Ligand Exchange Processes on Solvated Lithium Cations · • “Ligand Exchange Processes on...
Transcript of Ligand Exchange Processes on Solvated Lithium Cations · • “Ligand Exchange Processes on...
Ligand Exchange Processes on Solvated
Lithium Cations
Den Naturwissenschaftlichen Fakultäten
der Friedrich-Alexander-Universität Erlangen-Nürnberg
zur
Erlangung des Doktorgrades
vorgelegt von
Ewa Maria Pasgreta
aus Więcbork (Polen)
Als Dissertation genehmigt von den Naturwissenschaftlichen Fakultäten der Universität Erlangen-Nürnberg
Tag der mündlichen Prüfung:
Vorsitzender der Promotionskommission:
Erstberichterstatter:
Zweitberichterstatter:
23. 3. 2007
Prof. Dr. Eberhard Bänsch
Prof. Dr. Dr. h.c. mult. Rudi van Eldik
Prof. Dr. Lutz Dahlenburg
Meiner Familie
5
Acknowledgements
I wish to thank Prof. Dr. Dr. h.c. mult. Rudi van Eldik for the supervision of
this work and many fruitful discussions. I am particularly grateful for his
advice, encouragement, his trust in me, and helping me to present my projects
internationally. I appreciate very much the good working conditions and the
support that I could always count with.
Dr. Achim Zahl is gratefully acknowledged for all, sometimes very demanding,
NMR measurements.
I would like to thank Prof. Dr. Lothar Helm (EPFL Lausanne) for introducing
me to NMRICMA 2.7 program and for vary fruitful discussions.
I am very grateful towards Dr. Ralph Puchta for theoretical calculations, for
correcting the German version of the summary and for his constant help and
encouragement.
I would like to thank Dr. Maria Wolak and Joanna Procelewska for their help,
in particular, at the beginning of my stay in Erlangen, and for their continuous
support. I also thank Dr. Ivana Ivanović-Burmazović for many interesting
discussions and Dr. Immo Weber for his help in the laboratory work.
I am also highly indebted to my colleagues from the practical courses: Dr. Kai
Peters and Dr. Hans Hanauer for their help at the beginning and Dr. Jelena
Galinkina and Klaus Pokorny for their support in the last months.
6
Many thanks to my room-mates: Christian, Erika, Alex, Fanny and Sabine as
well as to all members of the group and my friends outside the institute.
I would like to thank people involved in the European project which financed
part of my work: Kersti, Jan, Casey, Michael, Rossend, Wiktor, Klaus, Marco,
Daniel, Sami, Bruno, Gregor, and Barbara.
I am very grateful towards my family for their support and encouragement
during all these years.
Last but not least, I would like to thank Radim for everything. Dziękuję Ci
bardzo!
Die vorliegende Arbeit entstand in der Zeit von November 2001 bis Dezember
2006 am Institut für Anorganische Chemie der Friedrich-Alexander-
Universtität Erlangen-Nürnberg.
7
Publications and Conference Contributions
Publications
• Ralph Puchta, Michael Galle, Nico van Eikema Hommes, Ewa
Pasgreta, Rudi van Eldik, „Evidence for Associative Ligand Exchange
Processes on Solvated Lithium Cations“, Inorg. Chem., 2004, 43, 8227-
8229
• Ewa Pasgreta, Ralph Puchta, Michael Galle, Nico van Eikema
Hommes, Achim Zahl, Rudi van Eldik, „Ligand Exchange Processes on
Solvated Lithium Cations Part II – Complexation by Cryptands in γ-
Butyrolactone as Solvent“, J. Inclusion Phenom. Macrocyclic Chem.,
2007, 58, 81-88.
• Ewa Pasgreta, Ralph Puchta, Michael Galle, Nico van Eikema
Hommes, Achim Zahl, Rudi van Eldik, „Ligand Exchange Processes on
Solvated Lithium Cations Part III – DMSO and Water/DMSO Mixtures“,
ChemPhysChem, 2007, 8, 1315-1320.
• Ewa Pasgreta, Ralph Puchta, Achim Zahl, Rudi van Eldik, „Ligand
Exchange Processes on Solvated Lithium Cations Part IV – Acetonitrile
and Hydrogen Cyanide“, Eur. J. Inorg. Chem., 2007, 1815-1822.
• Ewa Pasgreta, Ralph Puchta, Achim Zahl, Rudi van Eldik, „Ligand
Exchange Processes on Solvated Lithium Cations Part V – Complexation
by Cryptands in Acetone as Solvent“, Eur. J. Inorg. Chem., 2007, 3067-
3076.
8
Conference Contributions
• “7Li NMR Study of the Kinetics of Lithium Ion Complexation by C222, C221
and C211 in different solvents”, 35th International Conference on Coordination
Chemistry, Heidelberg, Germany, July 2002, poster presentation.
• “7Li NMR Study of Lithium Ion Complexation by C222, C221 and C211 in
different solvents”, Graduiertenkolleg Symposium, Hirschegg-Kleinwalsertal,
Austria, March 2003, oral presentation.
• „High Pressure 7Li NMR Kinetic Studies of Li+ Complexation by Cryptands in
Acetone“, Dalton Discussion Meeting on Inorganic Reaction Mechanisms,
Athens, Greece, January 2004, poster presentation.
• “The Mechanism of Ligand Exchange on [Li(H2O)n]+ and [Li(NH3)n]+”,
Chemistry Symposium Erlangen-Rennes, Erlangen, Germany, June 2004,
poster presentation.
• “7Li NMR Studies of the Kinetics of Lithium Ion Complexation by Cryptands in
Gamma-Butyrolactone as Solvent”, The 2004 Younger European Chemists’
Conference, Torino, Italy, August 2004, poster presentation.
• „7Li NMR Studies of the Kinetics of Lithium Ion Complexation by Cryptands in
Gamma-Butyrolactone as Solvent“, Catalysis Summer School, Liverpool, Great
Britain, September 2005, poster presentation.
• “High Pressure 7Li NMR Kinetic Study of Li+ Complexation by Cryptands in
Gamma-Butyrolactone”, Dalton Discussion Meeting on Inorganic Reaction
Mechanisms, Cracow, Poland, January 2006, poster presentation.
• “Ligand Exchange Processes on Solvated Lithium Cations. Complexation by
Cryptands in γ-Butyrolactone as Solvent”, Chemistry Symposium Erlangen-
Rennes, Rennes, France, June 2006, poster presentation.
9
Abbreviations
∆H# activation enthalpy
∆S# activation entropy
∆V# activation volume
h Planck constant, 6.6261 × 10-34 J s
kb Boltzmann constant, 1.3807 × 10-23 J K-1
k rate constant
kobs observed rate constant
kon forward reaction rate constant
koff backward reaction rate constant
R gas constant, 8.3145 J K-1 mol-1
T temperature
t time
A associative reaction mechanism
D dissociative reaction mechanism
I interchange reaction mechanism
IA interchange associative reaction mechanism
ID interchange dissociative reaction mechanism
C211 4,7,13,18-Tetraoxa-1,10-diazabicyclo[8.5.5]eicosane
C221 4,7,13,16,21-Pentaoxa-1,10-diazabicyclo[8.8.5]tricosane
C222 4,7,13,16,21,24-Hexaoxa-1,10-diazabicyclo[8.8.8]haxacosane
LiOTf Lithium Trifluoromethanesulfonate
CH3CN acetonitrile
10
MeOH methanol
EC ethylene carbonate
PC propylene carbonate
DMF N, N-dimethylformamid
DMSO dimethylsulfoxide
GBL γ-butyrolactone
DFT density functional theory
B3LYP Becke’s 3 Parameter Hybrid Functional using Lee-Yang-Paar
Correlation
HF Hartree-Fock
MP Møller-Plesset Perturbation
NMR nuclear magnetic resonance
DNMR dynamic nuclear magnetic resonance
11
Table of Contents
1. INTRODUCTION ............................................................................... 13
1.1. A EUROPEAN PROJECT ...................................................................................... 14 1.2. OUTLINE OF THIS THESIS .................................................................................. 14 1.3. LITHIUM........................................................................................................... 17 1.4. SOLVATION .......................................................................................................18
1.4.1. Conventional Solvents ........................................................................................19 1.4.2. Plasticizers ......................................................................................................... 20
1.5. CRYPTANDS...................................................................................................... 22 1.6. EXCHANGE PROCESSES..................................................................................... 26 1.7. KINETIC APPLICATIONS OF NMR SPECTROSCOPY .............................................. 29 1.8. ALKALI METAL NMR SPECTROSCOPY ............................................................... 32 1.9. PROGRAMS ...................................................................................................... 34 1.10. COMPUTATIONAL METHODS ......................................................................... 36
1.10.1. Ab-initio Methods...............................................................................................37 1.10.2. Density Functional Theory ................................................................................ 38
1.11. REFERENCES.................................................................................................... 40
2. LIGAND EXCHANGE PROCESSES ON SOLVATED LITHIUM CATIONS. DMSO AND WATER/DMSO MIXTURES .............................43
2.1. GENERAL REMARK ........................................................................................... 43 2.2. ABSTRACT........................................................................................................ 43 2.3. INTRODUCTION ................................................................................................ 44 2.4. EXPERIMENTAL SECTION.................................................................................. 46
2.4.1. General Procedures............................................................................................ 46 2.4.2. 7Li NMR Spectroscopy....................................................................................... 46 2.4.3. Quantum-Chemical Calculations .......................................................................47
2.5. RESULTS AND DISCUSSION ................................................................................47 2.5.1. Coordination Number of Li+ in DMSO ..............................................................47 2.5.2. Selective Solvation of Li+ ................................................................................... 49 2.5.3. Ligand Exchange on Solvated Li+...................................................................... 52 2.5.4. Computed Mechanisms of Solvent Exchange................................................... 52
2.6. CONCLUSIONS ...................................................................................................55 2.7. REFERENCES.....................................................................................................57
3. LIGAND EXCHANGE PROCESSES ON SOLVATED LITHIUM CATIONS. ACETONITRILE AND HYDROGEN CYANIDE ....................63
3.1. GENERAL REMARK ........................................................................................... 63 3.2. ABSTRACT........................................................................................................ 63 3.3. INTRODUCTION ................................................................................................ 64 3.4. EXPERIMENTAL SECTION...................................................................................65
3.4.1. General Procedures............................................................................................ 65 3.4.2. 7Li NMR Spectroscopy....................................................................................... 66 3.4.3. Quantum-Chemical Calculations ...................................................................... 66
3.5. RESULTS AND DISCUSSION ............................................................................... 66 3.5.1. Coordination Number of Li+.............................................................................. 66 3.5.2. Selective Coordination of Li+............................................................................. 69 3.5.3. Ligand Exchange on Solvated Li+....................................................................... 71 3.5.4. Quantum-Chemical Calculations ....................................................................... 71
12
3.6. CONCLUSIONS................................................................................................... 77 3.7. REFERENCES ....................................................................................................78
4. LIGAND EXCHANGE PROCESSES ON SOLVATED LITHIUM CATIONS. COMPLEXATION BY CRYPTANDS IN ACETONE AS SOLVENT ................................................................................................. 84
