complexes.[360,361] In particular, we used the reactivity of the [RuCl2(p-cym)]2
dimer towards triphenylphosphine to form the [RuCl2(p-cym)(PPh3)] piano-stool shaped
mono-ruthenium complex. The synthesis of [RuCl2(p-cym)(LSK-15)] complex (Ru-LSK-15a,
Figure 5.1, b,c) was performed by reacting LSK-15 with [RuCl2(p-cym)]2 in DCM for 24 h at
room temperature. After washing, the orange material retained its crystalline structure upon
coordination (Figure 5.2, a) and showed a decrease in BET surface area from 2355 to
1075 m2/g and in pore volume from 1.26 to 0.6 cm3/g. This space-filling effect was already
described in a similar system[127] and is attributed to the presence of the ruthenium complex
inside the cage of LSK-15.
31
The P NMR spectrum of the impregnated material (Figure 4.7, Chapter 4) showed two
additional features at +19 ppm and +33 ppm not present in LSK-15, which are in the region of
phosphine-ruthenium coordination and phosphine oxide species.[360] The presence of free
phosphine signals at -6 and -10 ppm confirms the partial coordination of phosphine groups,
which is in agreement with a ruthenium loading of 0.4 ± 0.2 wt.%, as measured by atomic
absorption spectroscopy. A low ruthenium loading was preferred as it presents the
advantages of enhanced diffusivity within the pore system and maximizes the access to the
catalytic sites.

89
Chapter 5

Figure 5.2 X-ray diffraction (a) and nitrogen physisorption isotherm at 77 K (b) of Ru-LSK-
15a (□), LSK-15 (●), and MIL-101(Al)-NH2 ( ).

The 1H-13C CP MAS NMR spectrum displayed features from the coordinated para-cymene
from +17 to +35 ppm, thus confirming the presence of the ruthenium sites in a
chemical environment analogous to [RuCl2(p-cym)(PPh3)].[360,361] To confirm the ruthenium
environment, XAS at the Cl and Ru K-edges was performed (Figure 4.10, Chapter 4).
Spectra at the Cl K-edge of [RuCl2(p-cym)]2 and [RuCl2(p-cym)(PPh3)]2 confirmed that the
signal pertaining to bridging chlorine ligands disappears upon coordination of PPh3, leaving
intact the signal of the terminal chlorine. A similar effect was observed in Ru-LSK-15a, which
confirmed the coordination of the complex onto the phosphine groups. Comparison of Ru
K-edge spectra also confirmed the isoelectronic configuration of [RuCl2(p-cym)(PPh3)]2 and
Ru-LSK-15a.

5.3 Catalytic decomposition of formic acid

Continuous dehydrogenation of formic acid was carried out in a fixed-bed reactor setup[234] at
different temperatures (125, 145 and 165 °C) to investigate the catalytic performance of
Ru-LSK-15a in the vapor-phase. Figure 5.3 displays the formic acid conversion and
selectivities to hydrogen and carbon monoxide as the reactor temperature was switched
between 125, 145 and 165 °C in 24 h allotments. During the first 24 h at 125 °C an induction
period was observed, in which the conversion of formic acid increased from 7 to 34 ± 13%.
The conversion stabilized at 75 ± 5% (TOF = 2,000 ± 440 h-1) once the temperature was
increased to 145 °C. Conversion increased further to 88 ± 8% (TOF = 2,400 ± 580 h-1) after
reaching 165 °C and remained constant for 24 h. No sign of deactivation was observed within
this period and an overall turnover number (TON) of 121,300 was calculated. The hydrogen
selectivity was recorded above 99% for the major part of the experiment, with a decrease to
97 ± 2% in the last few hours.

90
 Decomposition of formic acid over a ruthenium-functionalized P-MOF

Neither residual DCM nor carbon monoxide was detected, but traces of methanol were found
by gas chromatography. Conversion values with large standard deviations were observed
over the first 24 h prior to the system attaining steady-state conditions. As the reactor was
heated to 145 and 165 °C, the system evolved hydrogen and carbon dioxide at a constant
rate. The induction period was attributed to the elimination of a chloride ligand to form a
16-electron complex before the coordination of formates, following the proposal made by
Morris et al. for organometallic ruthenium complexes.[362] Amines have been reported to act
as proton scavengers in the liquid phase by deprotonating formic acid to create formate.[357]
It is possible that the amine moieties on LSK-15 fulfill a similar role in the decomposition
mechanism as we observed their partial protonation after catalytic testing (vide infra).

Figure 5.3 (a) formic acid conversion and (b) selectivity to hydrogen (black) and carbon
monoxide (red) with time while switching the reaction temperature between 125, 145 and
165 °C. Error bars represent the standard deviation of four consecutive measurements.

To investigate the role of the phosphine groups in LSK-15, we performed the reaction in the
presence of ruthenium with MIL-101(Al)-NH2 as support. Using the same post-synthetic
impregnation procedure as for Ru-LSK-15a only resulted in considerable leaching during the
washing step, suggesting weak ruthenium-amine coordination. Instead, we prepared
Ru@MIL-101(Al)-NH2 by incipient-wetness impregnation of MIL-101(Al)-NH2 with a
[RuCl2(p-cym)]2 solution containing a similar amount of ruthenium as in Ru-LSK-15a. With
this material, a maximum formic acid conversion of only 11 ± 3% at 165 °C was obtained
(Figure 5.4, b). A similar experiment carried out with Ru-LSK-15a led to a conversion level of
64 ± 2% at 165 °C (Figure 5.4, a), which indicates that phosphine groups are essential for
anchoring the ruthenium complex and establishing a defined and highly efficient active site.
This behavior is easily explained by the stronger ability of phosphines to coordinate
ruthenium, as compared to monodentate secondary amines.

91
Chapter 5

No conversion was detected for LSK-15 in the absence of ruthenium complex at a

temperature of 165 °C, thereby ruling out any intrinsic activity of the framework and auto-
decomposition of the substrate.

Figure 5.4 Formic acid decomposition over Ru-LSK-15a (a) and Ru@MIL-101(Al)-NH2 (b).
Temperature ramp was of 2 °C/min. Error bars indicate the standard deviation of four
consecutive measurements.

A long-term experiment was conducted to verify the stability of Ru-LSK-15a (ramp to 125 °C
and 24 hours dwell time before ramping to 145 °C, Figure 5.5). The vapor-phase formic acid
decomposition was performed for 23 days at 145 °C. Under these conditions, Ru-LSK-15a
gave a constant TOF of 2,300 ± 570 h-1. A TON of 1,290,000 was reached without
measurable deactivation. Additionally, no carbon monoxide was detected during the catalytic
run, suggesting that the catalyst does not suffer from a loss in performance and may continue
to operate for an even longer period of time.

Figure 5.5 Long-term stability testing of Ru-LSK-15a at 145 °C. (a) conversion of formic acid
and (b) hydrogen (black) and carbon monoxide (red) selectivities with time. Gray area
indicates induction period at 125 °C.

92
 Decomposition of formic acid over a ruthenium-functionalized P-MOF

Transmission electron microscopy (TEM) has been performed on the sample before and after
catalysis (Figure 5.6). Although electron micrographs can be difficult to assess due to the
interaction between the sample and the electron beam,[363] we do observe the conservation of
the molecular structure, as characterized by the absence of large ruthenium clusters.
Micrographs could not be differentiated from the as-synthesized material (Figure 5.6, a,b)
after both catalytic experiments described in Figure 5.3 and Figure 5.5 (Figure 5.6, d,e).
Energy-dispersive X-ray spectroscopy confirmed the presence of ruthenium in the observed
field of view. The stability of the ruthenium center is remarkable considering the relatively
harsh conditions of the reaction. We attribute the robustness of the catalyst to the
coordination strength of phosphine groups within the material, which prevents sintering thus
preserving the selectivity of the catalyst towards hydrogen production.

Figure 5.6 High-angle annual dark field scanning transmission electron microscopy
(HAADF-STEM) images of Ru-LSK-15a before (a-c) and after (d-f) various reaction
conditions.

In terms of stability, Ru-LSK-15a matches the highest reported TON values for the
dehydrogenation of formic acid, both in liquid and gas-phase. In comparison with supported
ruthenium nanoparticles in the gas-phase,[355,357] Ru-LSK-15a showed a large improvement in
terms of activity and selectivity. This highlights the advantages of immobilized complexes that
couple the high efficiency of homogeneous systems[364] with the convenience of
heterogeneous catalysts to expand their application to operation in the gas phase. A more
comprehensive comparison between Ru-LSK-15a and selected heterogeneous and
homogeneous catalysts reported in the literature for the dehydrogenation of formic acid is
summarized in Table 5.1.

93
94
Table 5.1 Catalytic performance of catalysts for the decomposition of formic acid based on this work and literature data.
Chapter 5

Entry Conditions Catalyst T (°C) TOF (h-1) TON Selectivity Ref.

1 0.2 wt.% Ru-LSK-15a, slow activation 125-165 2,410 121,300 SH2 ≈ 100%

2 0.7 wt.% Ru-LSK-15a, slow activation 145 2,260 1,290,000 SH2 ≈ 100%
Supported, this
Gas-phase work
3 0.1 wt.% Ru-LSK-15a, fast activation 165 4,220 2,540,000 SCO = 1%

4 0.2 wt.% Ru@MIL-101(Al)-NH2 165 308 n.r. SH2 ≈ 100%

[365]
5 Au/Al2O3 80 25,600 n.r. < 10 ppm CO
[357]
6 1 wt.% Ru/6.8 wt.%N-C 100 58 n.r. SH2 ≈ 90%
Supported,
Gas-phase [355]
7 2 wt.% Ru/Norit 150 911 n.r. SH2 ≈ 95%
[366]
8 Au/SiO2 200 7,495 n.r. SH2 ≈ 94%
[367]
9 Au/ZrO2 50 1,590[a] 118,400 < 5 ppm CO
[368]
10 IrCp*/CTF[b,c] 80 27,000 1,000,000 H2:CO2 = 1:1
Supported, [d] [356]
11 Ru-mTPPTS -MCM41 90 2780 71,000 < 3 ppm CO
Liquid-phase
[369]
12 Pd/C 100 255 n.r. SH2 ≈ 97%
[370]
13 PdAu/Al2O3 130 1000 n.r. SH2 ≈ 97%
[320]
14 RuCl2(PPh3)3 40 302 905 H2:CO2 = 1:1
[349]
15 [RuCl2(p-cym)]2 40 n.r. 932 H2:CO2 = 1:1
Homogeneous,
Liquid-phase [371]
16 [{Ir(Cp*)(OH2)}2(bpym)][c,e] 80 158,000 308,000 < 10 ppm CO
[354]
17 Fe(BF4)2·6H2O 80 5,390 92,417 < 20 ppm CO
[a] [b] [c] [d]
Initial TOF after 20 min. COF = covalent triazine framework, Cp* = cyclopentadienyl, mTPPTS = meta-trisulfonated triphenylphosphine,
[e]
bpym = bipyrimidine.
 Decomposition of formic acid over a ruthenium-functionalized P-MOF

A comparison of the X-ray diffraction powder patterns of the catalyst exposed to formic acid
at various temperatures (Figure 5.7, a) indicated a partial loss of the long-range order of the
MOF, which occurred gradually. The nitrogen physisorption isotherm after exposure to formic
acid at 145 °C revealed a considerable decrease in BET surface area and in total pore
volume after reaction from 1075 to 161 m2/g and 0.61 to 0.19 cm3/g, respectively, as
compared to the fresh sample (Figure 5.7, b). These values are consistent with the loss of
long-range order observed by X-ray diffraction. In spite of the apparent loss of order, the
constant activity and selectivity of the catalyst suggest that the ruthenium center remains
unaffected by the phase transformation of the support, which provides sufficient short-range
order to ensure molecular dispersion.

Figure 5.7 Characterization of Ru-LSK-15a after catalytic runs at indicated temperatures. (a)
XRD patterns and (b) typical nitrogen physisorption isotherm at 77 K of Ru-LSK-15a
as-synthesized (□) and after reaction at 145 °C (●).
31
A P MAS NMR spectrum of the material recovered after catalysis showed a substantial
decrease of signals related to free phosphine (Figure 5.8, a). Samples of LSK-15 exposed to
air or formic acid (data not shown here) reveal signals between +30 and +40 ppm that are
likely to originate from a mixture of phosphine oxide and the ruthenium-phosphine complex.
The 1H-13C CP MAS NMR spectrum (Figure 5.8, b) depicts clear signals arising from the
organic linkers and ligands. Carbon signals attributed to the amino linker[57] at +150 ppm and
+116 ppm declined markedly in intensity after the reaction. These observations are in
agreement with the chemical shifts of the protonated linker, which coalesce in the +130 ppm
region. While these measurements are not enough to unambiguously determine the role of
the amine groups within the catalytic cycle, they highlight a possible future improvement, in
which the basic environment of LSK-15 might be engineered to increase its reactivity towards
the deprotonation of the formic acid.

95
Chapter 5

A carbon signal at +172 ppm shows that the carboxylate bond responsible for the short-range
order within the framework is preserved. This is further confirmed by the exclusive presence
of octahedral aluminum, as observed in the 27Al NMR spectra (Figure 5.8, c).

Figure 5.8 31P (a), 1H-13C CP (b), and 27Al (c) MAS NMR spectra of Ru-LSK-15a
as-synthesized (bottom) and after formic acid decomposition at 145 °C for 23 days (top).

To compare our molecular catalyst to ruthenium nanoparticles on the same support, we

performed a catalytic run with Ru-LSK-15a under conditions that induce the decomposition of
the molecular complex. The material was exposed to a fast heating ramp to 165 °C and to a
higher flow of formic acid, which led to an initial TOF of 4’200 ± 800 h-1 after a short induction
period (Figure 5.9). Carbon monoxide was observed after 8 h, reaching a selectivity of
1.0 ± 0.2% after two days. High conversion values and a marked decrease of activity over
time were observed. TEM micrographs of the sample after reaction indeed confirmed the
formation of ruthenium clusters (Figure 5.6, c,f). This indicates that the molecularly dispersed
metal centers are essential for reaching high selectivity, which is required to compete against
homogeneous catalysts. Despite a lower selectivity, the performance of the catalyst under
these conditions is substantially better than other reported ruthenium nanoparticle
systems,[355,357] which suggests that the phosphine groups are still contributing to the high
reactivity of the catalyst.

96
 Decomposition of formic acid over a ruthenium-functionalized P-MOF

Figure 5.9 Decomposition of formic acid at 165 °C without induction period. (a) formic acid
conversion, (b) selectivity to hydrogen (black) and carbon monoxide (red).

5.4 Conclusion

In summary, we designed and prepared a multifunctional MOF catalyst containing a single-

site molecular ruthenium complex of the form [RuCl2(p-cym)(LSK-15)] and determined its
excellent performance in the vapor-phase dehydrogenation of formic acid. Phosphine
moieties drive the formation of ruthenium single-sites inside a heterogeneous support. In
addition, they ensure preservation of the original ligand coordination sphere and thus of the
catalytically active site. The resulting material displayed high selectivity for hydrogen and a
TON larger than 1,290,000 in the catalytic dehydrogenation of formic acid. We demonstrate
that MOFs with multiple functional groups are compelling candidates for establishing
single-site heterogeneous catalysts for vapor-phase reactions operated in continuous mode.

97
 Chapter 6

Supported Grubbs’ catalysts for liquid and gas

phase olefin metathesis
Chapter 6

6.1 Introduction

Originally reported by Eleuterio et al. in 1950s[372] and later investigated by Chauvin,[373]

Grubbs[8] and Schrock,[9] olefin metathesis is defined as the recombination of two olefins
forming two new olefins (Figure 6.1, a).[374] This reaction possesses two characteristic
attributes: 1) the recombination operates via a metallacyclobutane, formed by the
coordination of an olefin onto a metal carbene and 2) the different reactants and products
have a very similar enthalpy of formation. This leads to a statistical distribution of olefins at
the equilibrium,[375] unless additional entropic or enthalpic effects are taking place, such as
products removal, polymerization, and ring strain release.[376]

Figure 6.1 (a) General reaction of olefin metathesis. (b) Cross-metathesis and the role of the
metallacyclobutane in the mechanism described by Chauvin.[375]

Metathesis processes are usually separated into categories defined by the type of olefinic
substrates. Ring-opening metathesis polymerization (ROMP), ring-closing metathesis (RCM)
and cross-metathesis (CM) are among the most common variations at the preparative and
industrial scale.[376–379] CM is the reaction of two distinct olefins (Figure 6.1, b),[380–382] while
self-metathesis is a special case of CM that involves the same substrate twice. This process
often uses the reactivity of terminal olefins to produce an internal olefin and ethylene. The
reaction is driven to high conversion by stripping the ethylene off the reaction mixture using
reduced pressure or operating near the solvent boiling point.[383] Large-scale applications of
CM reactions are dominated by the production of propene from ethylene and butane, as well
as the reverse reaction depending on the current economical constrains, catalyzed by
WO3/SiO2 or Re2O7/Al2O3.[384] Applications of CM to the synthesis of functionalized olefins are
also emerging in pharmaceutical and fine chemistry.[380,383,385]

100
 Supported Grubbs’ catalyst for liquid and gas phase olefin metathesis

Functionalized olefins are however poorly compatible with rhenium, tungsten, and
molybdenum catalysts. Instead, ruthenium molecular complexes initially investigated by
Grubbs[386], Hoveyda[387], and Blechert[388] are nowadays routinely used. These catalysts
display spectacular performances in terms of activity and stability but their separation from
the product mixture remains a major concern.[389,390] Research has recently focused on their
immobilization onto solid supports such as organic polymers[391–393] and silica-based
materials[394–400] for the CM of non-functionalized olefins, as well as for RCM reactions.
Ruthenium complexes remain difficult to retain in solids due to the decoordination
of the supporting ligand involved in the catalyst activation.[393] In the case of
phosphine-functionalized materials, this translates to a “catch-and-release” mechanism, in
which the olefinic substrates compete with the phosphine moieties, often leading to ruthenium
leaching and product contamination.[400,401] We propose the use of P-MOFs that provide
well-defined constrained environments to durably retain an active catalyst within their pore
system. We present herein the immobilization of a Grubbs’ catalyst and its application in CM
reactions of un- and functionalized olefins to illustrate this strategy.