4.1. GENERAL REMARK ............................................................................................84 4.2. ABSTRACT.........................................................................................................84 4.3. INTRODUCTION.................................................................................................85 4.4. EXPERIMENTAL SECTION ..................................................................................87
4.4.1. General Procedures ............................................................................................87 4.4.2. 7Li NMR Spectroscopy........................................................................................87 4.4.3. Programs............................................................................................................ 88 4.4.4. Quantum-Chemical Calculations.......................................................................89
4.5. RESULTS AND DISCUSSION ................................................................................89 4.5.1. Spectroscopic and Kinetic Observations ...........................................................89 4.5.2. Suggested Exchange Mechanism.......................................................................93 4.5.3. High Pressure 7Li NMR Measurements.............................................................99 4.5.4. Theoretical Calculations...................................................................................100
4.6. CONCLUSIONS.................................................................................................104 4.7. REFERENCES ..................................................................................................106
5. LIGAND EXCHANGE PROCESSES ON SOLVATED LITHIUM CATIONS. COMPLEXATION BY CRYPTANDS IN γ-BUTYROLACTONE AS SOLVENT ........................................................................................... 112
5.1. GENERAL REMARK ...........................................................................................112 5.2. ABSTRACT........................................................................................................112 5.3. INTRODUCTION................................................................................................113 5.4. EXPERIMENTAL SECTION .................................................................................116
5.4.1. General Procedures ...........................................................................................116 5.4.2. 7Li NMR Spectroscopy.......................................................................................116 5.4.3. Programs............................................................................................................117 5.4.4. Quantum-Chemical Calculations......................................................................117
5.5. RESULTS AND DISCUSSION .............................................................................. 118 5.5.1. Spectroscopic Observations ............................................................................. 118 5.5.2. Suggested Exchange Mechanism..................................................................... 122 5.5.3. High Pressure 7Li NMR Measurements........................................................... 126 5.5.4. Influence of Water ............................................................................................ 127 5.5.5. Theoretical Calculations................................................................................... 127
5.6. CONCLUSIONS................................................................................................. 130 5.7. REFERENCES ...................................................................................................131
6. SUMMARY ........................................................................................137
7. ZUSAMMENFASSUNG ....................................................................144
1. Introduction
13
1. Introduction
Lithium, the lightest of all metals and the least reactive, has been the object
of numerous experimental and theoretical studies. One might say that lithium
chemistry is a branch of chemistry in itself. Anomalous properties of lithium,
because of its small size, include such phenomena as the lithium bond (similar
to the hydrogen bond), formation of aggregates, and complexation by crown
ethers or cryptands.1
Lithium was first discovered in 1817 by the Swede Johann August
Arfvedson and pure lithium metal was first prepared the following year by
Davy and Brande by electrolysis of lithium oxide. Lithium’s name originates
from Greek word “lithos” meaning stone and it is the lightest of all solids (0.53
g/cm3). The industrial process by which lithium is prepared today is
electrolysis of lithium chloride or lithium carbonate. Today, the amount of
lithium produced yearly is about 10.000 tons.2
Coordination chemistry of lithium and other alkali metals was viewed until
the 1960’s as a rather barren area, made so since the singly charged and
relatively large M+ ions were held to have rather poor coordinating ability. The
majority of the known metal salt complexes were hydrates. Two developments
have transformed this research area over past the 40 years. The first began
with the Nobel prize-winning discovery by Pedersen, Lehn and Cram of
macrocyclic ligands like “crown” ethers, cryptands, and spherands. Such
ligands wrap around the metal ion, encapsulating it to greater or lesser degree,
and are highly selective regarding which metal cation can be accommodated.
This feature has led to important application in cation separation, ion-specific
measuring devices, and phase-transfer catalysis, and as mimics of naturally
occurring macrocyclic antibiotics. The second major development was brought
about by extensive investigation of non-aqueous alkali metal chemistry
prompted largely by the importance of lithium reagents in organic synthesis.3
1. Introduction
14
1.1. A European Project
The research work presented in this thesis has been developed within the
framework of the European Research Training Network on Solvation
Dynamics and Ionic Mobility in Conventional Solvents and
Plasicizers – by Computational Chemistry and Experiment. The
overall goal of this Network (which involved the collaboration between six
European research groups) was to seek firm knowledge about the mechanisms
of the dynamical processes present, and their relative importance. The primary
scientific aims were:
• to discover unifying and differentiating features in ligand and solvent
exchange processes around metal ions in conventional and polymer
solvents;
• to study the connection between solvation dynamics and ionic
diffuson/conduction mechanism in these systems;
• to help in the search for new applied systems by providing a more
detailed understanding of the basic phenomena occurring in model
electrolytes
In this thesis these aims are approached from the microscopic point of
view, making use of both computer calculations and experimental techniques.
The way solvent molecules interact with the ion, and how this interaction
affects the physico-chemical properties of the system are studied.
1.2. Outline of this Thesis
The contributions of the present work to the network objectives can be
divided into three main topics which constitute the backbone of this thesis:
Solvation: how is the first coordination sphere around Li+ constituted;
how does the nature of the solvent influence solvation;
1. Introduction
15
Solvent exchange: the exchange process between the first and second
coordination spheres of the Li cation is studied; the interplay between
diffusion and the exchange mechanism is addressed.
Lithium complexation by cryptands and ligand exchange in
conventional solvents and plasticizers: how is Li+ complexed by
different cryptands; how do solvents of different properties influence the
process of Li+ exchange between complexed and solvated sites.
The Li+ cation remained the centre of our interest: its solvation shell and
the way it is solvated. Both experimental and theoretical methods were
employed. Solvents of different properties and their influence on the solvation
shell were investigated. Questions concerning selective solvation around Li+
and the solvation number of the Li+ cation were studied as a function of the
solvent properties. The Gutmann Donor Number was mostly considered as a
factor describing the donating abilities of the solvent. However, we were not
only interested in thermodynamic phenomena. Kinetic measurements were
also of primary importance. Once the solvation shell of the Li cation was
known, we wanted to investigate the exchange of certain solvent molecules
around Li+. As Li+ is a very labile cation, solvent exchange was found to be very
fast on the NMR time scale and this process was therefore also investigated by
computational methods.
Solvent exchange is the simplest ligand exchange process, where, in fact,
there is no net change in the composition of the first solvation shell. However,
this process helps to understand the exchange of Li+ between different
environments, i.e. between solvent and another ligand.
Macrocyclic ligands called cryptands were found to bind the alkali and
alkaline earth metal ions very efficiently. They have been extensively explored
with respect to their structure and binding properties. Although there is a
wealth of information about thermodynamic and structural studies on the
complexation of metal ions by macrocyclic ligands in different solvents, kinetic
and mechanistic studies have received less attention and hence are still poorly
understood. For a better understanding of the pronounced selectivities of
1. Introduction
16
macrocyclic ligands towards alkali and alkaline earth cations, it is necessary to
investigate their mechanisms of formation and dissociation. In this
contribution, we were particularly interested in the mechanistic aspects of the
Li+ exchange process between cryptands and solvent. In the case of many
solvents the exchange was very fast or, on the contrary, in the case of small
cavity ligands, there was almost no exchange observed. Therefore, two
solvents, viz. acetone and γ-butyrolactone, were chosen for more detailed
mechanistic investigations. They possess different properties and applications,
such that the obtained results could be of importance in different application
fields.
The experimental method used throughout was 7Li NMR. During the past
few decades it has been shown that multinuclear NMR is a very powerful
technique for the study of alkali complexes with macrocyclic ligands,
particularly in non-aqueous solutions. 7Li is active in NMR and possesses
features that make it convenient for this kind of investigation. Particular
attention was given to high pressure 7Li NMR, which is the first example
reported for this kind of research. High pressure NMR is a powerful tool
employed to elucidate details of the reaction mechanism. The determination of
activation volumes from the pressure dependence of exchange rates for
solvated metal ions is a useful method for the diagnosis of the intimate
mechanism.
The outline of this thesis is as follows: Part I is dedicated to introduce the
main topics studied and to give a brief introduction of the methods used. Parts
II, III, IV and V contain the results and discussion of specific studies on
Solvation of the Li Cation in DMSO and DMSO/water mixure, Solvation of
the Li Cation in CH3CN and CH3CN/water mixure, Ligand Exchange in
Conventional Solvent and Ligand Exchange in Plasticizer, respectively.
Finally, concluding remarks are made in Parts VI and VII (in German).
1. Introduction
17
1.3. Lithium
Lithium is the lightest of all metals. It has the highest specific heat of any
solid element and therefore is very useful in heat transfer applications. Lithium
is also a leading contender as battery anode material due to its high
electrochemical potential. The lithium ion is exceptionally small and has,
therefore, an exceptionally high charge to radius ratio. As a result, its
properties are considerably different from similar ions such as sodium and
potassium. Contrarily, its hydrated radius is much larger than similar ions and
therefore exhibits different solution properties such as very low vapor
pressure, freezing point, and other colligative properties. The large solvation
shell around the lithium ion also causes it to have lower mobility in solution.