6.2 Immobilization strategy

Grubbs’ catalysts are a class of compounds based on organoruthenium(II) complexes, first

reported in the form of [RuCl2(PR3)2(CHR’)] (R = Ph, PCy3; R’ = HC2Ph2, Ph, etc.) and
featuring a stable ruthenium carbene ligand.[402,403] The reactivity of these complexes was
later enhanced by replacing a PCy3 ligand with a N-heterocyclic carbene to yield the so-called
2nd generation Grubbs’ catalysts.[386] In this work, we selected [RuCl2(IMesH2)(PCy3)(CHPh)]
(IMesH2 = 1,3-dimesityl-4,5-dihydroimidazol-2-ylidene) as precursor, taking advantage of its
stable carbene moiety and its availability from commercial sources (Figure 6.2, 6a). The
substitution of PCy3 by PPh3 via ligand exchange was reported to be inefficient in mild
conditions due to the stronger electron-donating character of the PCy3 ligand.[232] Instead, we
synthesized [RuCl2(IMesH2)(C5H5N)2(CHPh)] (Figure 6.2, b) and subsequently replaced the
pyridine ligands by the phosphine moieties within the framework. LSK-15 was impregnated
with a solution of 6b in DCM at room temperature for 48 h and then washed with DCM to
yield the green-brown solid Ru-LSK-15b. The same procedure was extended to LSK-11 and
LSK-17 to yield Ru-LSK-11b and Ru-LSK-17b, respectively (Chapter 2).

101
Chapter 6

PPh3

MesN NMes MesN NMes MesN NMes

Cl Ph C5H5N Cl Ph PPh3 Cl Ph
Ru Ru Ru
Cl 100 eq. N Cl Cl
PCy3 N PPh3

Mes = mesityl
6a 6b MesN NMes
P-MOF Cl Ph
Ru
Cl
Ph P
Ph MOF

Ru-P-MOF

Figure 6.2 Catalyst immobilization strategy via pyridine substitution and subsequent
phosphine exchange

6.3 Characterization of Ru-LSK-15b

Solid-state MAS NMR spectroscopy was performed to confirm the incorporation of the
31
ruthenium complex within the MOF. P MAS NMR (Figure 6.3, left) highlighted a strong
signal at +35 ppm in addition to the two signals previously assigned to the free phosphine
groups (-6 and -11 ppm, see Chapter 3). The chemical shift of this new signal is similar to the
value reported for the triphenylphosphine ligand in [RuCl2(IMesH2)(PPh3)(CHPh)] at
+37.7 ppm[232] and does not correspond to residual PCy3 (+9 ppm) nor to POCy3
(+48 ppm).[404] The signal cannot be exclusively attributed to phosphine-ruthenium
coordination due its overlap with the phosphine oxide signal in LSK-15 (+35 ppm, Chapter 3).
We were unable to detect any signal from the carbene moiety by 1H-13C CP MAS NMR
spectroscopy (Figure 6.3, right); however, observation of carbene species in the solid-state
13
typically requires the use of C-labeled samples.[405] The carbon signal at +17 ppm was
attributed to the mesityl groups of the IMesH2 ligand[232] and the sharp signal at +136 ppm
possibly belongs to residual free pyridine ligands. The ruthenium-pyridine signal was not
observed, but might significantly overlap with the NH2-bdc linker signal at +151 ppm,[232]
preventing an accurate discrimination between 6b within the pore system and the
coordinated complex.

102
 Supported Grubbs’ catalyst for liquid and gas phase olefin metathesis

Figure 6.3 31P (left) and 1H-13C CP (right) MAS NMR spectra of Ru-LSK-15b recorded at
20 kHz spinning speed on a 700 MHz spectrometer. Spinning sidebands are marked with (*)
and residual solvent signals with (s).

Cl K-edge and Ru L2-edge X-ray absorption spectra were recorded for Ru-LSK-15b, 6a and
6b to determine the electronic state of the ruthenium in the immobilized complex (Figure 6.4).
All spectra at the Cl K-edge displayed features at 2821.4 eV, previously attributed to a
1s Cl → 4d Ru transition for similar complexes (Chapter 4). Since no bridging chlorine ligand
is present in this case, the evaluation of the immobilization strategy cannot be based solely
on the position of this spectral feature. Spectral differences were more marked between the
three compounds at the Ru L2-edge than at the Cl K-edge. The white line in the Ru-LSK-15b
spectra displayed a +0.6 eV shift compared to 6a, and an increase in relative broadness that
may be explained by the disorder within the complex. A direct comparison with literature data
is difficult, as Ru L2-edge XAS reports are only available for ruthenium(III)
complexes.[330,331,406,407] Ru L3-edge (~2838 eV) XAS has been investigated for ruthenium(II)
complexes,[327,329] but its proximity to the Cl K-edge makes the spectrum analysis
challenging;[408] the more so in the case of LSK-15, which intrinsically contains chlorine
(broad spectral feature at ~2827 eV).

103
Chapter 6

Figure 6.4 Cl K-edge (left) and Ru L2-edge (right) XAS of 6a, 6b and Ru-LSK-15b.

Scanning transmission electron microscopy (STEM) coupled to energy-dispersive

X-ray (EDX) spectroscopy was used to record a potential aggregation of the ruthenium
centers. No large ruthenium clusters were observed in the material, suggesting that the
impregnation yielded a molecular ruthenium catalyst (Figure 6.5, a). EDX spectroscopy
confirmed the presence of ruthenium in the field of view (Figure 6.5, b and c). A ruthenium
content of 0.6% w/w was measured by atomic emission spectroscopy of a digested sample.
Samples exposed to air showed clear signs of degradation with the color progressively
turning gray. In such samples, ruthenium clusters (20-50 nm) were observed in the electron
micrographs, demonstrating the high sensitivity of the catalyst towards oxygen and moisture.

Figure 6.5 (a) Secondary electron- and (b) HAADF-STEM images of Ru-LSK-15b with
region-of-interest for EDX spectroscopy marked by a red square. (c) EDX spectrum of the
sample on a copper plate.

104
 Supported Grubbs’ catalyst for liquid and gas phase olefin metathesis

6.4 Self-metathesis of terminal alkenes

6.4.1 1-Hexene self-metathesis in the liquid phase

Figure 6.6 Self-metathesis of 1-hexene to 5-decene.

The self-metathesis of 1-hexene to 5-decene at 25 °C in toluene (Figure 6.6) was used as

benchmark to explore the optimal reaction conditions. Catalytic performances of 6a, 6b
and Ru-LSK-15b in the self-metathesis of 1-hexene were tested in benzene at 25 °C
(Table 6.1, 1a-e). The activity of Ru-LSK-15b was relatively low (1c) compared to 6a (1a),
which may be explained by the difference in electron-donating character between the PCy3
ligand and the phosphine moiety within LSK-15. The similar activity between Ru-LSK-15b
and 6b suggests that the ruthenium catalyst may be present within LSK-15 as a pyridine
complex or that the reaction is strongly limited by diffusion. Ru-LSK-15b was also tested at
50 °C with a significant increase in activity (1d), albeit with a lower selectivity. The selectivity
towards 5-decene is decreased at high conversion for all cases expect for Ru-LSK-15b at
room temperature (1c), potentially due to the availability of internal olefins to form longer
metathesis products or to isomerization reactions.[409] To investigate the role of the phosphine
moieties, the impregnation procedure was applied to MIL-101(Al)-NH2. Considerable leaching
occurred during the washing step, and resulted in no observable ruthenium incorporation and
detectable catalytic activity (1e), which indicates that the presence of phosphine moieties is
required to retain the ruthenium catalyst.

Table 6.1 Catalytic performances of 6a, 6b and Ru-LSK-15b in the self-metathesis of

1-hexene.

Catalyst T °C t=1h t=4h

Conversion [b]
Selectivity Conversion Selectivity[b]
(%) (%) (%) (%)
1a 6a 25 62 88 73 76
1b 6b 25 33 82 68 74
1c Ru-LSK-15b 25 26 83 50 82
1d Ru-LSK-15b 50 83 60 96 42
1e Ru@MIL-101[a] 25 0 - 0 -
[a] [b]
Impregnation procedure applied to MIL-101(Al)-NH2 as described in Chapter 2. Measured by
GC-FID with dodecane as internal standard. Reaction conditions: [Ru] = 8.5  10-3 mmol/mL,
Ru / 1-hexene = 1 / 100 in benzene (0.7 mL), batch reactor under inert atmosphere.

105
Chapter 6

6.4.2 Propylene self-metathesis in the gas phase

[Ru]
2 +
Figure 6.7 Propene self-metathesis.

Following the results in the self-metathesis of 1-hexene, we attempted a similar reaction in

the gas-phase. The self-metathesis of propene (Figure 6.7) was performed because of the
low boiling points of both reactant and products that prevent excessive condensation within
the material and the setup. Ru-LSK-15b was tested at various temperatures using the
continuous flow setup described in Chapter 2. Under these conditions, a maximum turnover
frequency (TOF = 1708 h-1) was achieved at 110 °C, followed by a rapid deactivation of the
catalyst characterized by a strong decrease in the conversion in the downward
temperature ramp (Figure 6.8, left). The catalyst was subsequently tested at 100 °C
(Figure 6.8, right). Ru-LSK-15b displayed a moderate initial activity (TOF = 152 h-1 at 40 °C),
followed by a rapid deactivation in the following 4 h (TON = 970) and the same behavior was
observed at 80 °C over 5 h (TON = 1110). Despite a modest activity, this example serves as
a proof-of-concept for the immobilization of molecular ruthenium catalysts for gas-phase
metathesis reactions.

Figure 6.8 Continuous self-metathesis of propylene in the gas phase. (left) Conversion of
propene at various temperatures (0.5 °C/min ramp, 10 mol% propene, and 10 mL/min
total flow in argon). (right) Propene self-metathesis at 100 °C (2 °C/min ramp until 100 °C,
10 mol% propene, and 2 mL/min total flow in argon).

106
 Supported Grubbs’ catalyst for liquid and gas phase olefin metathesis

6.5 Cross-metathesis of functionalized olefins

6.5.1 Cross-metathesis of 1-hexene with 5-hexenyl acetate

Figure 6.9 CM reaction between 1-hexene and 5-hexenyl acetate.

We selected 5-hexenyl acetate for its reported use in the CM of functionalized olefins.[410]
In addition, the cross-metathesis of 5-hexenyl acetate with 1-hexene yields 5-decenyl acetate
(Figure 6.9), a key intermediate in the production of natural pheromones.[411] A marked
difference in activity was observed between the homogeneous complexes and the supported
catalysts (Table 6.2, 2a-f). In comparison to 6a and 6b (2a,b), Ru-SK-15b was significantly
less active, reaching 96% conversion of 5-hexenyl acetate in 96 h (2d,e), but produced
5-decenyl acetate in a similar selectivity. This was attributed to a lower metal loading in the
material (Ru / olefin = 180 compared to 1 / 100 in homogeneous conditions), to diffusion
limitations due to the relative bulkiness of the products, and to the electronic difference
between the ligand responsible for the initiation of the metathesis (vide supra). Metathesis
reactions in the presence of Ru-LSK-11b and Ru-LSK-17b did not yield any detectable
conversion (2c,f). The lack of reactivity suggests that no active ruthenium is leached out of
the frameworks during reaction and may be caused instead by pore size limitations.

Table 6.2 Conversion values for the CM of 1-hexene and 5-hexenyl acetate in the presence
of 6a, 6b and selected P-MOFs.

Catalyst Conversion (%)[a] Selectivity (%)[a]

1-hexene 5-hexenyl acetate 5-decenyl acetate
[b,c]
2a 6a 100 100 28
2b 6b[b,c] 100 100 25
2c Ru-LSK-11b[c] 0 0 -
2d Ru-LSK-15b[b,e] 27 38 32
2e Ru-LSK-15b[d,e] 66 96 25
2f Ru-LSK-17b[c] 0 0 -
[a] [b]
Recorded by GC-MS with dodecane as internal standard, measured after 24 h,
[c] -3 [d] [e]
[Ru] = 2.1  10 mmol/mL and Ru / olefin = 1 / 100, measured after 96 h, [Ru] = 1.2  10-3
mmol/mL and Ru / olefin = 1 / 180. Reaction conditions: 1-hexene / 5-hexenyl acetate = 1 / 1 at 45 °C
in DCM (2 mL), batch reactor under inert atmosphere.

Ru-LSK-17b underwent a noticeable color change during reaction, from green to light brown,
which may indicate a change of ligand coordination onto the ruthenium catalyst. This may be
imputable to additional reaction with the framework or possibly to a localized hindrance of the

107
Chapter 6

carbene moiety that prevents further olefin reaction. Framework-induced hindrance was
already reported for P-MOF organocatalysts,[412] and might explain the behavior of
Ru-LSK-17b.

While the selectivity towards 5-decenyl acetate was similar for the homogeneous catalysts
and Ru-LSK-15b, the distribution of secondary olefins was noticeably different
(Figure 6.10, a). This is a key aspect in CM, were the control of product distribution is often
limited.[383] In this case, self- and cross-metathesis occurred in parallel with isomerization
reactions. The selectivity of Ru-LSK-15b (68%) towards the self-metathesis product of
5-hexenyl acetate, 5-decenyl diacetate, was significantly higher than for 6a (36%) and 6b
(34%) at similar conversion level. Only small amounts of isomerization products were directly
observed (< 5% for all catalysts), but subsequent metathesis of the isomerized olefins led to
quantifiable amounts of by-products (Figure 6.10, b). Their analysis indicates that only
5-hexenyl acetate was converted into 4-hexenyl acetate, while 2-hexene and subsequent CM
reaction products with 5-hexenyl acetate were not detected. Instead, 1-hexene reacted with
4-hexenyl acetate to form 5-nonyl acetate. A fraction of unidentified by-products was also
observed in the case of the two homogeneous catalysts and are hypothetically attributed to
secondary metathesis reactions. These differences in product composition have a direct
impact on the reaction yield because self-metathesis products might potentially be converted
further into the desired CM product in the presence of an excess of reactants, while other
by-products are leading to a permanent loss.

Figure 6.10 (a) Products selectivity of the CM reaction of 1-hexene with 5-hexenyl acetate
and (b) observed by-products formed by CM reactions with 4-hexenyl acetate.

A hot filtration experiment was conducted on Ru-LSK-15b after 24 h of reaction. Both the
filtrate and the reaction mixture containing Ru-LSK-15b were left to react further for an
additional 42 h, while substrate conversion levels were monitored (Figure 6.11, left). No
conversion of 5-hexenyl acetate was observed in the filtrate, while it significantly increased in

108
 Supported Grubbs’ catalyst for liquid and gas phase olefin metathesis

the presence of the Ru-LSK-15b. This confirms that the reaction occurs within the pores of
LSK-15 and not via leaching of active species. We also noticed that the conversion of
1-hexene increased in the filtrate. However, the concentration of the two self- and
cross-metathesis products of 1-hexene remained stable after filtration (Figure 6.11, right), as
opposed to the vessel containing Ru-LSK-15b where the product concentrations increased
markedly over time. We attribute the observed conversion of 1-hexene in the filtrate to a
possible slow accumulation in the large setup overhead due the reaction conditions operating
close to the boiling point of the 1-hexene / DCM mixture.

Figure 6.11 Hot filtration experiment at 45 °C: (left) conversion of 1-hexene and 5-hexenyl
acetate and (right) selected product concentrations normalized by dodecane as internal
standard. Black = reactor containing Ru-LSK-15b, red = filtrate.

6.5.2 Cross-metathesis of 5-hexenyl acetate with protected olefin.

The activity of Ru-LSK-15b was further tested in the CM of more challenging olefins.
5-hexenyl acetate was reacted with 3-butene-2-ol, 2-benzolyloxy-3-butene, methyl acrylate
and acrolein dimethyl acetal under the same conditions as for 1-hexene. These substrates
are all functionalized, short olefins that are not prone to self-metathesis.[410] Their
cross-metathesis has been reported for Grubbs-type catalysts, albeit with moderate yields.[382]
In addition, they contain various functional groups, allowing the evaluation of their effect on
the metathesis within a constrained environment. However, no detectable conversion was
observed in the presence of Ru-LSK-15b for any of the substrates. Purification of the
substrates did not yield any conversion. Self-metathesis of 5-hexenyl acetate was not
observed in the presence of functionalized olefins, which may indicate a possible deactivation
of the catalyst by the substrates. Further enhancement of the phosphine basicity in LSK-15
may be required to perform cross-metathesis for this class of compounds.

109
Chapter 6

6.6 Conclusion

In this chapter, we successfully immobilized a 2nd generation Grubbs’ catalyst within the
pores of LSK-15 via a two-step ligand exchange strategy. The catalyst was active in CM
reactions in the liquid-phase with 1-hexene and 5-hexenyl acetate as substrates and in the
continuous gas-phase self-metathesis of propene. Despite a moderate activity and stability,
LSK-15 retained the ruthenium catalysts within its pore, as demonstrated by hot filtration
experiments, and influenced the olefin distribution, possibly via steric hindrance. This is highly
relevant for a class of catalyst that normally operates via a catch-and-release mechanism at
the phosphine ligand, for which the immobilization is retained by the constrained environment
provided by the MOF. This validates our strategy and provides important criteria for future
linker development to generate high-performance cross-metathesis catalysts within P-MOFs.

110
 Chapter 7

Rh-catalyzed asymmetric hydrogenation of olefins

in the presence of P-MOFs
Chapter 7

7.1 Introduction

The asymmetric hydrogenation of olefins, ketones and imines are key catalytic processes in
modern chemistry for the production of optically active molecules.[413] First reported by
Akabori et al. in 1956,[414] it was further developed by Knowles[265] and Noyori[415] with
rhodium- and ruthenium organometallic complexes as catalysts, gradually replacing the
kinetic resolution of enantiomers.[10,11] Nowadays, asymmetric hydrogenations are applied to
many key intermediates for the production of enantiomerically enriched drugs and fine
chemicals.[262,413,416] A popular class of catalysts for olefin hydrogenation is based on
[Rh(diene)(PP*)]X (X = BF4-, OTf-), where PP* is a chiral diphosphine ligand that control the
enantioselectivity of the hydrogenation.[265,417,418] Alternatively, monodentate chiral phosphines
can provide a more flexible approach to asymmetric hydrogenation in the form of
[Rh(diene)(P1*)(P2)]X,[264,419,420] which uses a combination of chiral and achiral phosphine
ligands to fine-tune the stereoselectivity of the catalysts.[267,421–425]

MOFs have been extensively investigated for asymmetric catalysis in the past decade.[158–160]
Among the most reported strategies, the use of homochiral frameworks supporting bidentate
ligands facilitates the immobilization of well-defined molecular complexes that promote high
enantioselectivity.[84,104,118,426,427] Bidentate P-MOFs were notably reported by Lin et al. as
support for rhodium and ruthenium precursors active in the asymmetric hydrogenation of
ketones.[206,215] However, controlling the contribution of the environment in enantioselective
reactions remains a major challenge and most selections of substrate/framework pairs still
rely heavily on experimental findings, often hindered by the complicated synthesis of novel
chiral frameworks. Alternatively, studies have suggested the use of achiral frameworks in
asymmetric catalysis. For example, Gascon and Bert reported the application of MIL-101(Cr)
in aldol condensation reactions, in which the framework was post-synthetically functionalized
by chiral modifiers.[428] Vilhanová et al. recently studied the positive effect of achiral zinc
MOFs on the rhodium-catalyzed asymmetric hydrogenation of prochiral olefins.[429]
Complexes could be recycled due to non-covalent interactions with the MOFs, promoting
their immobilization despite a lack of coordination sites. In this work, we propose to reinforce
these interactions with the use of rhodium complexes built from the combination of achiral
P-MOFs and monodentate chiral phosphine ligands. This combines the topological variety of
achiral MOFs and the chemical versatility of the combinatorial approach used in
homogeneous catalysis for the design of rhodium active sites. This chapter summarizes our
preliminary attempts and provides a critical view on hydrogenation reactions in the presence
of P-MOFs.

112
 Rh-catalyzed asymmetric hydrogenation of olefins in the presence of P-MOFs

7.2 Hydrogenation of acrolein with Rh/PPh3-LSK-3

Figure 7.1 Rhodium-catalyzed hydrogenation of acrolein.

The hydrogenation of acrolein in the presence of LSK-3 was performed to confirm the activity
of rhodium-functionalized P-MOFs in hydrogenation reactions. LSK-3 was previously used in
organocatalysis[218] and for the coordination of rhodium complexes (Chapter 4), but no
application in metal catalysis has been demonstrated so far. The catalyst was prepared by
suspending Rh-LSK-3 (Chapter 4) in a DCM solution containing PPh3 (1 equivalent to
rhodium). The crystals underwent a change in color from orange to yellow to form
Rh/PPh3-LSK-3 and no significant leaching was observed upon storage in solution. The
catalyst displayed only moderate performance (40% conversion in 24 h) and selectivity to
propanal (87%,Figure 7.1) as compared to [Rh(COD)(PPh3)2]BF4 in similar conditions
(87% conversion, 100% selectivity). However, this still attests to the potential use of P-MOFs
containing monodentate phosphines for the preparation of rhodium-phosphine catalysts.

7.3 Zinc P-MOFs and chiral monodentate phosphines as ligands for the
asymmetric hydrogenation of olefins

Figure 7.2 In-situ formation of the [Rh(COD)(P-MOF)((S)-NMDPP)] complex for the

asymmetric hydrogenation of olefins.

Hydrogenation in the presence of LSK-3 confirmed the potential of P-MOFs in the retention of
a rhodium catalyst for the hydrogenation of olefins. However, the P-MOFs reported in this
work are not intrinsically suited for the asymmetric hydrogenation of prochiral olefins due to
the achiral nature of the PPh2-bdc linker. While homochiral P-MOFs were reported by Lin et
al. for hydrogenation applications,[206,215] we opted instead for the in-situ generation of
rhodium catalysts using an achiral P-MOF and a chiral monodentate phosphine (Figure 7.2).
Combinatorial approaches using two monodentate phosphines have already been
demonstrated for homogeneous complexes,[267,422,424,425,430] but remain an exception for
supported phosphines.[431,432]

113
Chapter 7

We selected (S)-(+)-neomenthyldiphenylphosphine ((S)-NMDPP)[433] as the chiral

monodentate phosphine for its resemblance to PPh2-bdc and documented performance in the
hydrogenation of olefins.[422,434] Dimethyl itaconate was used to probe the reactivity and
selectivity of the system with various P-MOFs in comparison to a triphenylphosphine-
supported polymer (1.4-2.0 mmol PPh3/g on polystyrene). The materials were suspended in a
DCM solution containing [Rh(COD)2]BF4 (7.5 × 10-3 mmol) and (S)-NMDPP so that the ratio
Rh / (S)-NMDPP / P-MOF was 1 / 1.1 / 1.1 for each example. The phosphorus content in the
P-MOFs was derived from their linker ratio measured by 1H NMR spectroscopy after
digestion (Chapter 2). Conversion and enantiomeric excess (ee) were recorded
for two zinc(II) MOF topologies, i.e. MOF-5 (LSK-11 and LSK-13) and UMCM-1
(MIXUMCM-1-PPh2,[223] referred here as LSK-16) (Table 7.1, 1a-d). Both the presence of
LSK-11 (1a,b) and LSK-16 (1d) in the reaction mixture have a positive effect on the
hydrogenation selectivity, as compared to the phosphine-functionalized polymer support (1e)
but relatively little effect on the activity of the catalyst. While all the tested materials are
functionalized with similar triphenylphosphine groups, crystalline P-MOFs may provide a
well-defined environment that promotes a unique rhodium center, while the disordered nature
of the phosphine-functionalized polymer results in the formation of a mixture of catalytic
species, which may be detrimental to the overall selectivity. Recycling of the catalyst was
attempted for 1a but did not yield any significant conversion, presumably due to the sensitive
nature of the catalyst under hydrogen starvation conditions. The activity of the catalyst did not
significantly drop when the phosphine content was increased (1a,b) but the presence of
amine groups in LSK-13 seems to markedly affect the reactivity of the rhodium complex (1c).
This was not observed in LSK-16, which displayed an activity similar to LSK-11, indicating
that amine groups may influence the rhodium center in a specific geometry (vide infra).

Table 7.1 Hydrogenation of dimethyl itaconate using a combination of [Rh(COD)2]BF4 and

(S)-NMDPP in the presence of phosphine-functionalized supports.

MOF t=2h t = 24 h
Conversion (%) ee (%)[d] Conversion (%) ee (%)[d]
1a LSK-11 (1:2)[a] 12 51 57 69
1b LSK-11 (1:1) [a] 31 74 63 72
1c LSK-13 (1:10)[b] 1 48 18 19
1d LSK-16 (1:4)[b,c] 16 55 63 65
1e PPh3-polymer 19 28 67 22
[a] [b] [c]
PPh2-bdc to bdc molar ratio, PPh2-bdc to NH2-bdc molar ratio, the synthesis of
MIXUMCM-1-NH2-PPh2 (LSK-16) is reported elsewhere,[223] [d]ee recorded by GC-MS using a chiral
-3
column (see Chapter 2). Reaction conditions: [Rh] = 3.75 × 10 mmol/mL, Rh / (S)-NMDPP / P-MOF /
olefin = 1 / 1.1 / 1.1 / 100 in DCM (2 mL) at room temperature.

114
 Rh-catalyzed asymmetric hydrogenation of olefins in the presence of P-MOFs

For a combination of two non-identical monodentate phosphines, the formation of the catalyst
under hydrogenation conditions is influenced by the competition between the two ligands for
the coordination to the metal center. In the case of P-MOFs, the supported phosphines
groups may be less available for rhodium coordination than (S)-NMDPP due to
diffusion limitations within the crystals. This potentially favor the formation of a
[Rh(COD)((S)-NMDPP)2]BF4 complex instead of the desired [Rh(P-MOF)((S)-NMDPP)]BF4
that may impact the selectivity of the reaction. We investigated this effect using
homogeneous catalysts formed in-situ from mixtures of PPh3 and (S)-NMDPP in various
amounts (Table 7.2). Moderate ee were observed only in the case where (S)-NMDPP / PPh3
ratio > 1 (20-40% ee, 2c-2e), while the equimolar mixture of the two ligands (2b) led to a non-
selective complex. This suggests that PPh3 preferentially coordinates the rhodium over
(S)-NMDPP, which is an argument in favor of using immobilized triphenylphosphine ligands
for combinatorial experiments. The use of 2.2 equivalents of the chiral phosphine did not yield
a significant increase in selectivity (36% ee, 2f), but surprisingly, 1.1 equivalent of
(S)-NMDPP (58% ee, 2g) resulted in enantiomeric excess comparable to the experiments in
the presence of P-MOFs.

Table 7.2 Homogeneous hydrogenation of dimethyl itaconate using [Rh(COD)2]BF4 and

various PPh3 and (S)-NMDPP mixtures reacted in-situ.

Rh / P ratio t=2h t = 24 h
Conversion Conversion
PPh3 (S)-NMDPP ee (%)[a] ee (%)[a]
(%) (%)
2a 2.2 - n.d n.d 100 0
2b 1.1 1.1 19 2 100 1
2c 0.7 1.5 53 11 91 20
2d 0.4 1.8 64 22 100 19
2e 0.2 2 60 37 93 40
2f - 2.2 20 39 81 36
2g - 1.1 n.d. n.d. 100 58
[a]
ee recorded by GC-MS using a chiral column (see Chapter 2). Reaction conditions: [Rh] = 3.75
× 10-3 mmol/mL, Rh / (S)-NMDPP / P-MOF / olefin = 1 / 1.1 / 1.1 / 100 in DCM (2 mL) at room
temperature.

The same effect was observed in the hydrogenation of other typical prochiral olefins
(Table 7.3, 3a-d), where the enantioselectivity was radically enhanced in the presence of
1.1 ligand equivalents, both in the absence and presence of LSK-11. In particular, very high
enantioselectivity (> 99%) was achieved in the hydrogenation of (z)-methyl-2-acetamido-3-(4-
nitrophenyl)-acrylate (3d), while the typical rhodium-to-ligand ratio of 2.2 did not yield an
optically active product mixture.

115
Chapter 7

Table 7.3 Hydrogenation of prochiral olefins catalyzed by [Rh(COD)2]BF4, in combination with

(S)-NMDPP and LSK-11.

LSK-11 (1.1 eq.)

(S)-NMDPP (1.1 (S)-NMDPP (2.2
(S)-NMDPP (1.1
eq.) eq.)
eq.)
Conversio ee Conversio ee Conversio ee
Substrate
n (%) (%)[a] n (%) (%)[a] n (%) (%)[a]

3a 100 47 100 57 100 32

3b 100 56 100 66 100 22

3c 94 -59 89 -62 37 -16

3d 100 > 99 100 > 99 100 -1

[a]
ee recorded by GC-MS using a chiral column. Reaction conditions: [Rh] = 3.75 × 10-3 mmol/mL,
Rh / olefin = 1 / 100 in DCM (2 mL), 24 h at room temperature.

The main complication for hydrogenation catalysts prepared in-situ from a mixture of ligands
is the competition for the coordination to the rhodium center. While experiments with PPh3
show that triarylphosphine preferentially coordinates rhodium in the presence of (S)-NMDPP,
presumably saturating the catalytic centers and leading to low enantioselectivity, P-MOFs
displayed a more intricate behavior. The close similarity between the hydrogenation results in
the presence of LSK-11 and 1.1 equivalent of (S)-NMDPP (Table 7.3) makes the P-MOF
contribution difficult to rationalize. While P-MOFs provide isolated phosphine moieties,
preventing the saturation of the rhodium center, diffusion limitations may promote the
formation of rhodium-(S)-NMDPP complexes outside the framework that contributes
significantly to the catalytic activity and selectivity. Comparison with the literature is however
difficult as hydrogenation experiments with monodentate phosphine is usually carried out with
a rhodium-to-phosphine ratio equal to 2 or higher.

116
 Rh-catalyzed asymmetric hydrogenation of olefins in the presence of P-MOFs

7.4 Hydrogenation in the presence of amino-functionalized MIL-101

frameworks

The hydrogenation of dimethyl itaconate in the presence of functionalized MIL-101(Al)

frameworks displayed a behavior that contrasted strongly with our observations for zinc(II)
MOFs (Table 7.4). No product conversion was observed when [Rh(COD)2]BF4 or
[RhCl(COD)]2 were reacted with (S)-NMDPP in the presence of LSK-15 (4a,b) or
MIL-101(Al)-NH2 (4e). This suggests that no phosphine-rhodium complex is formed outside
the framework and that amine moieties in the MIL-101 topology are detrimental to the
catalytic activity. This is unexpected as primary amines are well-known substrates for a wide
range of rhodium-catalyzed reactions,[435] but their role as ligands for transition metals is far
less common.[436]

We attempted to mitigate this effect in LSK-15 by reacting NH2-bdc post-synthetically with

benzyl isothiocyanate and acetic anhydride[437] to form the thiourea (LSK-15-Tu) and amide
(LSK-15-Am) derivatives, respectively. The activity of the rhodium catalyst was partially
recovered in the case of LSK-15-Am (3d) but resulted in a racemic mixture, while the use
LSK-15-Tu led to no significant activity (3c). While the exact role played by the amine groups
has yet to be investigated systematically, these experiments demonstrate the importance of
the MOF environment on catalytic processes.

Table 7.4 Summary of dimethyl itaconate hydrogenation using [Rh(COD)2]BF4 in the

presence of (S)-NMDPP and MIL-101 derivatives.

MOF t=2h t = 24 h
Conversion (%) ee (%) [c]
Conversion (%) ee (%)[c]
4a LSK-15 0 - 0 -
4b LSK-15[a] 0 - 0 -
4c LSK-15-Tu[b] 3 40 4 36
4d LSK-15-Am[c] 2 29 43 3
4e MIL-101(Al)-NH2 0 - 0 -
[a] [b] [c]
[RhCl(COD)]2 used instead, thiourea-functionalized LSK-15 , amide-functionalized LSK-15.
-3
Reactions conditions: [Rh] = 3.75 × 10 mmol/mL, Rh / (S)-NMDPP / P-MOF / olefin = 1 / 1.1 / 1.1 /
100 in DCM (2 mL) at room temperature.

117
Chapter 7

7.5 Conclusion

This chapter summarizes our preliminary attempts in rhodium-catalyzed asymmetric

hydrogenation of olefins. Using LSK-3 as an example, we demonstrated that rhodium
catalysts can be immobilized by post-synthetic metalation to yield a center active in the
hydrogenation of acrolein, opening new perspectives in the incorporation of molecular
catalysts for future gas-phase applications. Due to the achiral nature of the P-MOFs
developed in this work, the presence of a chiral monodenate phosphine is required for
asymmetric hydrogenations of prochiral compounds. This resulted in a competitive
coordination behavior between the immobilized phosphine groups and the phosphine
adducts, similar to the combinatorial approach in homogeneous catalysis. This ligand
competition resulted mostly in non-selective complexes for the case of triphenylphosphine in
solution, and the commercial triphenylphosphine-functionalized polymer. Remarkably, the
phosphine site isolation provided by the P-MOFs prevented the saturation of the rhodium
center, yielding moderate enantioselectivity enhancements. However, the roles of the
framework topology and the amine groups were difficult to rationalize based on preliminary
results only. This illustrates the complexity surrounding catalytic systems that are present in
both the liquid and the solid phase, for which future systematic study of the ligand-metal
interactions should be supported by ab-initio calculations and in-situ characterization.

118
 Chapter 8

Conclusion and Outlook

Chapter 8

8.1 Conclusion

The scope of this thesis is to design novel P-MOFs based on linear dicarboxylate linkers and
evaluate their applications as supports for transition metal catalysts. To this aim, a series of
P-MOFs based on the diphenylphosphine-functionalized terephthalic acid, PPh2-bdc, was
synthesized using a mixed-linker approach. We demonstrated that the PPh2-bdc linker is
compatible with a variety of organic linkers, metal ions, and MOF topologies. All materials
exhibited permanent porosity and crystallinity comparable to their respective parent materials,
which represents a significant advance in the production of stable P-MOFs. Additionally, the
content of phosphines and other functional groups can be tuned by varying the linker ratios
without significant modification of the synthesis protocols. These features validate our
strategy for the incorporation of phosphines and provide a versatile method for the design
and optimization of future structures, including multi-functional supports.

Throughout this work, characterization of P-MOFs in the solid-state was used to access
chemical information that is often lost when applying traditional techniques based on sample
digestion. Element-specific techniques, such as 31P MAS NMR spectroscopy and P K-edge
XAS, provide an accurate description of electronic properties of the phosphine groups within
P-MOFs. Additionally, a combination of multi-edge XAS and MAS NMR spectroscopy yields a
powerful methodology to characterize the local coordination shell of organometallic
complexes coordinated to P-MOFs. This work highlights a unique aspect of P-MOFs, namely
that the hybrid nature of the framework, the high crystallinity, and the presence of phosphorus
provide the opportunity for the direct and indirect characterization of metal complexes at
various length scales with unparalleled chemical precision.

Three distinct catalytic applications were tested in this work to illustrate the chemical
versatility of P-MOFs. These materials were able to immobilize and durably retain molecular
complexes under different reaction conditions. This work successfully demonstrates that
P-MOFs are true solid ligands, which are closely related to their homogeneous counterparts.
While this led to promising results—in particular for Ru-LSK-15a that challenges nanoparticle-
based systems and homogeneous catalysts—it also highlighted some of the limitations
surrounding this type of materials. Phosphines are one of the most popular classes of ligands
in homogeneous catalysis due to the large number of structures available. Each complex can
thus be optimized by screening a collection of potential phosphine candidates, yielding a
high-performant catalyst for each specific organic transformation. However, this is not yet
achievable in P-MOFs as the number of phosphine moieties and P-MOF structures known is
still too limited.

120
 Conclusion and Outlook

Nonetheless, the results presented herein highlight the enormous potential of P-MOFs in the
incorporation of molecular complexes. Stable, porous, and highly ordered pockets containing
phosphines were achieved, which significantly extend the field of P-MOF synthesis.
In addition, this work provides fundamental guidelines for the characterization and future
applications of P-MOFs in metal-supported catalysis.