The aforementioned properties of the lithium ion contribute to its unique
behavior and effects in a variety of environments. The chloride salt of lithium
is one of the most hygroscopic materials known and is therefore used in many
air conditioning and drying systems. In general, the lithium ion has been well
documented to have profound biological effects of varying intensity on many
life forms. Numerous publications confirmed its influence on enzyme activity,
metabolism, respiration, and active transport. A prominent effect of lithium on
the nervous system can be found in the field of medicine.4, 5 Here it is used to
treat severe mental disorders such as manic-depression by controlling mania
and stabilizing mood swings.6-16 Although its therapeutic value has been
debated, there is no doubt that a number of manic patients have benefited
from lithium treatment. The lithium ion seems to counteract the manic
symptoms in a specific way, however, little is understood about the mechanism
of its therapeutic action. In conclusion, it is evident that lithium can have some
profound effects on a variety of chemical and biological systems. Although it
has been intensively studied, little is known about the details of the action of
lithium including required concentrations and mechanisms by which changes
occur.
1. Introduction
18
1.4. Solvation
Solute-solvents studies involving electrolytes have been a subject of interest
to solution chemists for quite sometime. Such studies on the transport
properties of electrolytes in aqueous and non-aqueous solvents are of interest
in various technologies like high energy density batteries, photo-
electrochemical cells, electrodeposition, wet electrolytic capacitors and in
electro-organic synthesis.
Ion solvation is a phenomenon of fundamental interest in many aspects of
chemistry because solvated ions are almost omnipresent on Earth. Hydrated
ions occur in aqueous solution in many chemical and biological systems. Ions
solvated in organic solvents or mixtures of water and organic solvents are also
very common. Selective solvation occurs when the solvate prefers one of the
solvents over the other. This phenomenon can be used to separate different
ions from each other, a process known as solvent separation. The exchange of
solvent molecules around ions in solution is fundamental to the understanding
of the reactivity of ions in solution. Solvated ions also play a key role in
electrochemical applications, where for instance the conductivity of
electrolytes depends on the ion-solvent interactions.17
A literature survey suggests that the use of mixed solvents involving high
permittivity, low viscosity, and a wide liquid range is practicable for use in high
energy density batteries, especially lithium batteries. The advantage of using
mixed solvents over their pure counterparts in lithium batteries is to
manipulate the suitable properties of these solvents for fulfilling the above
conditions.
It is known that aprotic solvents show high permittivities, low viscosities,
and strong temperature dependences. Hence, in order to develop suitable
mixed solvents fulfilling the conditions mentioned above, mixtures of solvents
must be prepared to balance the drawbacks of pure solvents. The
conductivities of lithium perchlorate in different solvent systems are explained
1. Introduction
19
in terms of solution viscosity and the donor-acceptor properties of the
solvents, which mainly govern the ion-solvent interaction.
1.4.1. Conventional Solvents
Solvents and solutes can be broadly classified into polar and non-polar.
There are a number of factors describing solvent properties. Polarity can be
measured as the dielectric constant or the dipole moment of a compound.
Polar solvents can be further subdivided into polar protic solvents (e.g. water
or methanol) and polar aprotic solvents (e.g. acetone, acetonitrile,
dimethylformamide or dimthylsulfoxide).
Table 1.1. Gutmann Donor Number values for common organic solvents.
Solvent DN, kcal mol-1
1,2-Dichloroethane (reference solvent) 0.0
Nitromethane 2.7
Acetonitrile 14.1
Propylene carbonate 15.1
Ethylene carbonate 16.4
Acetone 17.0
γ-Butyrolactone 18.0
Diethyl ether 19.2
Tetrahydrofuran 20.0
N,N-Dimethylformamide 26.6
Dimethyl sulfoxide 29.8
Methanol 30.0
Water 33.0
Triethylamine 61.0
1. Introduction
20
Donor Number (DN) is a qualitative measure of Lewis basicity. It is defined
as the negative enthalpy value for 1:1 adduct formation between a Lewis base
and the standard Lewis acid SbCl5, in dilute solution in the non-coordinating
solvent 1,2-dichloroethane with a zero DN. The Donor Number describes the
ability of a solvent to solvate cations and Lewis acids. The method was
developed by V. Gutmann in 1976.18 Values of DN for common organic solvents
are summarized in Table 1.1.
Solvents undergo various weak chemical interactions with the solute. The
most usual interactions are the relatively weak van der Waals interactions
(induced dipole interactions), the stronger dipole-dipole interactions, and even
stronger hydrogen bonds.
The role of a solvent on organometallic substitution reactions cannot be
overestimated; even a small modification of the solvent structure may cause a
large change in the rate and in the pattern of the processes in solution.18
1.4.2. Plasticizers
Battery technology has achieved spectacular progress in recent years. A
most successful product is the rechargeable Lithium Ion Battery (LIB), which
has reached an established commercial status with a production rate of several
millions of units per month. Currently, the lithium salt electrolyte is not held in
an organic solvent like in the past models, but in a solid polymer gel electrolyte
such as polyacrylonitrile, polyvinylidine fluoride, polyethylene oxide, etc.
There are many advantages of this design, for instance, the solid polymer
electrolyte is not flammable as the organic solvent that the Li-ion cell uses.
Thus, these batteries are less hazardous if mistreated.19
The vast majority of electrolytes are the electrolytic solution-types that
consist of salts (also called “electrolyte solutes”) dissolved in the solvents (also
called plasticizers), either water (aqueous) or organic molecules (non-
aqueous), and remain in the liquid state within the service-temperature range.
The most used solvents are ethylene carbonate, propylene carbonate and γ-
butyrolactone. (Figure 1.1)
1. Introduction
21
Figure 1.1. Structural formulas of ethylene carbonate (a), propylene carbonate (b) and γ-butyrolactone (c).
Experimentally it has been found that a mixture of two or more plasticizers
is more convenient, as it allows optimization of the balance between different
features (such as dielectric constant, viscosity, ionic diffusion, salt dissociation,
and chemical stability) and thus to enhance the battery performance and
cyclability.
The overall goal of the project was to study the solvation dynamics and
ionic mobility in conventional solvents and to seek firm knowledge about the
mechanisms of the dynamical processes present, and their relative importance.
The importance of this mechanism can be illustrated with an example from the
polymer field. High molecular-weight poly(ethylene oxide), PEO, with
dissolved ions is a polymeric system with low ionic conductivity at room
temperature. On the other hand, PEO with dissolved ions and a second, low-
molecular weight organic solvent added is an example of a polymeric system
with high ionic conductivity at room temperature. The addition of even modest
amount of such a small molecule to the pure polymer electrolyte drastically
increases the conductivity. The low-molecular weight organic solvent is called
plasticizer.
O O
O
O O
O
O
O
(a) (b) (c)
O O
O
O O
O
O
O
O O
O
O O
O
O
O
(a) (b) (c)
1. Introduction
22
1.5. Cryptands
Whereas coordination chemistry was almost exclusively concerned with the
complexes formed between organic or inorganic ligands and transition metal
ions, only scattered examples of alkali cation (AC) complexes were
documented. The situation changed entirely with the discovery (around the
middle 1960’s in short sequence) of three classes of organic ligands capable of
forming strong and selective complexes with alkali, as well as alkaline-earth,
cations. First, natural macrocyclic antibiotics (in particular valinomycin) were
found to bind and to make membranes permeable to AC’s (namely K+), acting
as ionophores. Then, it was discovered that macrocyclic polyethers, the crown
ethers, formed complexes with AC’s. Thirdly, macrobicyclic receptors, the
cryptands, were designed and shown to form even much more stable AC
cryptate complexes by inclusion into the three-dimensional molecular cavity.
Numerous other ligands were then developed in particular the spherands.
Examples of mycrocyclic ligands are illustrated in Figure 1.2. A rich
coordination chemistry of AC’s has thus been established, displaying a whole
range of complexes of varied structural, thermodynamic and kinetic
properties. Strong and selective complexation of AC’s was achieved, yielding
for instance stability constants higher than 105 l mol-1 in aqueous solution for
the Li+, Na+ and K+ cryptates of the optimal macrobicyclic receptors. Very
stable and selective alkaline-earth complexes of macrocyclic and
macropolycyclic ligands were also obtained.20
1. Introduction
23
Figure 1.2. Structural formulas of crown ethers (a), cryptands (b), and spherands (c).
The chelate effect of acyclic ligands was extended to macrocyclic and
cryptate effects defined by special features (such as substrate inclusion, high
stability, high selectivity, slow exchange rates, shielding from the environment)
presented by macrocyclic and macropolycyclic ligands and their metal ion
complexes.
The field of macrocyclic and macropolycyclic chemical compounds has
greatly advanced over the last years mainly with the goal of developing
complexing-agents, or molecular receptors, capable of binding very efficiently
and very selectively a given substrate. Thus, by taking into account specific
molecular interactions and the relationship between geometrical structure and
binding sites, the chemistry of molecular recognition has been developed, this
being a major part of supramolecular chemistry.21
1. Introduction
24
Compared to the usual macrocycles of the crown type which have a two
dimensional framework, the cryptands are three dimensional in structure
containing an intramolecular cavity (or “crypt”). Cryptands form inclusion
complexes, the cryptates, in which the substrate is located inside the cavity.
Several reviews and monographs which contain important information on
the structures and applications of macrocyclic complexes with alkali metal ions
are available.22-29
Many natural substances have a macrocyclic structure whether or not they
exhibit complexing properties, e.g., cyclic antibiotics, cyclodextrins. Since the
macrocyclic structure is so widespread in nature, one can conclude that it
offers over the linear arrangement certain “useful” properties with respect to
the functions performed. The fundamental observation has been incorporated
in the chemistry of molecular recognition.
The alkali metal cations are spherical species bearing a positive charge. The
recognition of these species involves the selection of a sphere from a collection
of spheres of different radii.
In 1975 Lehn and Sauvage30 reported stability contants and selectivities for
complexation of alkali metal and alkaline earth cations by cryptands. In these
multidentate ligands, two nitrogen atoms were linked by three oxygen atom
containing bridges. Diameters of the alkali metal cations and the cryptand
cavities are recorded in Table 1.2. The cavity size of the cryptand C211 is
appropriate for strong complexation of Li+, but too small to accommodate Na+
or any of the other alkali metal cations. Stability constants for complexation of
Li+, Na+, and K+ by cryptands C222, C221, and C211 are summarized in Table
1.3.29
1. Introduction
25
Table 1.2. Diameters of alkali metal cations and cryptand cavities.
Cation Diameter (Å) Cryptand Diameter (Å)
Li+ 1.48 C211 1.6
Na+ 2.04 C221 2.2
K+ 2.76 C222 2.8
Table 1.3. Stability constants for formation of alkali metal ion complexes with cryptands in water at 25 °C.
log K
Cryptand Li+ Na+ K+
C211 5.5 3.2 < 2
C221 2.5 5.4 4.0
C222 1.0 4.0 5.5
A cation is characterized by its ionic radius, charge and charge density,
polarizability, solvation energy, and also the ionization potential of the
corresponding metal. Cation complexation is governed by coordination
chemistry. The selection of the cation is ensured by a cavity of adequate size,
and the bond to the macrocycle arises through the presence of polar
interaction sites lining the cavity.
The first macrobicyclic diamines were synthesized by Simmons and Park.31
These ligands display a particular form of isomerism since the pair at the
bridgehead nitrogen may be directed either inwards or outwards with respect
to the cavity thus giving rise to three forms: exo-exo or out,out; exo-endo or
out,in; endo-endo or in,in. The predominant form depends on the size of the
bicyclic system.
Inclusion of a Mn+ metal cation in the intramolecular cavity of the three
possible forms of cryptands should give rise, in principle, to the following three
topographies: exo-exo, exo-endo and endo-endo. The endo-endo conformation
1. Introduction
26
is greatly favored because in this case both amine groups participate in
stabilizing the complex.
In this study we emphasize the analogy in structure between ion-polymer
complexes and ion-macrocyclic ligand complexes. It is suggested that the two
types of systems are complementary, in that studies of polymer-salt systems
can give information about interactions in mycrocyclic ligand-salt systems, and
vice versa.32 Crown ethers have been found to improve the electrical
conductivity of solid polymer electrolytes that might be used in lithium
batteries.33 However, there is another way in that the above case makes much
closer contact with the practical world of lithium batteries. Poly(ethylene
oxide), PEO, has been a prototypical polymer used in studies of lithium battery
solid electrolytes. The same –CH2CH2O– moiety that appears in the crown
ethers and cryptands is the building block of PEO. Thus it is hardly surprising
that PEO has a high affinity for alkali metal cations just as crown ethers do.34
1.6. Exchange Processes
The exchange of solvent molecules around a metal ion is the simplest
ligand substitution process and is fundamental to the understanding of the
reactivity of the metal ions in solution. The mechanism of ligand exchange is
often discussed in terms of the scheme developed by Merbach et al.,35 which
emphasizes a structural definition of the different exchange mechanisms, i.e.
the mechanism is defined by the positions of the incoming and leaving ligands
relative to the metal ion, as can be seen in Figure 1.3.
1. Introduction
27
Figure 1.3. Schematic volume profiles for different solvent exchange mechanisms: kP is the observed rate constant at a given pressure P; kP is decelerated or accelerated on increasing pressure for a D or A mechanism, respectively.36
The prevailing exchange mechanism can experimentally be deduced from
the activation parameters ∆S# and ∆V#. Large negative activation entropies or
activation volumes are interpreted as corresponding to an associative
mechanism (A) and large positive values as dissociative (D), whereas small
values are interpreted as concerted mechanisms (Ia, I, or Id).
The pressure dependence of the exchange rate constant leads to the
activation volume, ∆V#, which has become the major tool for the experimental
determination of the exchange mechanisms.37, 38 The volume of activation,
∆V#, is defined as the difference between the partial molar volumes of the
1. Introduction
28
transition state for the reaction and the reactants and is related to the pressure
variation of the rate constant at a constant temperature T by Equation (1.1).
RT
V
dP
kd
T
#ln Δ−=⎟
⎠⎞
⎜⎝⎛
(1.1)
An approximate solution of the above differential Equation (1.1) is Equation
(1.2).
PRTV
kkP
P
00
=
= ⎟⎟⎠
⎞⎜⎜⎝
⎛ Δ−=
#
lnln (1.2)
Solvent and ligand exchange and substitution reactions in liquid solutions
are basic chemical processes that in many cases control the rate and efficiency
of chemical reactions. This research topic is related to various technological
applications both in the polymer electrolyte field and for basic chemical
processes. Solvent dynamics is one of the phenomena which enable ionic
conductivity in a polymeric salt.
Exchange of a solvent molecule between the first and the second
coordination sphere is the simplest reaction that can occur on a metal ion in
aqueous and non-aqueous solution. This reaction is fundamental in
understanding not only the reactivity of metal ions in chemical and biological
systems but also the interaction between the metal ion and the solvent
molecules. The displacement of a solvent molecule from the first coordination
shell represents an important step in complex-formation reactions of metal
cations and in many redox processes.39 The measured rates of solvent
exchange vary extensively with the nature of the metal ion and, to a lesser
extend, with that of the solvent. Metal ions of main group elements have filled
electron shells, and they differ mainly in electric charge and ionic radius. The
number of solvent molecules in the first shell around the ion, commonly called
the coordination number, CN, ranges from 4 up to 10.40 Alkali group ions are
very labile, due to the low surface charge density and the absence of ligand
field stabilization effects.36 Solvent exchange at Li+ in water is extremely fast,
making it difficult to study directly. An exchange rate constant of
1. Introduction
29
approximately 109 s-1 at 25 °C was predicted on the basis of complex-formation
data.36 (Figure 1.4)
Figure 1.4. Mean lifetimes, τH2O, of a particular water molecule in the first coordination shell of a given metal ion and the corresponding water exchange rate kH2O at 298 K. (taken from the recent review36)
1.7. Kinetic Applications of NMR Spectroscopy
The rates of chemical reactions vary over many orders of magnitude. Some
reactions are so slow that they practically do not occur, even though the free
energy would decrease considerably. On the other hand, many chemical
reactions are so fast that until recently they have been called “immeasurably
fast”. During the past 50 years, however, a wealth of techniques have been
developed to study reaction times much shorter than one second and well
outside the range of conventional analytical methods.41
Figure 1.5 shows various experimental methods and their time scales.
1. Introduction
30
Figure 1.5. Various experimental methods and their time scales.
Spectroscopic techniques such as optical spectrophotometry or NMR
spectroscopy have often been used to monitor concentration changes as a
function of time, thereby allowing the kinetic parameters to be determined.
Fast chemical exchange may play a direct role in the optical (or NMR)
absorption process itself, changing the line shape of the observed spectra. This
kinetic modification of line shapes occurs at complete equilibrium, whereas all
other fast reaction techniques require finite deviations from equilibrium.
Kinetic information is available from the resulting line broadening. The older
CW (continuous wave) NMR technique has been superseded by FT (Fourier
transform) NMR.41
Nuclear magnetic resonance (NMR) is one of the most widely used methods
for studying rapid intra- and intermolecular equilibrium processes of
inorganic, organic, and biochemical systems involving the transfer of
1. Introduction
31
components of ions and molecules or the complete entities between different
chemical environments in the liquid state. The only requirement of the method
is that the chemical equilibration process must involve the exchange of nuclear
spins between different magnetic environments, under which circumstances
first order rate constant in the range from 5×10-3 s-1 to 106 s-1 are accessible
using commercially available NMR spectrometers. In the vast majority of
applications, the chemical system studied is at thermal equilibrium and the
chemical kinetic parameters are determined from the modification of the NMR
spectrum induced by the nuclear spin exchange characterizing the chemical
equilibration processes.42
The derivation of the chemical parameters from NMR data requires
knowledge of the factors determining spectral line shapes and the sources of
spectral modifications induced by nuclear spin exchange accompanying
chemical equilibration processes.
Among other spectroscopic techniques, high-pressure NMR especially has
found remarkable applications in the investigation of reaction mechanisms. A
high-pressure NMR probe designed for a 400 MHz narrow bore NMR
spectrometer was developed in our laboratory43 and is presented in Figure
1.6(a) The aim of the proposed design was to allow high resolution, high-
pressure NMR experiments in a narrow bore magnet system without any
modification of the magnet. The probe should be easy and safe to handle. More
details are reported elsewhere.43 The NMR tubes used for high pressure
measurements are presented in Figure 1.6(b)
1. Introduction
32
Figure 1.6. Narrow bore NMR probeheads (a) and NMR high pressure tubes (b).
1.8. Alkali Metal NMR Spectroscopy
The magnitudes of chemical shifts for alkali metal resonances increase
markedly from lithium to caesium, and are a further order of magnitude larger
for thalium. However the sensitivity of the alkali metal chemical shifts to the
environment, specifically the solvation variation, increases in the opposite
direction. Chemical shifts depend both on the nucleus and the nature of the
solvent. They also, for a given alkali metal cation, depend on the nature of the
counterion and the concentration of the salt.44
The properties of 7Li nucleus are quite favorable for NMR studies. (see
Table 1.4) The resonance lines of Li+ ion in solutions are exceptionally narrow
and chemical shift can be measured with considerable accuracy.45
Lithium consists of two stable isotopes: 6Li and 7Li. The magnetogyric ratio,
γ, of 7Li is comparatively high. At a magnetic field strength of 9.4 T (1H: 400
MHz), the correlate Larmor frequency of 7Li is 155.45 MHz. Because of its high
1. Introduction
33
γ number and its high natural abundance, 7Li is a sensitive and easy to detect
nucleus. However, many NMR studies on organolithium compounds are now
frequently carried out by observing the 6Li nucleus, preferably in 6Li isotope-
enriched samples.46
Table 1.4. Nuclear properties of key isotopes of lithium.46, 47
Isotope 6Li 7Li
Atomic mass 6.01512 7.01600
Natural abundance (%) 7.42 92.58
Nuclear spin 1 3/2
υo at 9.4 T (1H = 40 MHz) 58.862 155.454
Nuclear magnetic moment μ +0.82203 +3.25636
Quadrupole magnetic
moment Q (10-28 m2) -8 × 10-4 -4.5 × 10-2
Receptivity Dc (13C = 1.00) 3.58 1540
Width factor (7Li = 1.0) 2.0 × 10-3 1.00
Selective solvation of alkali metal ions in mixed solvents can be monitored
from the extent to which the chemical shift of the alkali metal nucleus varies
with solvent composition.
The linewidth measurements may also give some idea of solvation in mixed
solvents. The dependence of the resonance linewidths on solvent composition
is investigated in the mixtures containing one good and one poor solvent. The
difference in pattern is striking and is attributed to relatively long-lived cation-
solvent complexes formed by the strongly coordinating solvents. Such a
difference in behavior is reasonable. It is consistent with solvating abilities of
the solvents, and with their respective Gutmann donor numbers.