8.2 Outlook

8.2.1 Synthetic strategies

Based on the results reported herein, the design of P-MOFs will combine isoreticular
chemistry with the tuning of phosphine groups (Figure 8.1). The protocol for the synthesis of
PPh2-bdc described in this work may be adapted to generate a series of organophosphine
linkers of the type PR2-bdc (R = methyl, isopropyl, cyclohexyl, etc.). The local coordination
sphere of the phosphine can thus be adjusted by its carbon substituents, independent from
the MOF topology. This allows designing a family of solid porous ligands with similar physical
attributes but distinct properties for the coordination of transition metals. Besides phosphine
modification, isoreticular chemistry and a mixed-linker approach can be used in parallel to
control the framework properties. This includes the framework order, such as the topology,
linker ratio, and spatial distribution of linkers, and also its physical properties
(crystal size, diffusivity, and chemical compatibility). Future materials may therefore be
tailored for applications that require well-defined catalytic pockets, such as regio-selective
and asymmetric reactions, combining existent knowledge on organometallic complexes with
computational chemistry tools such as density functional theory (DFT), to mimic highly
performing homogeneous catalysts or design novel reactive environments.

Figure 8.1 Design concept using both phosphine tuning and isoreticular chemistry to
generate a library of P-MOFs.

121
Chapter 8

During this work, the incorporation of the PPh2-bdc linker was only performed at the synthesis
step, successfully creating MIXMOFs with phosphine functional groups. The compatibility of
the phosphines was demonstrated with several metal ions and organic linkers; however, a
major drawback of this method is the partial oxidation of the phosphines when exposed to
high temperature. The growing interest in microwave-assisted synthesis of MOFs represents
a suitable alternative to the synthesis of P-MOFs by conventional heating.[79,438–440]
In particular, microwave heating has been shown to drastically reduce crystallization
times,[441–443] which may be beneficial to phosphine integrity. Frameworks requiring high
synthesis temperatures, such as MIL-101(Cr), may then be synthesized with low phosphine
oxide content, as oxidation by traces of oxygen is often limited kinetically.

Oxidation may also be prevented by using post-synthetic functionalization strategies, such as

post-synthetic reduction and post-synthetic ligand exchange (PSE). Not yet reported for
P-MOFs, post-synthetic reduction may be achieved in chemically resilient frameworks by
using mild reducing agents. If successful, it would provide a method to protect and
regenerate phosphines after synthesis conditions. Alternatively, PSE may be used to
incorporate the phosphine groups post-synthetically, allowing for a finer control of
phosphorus loading and preventing phosphine oxidation thanks to the mild thermal conditions
required. PSE may also provide a way to generate P-MOFs for which synthetic conditions are
not compatible with phosphine moieties. It is likely that, in the future, the synthesis of P-MOFs
will encompass a broader range of synthetic techniques that are currently being investigated
for other MOFs, such as the ones described in this work, selected on the basis of the relative
stability of the targeted phosphine group. Stability and chemical compatibility criteria can
notably be drawn from the extensive knowledge of phosphine ligands in the literature.

Additionally, there are other classes of porous materials that may be compatible with our
phosphine incorporation strategy to extend the scope of porous solid ligands. For example,
Jiang et al. reported a covalent organic framework (COF) formed by condensation of
2,5-dimetoxyterephthaldehyde with 1,3,5-tris(4-aminophenyl)benzene in synthetic conditions
that are compatible with phosphine moieties.[444] This suggests that an aldehyde version of
PPh2-bdc might be used for the synthesis of P-COFs (Figure 8.2). While P-MOFs benefit from
the inorganic units to provide a wide range of structures, P-COFs may be used in the future
to incorporate organometallic species that are active in media incompatible with MOF
structures, e.g. in an aqueous basic environment.

122
 Conclusion and Outlook

Figure 8.2 Potential P-COF synthesis based on the condensation of diphenylphosphine-

substituted terephthaldehyde with 1,3,5-tris(4-aminophenyl)benzene.

8.2.2 Characterization
31
A key part in this thesis is the demonstration of the ability of solid-state P MAS NMR
spectroscopy and XAS to provide the phosphorus electronic state within P-MOFs. While a
short-range order is critical for the design of active organometallic catalysts, further
phosphorus characterization is required at various length scales to yield optimized materials.
Very little is known about the distribution of phosphine linkers within the cage system of
MIXMOFs. In Chapter 3 and Chapter 4, we observed that complex topologies such as MIL-
101 and UMCM-1 resulted in a mixture of P(III) sites that could not be discriminated using
XAS at the P K-edge. The localization of these P(III) sites within the cage may be achieved in
the future by using a combination of space-dependent NMR techniques such as rotational
echo double resonance (REDOR) and two-dimensional spectrum acquisition methods based
on spin correlations.[445] For example, the use of REDOR, in combination with ab-initio
calculations, was recently used to characterize the local distribution of linkers within
zinc-based MIXMOFs.[446] The acquisition of high-resolution 13C and 15N spectra may also be
achieved using dynamic nuclear polarization NMR spectroscopy,[447,448] which might be
potentially applied to the characterization of immobilized organometallic complexes in
P-MOFs. Characterization at the scale of a single-crystal was briefly explored in Rh-LSK-11
as a proof-of-concept using a micro-focused X-ray beam. In future applications, combined
fluorescence mapping and XAS at the P K-edge may be complemented by other
synchrotron-based techniques such as high-resolution X-ray tomography to understand the
effect of phosphorus special distribution on metal loading and reactivity.[449]

123
Chapter 8

8.2.3 Catalysis

During this work, rhodium catalysts in the presence of P-MOFs were investigated for the
asymmetric hydrogenation of olefins. Alternatively, these systems might be potentially
applied in other rhodium-catalyzed C=C activation reactions, such as hydroformylation.[192]
Hydroformylation of terminal olefins was recently reported using Rh@MIL-101(Cr)
nanoparticles at 100 °C under 50 bar with a mixture of hydrogen and carbon monoxide
(1:1).[138] Preliminary results in the hydroformylation of 1-hexene with [Rh(acac)(COD)] and
LSK-15 displayed enhanced selectivity towards aldehyde products, in comparison to the
nanoparticle-based system (Table 8.1). Despite a lower ratio of linear/branched (l/b)
aldehydes, the system was recycled 4 times with minor loss of activity, highlighting the
potential of P-MOFs in the immobilization of rhodium complexes under hydroformylation
conditions.

Table 8.1 Preliminary results in the hydroformylation of 1-hexene in the presence of

[Rh(acac)(COD)] and LSK-15.

[Rh][a]
[Rh][a] + LSK-15 Rh@MIL-101(Cr)[b]
+ 2 PPh3
1st run 2nd run 3rd run 4th run
Conversion (%) 96.9 98.0 98.3 93.1 97.7 96.4
Aldehyde 20.3
87.0 83.0 84.2 92.7 93.2
selectivity (%)
l/b ratio 0.9 1.1 0.9 0.8 1.9 2.3
olefin/Rh 1000 1000 1000 1000 1000 70,000
[a]
[Rh] = [Rh(acac)(COD)], p = 50 bar H2/CO mixture, T = 100 °C, t = 4 h, [b]as reported by Schulz
et al.[138]

Throughout this work, we demonstrated that P-MOFs can act as strong coordination points
for the immobilization of transition metal nuclei, yielding heterogeneous molecular catalysts.
However, the activity of the immobilized complexes presented in this work dramatically varied
from one application to another, highlighting the need for further optimizations. An improved
strategy that encompasses both the phosphine basicity (vide supra) and the MOF
environment will be required to reach highly performant catalysts. To take the example of
Ru-LSK-15a (Chapter 5), a systematic variation of the amine groups by pre-synthesis
functionalization of the aminoterephthalic acid (Figure 8.3, a) may be performed to study the
effect of the basic environment on the rate of activation of formic acid. This strategy may be
generalized to other [RuX2(P-N-MIXMOF)(η6-arene)] immobilized complexes, as well as other
reactions catalyzed by a ruthenium-based system. For example, transfer hydrogenation,[231]
formic acid production from carbonates,[450] and the conversion of ethanol to n-butanol[451]
among others,[318,452,453] might potentially be catalyzed without the need for basic additives.

124
 Conclusion and Outlook

Figure 8.3 Examples of pore tuning strategies in P-N-MIXMOFs. (a) Optimization of amine
basicity for the dehydrogenation of formic acid and (b) orthogonal PSM strategies to induce
pore-controlled steric effects.

Another outlook for mixed phosphine-amine frameworks is their potential use in “orthogonal”
post-synthetic modifications (PSMs). Taking advantage of the reactivity differences between
amines and phosphines, the two groups may be independently modified, using the
phosphines as anchoring points for metallic species, and the amines as precursors for
organic reactions. Amino-MOFs are already described as suitable substrates for organic
modifications, such as condensation reactions with aldehydes and anhydrides, and
precursors of azides that can undergo “click chemistry”.[88–90] The distance between the
phosphine and amine groups can thus be systematically optimized using PSM to generate
organic tethers with various lengths and rigidities (Figure 8.3, b). This allows tuning the steric
environment of P-MOFs independently from the inner coordination shell of the organometallic
catalyst, providing an approach that would be similar to the design of artificial enzymes for
regio- and enantio-selective reactions.[454]

Finally, the scope of P-MOFs, and in particular of ruthenium-phosphine MOF systems, may
be expanded beyond catalytic applications. For example, half-sandwich ruthenium complexes
have been reported as potential antitumoral compounds and are thought to be the next
alternative to platinum-based drugs based on an increase in DNA intercalation potential in
tumor cells and a lower cytotoxicity.[455] The targeted delivery of ruthenium compounds using
MOFs has recently been investigated using the “ship-in-a-bottle approach”.[456,457] The use of
phosphine groups may increase the retention properties of the framework, potentially leading
to long-term or controlled release of the organometallic-based drugs. In conclusion, the
structural and chemical diversity of MOFs combined with the coordination properties of
phosphine moieties provide a flexible and promising platform, the full potential of which has
yet to be explored.

125
List of publications

Journal articles
A. Beloqui Redondo,* F. L. Morel,* M. Ranocchiari, J. A. van Bokhoven, Functionalized
ruthenium-phosphine metal-organic framework for continuous vapor-phase
dehydrogenation of formic acid, ACS Catal. 2015, 5, 7099–7103. *equal contribution

A. Schütrumpf, E. Kirpi, A. Bulut, F. L. Morel, M. Ranocchiari, E. Lork, Y. Zorlu,

S. Grabowsky, G. Yücesan, J. Beckmann, Tetrahedral tetraphosphonic acids. New building
blocks in supramolecular chemistry, Cryst. Growth Des. 2015, 15, 4925–4931.

F. L. Morel, S. Pin, T. Huthwelker, M. Ranocchiari, J. A. van Bokhoven, Phosphine and

phosphine oxide groups in metal–organic frameworks detected by P K-edge XAS, Phys.
Chem. Chem. Phys. 2015, 17, 3326–3331.

X. Xu, S. M. Rummelt, F. L. Morel, M. Ranocchiari, J. A. van Bokhoven, Selective catalytic

behavior of a phosphine-tagged metal-organic framework organocatalyst, Chem. - A Eur.
J. 2014, 20, 15467–15472.

F. L. Morel, M. Ranocchiari, J. A. van Bokhoven, Synthesis and characterization of

phosphine-functionalized metal-organic frameworks based on MOF-5 and MIL-101
topologies,
Ind. Eng. Chem. Res. 2014, 53, 9120–9127.

Book chapter
F. L. Morel,* X. Xu,* M. Ranocchiari, J. A. van Bokhoven, Functional metal-organic
frameworks: synthesis and reactivity in Chemistry of organo-hybrids: synthesis and
characterization of functional nano-objects, John Wiley & Sons, Inc., 2015, pp. 200–232.
*equal contribution

127
Oral presentations
F. L. Morel, A. Beloqui Redondo, M. Ranocchiari, J. A. van Bokhoven, A functionalized
ruthenium-phosphine MOF for the selective dehydrogenation of formic acid,
EuroMOF 2015, Potsdam (DE), 2015.

F. L. Morel, A. Beloqui Redondo, M. Ranocchiari, J. A. van Bokhoven, Incorporation of

ruthenium catalytic centers in phosphine-functionalized metal-organic frameworks,
Swiss Chemical Society Fall Meeting, Lausanne (CH), 2015.

F. L. Morel, M. Ranocchiari, J. A. van Bokhoven, Novel metal-organic frameworks with

phosphine functional groups: synthesis, characterization and applications in metal-
supported catalysis, SLS Symposium on Novel Materials, PSI Villigen (CH), 2014.

Poster presentations
F. L. Morel, A. Beloqui Redondo, M. Ranocchiari, J. A. van Bokhoven, Incorporation of a
ruthenium catalyst in a phosphine-MOF for the selective dehydrogenation of formic acid,
1st PSI Catalysis Seminar, Villigen (CH), 2015.

F. L. Morel, M. Ranocchiari, J. A. van Bokhoven, Novel phosphine MOFs based on MOF-5

and MIL-101 topologies: synthesis, characterization and applications in metal-supported
catalysis, 4th International Conference on Metal Organic Frameworks & Open
Framework Compounds, Kobe (JPN), 2014.

F. L. Morel, X. Xu, M. Ranocchiari, J. A. van Bokhoven, Towards heterogeneous

asymmetric hydrogenation of olefins using phosphine-substituted metal-organic framework,
Swiss Chemical Society Fall Meeting, Lausanne (CH), 2014.

F. L. Morel, M. Ranocchiari, J. A. van Bokhoven, Opportunities and challenges in

characterization of metal-organic frameworks by NMR spectroscopy, EMBO Summer
School on Solution and Solid-state NMR of Paramagnetic molecules, Firenze (I),
2014.

F. L. Morel, M. Ranocchiari, J. A. van Bokhoven, Expanding the series of phosphine-

substituted metal-organic frameworks: towards a suitable platform for heterogeneous
hydrogenation, Swiss Chemical Society Fall Meeting, Lausanne (CH), 2013.

F. L. Morel, M. Ranocchiari, J. A. van Bokhoven, Synthesis of novel phosphine-substituted

metal-organic frameworks and their use in enantioselective hydrogenation, Swiss
Chemical Society Fall Meeting, Lausanne (CH), 2012.

128
129
Acknowledgements
Looking back at the past four years, I realized that I could only be successful thanks to the
many people that supported me along the difficult journey that is a doctoral thesis. First, I
would like to thank Jeroen for accepting me in his group, for his guidance throughout the
project and for his friendliness that helped me to stay motivated all these years. I would like to
thank Marco for his never-ending patience in teaching me a great deal about chemistry and
helping me to become the scientist that I am today. Thank you both for the challenging
project and for all the interesting discussions that ensued. I also would like to thank Prof. Dr.
L. Kiwi, Prof. Dr. V. Hatzimanikatis and Prof. Dr. M. E. Davis who have encouraged me with
their support and teaching to pursue a Ph.D. thesis. Finally I would like to thank Prof. Dr. A.
Togni for accepting to take part in my defense committee.

Most of the work presented herein has been realized in collaboration with people that I would
like to thank for their technical help and insightful discussions. I would like to thank Amaia for
her expertise on formic acid decomposition that led us to a successful common project. I
would like to thank Thomas, Sonia and Camelia at PHOENIX, Maarten at SuperXAS, Nicola
at MS, and everybody at LSK for introducing me to the world of synchrotron spectroscopy.
Thanks to Rene, Doris and Maxence for their help with the NMR spectrometers. Thanks to
Frank and Erich for the electron microscopy tomograms. I would like to express my gratitude
to Kim, who helped me many times with manuscript proof-reading and gave me many tips to
improve my scientific writing skills. In addition, I would like to thank Xiaoying, Amaia and Jin-
Hee for their corrections and critical comments on various manuscripts and thesis chapters.
Finally I would like to thank Rifat for his good work during his Master thesis.

There are many past and present group members that I would like to thank for creating a
wonderful work atmosphere: Stephan, Christian L., Jan, Maria, Petr, Simone, Andrew, Luke,
Selmi, Patrick, Thanh-Binh, Lorenz, Erich, Thomas, Roman, Martin B., Bradley, Matthäus,
Waiz, Min Beom, Marco T., Manasa, Kim, Daniel S., Jin Hee, Andrey, Ali, Rosh, Rok,
Christian P., Arno, Zhiqiang, René, Amaia, Maxime, Vicky, Urs, Bahir, Sotiria, Piotr, Daniel
F., Cristina, Matthew, Martin M., Charlotte and Daniela. Special thanks to Beáta, Jirka and
Evalyn for making me feel welcome at my arrival at PSI. Super-special thanks to Xiaoying,
who shared my highlights and daily struggles for almost four years, and who was a
passionate colleague and a great friend. It was a pleasure and an honor to work alongside.

I would like to thank my family, especially my parents that have supported me throughout my
life and studies. At last but not least, I would like to thank Helena for all the wonderful years
spent together, for her invaluable support and for being the best thing that has ever
happened in my life.

131
Appendix
Table of abbreviations
acac Acetylacetonate
bdc 1,4-benzenedicarboxylate
bpdc Biphenyl-4-,4’-dicarboxylate
BET Brunauer-Emmett-Teller
COD 1,5-cyclooctadiene
CP Cross-polarization
DCM Dichloromethane
DEF N,N-diethylformamide
DMF N,N-dimethylformamide
DMSO Dimethyl sulfoxide
ee Enantiomeric excess
EXAFS Extended X-ray absorption fine structure
H3btc Benzene-1,3,5-tricarboxylate
HKUST Hong Kong University of Science and Technology
IMesH2 1,3-dimesityl-4,5-dihydroimidazol-2-ylidene
IRMOF Isoreticular metal-organic framework
LSK Laboratory for Catalysis and Sustainable Chemistry
MAS Magic angle spinning
MIL Material Institut Lavoisier
MOF Metal-organic framework
NMR Nuclear magnetic resonance
p-cym para-cymene
PSE Post-synthetic ligand exchange
PSM Post-synthetic modification
PTFE Polytetrafluoroethylene
TGA Thermogravimetric analysis
UiO University of Oslo
UMCM University of Michigan crystalline material
XAS X-ray absorption spectroscopy
XANES X-ray absorption near-edge structure
XRD X-ray diffraction

A-II
 Appendix

References
[1] J. Hagen, Industrial Catalysis, Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim,
FRG, 2005.
[2] G. C. Bond, C. Louis, D. T. Thompson, Catalysis by Gold, Imperial College Press,
2006.
[3] J. Garside, F. Shintaro, The Expanding World of Chemical Engineering, Taylor &
Francis, 2001.
[4] B. Cornils, W. A. Herrmann, M. Rasch, Angew. Chem. Int. Ed. 1994, 33, 2144–2163.
[5] K. Ziegler, Rubber Chem. Technol. 1965, 38, 23–36.
[6] G. Natta, Rubber Chem. Technol. 1965, 38, 37–60.
[7] G. Wilkinson, Science 1974, 185, 109–112.
[8] R. H. Grubbs, Adv. Synth. Catal. 2007, 349, 34–40.
[9] R. R. Schrock, Angew. Chem. Int. Ed. 2006, 45, 3748–3759.
[10] W. S. Knowles, Angew. Chem. Int. Ed. 2002, 41, 1999–2007.
[11] R. Noyori, Angew. Chem. Int. Ed. 2002, 41, 2008–2022.
[12] H. Davy, Philos. Trans. R. Soc. London 1817, 107, 77–85.
[13] A. J. B. Robertson, Platin. Met. Rev. 1983, 27, 31–39.
[14] R. Van Santen, in Catal. from Princ. to Appl. (Eds.: M. Beller, A. Renken, R. Van
Santen), Wiley-VCH Verlag GmbH, 2012, pp. 3–19.
[15] A. Zecchina, S. Bordiga, E. Groppo, in Sel. Nanocatalysts Nanosci. Concepts Heterog.
Homog. Catal. (Eds.: A. Zecchina, S. Bordiga, grop), Wiley-VCH Verlag GmbH & Co.
KGaA, 2011, pp. 1–27.
[16] J. M. Thomas, R. Raja, D. W. Lewis, Angew. Chem. Int. Ed. 2005, 44, 6456–6482.
[17] M. W. McKittrick, C. W. Jones, J. Am. Chem. Soc. 2004, 126, 3052–3053.
[18] C. Bianchini, D. G. Burnaby, J. Evans, P. Frediani, A. Meli, W. Oberhauser, R. Psaro,
L. Sordelli, F. Vizza, J. Am. Chem. Soc. 1999, 121, 5961–5971.
[19] C. Copéret, M. Chabanas, R. Petroff Saint-Arroman, J.-M. Basset, Angew. Chem. Int.
Ed. 2003, 42, 156–181.
[20] A. G. Slater, A. I. Cooper, Science 2015, 348, aaa80751–aaa80758.
[21] J. W. Johnson, A. J. Jacobson, W. M. Butler, S. E. Rosenthal, J. F. Brody, J. T.
Lewandowski, J. Am. Chem. Soc. 1989, 111, 381–383.
[22] B. F. Abrahams, B. F. Hoskins, D. M. Michail, R. Robson, Nature 1994, 369, 727–729.
[23] O. M. Yaghi, H. Li, M. Eddaoudi, M. O’Keeffe, Nature 1999, 402, 276–279.
[24] O. M. Yaghi, G. Li, H. Li, Nature 1995, 378, 703–706.
[25] O. M. Yaghi, H. Li, J. Am. Chem. Soc. 1995, 117, 10401–10402.
[26] S. Kitagawa, R. Kitaura, S. Noro, Angew. Chem. Int. Ed. 2004, 43, 2334–2375.

A-III
[27] J. Rowsell, O. M. Yaghi, Microporous Mesoporous Mater. 2004, 73, 3–14.
[28] M. J. Rosseinsky, Microporous Mesoporous Mater. 2004, 73, 15–30.
[29] W. Lu, Z. Wei, Z.-Y. Gu, T.-F. Liu, J. Park, J. Park, J. Tian, M. Zhang, Q. Zhang, T.
Gentle Iii, et al., Chem. Soc. Rev. 2014, DOI 10.1039/c4cs00003j.
[30] H. Chun, D. N. Dybtsev, H. Kim, K. Kim, Chem. - A Eur. J. 2005, 11, 3521–3529.
[31] M. Eddaoudi, J. Kim, N. Rosi, D. Vodak, J. Wachter, M. O’Keeffe, O. M. Yaghi,
Science 2002, 295, 469–472.
[32] S. S. Chui, S. M.-F. Lo, J. P. H. Charmant, A. G. Orpen, I. D. Williams, Science 1999,
283, 1148–1150.
[33] M. P. Byrn, C. J. Curtis, Y. Hsiou, S. I. Khan, P. A. Sawin, S. K. Tendick, A. Terzis, C.
E. Strouse, J. Am. Chem. Soc. 1993, 115, 9480–9497.
[34] S. Drumel, P. Janvier, P. Barboux, M. Bujoli-Doeuff, B. Bujoli, Inorg. Chem. 1995, 34,
148–156.
[35] D. Zhao, D. Yuan, H.-C. Zhou, Energy Environ. Sci. 2008, 1, 222–235.
[36] R. J. Kuppler, D. J. Timmons, Q.-R. Fang, J.-R. Li, T. A. Makal, M. D. Young, D. Yuan,
D. Zhao, W. Zhuang, H.-C. Zhou, Coord. Chem. Rev. 2009, 253, 3042–3066.
[37] J.-R. Li, R. J. Kuppler, H.-C. Zhou, Chem. Soc. Rev. 2009, 38, 1477–504.
[38] W. Xuan, C. Zhu, Y. Liu, Y. Cui, Chem. Soc. Rev. 2012, 41, 1677–1695.
[39] K. Koh, A. G. Wong-Foy, A. J. Matzger, Angew. Chem. Int. Ed. 2008, 47, 677–680.
[40] H. K. Chae, D. Y. Siberio-Pérez, J. Kim, Y. Go, M. Eddaoudi, A. J. Matzger, M.
O’Keeffe, O. M. Yaghi, Nature 2004, 427, 523–527.
[41] H. Furukawa, N. Ko, Y. B. Go, N. Aratani, S. B. Choi, E. Choi, A. O. Yazaydin, R. Q.
Snurr, M. O’Keeffe, J. Kim, et al., Science 2010, 329, 424–428.
[42] O. K. Farha, A. Ö. Yazaydın, I. Eryazici, C. D. Malliakas, B. G. Hauser, M. G.
Kanatzidis, S. T. Nguyen, R. Q. Snurr, J. T. Hupp, Nat. Chem. 2010, 2, 944–948.
[43] D. Yuan, D. Zhao, H.-C. Zhou, Inorg. Chem. 2011, 50, 10528–10530.
[44] O. K. Farha, C. E. Wilmer, I. Eryazici, B. G. Hauser, P. A. Parilla, K. O’Neill, A. A.
Sarjeant, S. T. Nguyen, R. Q. Snurr, J. T. Hupp, J. Am. Chem. Soc. 2012, 134, 9860–
9863.
[45] O. K. Farha, I. Eryazici, N. C. Jeong, B. G. Hauser, C. E. Wilmer, A. A. Sarjeant, R. Q.
Snurr, S. T. Nguyen, A. Ö. Yazaydın, J. T. Hupp, J. Am. Chem. Soc. 2012, 134,
15016–15021.
[46] N. L. Rosi, M. Eddaoudi, J. Kim, M. O’Keeffe, O. M. Yaghi, Angew. Chemie 2002, 114,
294–297.
[47] S. Ahrland, J. Chatt, N. R. Davies, Q. Rev. Chem. Soc. 1958, 12, 265–276.
[48] V. Colombo, S. Galli, H. J. Choi, G. D. Han, A. Maspero, G. Palmisano, N. Masciocchi,
J. R. Long, Chem. Sci. 2011, 2, 1311.

A-IV
 Appendix

[49] S. Bauer, C. Serre, T. Devic, P. Horcajada, J. Marrot, G. Férey, N. Stock, Inorg. Chem.
2008, 47, 7568–7576.
[50] S. Surblé, C. Serre, C. Mellot-Draznieks, F. Millange, G. Férey, Chem. Commun.
2006, 284–286.
[51] J. H. Cavka, S. Jakobsen, U. Olsbye, N. Guillou, C. Lamberti, S. Bordiga, K. P.
Lillerud, J. Am. Chem. Soc. 2008, 130, 13850–13851.
[52] S. J. Garibay, S. M. Cohen, Chem. Commun. 2010, 46, 7700–7702.
[53] H. Fei, S. M. Cohen, Chem. Commun. 2014, 4810–4812.
[54] W. Morris, B. Volosskiy, S. Demir, F. Gándara, P. L. McGrier, H. Furukawa, D. Cascio,
J. F. Stoddart, O. M. Yaghi, Inorg. Chem. 2012, 51, 6443–6445.
[55] S. Biswas, T. Ahnfeldt, N. Stock, Inorg. Chem. 2011, 50, 9518–9526.
[56] D. Jiang, L. L. Keenan, A. D. Burrows, K. J. Edler, Chem. Commun. 2012, 48, 12053–
12055.
[57] P. Serra-Crespo, E. V. Ramos-Fernandez, J. Gascon, F. Kapteijn, Chem. Mater. 2011,
23, 2565–2572.
[58] A. Buragohain, P. Van Der Voort, S. Biswas, Microporous Mesoporous Mater. 2015,
215, 91–97.
[59] T. Loiseau, C. Serre, C. Huguenard, G. Fink, F. Taulelle, M. Henry, T. Bataille, G.
Férey, Chemistry 2004, 10, 1373–1382.
[60] C. Serre, F. Millange, C. Thouvenot, M. Noguès, G. Marsolier, D. Louër, G. Férey, J.
Am. Chem. Soc. 2002, 124, 13519–13526.
[61] A. Lieb, H. Leclerc, T. Devic, C. Serre, I. Margiolaki, F. Mahjoubi, J. S. Lee, A. Vimont,
M. Daturi, J.-S. Chang, Microporous Mesoporous Mater. 2012, 157, 18–23.
[62] P. Horcajada, S. Surblé, C. Serre, D.-Y. Hong, Y.-K. Seo, J.-S. Chang, J.-M.
Grenèche, I. Margiolaki, G. Férey, Chem. Commun. 2007, 100, 2820–2822.
[63] G. Férey, C. Mellot-Draznieks, C. Serre, F. Millange, J. Dutour, S. Surblé, I.
Margiolaki, Science 2005, 309, 2040–2042.
[64] M. Hartmann, M. Fischer, Microporous Mesoporous Mater. 2012, 164, 38–43.
[65] M. Díaz-García, Á. Mayoral, I. Díaz, M. Sánchez-Sánchez, Cryst. Growth Des. 2014,
14, 2479–2487.
[66] S. R. Caskey, A. G. Wong-Foy, A. J. Matzger, J. Am. Chem. Soc. 2008, 130, 10870–
10871.
[67] N. L. Rosi, J. Kim, M. Eddaoudi, B. Chen, M. O’Keeffe, O. M. Yaghi, J. Am. Chem.
Soc. 2005, 127, 1504–1518.
[68] P. D. C. Dietzel, B. Panella, M. Hirscher, R. Blom, H. Fjellvåg, Chem. Commun. 2006,
1, 959–961.
[69] P. D. C. Dietzel, Y. Morita, R. Blom, H. Fjellvåg, Angew. Chem. Int. Ed. 2005, 44,

A-V
 6354–6358.
[70] T. Fukushima, S. Horike, Y. Inubushi, K. Nakagawa, Y. Kubota, M. Takata, S.
Kitagawa, Angew. Chem. Int. Ed. 2010, 49, 4820–4824.
[71] W. Kleist, M. Maciejewski, A. Baiker, Thermochim. Acta 2010, 499, 71–78.
[72] H. Deng, C. J. Doonan, H. Furukawa, R. B. Ferreira, J. Towne, C. B. Knobler, B.
Wang, O. M. Yaghi, Science 2010, 327, 846–850.
[73] Z. Wang, K. K. Tanabe, S. M. Cohen, Inorg. Chem. 2009, 48, 296–306.
[74] S. Marx, W. Kleist, J. Huang, M. Maciejewski, A. Baiker, Dalton Trans. 2010, 39,
3795–3798.
[75] Z. Wang, K. K. Tanabe, S. M. Cohen, Inorg. Chem. 2009, 48, 296–306.
[76] M. Ranocchiari, J. A. van Bokhoven, Chimia 2013, 67, 397–402.
[77] A. Rabenau, Angew. Chem. Int. Ed. 1985, 24, 1026–1040.
[78] Y.-R. Lee, J. Kim, W.-S. Ahn, Korean J. Chem. Eng. 2013, 30, 1667–1680.
[79] J. Klinowski, F. A. A. Paz, P. Silva, J. Rocha, Dalton Trans. 2011, 40, 321–330.
[80] N. Stock, S. Biswas, Chem. Rev. 2012, 112, 933–969.
[81] B. F. Hoskins, R. Robson, J. Am. Chem. Soc. 1990, 112, 1546–1554.
[82] Y.-H. Kiang, G. B. Gardner, S. Lee, Z. Xu, E. B. Lobkovsky, J. Am. Chem. Soc. 1999,
121, 8204–8215.
[83] Y.-H. Kiang, G. B. Gardner, S. Lee, Z. Xu, J. Am. Chem. Soc. 2000, 122, 6871–6883.
[84] J. Seo, D. Whang, H. Lee, S. Jun, J. Oh, Y. Jeon, K. Kim, Nature 2000, 404, 982–986.
[85] C.-D. Wu, A. Hu, L. Zhang, W. Lin, J. Am. Chem. Soc. 2005, 127, 8940–8941.
[86] Z. Wang, S. M. Cohen, J. Am. Chem. Soc. 2007, 129, 12368–12369.
[87] K. K. Tanabe, S. M. Cohen, Chem. Soc. Rev. 2011, 40, 498–519.
[88] S. M. Cohen, Chem. Rev. 2012, 112, 970–1000.
[89] J. D. Evans, C. J. Sumby, C. J. Doonan, Chem. Soc. Rev. 2014, 5933–5951.
[90] F. L. Morel, X. Xu, M. Ranocchiari, J. A. van Bokhoven, in Chem. Organo-Hybrids
Synth. Charact. Funct. Nano-Objects (Eds.: B. Charleux, C. Copéret, E. Lacôte), John
Wiley & Sons, Inc., Hoboken, NJ, USA, 2015, pp. 200–232.
[91] A. D. Burrows, C. G. Frost, M. F. Mahon, C. Richardson, Angew. Chem. Int. Ed. 2008,
47, 8482–8486.
[92] Y.-F. Song, L. Cronin, Angew. Chem. Int. Ed. 2008, 47, 4635–4637.
[93] Z. Wang, S. M. Cohen, Angew. Chemie 2008, 120, 4777–4780.
[94] J. S. Costa, P. Gamez, C. A. Black, O. Roubeau, S. J. Teat, J. Reedijk, Eur. J. Inorg.
Chem. 2008, 2008, 1551–1554.
[95] Y. Goto, H. Sato, S. Shinkai, K. Sada, J. Am. Chem. Soc. 2008, 130, 14354–14355.
[96] M. Savonnet, A. Camarata, J. Canivet, D. Bazer-Bachi, N. Bats, V. Lecocq, C. Pinel,
D. Farrusseng, Dalt. Trans. 2012, 41, 3945–3948.

A-VI
 Appendix

[97] M. Servalli, M. Ranocchiari, J. A. van Bokhoven, Chem. Commun. 2012, 48, 1904–
1906.
[98] P. García-García, M. Müller, A. Corma, Chem. Sci. 2014, 5, 2979–3007.
[99] S. S. Kaye, J. R. Long, J. Am. Chem. Soc. 2008, 130, 806–807.
[100] J. G. Vitillo, E. Groppo, S. Bordiga, S. Chavan, G. Ricchiardi, A. Zecchina, Inorg.
Chem. 2009, 48, 5439–5448.
[101] S. Chavan, J. G. Vitillo, M. J. Uddin, F. Bonino, C. Lamberti, E. Groppo, K.-P. Lillerud,
S. Bordiga, Chem. Mater. 2010, 22, 4602–4611.
[102] S. Chavan, J. G. Vitillo, C. Larabi, E. Alessandra Quadrelli, P. D. C. Dietzel, S.
Bordiga, Microporous Mesoporous Mater. 2012, 157, 56–61.
[103] C. Larabi, P. K. Nielsen, S. Helveg, C. Thieuleux, F. B. Johansson, M. Brorson, E. A.
Quadrelli, ACS Catal. 2012, 2, 695–700.
[104] L. Ma, J. M. Falkowski, C. Abney, W. Lin, Nat. Chem. 2010, 2, 838–846.
[105] L. Ma, C.-D. Wu, M. M. Wanderley, W. Lin, Angew. Chem. Int. Ed. 2010, 49, 8244–
8248.
[106] E. D. Bloch, D. Britt, C. Lee, C. J. Doonan, F. J. Uribe-Romo, H. Furukawa, J. R. Long,
O. M. Yaghi, J. Am. Chem. Soc. 2010, 132, 14382–14384.
[107] B. Chen, L. Wang, Y. Xiao, F. R. Fronczek, M. Xue, Y. Cui, G. Qian, Angew. Chem.
Int. Ed. 2009, 48, 500–503.
[108] P. V. Dau, M. Kim, S. M. Cohen, Chem. Sci. 2013, 4, 601–605.
[109] P. V Dau, S. M. Cohen, Chem. Commun. 2013, 49, 6128–6130.
[110] C. Wang, Z. Xie, K. E. DeKrafft, W. Lin, J. Am. Chem. Soc. 2011, 133, 13445–13454.
[111] W. Morris, B. Volosskiy, S. Demir, F. Gándara, P. L. McGrier, H. Furukawa, D. Cascio,
J. F. Stoddart, O. M. Yaghi, Inorg. Chem. 2012, 51, 6443–6445.
[112] M. E. Kosal, J.-H. Chou, S. R. Wilson, K. S. Suslick, Nat. Mater. 2002, 1, 118–121.
[113] K. S. Suslick, P. Bhyrappa, J.-H. Chou, M. E. Kosal, S. Nakagaki, D. W. Smithenry, S.
R. Wilson, Acc. Chem. Res. 2005, 38, 283–291.
[114] O. K. Farha, A. M. Shultz, A. A. Sarjeant, S. T. Nguyen, J. T. Hupp, J. Am. Chem. Soc.
2011, 133, 5652–5655.
[115] W.-Y. Gao, M. Chrzanowski, S. Ma, Chem. Soc. Rev. 2014, 43, 5841–5866.
[116] A. M. Shultz, O. K. Farha, D. Adhikari, A. A. Sarjeant, J. T. Hupp, S. T. Nguyen, Inorg.
Chem. 2011, 50, 3174–3176.
[117] F. Song, C. Wang, J. M. Falkowski, L. Ma, W. Lin, J. Am. Chem. Soc. 2010, 132,
15390–8.
[118] F. Song, T. Zhang, C. Wang, W. Lin, Proc. R. Soc. A Math. Phys. Eng. Sci. 2012, 468,
2035–2052.
[119] F. Zadehahmadi, F. Ahmadi, S. Tangestaninejad, M. Moghadam, V. Mirkhani, I.