1. Introduction
34
The exchange of ions between different environments is usually rapid with
respect to the NMR time scale, resulting in only one resonance signal at an
average frequency determined by the magnetic shielding and lifetime of the
nucleus in each of the sites. Alteration of parameters such as concentration,
counterions, and solvent produces changes in the relative proportion and type
of environment which may be reflected by alteration of the NMR spectra on
chemical shift and/or line shape, and/or line width of the observed
resonance.45
1.9. Programs
The line widths of the NMR signals can be obtained by fitting Lorentzian
functions to the experimental spectra using the NMRICMA 2.7 program.48
NMRICMA is a program written for MATLAB to evaluate 1D NMR and EPR
spectra. It reads experimental NMR spectra treated with 1D WIN-NMR from
Bruker. NMRICMA can fit Lorentzian resonance lines to experimental spectra,
fit NMR multiplets (formed by Lorentzian lines) to experimental spectra, fit
exchange broadened NMR spectra using Kubo-Sack formalism and simulate
exchange broadened NMR spectra.
Complete line-shape analysis based on the Kubo-Sack formalism using
modified Bloch equations49 was performed with NMRICMA 2.7 to extract rate
constants from experimental spectra. In order to elucidate the exchange rate
constant between two species, the exchange matrix must be created. In case of
a two site exchange, a 2 × 2 exchange matrix was created.
⎥⎥⎥⎥
⎦
⎤
⎢⎢⎢⎢
⎣
⎡
−
−
=
ABB
AAB
B
A
ABAB
kp
pk
p
p
kk
D
1. Introduction
35
where: kAB refers to the rate constant between species A and B,
pA and pA represent the populations of the NMR signals referring
to the species A and B, respectively.
The exchange matrix is created and saved as the matex file in Matlab:
function m = matex2(p,k) %---------------------------------------------------------------------------- % funtion subroutine for exchange %---------------------------------------------------------------------------- % % alpha - version; started on january 1994 % © lothar helm, icma université lausanne % % rate constant(s) = k % populations = p % % global xtitle matexfile si td sw sf limits nlimits; nsites = length(p); m = zeros(nsites); m(1,1) = -k(1); m(1,2) = +k(1); m(2,1) = p(1)/p(2)*k(1); m(2,2) = -m(2,1);
The adjustable parameters are: resonance frequency, intensity
(population), line width and baseline. The temperature and pressure
dependent 7Li line widths and chemical shifts employed in the line shape
analysis were extrapolated from low temperatures where no exchange-induced
modification occurred.
The fitting method used throughout was the Lavenberg-Marquard method.
The progress of fitting can be followed by observing the MATLAB window. The
iteration stops when the maximum number of iterations is reached. The
spectrum plot window of NMRICMA shows the result of the fit: experimental
and fitted spectra on an overlay plot, difference between fitted and
experimental spectrum, and fitting results are given as well as some other
1. Introduction
36
useful information. One example of the output spectrum can be seen in Figure
1.7.
Figure 1.7. The output 7Li NMR spectrum from the NMRICMA fitting program.
1.10. Computational Methods
The theoretical calculations employed in this work were mostly based on
quantum chemical methods. They were based either on the wave function or
on the electron density of the investigated molecule to calculate its structure
and properties. The significance of these two methods can be recognized from
two Nobel Prize Awards given in 1998 to J. A. Pople for the development of the
computational methods in quantum chemistry and to W. Kohn for the
development of the density functional theory.
1. Introduction
37
The quantum chemical methods based on the wave function come from the
Schrödinger equation and describe all relevant electronic, time dependent and
relativistic effects. For the purpose of many chemical problems one can often
neglect the time dependent and relativistic effects to obtain in this way the
time-independent, non-relativistic Schrödinger equation. The fundamental
assumption (known as the Born-Oppenheimer approximation) is that, in the
adiabatic limit, we can decouple the nuclear and the electronic degrees of
freedom; in this way the time-independent electronic Schrödinger equation is
obtained.
1.10.1. Ab-initio Methods
In order to solve the electronic Schrödinger equation, an additional
approximation is introduced: Hartree-Fock (HF) approximation.
The wave function which describes both the single electron spatial
distribution and its spin is called spin orbital. The fundamental idea of the HF-
theory is that the N-electrons wave function (N is the number of electrons) is
displaced by more single electron wave functions. To a first approximation we
can consider that the electrons do not interact. The total Hamiltonian will be
the sum of N individual Hamiltonians. The total wave function is the product
of N spin orbitals. Since electrons in fact interact with each other, in the
Hartree-Fock approximation, the Coulomb interaction among electrons is
averaged and the many electrons wave function (and energies) is obtained in
an iterative way. In this model one electron is influenced by the presence of
Quantum Chemical Calculations
electron density based methodswave function based methods
density functional theory (DFT)
e.g. B3LYP, BLYP, PBE, BP86…
Ab-initio methods
e.g. HF, MP2, MP4…
semi empirical methods
Quantum Chemical Calculations
electron density based methodswave function based methods
density functional theory (DFT)
e.g. B3LYP, BLYP, PBE, BP86…
Ab-initio methods
e.g. HF, MP2, MP4…
semi empirical methods
1. Introduction
38
other electrons. For practical purposes, the unknown spatial molecular orbitals
are often transformed into one set of known spatial functions, viz. the basis set.
In practice the Hartree-Fock equation is solved by introducing a finite set of K
basis functions, which lead to a set of N occupied spin orbitals and K-N virtual
ones.
Since there is no correlation in the electron movement in the HF-
approximation, this must be considered by means of some other methods. A
different procedure to find the correlation energy, which is not variational but
size consistent at each level, is the Perturbation Theory (PT). In 1934 Møller
and Plesset proposed a perturbation treatment of atoms and molecules in
which the unperturbated wave function is the Hartree-Fock function; this is
the so called Møller-Plesset (MP) perturbation theory (general: many body
perturbation theory). It is based on Rayleigh-Schrödinger perturbation theory
where the total Hamiltonian of the system is divided into two parts: a zero-
order part which has known eigenvalues and eigenfunctions, and a
perturbation operator. Second (MP2) and fourth (MP4) order Møller-Plesset
calculations (where we refer to the order of the energies) are standard levels
used for small systems and are implemented in many computational chemistry
codes.
1.10.2. Density Functional Theory
Hohenberg and Kohn in 1964 suggested that the many-electron wave
function was a too complicated entity to deal with as the fundamental variable
in a variational approach. Firstly, it cannot adequately be described without ~
1023 parameters, and secondly it has the complication of possessing a phase as
well as a magnitude. They chose instead to use the electron density as their
fundamental variable.
The advent of modern DFT methods for molecular systems has
revolutionised computational chemistry, especially for the transition metals.50
In a nutshell, DFT is based on the fact that all molecular electronic properties
can be calculated if the electron density is known. The only blemish is that the
exact form of all the components of the functional that gives the energy from
1. Introduction
39
the electron density is unknown. There are, however, a number of approximate
functionals available. These can be divided into three classes:
1. Functionals based on the local density approximation (LDA) only depend
on the value of the electron density at any given point in space.
2. Functionals that use not only the value of the electron density, but also
its gradient (how fast it changes in space) are based on the generalised
gradient approximation (GGA).
3. Hybrid density functional techniques combine GGA functionals with a
parameterised proportion of the exchange energy calculated by Hartree–Fock
(HF) theory. They are thus a cross between DFT and conventional HF ab initio
theory. However, hybrid DFT methods are the most accurate of the three
classes and, therefore, enjoy great popularity. The best-known hybrid level of
theory is Becke's three-parameter technique with the Lee–Yang–Parr
correlation functional (B3LYP).51, 52
The two major advantages of DFT calculations are that they give good to
excellent results for metal systems, including transition metals, and that they
scale relatively well with the size of the system to be calculated, so that large
complexes and clusters can be treated. However, DFT shows some serious
limitations for studying water-exchange reactions. Hydrogen-bonding and
proton-transfer reactions are generally not treated well by DFT.53 Compared
with MP2, even hybrid DFT with an adequate basis set also favors lower
coordination numbers.54
1. Introduction
40
1.11. References
1. Sapse, A.-M.; Schleyer, P. V. R.; Editors Lithium Chemistry: A
Theoretical and Experimental Overview; John Wiley & Sons, Inc.: New
York, 1995; p 595.
2. Deberitz, J.; Boche, G. Chem. unserer Zeit 2003, 37, 258.
3. Snaith, R.; Wright, D. S. In Lithium Chemistry; Sapse, A.-M.; Schleyer,
P. V. R., Eds.; John Wiley & Sons, Inc.: New York, 1995; pp 227.
4. Rosenthal, N. E.; Goodwin, F. K. Annu. Rev. Med. 1982, 33, 555.
5. Wood, A. J.; Goodwin, G. M. Psychol. Med. 1987, 17, 579.
6. Soares, J. C.; Gershon, S. Neuropsychopharmacol. 1998, 19, 167.
7. Bhagwagar, Z.; Goodwin, G. M. Clin. Neurosci. Res. 2002, 2, 222.
8. Fountoulakis, K. N.; Vieta, E.; Sanchez-Moreno, J.; Kaprinis, S. G.;
Goikolea, J. M.; Kaprinis, G. S. J. Affect. Disorders 2005, 86, 1.
9. Shaw, M. The role of lithium clinics in the treatment of bipolar disorder;
South West Yorkshire Mental Health NHS Trust: England: United
Kingdom, 2004; pp 42.
10. Salzman, C. Harvard review of psychiatry 2003, 11, 230.
11. Cowen, P. J. Br. Med. Bull. 1996, 52, 539.
12. Goodwin Guy, M.; Young Allan, H. J. Psychopharmacol. 2003, 17, 3.
13. Thase, M. E.; Sachs, G. S. Biol. Psychiat. 2000, 48, 558.
14. Fetter Jeffrey, C.; Askland Kathleen, D. Am. J. Psychiatry 2005, 162,
1546; author reply 1547.
15. Gijsman Harm, J.; Geddes John, R.; Rendell Jennifer, M.; Nolen
Willem, A.; Goodwin Guy, M. Am. J. Psychiatry 2004, 161, 1537.