A-VII
 Mohammadpoor-Baltork, R. Kardanpour, J. Mol. Catal. A Chem. 2015, 398, 1–10.
[120] S. Bhattacharjee, D.-A. Yang, W.-S. Ahn, Chem. Commun. 2011, 47, 3637–3639.
[121] M. J. Ingleson, J. Perez Barrio, J.-B. Guilbaud, Y. Z. Khimyak, M. J. Rosseinsky,
Chem. Commun. 2008, 2680–2682.
[122] X. Zhang, F. X. Llabrés i Xamena, A. Corma, J. Catal. 2009, 265, 155–160.
[123] K. M. L. Taylor-Pashow, J. Della Rocca, Z. Xie, S. Tran, W. Lin, J. Am. Chem. Soc.
2009, 131, 14261–14263.
[124] K. K. Tanabe, S. M. Cohen, Angew. Chemie 2009, 121, 7560–7563.
[125] K. K. Tanabe, S. M. Cohen, Inorg. Chem. 2010, 49, 6766–6774.
[126] C. J. Doonan, W. Morris, H. Furukawa, O. M. Yaghi, J. Am. Chem. Soc. 2009, 131,
9492–9493.
[127] J. Canivet, S. Aguado, Y. Schuurman, D. Farrusseng, J. Am. Chem. Soc. 2013, 135,
4195–4198.
[128] Y. K. Hwang, D.-Y. Hong, J.-S. Chang, S. H. Jhung, Y.-K. Seo, J. Kim, A. Vimont, M.
Daturi, C. Serre, G. Férey, Angew. Chem. Int. Ed. 2008, 47, 4144–4148.
[129] H. G. T. Nguyen, M. H. Weston, O. K. Farha, J. T. Hupp, S. T. Nguyen,
CrystEngComm 2012, 14, 4115–4118.
[130] A. Demessence, D. M. D’Alessandro, M. L. Foo, J. R. Long, J. Am. Chem. Soc. 2009,
131, 8784–8786.
[131] M. Meilikhov, K. Yusenko, R. A. Fischer, J. Am. Chem. Soc. 2009, 131, 9644–9645.
[132] M. Banerjee, S. Das, M. Yoon, H. J. Choi, M. H. Hyun, S. M. Park, G. Seo, K. Kim, J.
Am. Chem. Soc. 2009, 131, 7524–7525.
[133] S. Hermes, M.-K. Schröter, R. Schmid, L. Khodeir, M. Muhler, A. Tissler, R. W.
Fischer, R. A. Fischer, Angew. Chem. Int. Ed. 2005, 44, 6237–6241.
[134] H. Kim, H. Chun, G.-H. Kim, H.-S. Lee, K. Kim, Chem. Commun. 2006, 2759–2761.
[135] M. S. El-Shall, V. Abdelsayed, A. E. R. S. Khder, H. M. A. Hassan, H. M. El-Kaderi, T.
E. Reich, J. Mater. Chem. 2009, 19, 7625–7631.
[136] M. Sabo, A. Henschel, H. Fröde, E. Klemm, S. Kaskel, J. Mater. Chem. 2007, 17,
3827–3832.
[137] F. G. Cirujano, F. X. Llabrés i Xamena, A. Corma, Dalton Trans. 2012, 41, 4249–4254.
[138] T. Van Vu, H. Kosslick, A. Schulz, J. Harloff, E. Paetzold, M. Schneider, J. Radnik, N.
Steinfeldt, G. Fulda, U. Kragl, Appl. Catal. A Gen. 2013, 468, 410–417.
[139] Z. Guo, C. Xiao, R. V. Maligal-Ganesh, L. Zhou, T. W. Goh, X. Li, D. Tesfagaber, A.
Thiel, W. Huang, ACS Catal. 2014, 4, 1340–1348.
[140] X. Li, Z. Guo, C. Xiao, T. W. Goh, D. Tesfagaber, W. Huang, ACS Catal. 2014, 4,
3490–3497.
[141] Y. Liu, W. Zhang, S. Li, C. Cui, J. Wu, H. Chen, F. Huo, Chem. Mater. 2014, 26, 1119–

A-VIII
 Appendix

1125.
[142] P. Hu, J. V Morabito, C. Tsung, ACS Catal. 2014, 4, 4409–4419.
[143] A. Aijaz, Q. Xu, J. Phys. Chem. Lett. 2014, 5, 1400–1411.
[144] V. Lykourinou, Y. Chen, X. Wang, L. Meng, T. Hoang, L. Ming, R. L. Musselman, S.
Ma, J. Am. Chem. Soc. 2011, 133, 10382–10385.
[145] P. Deria, J. E. Mondloch, O. Karagiaridi, W. Bury, J. T. Hupp, O. K. Farha, Chem. Soc.
Rev. 2014, 41–44.
[146] J. Li, D. J. Timmons, H. Zhou, J. Am. Chem. Soc. 2009, 131, 6368–6369.
[147] O. Karagiaridi, W. Bury, J. E. Mondloch, J. T. Hupp, O. K. Farha, Angew. Chem. Int.
Ed. 2014, 53, 4530–4540.
[148] M. Kim, J. F. Cahill, Y. Su, K. A. Prather, S. M. Cohen, Chem. Sci. 2012, 3, 126–130.
[149] M. Kim, J. F. Cahill, H. Fei, K. A. Prather, S. M. Cohen, J. Am. Chem. Soc. 2012, 134,
18082–18088.
[150] O. Karagiaridi, M. B. Lalonde, W. Bury, A. A. Sarjeant, O. K. Farha, J. T. Hupp, J. Am.
Chem. Soc. 2012, 134, 18790–18796.
[151] H. Fei, J. F. Cahill, K. A. Prather, S. M. Cohen, Inorg. Chem. 2013, 52, 4011–4016.
[152] M. B. Lalonde, J. E. Mondloch, P. Deria, A. A. Sarjeant, S. S. Al-Juaid, O. I. Osman,
O. K. Farha, J. T. Hupp, Inorg. Chem. 2015, 54, 7142–7144.
[153] S. Pullen, H. Fei, A. Orthaber, S. M. Cohen, S. Ott, J. Am. Chem. Soc. 2013, 135,
16997–17003.
[154] T. K. Prasad, D. H. Hong, M. P. Suh, Chemistry 2010, 16, 14043–14050.
[155] K. Sumida, D. L. Rogow, J. A. Mason, T. M. McDonald, E. D. Bloch, Z. R. Herm, T.
Bae, J. R. Long, Chem. Rev. 2012, 112, 724–781.
[156] Z. Y. Gu, J. Park, A. Raiff, Z. Wei, H. C. Zhou, ChemCatChem 2014, 6, 67–75.
[157] J. Liu, L. Chen, H. Cui, J. Zhang, L. Zhang, C.-Y. Su, Chem. Soc. Rev. 2014, 6011–
6061.
[158] M. Yoon, R. Srirambalaji, K. Kim, Chem. Rev. 2012, 112, 1196–231.
[159] L. Ma, C. Abney, W. Lin, Chem. Soc. Rev. 2009, 38, 1248–1256.
[160] C. Wang, M. Zheng, W. Lin, J. Phys. Chem. Lett. 2011, 2, 1701–1709.
[161] A. Corma, H. García, F. X. Llabrés i Xamena, Chem. Rev. 2010, 110, 4606–4655.
[162] J. Gascon, A. Corma, F. Kapteijn, F. X. Llabrés i Xamena, ACS Catal. 2014, 4, 361–
378.
[163] A. Dhakshinamoorthy, M. Alvaro, H. Garcia, Catal. Sci. Technol. 2011, 856–867.
[164] D. Farrusseng, S. Aguado, C. Pinel, Angew. Chem. Int. Ed. 2009, 48, 7502–7513.
[165] M. Ranocchiari, J. A. van Bokhoven, Phys. Chem. Chem. Phys. 2011, 13, 6388–6396.
[166] M. Ranocchiari, C. Lothschutz, D. Grolimund, J. A. van Bokhoven, Proc. R. Soc. A
Math. Phys. Eng. Sci. 2012, 1985–1999.

A-IX
[167] C. I. Ezugwu, N. A. Kabir, M. Yusubov, F. Verpoort, Coord. Chem. Rev. 2016, 307,
188–210.
[168] P. Horcajada, R. Gref, T. Baati, P. K. Allan, G. Maurin, P. Couvreur, G. Férey, R. E.
Morris, C. Serre, Chem. Rev. 2012, 112, 1232–1268.
[169] S. Keskin, S. Kızılel, Ind. Eng. Chem. Res. 2011, 50, 1799–1812.
[170] M. D. Allendorf, C. A. Bauer, R. K. Bhakta, R. J. T. Houk, Chem. Soc. Rev. 2009, 38,
1330–1352.
[171] L. J. Murray, M. Dincă, J. R. Long, Chem. Soc. Rev. 2009, 38, 1294–1314.
[172] J. Li, J. Sculley, H. Zhou, Chem. Rev. 2012, 112, 869–932.
[173] N. A. Khan, Z. Hasan, S. H. Jhung, J. Hazard. Mater. 2013, 244-245, 444–456.
[174] D. Britt, D. Tranchemontagne, O. M. Yaghi, Proc. Natl. Acad. Sci. U. S. A. 2008, 105,
11623–11627.
[175] L. Bing, J. Suyun, L. Bogen, Prog. Chem. 2013, 1, 36–45.
[176] B. Kesanli, W. Lin, Coord. Chem. Rev. 2003, 246, 305–326.
[177] M. Yoon, R. Srirambalaji, K. Kim, Chem. Rev. 2012, 112, 1196–231.
[178] M. Kurmoo, Chem. Soc. Rev. 2009, 38, 1353–1379.
[179] L. Alaerts, M. Maes, L. Giebeler, P. A. Jacobs, J. A. Martens, J. F. M. Denayer, C. E.
A. Kirschhock, D. E. De Vos, J. Am. Chem. Soc. 2008, 130, 14170–14178.
[180] N. Chang, Z. Y. Gu, X. P. Yan, J. Am. Chem. Soc. 2010, 132, 13645–13647.
[181] A. J. Howarth, Y. Liu, J. T. Hupp, O. K. Farha, CrystEngComm 2015, 17, 7245–7253.
[182] M. Nič, J. Jirát, B. Košata, A. Jenkins, A. McNaught, Eds. , IUPAC Compendium of
Chemical Terminology, IUPAC, Research Triagle Park, NC, 2009.
[183] L. Maier, in Prog. Inorg. Chem. (Ed.: F.A. Cotton), John Wiley & Sons, Inc., 1963, pp.
27–210.
[184] S. A. Buckler, J. Am. Chem. Soc. 1962, 84, 3093–3097.
[185] K. D. Berlin, G. B. Butler, Chem. Rev. 1960, 60, 243–260.
[186] J. E. Cobb, C. M. Cribbs, B. R. Henke, D. E. Uehling, A. G. Hernan, C. Martin, C. M.
Rayner, in Encycl. Reagents Org. Synth., John Wiley & Sons, Ltd, Chichester, UK,
2005, pp. 1–6.
[187] S. Trippett, Pure Appl. Chem. 1964, 9, 255–270.
[188] H. R. Hudson, in Chem. Organophosphorus Compd., John Wiley & Sons, Inc.,
Hoboken, NJ, USA, 1990, pp. 385–471.
[189] X. Lu, C. Zhang, Z. Xu, Acc. Chem. Res. 2001, 34, 535–544.
[190] S. E. Denmark, G. L. Beutner, Angew. Chem. Int. Ed. 2008, 47, 1560–1638.
[191] J. A. Osborn, F. H. Jardine, J. F. Young, G. Wilkinson, J. Chem. Soc. A Inorganic,
Phys. Theor. 1966, 1711–1732.
[192] R. Franke, D. Selent, A. Börner, Chem. Rev. 2012, 112, 5675–732.

A-X
 Appendix

[193] R. H. Grubbs, Tetrahedron 2004, 60, 7117–7140.

[194] E. I. Negishi, Angew. Chem. Int. Ed. 2011, 50, 6738–6764.
[195] A. Suzuki, Angew. Chem. Int. Ed. 2011, 50, 6722–6737.
[196] X. Wang, K. Ding, J. Am. Chem. Soc. 2004, 126, 10524–10525.
[197] M. Guinó, K. K. M. Hii, Chem. Soc. Rev. 2007, 36, 608–617.
[198] A. S. H. King, L. J. Twyman, J. Chem. Soc. Perkin Trans. 1 2002, 2209–2218.
[199] C. A. Parrish, S. L. Buchwald, J. Org. Chem. 2001, 66, 3820–3827.
[200] I. W. Davies, L. Matty, D. L. Hughes, P. J. Reider, J. Am. Chem. Soc. 2001, 123,
10139–10140.
[201] D. Drewry, D. Coe, S. Poon, Med. Res. Rev. 1999, 97–148.
[202] Y. Uozumi, H. Danjo, T. Hayashi, Tetrahedron Lett. 1997, 38, 3557–3560.
[203] M. J. Plater, M. R. S. J. Foreman, E. Coronado, C. J. Gómez-García, A. M. Z. Slawin,
J. Chem. Soc. Dalt. Trans. 1999, 4209–4216.
[204] M. Plater, M. Foreman, J. Skakle, J. Chem. Crystallogr. 2000, 30, 449–453.
[205] X. Xu, M. Nieuwenhuyzen, S. L. James, Angew. Chem. Int. Ed. 2002, 41, 764–767.
[206] J. M. Falkowski, T. Sawano, T. Zhang, G. Tsun, Y. Chen, J. V Lockard, W. Lin, J. Am.
Chem. Soc. 2014, 136, 5213–5216.
[207] J. Václavík, M. Servalli, C. Lothschütz, J. Szlachetko, M. Ranocchiari, J. A. van
Bokhoven, ChemCatChem 2013, 5, 692–696.
[208] X. Tan, L. Li, J. Zhang, X. Han, L. Jiang, F. Li, C. Su, Chem. Mater. 2012, 24, 480–
485.
[209] A. J. Nuñez, L. N. Shear, N. Dahal, I. A. Ibarra, J. Yoon, Y. K. Hwang, J.-S. Chang, S.
M. Humphrey, Chem. Commun. 2011, 47, 11855–11857.
[210] A. M. Bohnsack, I. A. Ibarra, P. W. Hatfield, J. W. Yoon, Y. K. Hwang, J.-S. Chang, S.
M. Humphrey, Chem. Commun. 2011, 47, 4899–4901.
[211] S. M. Humphrey, S. E. Oungoulian, J. W. Yoon, Y. K. Hwang, E. R. Wise, J.-S. Chang,
Chem. Commun. 2008, 2891–2893.
[212] K. J. Gagnon, H. P. Perry, A. Clearfield, Chem. Rev. 2012, 112, 1034–1054.
[213] S. M. Humphrey, P. K. Allan, S. E. Oungoulian, M. S. Ironside, E. R. Wise, Dalton
Trans. 2009, 2298–305.
[214] I. A. Ibarra, K. E. Tan, V. M. Lynch, S. M. Humphrey, Dalton Trans. 2012, 41, 3920–
3923.
[215] A. Hu, H. L. Ngo, W. Lin, J. Am. Chem. Soc. 2003, 125, 11490–11491.
[216] I. A. Ibarra, T. W. Hesterberg, B. J. Holliday, V. M. Lynch, S. M. Humphrey, Dalt.
Trans. 2012, 41, 8003–8009.
[217] X. Xu, J. A. van Bokhoven, M. Ranocchiari, ChemCatChem 2014, 6, 1887–1891.
[218] X. Xu, S. M. Rummelt, F. L. Morel, M. Ranocchiari, J. A. van Bokhoven, Chemistry

A-XI
 2014, 20, 15467–15472.
[219] M. G. Goesten, K. B. Sai Sankar Gupta, E. V. Ramos-Fernandez, H. Khajavi, J.
Gascon, F. Kapteijn, CrystEngComm 2012, 14, 4109–4111.
[220] J. Fritsch, F. Drache, G. Nickerl, W. Böhlmann, S. Kaskel, Microporous Mesoporous
Mater. 2013, 172, 167–173.
[221] M. Ranocchiari, S. M. Rummelt, J. A. van Bokhoven, Synthesis of Water-Soluble
Phosphine Oxides by Pd/C-Catalyzed P-C Coupling in Water, 2013, EP2626362A1.
[222] S. M. Rummelt, M. Ranocchiari, J. A. van Bokhoven, Org. Lett. 2012, 14, 2188–2190.
[223] X. Xu, Design of Amino and Phosphine Functionalized Metal-Organic Frameworks for
Organocatalysis, ETH Zurich, 2015.
[224] L. Bromberg, Y. Diao, H. Wu, S. A. Speakman, T. A. Hatton, Chem. Mater. 2012, 24,
1664–1675.
[225] S. Wang, L. Bromberg, H. Schreuder-Gibson, T. A. Hatton, ACS Appl. Mater.
Interfaces 2013, 5, 1269–1278.
[226] G. Giordano, R. H. Crabtree, Inorg. Synth. 1979, 19, 218–220.
[227] K. Collins, SyntheticPage 2007, 10–11.
[228] T. G. Schenck, J. M. Downes, C. R. C. Milne, P. B. Mackenzie, T. G. Boucher, J.
Whelan, B. Bosnich, Inorg. Chem. 1985, 24, 2334–2337.
[229] J. Albers, E. Dinjus, S. Pitter, O. Walter, J. Mol. Catal. A Chem. 2004, 219, 41–46.
[230] M. A. Bennett, T.-N. Huang, T. W. Matheson, A. K. Smith, in Inorg. Synth. (Ed.: J.P.
Fackler), John Wiley & Sons, Inc., Hoboken, NJ, USA, 1982, pp. 74–78.
[231] A. Grabulosa, A. Mannu, E. Alberico, S. Denurra, S. Gladiali, G. Muller, J. Mol. Catal.
A Chem. 2012, 363-364, 49–57.
[232] M. S. Sanford, J. A. Love, R. H. Grubbs, Organometallics 2001, 20, 5314–5318.
[233] K. Nakamura, K. Takenaka, Tetrahedron Asymmetry 2002, 13, 415–422.
[234] A. Beloqui Redondo, D. Fodor, M. A. Brown, J. A. van Bokhoven, Catal. Commun.
2014, 56, 128–133.
[235] W. H. Bragg, W. L. Bragg, Proc. R. Soc. A Math. Phys. Eng. Sci. 1913, 88, 428–438.
[236] C. P. Gómez, R. A. Jacobson, in Charact. Mater., John Wiley & Sons, Inc., Hoboken,
NJ, USA, 2012.
[237] L. J. Farrugia, J. Appl. Crystallogr. 1999, 32, 837–838.
[238] G. M. Sheldrick, T. R. Schneider, Methods Enzymol. 1997, 277, 319–343.
[239] G. M. Sheldrick, Acta Crystallogr. Sect. C Struct. Chem. 2015, 71, 3–8.
[240] O. V. Dolomanov, L. J. Bourhis, R. J. Gildea, J. A. K. Howard, H. Puschmann, J. Appl.
Crystallogr. 2009, 42, 339–341.
[241] A. L. Spek, Acta Crystallogr. Sect. C Struct. Chem. 2015, 71, 9–18.
[242] R. B. Von Dreele, B. H. Toby, in Charact. Mater., John Wiley & Sons, Inc., Hoboken,

A-XII
 Appendix

NJ, USA, 2012, pp. 1340–1360.

[243] J. A. van Bokhoven, C. Lamberti, X-Ray Absorption and X-Ray Emission
Spectroscopy: Theory and Applications, Wiley-VCH Verlag GmbH, 2016.
[244] S. Bordiga, E. Groppo, G. Agostini, J. A. van Bokhoven, C. Lamberti, Chem. Rev.
2013, 113, 1736–1850.
[245] “NIST X-ray absorption coefficient webpage,” can be found under
http://physics.nist.gov/PhysRefData/XrayMassCoef/tab3.html.
[246] R. Frahm, M. Nachtegaal, J. Stötzel, M. Harfouche, J. A. van Bokhoven, J.-D.
Grunwaldt, R. Garrett, I. Gentle, K. Nugent, S. Wilkins, in AIP Conf. Proc., 2010, pp.
251–255.
[247] R. Frahm, M. Richwin, D. LtzenkirchenHecht, Phys. Scr. 2005, 115, 974–976.
[248] “JAQ,” can be found under http://jaq-online.forumprofi.de/index.php.
[249] D. Haskel, “FLUO webpage,” can be found under
http://www.aps.anl.gov/~haskel/fluo.html.
[250] B. Ravel, M. Newville, J. Synchrotron Radiat. 2005, 12, 537–541.
[251] F. A. Bovey, Nuclear Magnetic Resonance Spectroscopy, Academic Press Inc.
(London) Ltd., San Diego, 1988.
[252] J. B. Lambert, E. P. Mazzola, Nuclear Magnetic Resonance Spectroscopy: An
Introduction to Principles, Applications, and Experimental Methods, Pearson
Education, 2004.
[253] E. R. Andrew, Prog. Nucl. Magn. Reson. Spectrosc. 1971, 8, 1–39.
[254] S. R. Hartmann, E. L. Hahn, Phys. Rev. 1962, 128, 2042–2053.
[255] K. S. W. Sing, D. H. Everett, R. A. W. Haul, L. Moscou, R. A. Pierotti, J. Rouquérol, T.
Siemieniewska, Pure Appl. Chem. 1985, 57, 603–619.
[256] S. Brunauer, P. H. Emmett, E. Teller, J. Am. Chem. Soc. 1938, 60, 309–319.
[257] G. Papp, H. Horváth, G. Laurenczy, I. Szatmári, Á. Kathó, F. Joó, Dalton Trans. 2013,
42, 521–519.
[258] H. Fernández-Pérez, P. Etayo, A. Panossian, A. Vidal-Ferran, Chem. Rev. 2011, 111,
2119–2176.
[259] M. J. Burk, Acc. Chem. Res. 2000, 33, 363–372.
[260] R. Noyori, S. Hashiguchi, Acc. Chem. Res. 1997, 30, 97–102.
[261] F. Lagasse, H. B. Kagan, Chem. Pharm. Bull. 2000, 48, 315–324.
[262] W. Zhang, Y. Chi, X. Zhang, Acc. Chem. Res. 2007, 40, 1278–1290.
[263] P. D. Achord, P. Kiprof, B. Barker, J. Mol. Struct. THEOCHEM 2008, 849, 103–111.
[264] M. van den Berg, A. J. Minnaard, E. P. Schudde, J. van Esch, A. H. M. de Vries, J. G.
de Vries, B. L. Feringa, J. Am. Chem. Soc. 2000, 122, 11539–11540.
[265] B. D. Vineyard, W. S. Knowles, M. J. Sabacky, G. L. Bachman, D. J. Weinkauff, J. Am.

A-XIII
 Chem. Soc. 1977, 99, 5946–5952.
[266] C. J. den Reijer, The Synthesis, Characterization, and Reactivity of Ru(II)-Arene
Complexes with MeO-Biphep and Binap, ETH Zürich, 2001.
[267] D. W. Norman, C. A. Carraz, D. J. Hyett, P. G. Pringle, J. B. Sweeney, A. G. Orpen, H.
Phetmung, R. L. Wingad, J. Am. Chem. Soc. 2008, 130, 6840–6847.
[268] S. Bhattacharjee, C. Chen, W.-S. Ahn, RSC Adv. 2014, 4, 52500–52525.
[269] O. Herd, A. Heßler, M. Hingst, P. Machnitzki, M. Tepper, O. Stelzer, Catal. Today
1998, 42, 413–420.
[270] H. Zhang, Y. Ma, X.-L. Sun, Chem. Commun. 2009, 3032–3034.
[271] S. T. Meek, J. J. Perry, S. L. Teich-McGoldrick, J. A. Greathouse, M. D. Allendorf,
Cryst. Growth Des. 2011, 11, 4309–4312.
[272] P. W. N. M. van Leeuwen, J. C. Chadwick, in Homog. Catal. Act. - Stab. - Deactiv.
(Eds.: P.W.N.M. van Leeuwen, J.C. Chadwick), Wiley-VCH Verlag GmbH & Co.
KGaA, 2011.
[273] K. Naumann, G. Zon, K. Mislow, J. Am. Chem. Soc. 1969, 91, 7012–7023.
[274] M. N. Chumachenko, I. F. Serebryakova, Bull. Acad. Sci. USSR Div. Chem. Sci. 1971,
20, 2191–2193.
[275] C. A. Busacca, R. Raju, N. Grinberg, N. Haddad, P. James-Jones, H. Lee, J. C.
Lorenz, A. Saha, C. H. Senanayake, J. Org. Chem. 2008, 73, 1524–1531.
[276] Y. Li, S. Das, S. Zhou, K. Junge, M. Beller, J. Am. Chem. Soc. 2012, 134, 9727–9732.
[277] H. A. van Kalkeren, S. H. Leenders, C. R. A. Hommersom, F. P. J. T. Rutjes, F. L. van
Delft, Chemistry 2011, 17, 11290–11295.
[278] Y. Li, L.-Q. Lu, S. Das, S. Pisiewicz, K. Junge, M. Beller, J. Am. Chem. Soc. 2012,
134, 18325–18329.
[279] A. P. Nelson, O. K. Farha, K. L. Mulfort, J. T. Hupp, J. Am. Chem. Soc. 2009, 131,
458–460.
[280] P. W. N. M. van Leeuwen, Homogeneous Catalysis, Springer-Verlag,
Berlin/Heidelberg, 2004.
[281] S. D. Burke, R. L. Danheiser, in Handb. Reagents Org. Synth., John Wiley & Sons,
1999, pp. 493–498.
[282] P. Conte, R. Spaccini, A. Piccolo, Prog. Nucl. Magn. Reson. Spectrosc. 2004, 44,
215–223.
[283] M. Padmanaban, P. Müller, C. Lieder, K. Gedrich, R. Grünker, V. Bon, I. Senkovska,
S. Baumgärtner, S. Opelt, S. Paasch, et al., Chem. Commun. 2011, 47, 12089–12091.
[284] M. Kim, J. A. Boissonnault, C. A. Allen, P. V Dau, S. M. Cohen, Dalton Trans. 2012,
41, 6277–6282.
[285] O. I. Lebedev, F. Millange, C. Serre, G. Van Tendeloo, G. Férey, Chem. Mater. 2005,

A-XIV
 Appendix

17, 6525–6527.
[286] Z. Hasan, E.-J. Choi, S. H. Jhung, Chem. Eng. J. 2013, 219, 537–544.
[287] K.-S. Lin, A. K. Adhikari, Y.-H. Su, C.-W. Shu, H.-Y. Chan, Adsorption 2012, 18, 483–
491.
[288] Z. Saedi, S. Tangestaninejad, M. Moghadam, V. Mirkhani, I. Mohammadpoor-Baltork,
Catal. Commun. 2012, 17, 18–22.
[289] N. V. Maksimchuk, O. V. Zalomaeva, I. Y. Skobelev, K. A. Kovalenko, V. P. Fedin, O.
A. Kholdeeva, Proc. R. Soc. A Math. Phys. Eng. Sci. 2012, 468, 2017–2034.
[290] A. Henschel, K. Gedrich, R. Kraehnert, S. Kaskel, Chem. Commun. 2008, 4192–4.
[291] N. A. Khan, J. W. Jun, S. H. Jhung, Eur. J. Inorg. Chem. 2010, 1043–1048.
[292] M. G. Goesten, P. C. M. M. Magusin, E. A. Pidko, B. Mezari, E. J. M. Hensen, F.
Kapteijn, J. Gascon, Inorg. Chem. 2014, 53, 882–887.
[293] J. Klinowski, Annu. Rev. Mater. Sci. 1988, 18, 189–218.
[294] B. Stewart, A. Harriman, L. J. Higham, Organometallics 2011, 30, 5338–5343.
[295] J. Tong, S. Liu, S. Zhang, S. Z. Li, Spectrochim. Acta. A. Mol. Biomol. Spectrosc.
2007, 67, 837–846.
[296] L. D. Quin, A. J. Williams, Practical Interpretation of P-31 NMR Spectra and Computer
Assisted Structure Verification, Advanced Chemistry Development, Inc. Toronto,
Canada, 2004.
[297] O. Kühl, Phosphorus-31 NMR Spectroscopy : A Concise Introduction for the Synthetic
Organic and Organometallic Chemist, Berlin, 2008.
[298] J. A. van Bokhoven, T. Ressler, F. M. F. de Groot, G. Knopp-Gericke, in In-Situ
Spectrosc. Catal. (Ed.: B.M. Weckhuysen), 2004, pp. 123–144.
[299] S. Bordiga, F. Bonino, K. P. Lillerud, C. Lamberti, Chem. Soc. Rev. 2010, 39, 4885–
4927.
[300] C. Prestipino, L. Regli, J. G. Vitillo, F. Bonino, A. Damin, C. Lamberti, A. Zecchina, P.
L. Solari, K. O. Kongshaug, S. Bordiga, Chem. Mater. 2006, 18, 1337–1346.
[301] L. Valenzano, B. Civalleri, S. Chavan, S. Bordiga, M. H. Nilsen, S. Jakobsen, K. P.
Lillerud, C. Lamberti, Chem. Mater. 2011, 23, 1700–1718.
[302] L. J. Murray, M. Dinca, J. Yano, S. Chavan, S. Bordiga, C. M. Brown, J. R. Long, J.
Am. Chem. Soc. 2010, 132, 7856–7857.
[303] F. Bonino, S. Chavan, J. G. Vitillo, E. Groppo, G. Agostini, C. Lamberti, P. D. C.
Dietzel, C. Prestipino, S. Bordiga, Chem. Mater. 2008, 20, 4957–4968.
[304] M. Müller, S. Hermes, K. Kähler, M. W. E. van den Berg, M. Muhler, R. A. Fischer,
Chem. Mater. 2008, 20, 4576–4587.
[305] Z. Guo, T. Kobayashi, L.-L. Wang, T. W. Goh, C. Xiao, M. A. Caporini, M. Rosay, D. D.
Johnson, M. Pruski, W. Huang, Chem. - A Eur. J. 2014, 20, 16308–16313.

A-XV
[306] C. Xiao, T. W. Goh, K. Brashler, Y. Pei, Z. Guo, W. Huang, J. Phys. Chem. B 2014,
118, 14168–14176.
[307] F. L. Morel, M. Ranocchiari, J. A. van Bokhoven, Ind. Eng. Chem. Res. 2014, 53,
9120–9127.
[308] F. L. Morel, S. Pin, T. Huthwelker, M. Ranocchiari, J. A. van Bokhoven, Phys. Chem.
Chem. Phys. 2015, 17, 3326–3331.
[309] L. Thomas, Spectrochim. Acta 1965, 21, 1905–1914.
[310] A. E. Senear, W. Valient, J. Wirth, J. Org. Chem. 1960, 25, 2001–2006.
[311] L. C. Thomas, R. A. Chittenden, Spectrochim. Acta 1964, 20, 467–487.
[312] Z. Wang, S. M. Cohen, Chem. Soc. Rev. 2009, 38, 1315–1329.
[313] J. A. Davies, S. Dutremez, A. A. Pinkerton, Inorg. Chem. 1991, 30, 2380–2387.
[314] R. R. Ernst, Rev. Sci. Instrum. 1966, 37, 93–102.
[315] C. Engemann, R. Franke, J. Hormes, C. Lauterbach, E. Hartmann, J. Clade, M.
Jansen, Chem. Phys. 1999, 243, 61–75.
[316] E. Hodson, S. J. Simpson, Polyhedron 2004, 23, 2695–2707.
[317] S. A. Serron, S. P. Nolan, Organometallics 1995, 14, 4611–4616.
[318] P. Kumar, R. K. Gupta, D. S. Pandey, Chem. Soc. Rev. 2014, 43, 707–733.
[319] J. Václavík, P. Sot, B. Vilhanová, J. Pecháček, M. Kuzma, P. Kačer, Molecules 2013,
18, 6804–6828.
[320] B. Loges, A. Boddien, H. Junge, M. Beller, Angew. Chem. Int. Ed. 2008, 47, 3962–
3965.
[321] C. Fellay, P. J. Dyson, G. Laurenczy, Angew. Chem. Int. Ed. 2008, 47, 3966–3968.
[322] A. Boddien, D. Mellmann, F. Gärtner, R. Jackstell, H. Junge, P. J. Dyson, G.
Laurenczy, R. Ludwig, M. Beller, Science 2011, 333, 1733–1736.
[323] A. Boddien, F. Gärtner, D. Mellmann, P. Sponholz, H. Junge, G. Laurenczy, M. Beller,
Chim. Int. J. Chem. 2011, 65, 214–218.
[324] A. Thevenon, E. Frost-Pennington, G. Weijia, A. F. Dalebrook, G. Laurenczy,
ChemCatChem 2014, 6, 3146–3152.
[325] H. Junge, A. Boddien, F. Capitta, B. Loges, J. R. Noyes, S. Gladiali, M. Beller,
Tetrahedron Lett. 2009, 50, 1603–1606.
[326] E. I. Solomon, B. Hedman, K. O. Hodgson, A. Dey, R. K. Szilagyi, Coord. Chem. Rev.
2005, 249, 97–129.
[327] M. U. Delgado-Jaime, J. C. Conrad, D. E. Fogg, P. Kennepohl, Inorganica Chim. Acta
2006, 359, 3042–3047.
[328] S. A. Kozimor, P. Yang, E. R. Batista, K. S. Boland, C. J. Burns, D. L. Clark, S. D.
Conradson, R. L. Martin, M. P. Wilkerson, L. E. Wolfsberg, J. Am. Chem. Soc. 2009,
131, 12125–12136.