16. Geddes John, R.; Burgess, S.; Hawton, K.; Jamison, K.; Goodwin Guy,
M. Am. J. Psychiatry 2004, 161, 217.
17. Spångberg, D., Ph.D. Thesis, Uppsala University, Uppsala, 2003.
18. Gutmann, V. Electrochim. Acta 1976, 21, 661.
19. Masia, M., Ph.D. Thesis, Universitat Politècnica de Catalunya,
Barcelona, 2005.
1. Introduction
41
20. Lehn, J. M. In Perspect. Coord. Chem.; Williams, A. F.; Floriani, C.;
Merbach, A. E., Eds.; VCH Verlagsgesellschaft/Verlag Helvetica Chimica
Acta: Weinheim/Basel, 1992; pp 447.
21. Dietrich, B.; Viout, P.; Lehn, J. M. Macrocyclic Chemistry: Aspects of
Organic and Inorganic Supra Molecular Chemistry; 1993; p 384.
22. Hubberstey, P. Coord. Chem. Rev. 1985, 66, 1.
23. Izatt, R. M.; Bradshaw, J. S.; Nielsen, S. A.; Lamb, J. D.; Christensen, J.
J.; Sen, D. Chem. Rev. 1985, 85, 271.
24. Hubberstey, P. Coord. Chem. Rev. 1988, 85, 1.
25. Bajaj, A. V.; Poonia, N. S. Coord. Chem. Rev. 1988, 87, 55.
26. Olsher, U.; Izatt, R. M.; Bradshaw, J. S.; Dalley, N. K. Chem. Rev. 1991,
91, 137.
27. Izatt, R. M.; Pawlak, K.; Bradshaw, J. S.; Bruening, R. L. Chem. Rev.
1991, 91, 1721.
28. Izatt, R. M.; Pawlak, K.; Bradshaw, J. S.; Bruening, R. L. Chemical
Reviews 1995, 95, 2529.
29. Bartsch, R. A.; Ramesh, V.; Bach, R. O.; Shono, T.; Kimura, K. In
Lithium Chemistry; Sapse, A.-M.; Schleyer, P. V. R., Eds.; John Wiley &
Sons, Inc.: New York, 1995; pp 393.
30. Lehn, J. M.; Sauvage, J. P. J. Am. Chem. Soc. 1975, 97, 6700.
31. Park, C. H.; Simmons, H. E. J. Am. Chem. Soc. 1968, 90, 2429.
32. Lorimer, J. W. Pure Appl. Chem. 1993, 65, 1499.
33. Kaplan, M. L.; Reitman, E. A.; Cava, R. J. Polymer 1989, 30, 504.
34. Eyring, E. M.; Petrucci, S. In Cation Binding Macrocycles; Inoue, Y,;
Gokel, G. W., Eds.; Marcel Dekker, Inc.: New York, 1990, pp 179.
35. Helm, L.; Merbach, A. E. Coord. Chem. Rev. 1999, 187, 151.
36. Helm, L.; Merbach, A. E. Chem. Rev. 2005, 105, 1923.
37. Van Eldik, R. Studies in Inorganic Chemistry 1986, 7, 1.
38. van Eldik, R.; Klarner, F.-G. High Pressure Chemistry: Synthetic,
Mechanistic, and Supercritical Applications; 2003; p 458.
39. Richens, D. T. The Chemistry of Aqua Ions; Wiley&Sons: Chichester,
1997.
1. Introduction
42
40. Radnai, T. J. Mol. Liq. 1995, 65/66, 229.
41. Strehlow, H.; Editor Rapid Reactions in Solution; VCH: Weinheim,
1992; p 341.
42. Lincoln, S. F. Prog. React. Kinet. 1977, 9, 91.
43. Zahl, A.; Igel, P.; Weller, M.; van Eldik, R. Rev. Sci. Instrum. 2004, 75,
3152.
44. Burgess, J. Metal Ions in Solution; Ellis Horwood: Chichester, 1978; p
464.
45. Cahen, Y. M.; Handy, P. R.; Roach, E. T.; Popov, A. I. J. Phys. Chem.
1975, 79, 80.
46. Bauer, W. In Lithium Chemistry; Sapse, A.-M.; Schleyer, P. V. R., Eds.;
John Wiley & Sons, Inc.: New York, 1995; pp 125.
47. Emsley, J. The Oxford Book of the Elements; Oxford University Press:
Oxford, 2001; p 448.
48. Helm, L.; Borel, A. NMRICMA 2.7, University of Lausanne, 2000.
49. Pople, J. A.; Schneider, W. G.; Bernstein, H. J. High-Resolution Nuclear
Magnetic Resonance; McGrow-Hill Book Co.: New York, 1959; p 513.
50. Koch, W.; Holthausen, M. C. A Chemist's Guide to Density Functional
Theory, 2nd Edition; John Wiley & Sons, Inc.: Chichester, UK, 2001; p
528.
51. Becke, A. D. J. Chem. Phys. 1993, 98, 5648.
52. Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B: Condens. Matter Mater.
Phys. 1988, 37, 785.
53. Pavese, M.; Chawla, S.; Lu, D.; Lobaugh, J.; Voth, G. A. J. Chem. Phys.
1997, 107, 7428.
54. Diaz, N.; Suarez, D.; Merz, K. M., Jr. Chem. Phys. Lett. 2000, 326, 288.
2. DMSO and Water/DMSO Mixtures
43
2. Ligand Exchange Processes on Solvated
Lithium Cations. DMSO and Water/DMSO
Mixtures
2.1. General remark
The work presented in this chapter was conducted by several authors within a
joint project and has been published as:
Pasgreta, E.; Puchta, R.; Galle, M.; van Eikema-Hommes N.; Zahl A.; van Eldik
R. ChemPhysChem 2007, 8, 1315-1320.
2.2. Abstract
Solutions of LiClO4 in solvent mixtures consisting of dimethylsulfoxide
(DMSO) and water, or DMSO and γ-butyrolactone, were studied by 7Li NMR
spectroscopy. Chemical shifts indicate that the Li+ ion is coordinated by four
DMSO molecules. In the binary solvent mixture of water and DMSO, no
selective solvation is detected, thus indicating that on increasing the water
content of the solvent mixture, DMSO is gradually displaced by water in the
coordination sphere of Li+. The ligand-exchange mechanism of Li+ ions
solvated by DMSO and water/DMSO mixtures was studied using DFT
calculations. Ligand exchange on [Li(DMSO)4]+ was found to follow a limiting
associative (A) mechanism. The displacement of coordinated H2O by DMSO in
[Li(H2O)4]+ follows an associative interchange mechanism. The suggested
mechanisms are discussed in reference to available experimental and
theoretical data.
2. DMSO and Water/DMSO Mixtures
44
2.3. Introduction
Lithium salts are widely used in industrial applications, synthesis, and
medicine, for example, in psychiatry.1 Therefore, detailed understanding of
fundamental reactions of solvated lithium cations, for example ligand-
exchange reactions in different solvents, is essential.
Knowledge of elementary reactions, such as solvent exchange, in different
solvents is an important prerequisite for selecting, optimizing and tailoring the
ideal solvent and solvent mixture. Whereas experimental methods investigate
reactions under real conditions in a whole ensemble of molecules, quantum-
chemical calculations can focus on a single molecule in the presence of a small
number of solvent molecules, for example, those of the first solvation shell.2-7
This complementary approach allows further mechanistic insight to be gained.
Dimethylsulfoxide (DMSO) is a common solvent in chemical laboratories,8-
11 even though industry tries to avoid it. It is widely used in laboratories due to
its ability to dissolve a wide range of chemical compounds (from ionic
inorganic salts to purely covalent compounds), because it is an aprotic but
highly polar polyfunctional molecule with both a hydrophobic and a
hydrophilic site. Moreover, it has the highest dipole moment (4.3 Debye12) and
highest dielectric constant (48 at 20 °C) of any dipolar aprotic liquid.
Furthermore, DMSO is also used as a valuable reactant for example in a
number of oxidation reactions,13, 14 like the Swern oxidation of alcohols to
aldehydes and ketones using activated DMSO,15, 16 or in a Pummerer-type
reaction17, 18 to obtain the Tröger’s bases.19
The use of DMSO in human and veterinary medicine is based on its ability
to penetrate deeply through skin and other membranes without damaging
them, while carrying other compounds deep into a biological system.20-24 On
the other hand, this effect causes the biggest toxicological concerns.25-28
DMSO possesses very strong donating abilities. This property has also been
confirmed by various studies of stability constants of alkali-metal complexes.29
It is well-known that the solvating ability of the solvent, as expressed by the
2. DMSO and Water/DMSO Mixtures
45
Guttmann donor number (DN)30 plays a fundamental role in different
complexation reactions. DMSO is a solvent of high solvating ability (DN =
29.8) and it can strongly compete with ligands such as crown ethers or
cryptands for alkali-metal ion coordination.29
A variety of theoretical and experimental studies on liquid DMSO,31-41
water/DMSO mixtures,42-47 DMSO/nonaqueous electrolyte solutions32, 48-51 and
solutions of biomolecules in DMSO52-54 have been performed in recent years to
understand the physical properties of DMSO solutions. Solutions of alkali-
metal ions, mostly Li+, in DMSO have also been investigated experimentally55-
60 and theoretically,32, 61-69 as well as in water/DMSO mixtures70-73 and in other
solvent combinations.74-76 One of the first reports on the utilization of 7Li NMR
in the investigation of LiClO4 in various solvents (including DMSO) was
published by Maciel et al..77 In DMSO, no evidence for contact ion pairing was
observed,73, 78 as might be expected, for instance, in THF (DN = 7.4).79 It was
also shown that Li+ interacts preferentially with DMSO electrostatically
through the oxygen donor of the S=O group.76
The special interest in Li+ ions in DMSO and DMSO mixtures with other
solvents originates from the use of Li+ ions in batteries with high energy
densities.80-83 Such studies on the transport properties of electrolytes in
aqueous and nonaqueous solvents are of interest in various technologies, such
as high-energy-density batteries, photo-electrochemical cells,
electrodeposition, wet electrolytic capacitors and in electro-organic
synthesis.82 Electrochemical technologies search for electrolyte solutions with
optimized properties for the special tasks which they must fulfill in devices and
processes.83 A literature survey suggests that the use of mixed solvents
involving high permittivity, low viscosity and wide liquid range is practicable
for use in high-energy-density batteries, especially lithium batteries. The
advantage of using mixed solvents over using their pure counterparts in
lithium batteries is to be able to manipulate the suitable properties of these
solvents for fulfilling the above conditions.
2. DMSO and Water/DMSO Mixtures
46
While many studies explored the behaviour of Li+ in DMSO and
water/DMSO mixtures, to our knowledge, no attempts have been made to
investigate the ligand-exchange mechanism of solvated Li+ in DMSO and
water/DMSO mixtures. Herein, we extend our combined quantum-chemical
and experimental investigations of the ligand-exchange mechanism around
Li+7 and focus on DMSO and water/DMSO mixtures.
2.4. Experimental Section
2.4.1. General Procedures
All transformations were carried out in an Ar or N2 atmosphere using
standard Schlenk techniques, in an Ar atmosphere in a drybox, or under
vacuum. Lithium perchlorate (Aldrich, battery grade) was vacuum-dried at 120
°C for 48 h and then stored under dry nitrogen. DMSO (Acros, extra dry, with
molecular sieves, water < 50 ppm) was used as received. γ-Butyrolactone
(Aldrich) was dried over anhydrous CaSO4, then fractionally distilled under
vacuum and stored under nitrogen over Linde 3Å molecular sieves. Solutions
of LiClO4 were prepared under dry nitrogen. For 7Li NMR studies, these
solutions were sealed in Ar atmosphere in 5-mm NMR tubes.
2.4.2. 7Li NMR Spectroscopy
Fourier-transform 7Li NMR spectra were recorded at 155 MHz on a Bruker
Avance DRX 400WB spectrometer equipped with a superconducting BC-
94/89 magnet system. Spinning 5-mm tubes were used with a 1-mm outer
diameter melting-point capillary inserted coaxially in the spinning tube and
filled with an external reference solution (usually 1 M LiClO4 in DMF). The
experiments were conducted at room temperature and at ambient pressure.
For a typical kinetic run, 0.8 mL of solution was transferred under an Ar to a
Wilmad screw-cup NMR tube equipped with a poly(tetrafluoroethylene)
septum. All samples were prepared in the same fashion.
2. DMSO and Water/DMSO Mixtures
47
2.4.3. Quantum-Chemical Calculations
All structures were fully optimized using the B3LYP hybrid density
functional84-86 and the LANL2DZp basis set, that is, with pseudo-potentials on
sulfur and the valence basis set augmented with polarization functions,87-91 and
characterized as minima or transition-state structures by computation of
vibrational frequencies (for minima, all frequencies are positive, NImag = 0;
for transition structures, exactly one imaginary frequency is present, NImag =
1). Structures were subsequently refined at B3LYP/6-311+G**,84-86, 92 that is,
with diffuse functions on non-hydrogen atoms, in order to minimise the basis
set superposition error (BSSE).93 Zero-point energies and contributions to
enthalpy and entropy were computed for a temperature of 298 K at
B3LYP/LANL2DZp or B3LYP/6-311+G**, as indicated. The influence of the
bulk solvent was evaluated via single-point calculations using the CPCM
(conductor-like polarizable continuum model)94, 95 formalism, that is,
B3LYP(CPCM)/6-311+G**//B3LYP/6-311+G** and water as solvent. Structure
optimizations at this level, probed for selected system, were found to yield
essentially the same results. The approximation of solvent effects by the
polarizable continuum model (PCM)96 or the isodensity molecular surface
polarizable continuum model (IPCM)97 unfortunately failed for some systems,
[see 98, 99] but where successful, they agreed with CPCM results. The Gaussian
03 suite of programs was used throughout.100
2.5. Results and Discussion
2.5.1. Coordination Number of Li+ in DMSO
To determine the coordination number of Li+ in DMSO, the cation should
be dissolved in a solvent mixture that consists of a weakly and strongly
coordinating component. The method is based on monitoring the resonance
frequency of 7Li nucleus as a function of DMSO-to-metal ion molar ratio while
keeping the metal ion concentration constant and varying the concentration of
DMSO over a wide range. This kind of study might be hampered by limited
solubilities of Li salts in solvent mixtures or limited miscibility of mixture
2. DMSO and Water/DMSO Mixtures
48
components. There are a few conditions that must be obeyed: 1) A weakly
coordinating solvent must be available, which will dissolve all solution
components. 2) A measurable parameter must exist which is a function of
solvent composition. 3) The formation constant of the solvate must be
sufficiently large that a limiting value of the parameter may be obtained. In
this case, the parameter was the Li+ NMR signal. Its position in the NMR
spectrum depends on the solvation shell of the solvated ion. Typical alkali-
metal salts are not appreciably soluble in truly non-coordinating solvents.
Therefore, it was necessary to select a solvent with a weak interaction toward
DMSO and the alkali-metal ions.101
γ-Butyrolactone was found to represent a good compromise for the various
requirements. Contrary to DMSO, it is a rather weak donor (DN = 18.0)102 that
appreciably dissolves alkali-metal salts, and its interaction with DMSO is much
weaker than that between DMSO and a metal cation. The molar ratio data
confirmed the “inertness” of the solvent and indicated that it is in fact
sufficiently unreactive for the purpose of this investigation.
Figure 2.1. 7Li NMR chemical-shift variation as a function of the DMSO/LiClO4 molar
ratio in γ-butyrolactone solution.
2. DMSO and Water/DMSO Mixtures
49
Figure 2.1 shows the molar-ratio plot of the DMSO/LiClO4 system in γ-
butyrolactone solution, where the LiClO4 concentration was kept constant at
0.05 M. A clear break can be observed at the DMSO/Li+ ratio of 4:1, indicating
a solvation number of 4 for Li+ in DMSO.
2.5.2. Selective Solvation of Li+
Selective solvation of Li+ in H2O/DMSO mixture was studied by Emons et
al.,73 using Raman spectroscopy. Their results suggest that Li+-ion
coordination by DMSO is stronger than by water. On the other hand, the
Gutmann donor number for water (3330) is higher than for DMSO (29.8),
which would suggest the opposite.
We performed 7Li NMR measurements keeping the concentration of LiClO4
constant (0.1 M) and varying the composition of the mixture of H2O/DMSO by
increasing the percentage of H2O from 0 to 100 % (molar ratio). For each
sample, the 7Li NMR spectrum was recorded. The chemical shift varied
gradually from 4.35 ppm (DMSO) to 5.07 ppm (water; see Figure 2.2), and the
plot of chemical shift versus solvent composition (Figure 2.3) shows no clear
inflection like that observed for the DMSO/γ-butyrolactone system in Figure
2.1. Thus, the composition of the first coordination sphere around Li+ in
mixtures of H2O and DMSO depends only on the ratio of these solvents. DMSO
and water are equally strong ligands in the coordination of Li+.
2. DMSO and Water/DMSO Mixtures
50
Figure 2.2. 7Li NMR spectra of LiClO4 in H2O/DMSO for different mixture compositions.
Figure 2.3. Chemical-shift variation as a function of water content in LiClO4 solutions of
H2O/DMSO mixtures.
2. DMSO and Water/DMSO Mixtures
51
Model calculations (B3LYP(CPCM)/6-311+G**//B3LYP/6-311+G**)
corroborate these results (Table 2.1). The exchange reactions in the stepwise
conversion of [Li(H2O)4]+ + 4 DMSO to [Li(DMSO)4]+ + 4 water are computed
to be essentially thermoneutral. Only for the exchange of the last water
molecule by DMSO is the computed reaction enthalpy somewhat larger (–3.4
kcal/mol).
Table 2.1. Calculated energies for the stepwise conversion of [Li(H2O)4]+ to [Li(DMSO)4]+.[a]
ΔEDFT [b] ΔHCPCM [c] ΔGCPCM [c]
1 [Li(H2O)4]+ + 4 DMSO 0.0 0.0 0.0
2 [Li(DMSO)(H2O)3]+ + H2O + 3 DMSO –13.3 –0.5 2.6
3 [Li(DMSO)2(H2O)2]+ + 2 H2O + 2 DMSO –24.6 0.8 3.6
4 [Li(DMSO)3(H2O)]+ + 3 H2O + DMSO –33.4 –0.8 6.0
5 [Li(DMSO)4]+ + 4 H2O –42.6 –4.2 4.7
[a] Energies in kcal/mol relative to [Li(H2O)4]+ and 4 DMSO; [b] B3LYP/6-311+G**, including ∆ZPE computed at B3LYP/6-311+G**; [c] B3LYP(CPCM)/6 311+G**//B3LYP/6-311+G** with thermodynamic contributions calculated at B3LYP/6-311+G**
The Gibbs free energy values might be interpreted as predicting the
exchange reactions to be endergonic. However, due to inherent inaccuracies,
the error interval of the computed entropy contributions is too large to draw
any conclusions here. Note also the importance of including the effect of the
bulk solvent: in the gas-phase calculations, all solvent-exchange steps are
significantly exothermic.
Calculations at the B3LYP(IPCM)/6-311+G**//B3LYP/6-311+G** level,
that is, employing the isodensity surface polarizable continuum model, were
successful only for [Li(H2O)4]+ and [Li(DMSO)4]+ (Table 2.1, entries 1 and
5). However, the computed reaction enthalpy for the complete solvent
exchange, –5.3 kcal/mol, agrees well with the CPCM result.
2. DMSO and Water/DMSO Mixtures
52
2.5.3. Ligand Exchange on Solvated Li+
Exchange of a solvent molecule between the first and the second
coordination shell is the simplest reaction that can occur on a metal ion in
solution. This reaction is of fundamental importance in understanding not
only the reactivity of the metal ion in chemical and biological systems but also
the interaction between the metal ion and the solvent molecule. The
displacement of a solvent molecule from the first coordination shell presents
an important step in complex-formation reactions of metal cations and in
many redox processes.103 Mechanistically, associative (A), interchange (I) and
dissociative (D) pathways are conceivable. Solvent exchange on Li+ in water is
extremely fast, making it difficult to study directly. An exchange rate constant
of approximately 109 s-1 at 25 °C was predicted on the basis of complex-
formation data.104 As shown above, DMSO coordinates to Li+ with a similar
strength as water. Its exchange process is also very fast and therefore
impossible to be studied experimentally. All alkali-metal ions and the large
alkaline-earth-metal ions are very labile due to the low surface charge density
and the absence of ligand field stabilization effects.104
2.5.4. Computed Mechanisms of Solvent Exchange
Experimental (vide supra) and previous theoretical62 studies agree with the
present study that the first coordination shell around Li+ consists of four
tightly bound DMSO molecules. The computed Li-O distance is 1.96 Å. An
additional DMSO molecule does not coordinate to the lithium cation but
instead is only weakly bound, mainly by electrostatic interactions. The Li…O
distance is well over 5 Å (see Figure 2.4), and the computed gas-phase binding
energy for the fifth DMSO molecule in [Li(DMSO)4(DMSO)]+ is 7.6 kcal/mol.