A-XVI
 Appendix

[329] T. V Harris, R. K. Szilagyi, K. L. McFarlane Holman, J. Biol. Inorg. Chem. 2009, 14,
891–898.
[330] I. Alperovich, G. Smolentsev, D. Moonshiram, J. W. Jurss, J. J. Concepcion, T. J.
Meyer, A. Soldatov, Y. Pushkar, J. Am. Chem. Soc. 2011, 133, 15786–15794.
[331] I. Alperovich, D. Moonshiram, A. Soldatov, Y. Pushkar, Solid State Commun. 2012,
152, 1880–1884.
[332] C. J. Elsevier, B. Kowall, H. Kragten, Inorg. Chem. 1995, 34, 4836–4839.
[333] Y. Wu, E. D. E., J. Phys. Condens. Matter 1995, 7, 3973–3989.
[334] J. M. Praetorius, D. P. Allen, R. Wang, J. D. Webb, F. Grein, P. Kennepohl, C. M.
Crudden, J. Am. Chem. Soc. 2008, 130, 3724–3725.
[335] R. Ameloot, F. Vermoortele, J. Hofkens, F. C. De Schryver, D. E. De Vos, M. B. J.
Roeffaers, Angew. Chem. Int. Ed. 2013, 52, 401–405.
[336] A. M. Katzenmeyer, J. Canivet, G. Holland, D. Farrusseng, A. Centrone, Angew.
Chem. Int. Ed. 2014, 53, 2852–2856.
[337] M. Ranocchiari, C. Lothschutz, D. Grolimund, J. A. van Bokhoven, Proc. R. Soc. A
Math. Phys. Eng. Sci. 2012, 468, 1985–1999.
[338] G. W. Crabtree, M. S. Dresselhaus, M. V. Buchanan, Phys. Today 2004, 57, 39–44.
[339] F. Joó, ChemSusChem 2008, 1, 805–808.
[340] S. Dunn, Int. J. Hydrogen Energy 2002, 27, 235–264.
[341] W.-H. Wang, Y. Himeda, J. T. Muckerman, G. F. Manbeck, E. Fujita, Chem. Rev.
2015, DOI 10.1021/acs.chemrev.5b00197.
[342] T. He, Q. Pei, P. Chen, J. Energy Chem. 2015, DOI 10.1016/j.jechem.2015.08.007.
[343] X. Liu, S. Li, Y. Liu, Y. Cao, Chinese J. Catal. 2015, 36, 1461–1475.
[344] A. Boddien, B. Loges, H. Junge, M. Beller, ChemSusChem 2008, 1, 751–758.
[345] M. Martis, K. Mori, K. Fujiwara, W. S. Ahn, H. Yamashita, J. Phys. Chem. C 2013, 117,
22805–22810.
[346] X. Gu, Z. H. Lu, H. L. Jiang, T. Akita, Q. Xu, J. Am. Chem. Soc. 2011, 133, 11822–
11825.
[347] M. Jezequel, V. Dufaud, M. J. Ruiz-Garcia, F. Carrillo-Hermosilla, U. Neugebauer, G.
P. Niccolai, F. Lefebvre, F. Bayard, J. Corker, S. Fiddy, et al., J. Am. Chem. Soc.
2001, 123, 3520–3540.
[348] C. Fellay, P. J. Dyson, G. Laurenczy, Angew. Chem. Int. Ed. 2008, 47, 3966–3968.
[349] B. Loges, A. Boddien, H. Junge, M. Beller, Angew. Chem. Int. Ed. 2008, 47, 3962–
3965.
[350] A. Boddien, F. Gärtner, D. Mellmann, P. Sponholz, H. Junge, G. Laurenczy, M. Beller,
Chim. Int. J. Chem. 2011, 65, 214–218.
[351] Y. Himeda, N. Onozawa-Komatsuzaki, H. Sugihara, K. Kasuga, Organometallics 2007,

A-XVII
 26, 702–712.
[352] A. Boddien, B. Loges, F. Gärtner, C. Torborg, K. Fumino, H. Junge, R. Ludwig, M.
Beller, J. Am. Chem. Soc. 2010, 132, 8924–8934.
[353] E. A. Bielinski, P. O. Lagaditis, Y. Zhang, B. Q. Mercado, C. Würtele, W. H.
Bernskoetter, N. Hazari, S. Schneider, J. Am. Chem. Soc. 2014, 136, 10234–10237.
[354] A. Boddien, D. Mellmann, F. Gärtner, R. Jackstell, H. Junge, P. J. Dyson, G.
Laurenczy, R. Ludwig, M. Beller, Science 2011, 333, 1733–1736.
[355] F. Solymosi, Á. Koós, N. Liliom, I. Ugrai, J. Catal. 2011, 279, 213–219.
[356] W. Gan, P. J. Dyson, G. Laurenczy, ChemCatChem 2013, 5, 3124–3130.
[357] M. Zacharska, O. Y. Podyacheva, L. S. Kibis, A. I. Boronin, B. V. Senkovskiy, E. Y.
Gerasimov, O. P. Taran, A. B. Ayusheev, V. N. Parmon, J. J. Leahy, et al.,
ChemCatChem 2015, 7, 2910–2917.
[358] F. L. Morel, S. Pin, T. Huthwelker, M. Ranocchiari, J. A. van Bokhoven, Phys. Chem.
Chem. Phys. 2015, 17, 3326–3331.
[359] A. Beloqui Redondo, F. L. Morel, M. Ranocchiari, J. A. van Bokhoven, ACS Catal.
2015, 5, 7099–7103.
[360] E. Hodson, S. J. Simpson, Polyhedron 2004, 23, 2695–2707.
[361] S. A. Serron, S. P. Nolan, Organometallics 1995, 14, 4611–4616.
[362] D. J. Morris, G. J. Clarkson, M. Wills, Organometallics 2009, 28, 4133–4140.
[363] R. F. Egerton, P. Li, M. Malac, Micron 2004, 35, 399–409.
[364] P. Sponholz, D. Mellmann, H. Junge, M. Beller, ChemSusChem 2013, 6, 1172–1176.
[365] M. Ojeda, E. Iglesia, Angew. Chem. Int. Ed. 2009, 48, 4800–4803.
[366] A. Gazsi, T. Bánsági, F. Solymosi, J. Phys. Chem. C 2011, 115, 15459–15466.
[367] Q.-Y. Bi, X.-L. Du, Y.-M. Liu, Y. Cao, H.-Y. He, K.-N. Fan, J. Am. Chem. Soc. 2012,
134, 8926–8933.
[368] A. V. Bavykina, M. G. Goesten, F. Kapteijn, M. Makkee, J. Gascon, ChemSusChem
2015, 8, 809–812.
[369] D. A. Bulushev, S. Beloshapkin, J. R. H. Ross, Catal. Today 2010, 154, 7–12.
[370] D. A. Bulushev, S. Beloshapkin, P. E. Plyusnin, Y. V. Shubin, V. I. Bukhtiyarov, S. V.
Korenev, J. R. H. Ross, J. Catal. 2013, 299, 171–180.
[371] J. F. Hull, Y. Himeda, W.-H. Wang, B. Hashiguchi, R. Periana, D. J. Szalda, J. T.
Muckerman, E. Fujita, Nat. Chem. 2012, 4, 383–388.
[372] H. S. Eleuterio, J. Mol. Catal. 1991, 65, 55–61.
[373] Y. Chauvin, Angew. Chem. Int. Ed. 2006, 45, 3740–3747.
[374] N. Calderon, H. Y. Chen, K. W. Scott, Tetrahedron Lett. 1967, 8, 3327–3329.
[375] P. Jean-Louis Hérisson, Y. Chauvin, Die Makromol. Chemie 1971, 141, 161–176.
[376] S. Kotha, M. K. Dipak, Tetrahedron 2012, 68, 397–421.

A-XVIII
 Appendix

[377] M. Schuster, S. Blechert, Angew. Chem. Int. Ed. 1997, 36, 2036–2056.
[378] R. H. Grubbs, S. Chang, Tetrahedron 1998, 54, 4413–4450.
[379] J. Mol, J. Mol. Catal. A Chem. 2004, 213, 39–45.
[380] D. J. O’Leary, H. E. Blackwell, R. A. Washenfelder, R. H. Grubbs, Tetrahedron Lett.
1998, 39, 7427–7430.
[381] D. J. O’Leary, H. E. Blackwell, R. A. Washenfelder, K. Miura, R. H. Grubbs,
Tetrahedron Lett. 1999, 40, 1091–1094.
[382] H. E. Blackwell, D. J. O’Leary, A. K. Chatterjee, R. A. Washenfelder, D. A. Bussmann,
R. H. Grubbs, J. Am. Chem. Soc. 2000, 122, 58–71.
[383] S. J. Connon, S. Blechert, Angew. Chem. Int. Ed. 2003, 42, 1900–1923.
[384] S. Lwin, I. E. Wachs, ACS Catal. 2014, 4, 2505–2520.
[385] S. J. Meek, R. V O’Brien, J. Llaveria, R. R. Schrock, A. H. Hoveyda, Nature 2011, 471,
461–466.
[386] T. M. Trnka, R. H. Grubbs, Acc. Chem. Res. 2001, 34, 18–29.
[387] S. B. Garber, J. S. Kingsbury, B. L. Gray, A. H. Hoveyda, J. Am. Chem. Soc. 2000,
122, 8168–8179.
[388] S. Gessler, S. Randl, S. Blechert, Tetrahedron Lett. 2000, 41, 9973–9976.
[389] B. R. Galan, K. P. Kalbarczyk, S. Szczepankiewicz, J. B. Keister, S. T. Diver, Org.
Lett. 2007, 9, 1203–1206.
[390] J. M. French, C. A. Caras, S. T. Diver, Org. Lett. 2013, 15, 5416–5419.
[391] K. Mennecke, K. Grela, U. Kunz, A. Kirschning, Synlett 2005, 2948–2952.
[392] V. S. Thengarai, J. Keilitz, R. Haag, Inorganica Chim. Acta 2014, 409, 179–184.
[393] M. R. Buchmeiser, New J. Chem. 2004, 28, 549–557.
[394] J. Pastva, J. Čejka, N. Žilková, O. Mestek, M. Rangus, H. Balcar, J. Mol. Catal. A
Chem. 2013, 378, 184–192.
[395] H. Staub, F. Kleitz, F.-G. Fontaine, Microporous Mesoporous Mater. 2013, 175, 170–
177.
[396] H. Balcar, T. Shinde, N. Zilková, Z. Bastl, Beilstein J. Org. Chem. 2011, 7, 22–28.
[397] D. Bek, H. Balcar, N. Žilková, A. Zukal, M. Horáček, J. Čejka, ACS Catal. 2011, 1,
709–718.
[398] H. Balcar, D. Bek, J. Sedláček, J. Dědeček, Z. Bastl, M. Lamač, J. Mol. Catal. A
Chem. 2010, 332, 19–24.
[399] M. Mayr, M. R. Buchmeiser, K. Wurst, Adv. Synth. Catal. 2002, 344, 712–719.
[400] K. Melis, D. De Vos, P. Jacobs, F. Verpoort, J. Mol. Catal. A Chem. 2001, 169, 47–56.
[401] S. T. Nguyen, R. H. Grubbs, J. Organomet. Chem. 1995, 497, 195–200.
[402] S. T. Nguyen, L. K. Johnson, R. H. Grubbs, J. W. Ziller, J. Am. Chem. Soc. 1992, 114,
3974–3975.

A-XIX
[403] P. Schwab, M. B. France, J. W. Ziller, R. H. Grubbs, Angew. Chem. Int. Ed. 1995, 34,
2039–2041.
[404] R. Caraballo, M. Rahm, P. Vongvilai, T. Brinck, O. Ramström, Chem. Commun. 2008,
6603.
[405] S. Soignier, M. Taoufik, E. Le Roux, G. Saggio, C. Dablemont, A. Baudouin, F.
Lefebvre, A. de Mallmann, J. Thivolle-Cazat, J.-M. Basset, et al., Organometallics
2006, 25, 1569–1577.
[406] I. Alperovich, D. Moonshiram, J. J. Concepcion, Y. Pushkar, J. Phys. Chem. C 2013,
117, 18994–19001.
[407] F. M. F. de Groot, Z. W. Hu, M. F. Lopez, G. Kaindl, F. Guillot, M. Tronc, J. Chem.
Phys. 1994, 101, 6570.
[408] M. Saes, Picosecond X-Ray Absorption Spectroscopy : Application To Coordination
Chemistry Compounds in Solution, EPFL, 2004.
[409] C.-X. Bai, Z.-Q. Zhang, X.-B. LÜ, R. He, W.-Z. Zhang, S.-L. Lu, Chinese J. Chem.
2006, 24, 1639–1642.
[410] A. K. Chatterjee, T.-L. Choi, D. P. Sanders, R. H. Grubbs, J. Am. Chem. Soc. 2003,
125, 11360–11370.
[411] R. L. Pederson, R. H. Grubbs, Metathesis Syntheses of Pheromones or Their
Components, 2004, US 6,696,597.
[412] X. Xu, S. M. Rummelt, F. L. Morel, M. Ranocchiari, J. A. van Bokhoven, Chem. - A
Eur. J. 2014, 20, 15467–15472.
[413] H.-U. Blaser, Chem. Commun. 2003, 221, 293–6.
[414] S. Akabori, S. Sakurai, Y. Izumi, Y. Fuji, Nature 1956, 178, 323–324.
[415] R. Noyori, T. Ohkuma, Angew. Chem. Int. Ed. 2001, 40, 40–73.
[416] H.-U. Blaser, C. Malan, B. Pugin, F. Spindler, H. Steiner, M. Studer, Adv. Synth. Catal.
2003, 345, 103–151.
[417] M. J. Burk, J. Am. Chem. Soc. 1991, 113, 8518–8519.
[418] R. R. Schrock, J. A. Osborn, J. Am. Chem. Soc. 1976, 98, 2143–2147.
[419] W. Chen, J. Xiao, Tetrahedron Lett. 2001, 42, 2897–2899.
[420] I. V. Komarov, A. Börner, Angew. Chem. Int. Ed. 2001, 40, 1197–1200.
[421] G. Erre, S. Enthaler, K. Junge, S. Gladiali, M. Beller, Coord. Chem. Rev. 2008, 252,
471–491.
[422] J. Wassenaar, J. N. H. Reek, Org. Biomol. Chem. 2011, 9, 1704–1713.
[423] M. T. Reetz, X. Li, Tetrahedron 2004, 60, 9709–9714.
[424] M. T. Reetz, G. Mehler, Tetrahedron Lett. 2003, 44, 4593–4596.
[425] M. T. Reetz, T. Sell, A. Meiswinkel, G. Mehler, Angew. Chem. Int. Ed. 2003, 42, 790–
793.

A-XX
 Appendix

[426] K. Mo, Y. Yang, Y. Cui, J. Am. Chem. Soc. 2014, 1–4.

[427] A. M. Shultz, O. K. Farha, J. T. Hupp, S. T. Nguyen, J. Am. Chem. Soc. 2009, 131,
4204–4205.
[428] A. L. W. Demuynck, M. G. Goesten, E. V. Ramos-Fernandez, M. Dusselier, J.
Vanderleyden, F. Kapteijn, J. Gascon, B. F. Sels, ChemCatChem 2014, 6, 2211–2214.
[429] B. Vilhanová, M. Ranocchiari, J. A. van Bokhoven, ChemCatChem 2015, n/a–n/a.
[430] M. T. Reetz, G. Mehler, A. Meiswinkel, Tetrahedron: Asymmetry 2004, 15, 2165–2167.
[431] S. R. Gilbertson, X. Wang, Tetrahedron 1999, 55, 11609–11618.
[432] N. E. Leadbeater, M. Marco, Chem. Rev. 2002, 102, 3217–3274.
[433] J. D. Morrison, R. E. Burnett, A. M. Aguiar, C. J. Morrow, C. Phillips, J. Am. Chem.
Soc. 1971, 93, 1301–1303.
[434] L. A. Oro, D. Carmona, C. J. Elsevier, in Handb. Homog. Hydrog., Wiley-VCH Verlag
GmbH, 2006, pp. 2–30.
[435] L. D. Julian, in Top. Heterocycl. Chem. (Ed.: J.P. Wolfe), Springer Berlin Heidelberg,
2013, pp. 109–155.
[436] D. Heller, A. H. M. De Vries, J. G. De Vries, A. H. M. De Vries, J. G. De Vries, in
Handb. Homog. Hydrog. (Eds.: J.G. de Vries, C.J. Elsevier), Wiley-VCH Verlag GmbH,
Weinheim, Germany, 2006, pp. 1483–1516.
[437] K. K. Tanabe, Z. Wang, S. M. Cohen, J. Am. Chem. Soc. 2008, 130, 8508–8517.
[438] M. Taddei, P. V. Dau, S. M. Cohen, M. Ranocchiari, J. A. van Bokhoven, F.
Costantino, S. Sabatini, R. Vivani, Dalt. Trans. 2015, 44, 14019–14026.
[439] W. Liang, D. M. D’Alessandro, Chem. Commun. 2013, 49, 3706–3708.
[440] C.-M. Lu, J. Liu, K. Xiao, A. T. Harris, Chem. Eng. J. 2010, 156, 465–470.
[441] Z. Ni, R. I. Masel, J. Am. Chem. Soc. 2006, 128, 12394–12395.
[442] S. H. Jhung, J.-H. Lee, J.-S. Chang, Bull. Korean Chem. Soc. 2005, 26, 880–881.
[443] S. H. Jhung, J.-H. Lee, J. W. Yoon, C. Serre, G. Férey, J.-S. Chang, Adv. Mater. 2007,
19, 121–124.
[444] H. Xu, J. Gao, D. Jiang, Nat. Chem. 2015, 7, 905–912.
[445] S. E. Ashbrook, D. M. Dawson, V. R. Seymour, Phys. Chem. Chem. Phys. 2014,
8223–8242.
[446] X. Kong, H. Deng, F. Yan, J. Kim, J. A. Swisher, B. Smit, O. M. Yaghi, J. A. Reimer,
Science 2013, 341, 882–885.
[447] A. J. Rossini, A. Zagdoun, M. Lelli, J. Canivet, S. Aguado, O. Ouari, P. Tordo, M.
Rosay, W. E. Maas, C. Copéret, et al., Angew. Chem. Int. Ed. 2012, 51, 123–127.
[448] T. Kobayashi, F. A. Perras, I. I. Slowing, A. D. Sadow, M. Pruski, ACS Catal. 2015,
7055–7062.
[449] M. Dierolf, A. Menzel, P. Thibault, P. Schneider, C. M. Kewish, R. Wepf, O. Bunk, F.

A-XXI
 Pfeiffer, Nature 2010, 467, 436–439.
[450] S. Moret, P. J. Dyson, G. Laurenczy, Nat. Commun. 2014, 5, 4017.
[451] R. L. Wingad, P. J. Gates, S. T. G. Street, D. F. Wass, ACS Catal. 2015, 5, 5822–
5826.
[452] A. Grabulosa, A. Mannu, A. Mezzetti, G. Muller, J. Organomet. Chem. 2012, 696,
4221–4228.
[453] L. E. Heim, N. E. Schlörer, J.-H. Choi, M. H. G. Prechtl, Nat. Commun. 2014, 5, 3621.
[454] T. K. Hyster, L. Knorr, T. R. Ward, T. Rovis, Science 2012, 338, 500–503.
[455] K. J. Kilpin, S. M. Cammack, C. M. Clavel, P. J. Dyson, Dalton Trans. 2013, 42, 2008–
2014.
[456] E. Quartapelle Procopio, S. Rojas, N. M. Padial, S. Galli, N. Masciocchi, F. Linares, D.
Miguel, J. E. Oltra, J. A. R. Navarro, E. Barea, Chem. Commun. 2011, 47, 11751–
11753.
[457] S. Rojas, E. Quartapelle-Procopio, F. J. Carmona, M. A. Romero, J. A. R. Navarro, E.
Barea, J. Mater. Chem. B 2014, 2, 2473–2477.

A-XXII