When the influence of the bulk solvent is taken into account, at
B3LYP(CPCM)/6-311+G**//B3LYP/6-311+G**, this binding energy
(computed for the gas-phase structure) vanishes. Optimization of the
precursor complex at B3LYP(CPCM)/6-311+G** resulted in only minimal
changes in structure and energy.
2. DMSO and Water/DMSO Mixtures
53
Figure 2.4. Reaction profile for DMSO exchange on solvated Li+; enthalpy: B3LYP(CPCM)/6-311+G** including thermodynamic contributions at B3LYP/LANL2DZp; gas-phase energy values: B3LYP/6-311+G** including ΔZPE at B3LYP/LANL2DZp, in parentheses. T.S. = transition state.
In a previous study, we found a limiting associative (A) mechanism for
water exchange around Li+.7 Mechanistic details are very similar for DMSO
exchange. The reaction proceeds through a distorted trigonal-bipyramidal
reactive intermediate [Li(DMSO)5]+ that is reached via a late transition state.
The enthalpy profile (Figure 2.4, B3LYP(CPCM)/6-311+G**, thermodynamic
contributions computed at B3LYP/LANL2DZp) is in line with the experimental
observation (vide supra) of a very fast exchange process: the five-coordinate
intermediate is computed to be 7.9 kcal/mol less stable than [Li(DMSO)4]+ and
free DMSO, while an overall activation barrier of only 8.4 kcal/mol is
computed. Obviously, the intermediate will be extremely short-lived, since its
barrier toward loss of DMSO is only 0.5 kcal/mol. Note that with the obvious
exception of the binding energy for the fifth DMSO molecule in the precursor
complex (vide supra), the influence of the bulk solvent on the energy profile is
minor.
2. DMSO and Water/DMSO Mixtures
54
One might argue that this energy difference is smaller than the accuracy of
the computational method and thus meaningless. However, whereas the value
of the (small) barrier may indeed vary, the shape of the potential-energy
surface usually does not (see also ref.7). In the present case, zero-point energy
and thermodynamic corrections do not change the energetic order of the
structures. Inclusion of entropy contributions (we believe these to be
sufficiently accurate, in view of the similarity of the structures calculated for
intermediate and transition states) even stabilises the intermediate: a free
energy barrier of 1.8 kcal/mol is computed. We therefore consider the
mechanistic conclusion of associative (A) exchange to be reliable.
We also investigated in detail the mechanism of the exchange of a water
ligand by DMSO. The enthalpy profile, calculated at B3LYP(CPCM)/6-311+G**
with thermodynamic contributions computed at B3LYP/6-311+G**, is shown
in Figure 2.5. B3LYP/6-311+G** gas-phase energy values, corrected for ∆ZPE
at this level, are included for comparison.
Figure 2.5. Reaction profile for the water/DMSO exchange around [Li(H2O)4]+; enthalpy: B3LYP(CPCM)/6-311+G** including thermodynamic contributions at B3LYP/6-311+G**; gas-phase energy values: B3LYP/6-311+G** including ΔZPE, in parentheses. T.S. = transition state.
2. DMSO and Water/DMSO Mixtures
55
The reaction involves a precursor complex, in which the incoming DMSO
molecule is weakly bound to two water ligands through hydrogen bonds. As in
the case of water exchange7 and DMSO exchange (vide supra), the gas-phase
binding energy for this complex is exaggerated. Otherwise, the influence of the
bulk solvent on the energy profile is minor.
The two H2O ligands with hydrogen bonds to the DMSO oxygen move to
axial positions while DMSO approaches the lithium cation (Figure 2.5). In the
distorted trigonal-bipyramidal transition structure, one hydrogen bond to the
leaving water molecule remains. Note that the reaction proceeds with cis
stereochemistry. The activation barrier is 6.9 kcal/mol, relative to the
precursor complex, or 3.3 kcal/mol relative to separated [Li(H2O)4]+ and
DMSO. The computed normal mode vector for the imaginary frequency clearly
corresponds to a ligand substitution: the Li–Owater bond length increases upon
shortening of the Li–ODMSO bond (2.29 Å; 0.4 Å longer than in the product).
The Li–Owater bond is elongated by only 0.16 Å in the transition structure. We
conclude that the reaction will be very fast (cf. ref.105) and proceeds through an
associative interchange mechanism. No stable five-coordinate intermediate
structure could be found, despite extensive searches.
The reaction proceeds to a product complex in which the leaving water
molecule has hydrogen bonds to two water ligands and to DMSO; this
arrangement is 3.6 kcal/mol less stable than the precursor complex. Breaking
the bond to DMSO leads to a second, somewhat more stable complex, from
which the water molecule is released.
2.6. Conclusions
It was confirmed both experimentally and theoretically that Li+ has a
coordination number of four in pure DMSO and in water/DMSO mixtures. In
the binary mixture of water and DMSO, both solvents coordinate equally
strongly by the Li+ ion. Displacement of DMSO by water in the coordination
sphere of solvated Li+ occurs gradually, indicating that the two compounds
have similar donor strengths. In the solvent mixture consisting of DMSO and
2. DMSO and Water/DMSO Mixtures
56
γ-butyrolactone, the latter component is weakly coordinating and can be
considered as a diluent for DMSO and alkali-metal ions. Computational
studies suggest that ligand exchange on [Li(DMSO)4]+ follows a limiting
associative mechanism, whereas displacement of coordinated water by DMSO
seems to follow an associative interchange mechanism.
2. DMSO and Water/DMSO Mixtures
57
2.7. References
1. Birch, N. J. Chem. Rev. 1999, 99, 2659.
2. Erras-Hanauer, H.; Clark, T.; van Eldik, R. Coord. Chem. Rev. 2003,
238-239, 233.
3. Hartmann, M.; Clark, T.; van Eldik, R. J. Mol. Model. 1996, 2, 354.
4. Hartmann, M.; Clark, T.; van Eldik, R. J. Am. Chem. Soc. 1997, 119,
7843.
5. Hartmann, M.; Clark, T.; Van Eldik, R. J. Phys. Chem. A 1999, 103,
9899.
6. Puchta, R.; van Eikema Hommes, N.; van Eldik, R. Helv. Chim. Acta
2005, 88, 911.
7. Puchta, R.; Galle, M.; Van Eikema Hommes, N.; Pasgreta, E.; Van Eldik,
R. Inorg. Chem. 2004, 43, 8227.
8. Schlafer, H. L.; Schaffernicht, W. Angew. Chem. 1960, 72, 618.
9. Martin, D.; Weise, A.; Niclas, H. J. Angew. Chem., Int. Ed. Engl. 1967,
6, 318.
10. Mandal, P. K.; Dutta, P.; Roy, S. C. Tetrahedron Lett. 1997, 38, 7271.
11. Kiselev, V. D.; Kashaeva, E. A.; Luzanova, N. A.; Konovalov, A. I. Russ. J.
Gen. Chem. 1997, 67, 1361.
12. Higashigaki, Y.; Christensen, D. H.; Wang, C. H. J. Phys. Chem. 1981,
85, 2531.
13. Sweat, F. W.; Epstein, W. W. J. Org. Chem. 1967, 32, 835.
14. Epstein, W. W.; Sweat, F. W. Chem. Rev. 1967, 67, 247.
15. Tidwell, T. T. Synthesis 1990, 10, 857.
16. Mancuso, A. J.; Swern, D. Synthesis 1981, 3, 165.
17. Pummerer, R. Ber. Dtsch. Chem. Ges. 1909, 42, 2282.
18. Laleu, B.; Machado, M. S.; Lacour, J. Chem. Commun. 2006, 26, 2786.
19. Li, Z.; Xu, X.; Peng, Y.; Jiang, Z.; Ding, C.; Qian, X. Synthesis 2005, 8,
1228.
20. Shearer, A. G.; Hampton, R. Y. EMBO J. 2005, 24, 149.
2. DMSO and Water/DMSO Mixtures
58
21. Parry, J. M.; Fowler, P.; Quick, E.; Parry, E. M. Cytogenet. Genome Res.
2004, 104, 283.
22. Scholl, P. F.; McCoy, L.; Kensler, T. W.; Groopman, J. D. Chem. Res.
Toxicol. 2006, 19, 44.
23. Coluccia, M.; Sava, G.; Loseto, F.; Nassi, A.; Boccarelli, A.; Giordano, D.;
Alessio, E.; Mestroni, G. Eur. J. Cancer 1993, 29A, 1873.
24. Bardouki, H.; Barcellos da Rosa, M.; Mihalopoulos, N.; Palm, W. U.;
Zetzsch, C. Atmos. Environ. 2002, 36, 4627.
25. Haschek, W. M.; Baer, K. E.; Rutherford, J. E. Toxicology 1989, 54, 197.
26. Wood, D. C.; Weber, F. S.; Palmquist, M. A. J. Pharmacol. Exp. Ther.
1971, 177, 520.
27. Noel, P. R.; Barnett, K. C.; Davies, R. E.; Jolly, D. W.; Leahy, J. S.;
Mawdesley-Thomas, L. E.; Shillam, K. W.; Squires, P. F.; Street, A. E.;
Tucker, W. C.; Worden, A. N. Toxicology 1975, 3, 143.
28. Jacob, S. W.; Wood, D. C. Am. J. Surg. 1967, 114, 414.
29. Karkhaneei, E.; Afkhami, A.; Shamsipur, M. Polyhedron 1996, 15, 1989.
30. Gutmann, V. Electrochim. Acta 1976, 21, 661.
31. Luzar, A.; Soper, A. K.; Chandler, D. J. Chem. Phys. 1993, 99, 6836.
32. Rao, B. G.; Singh, U. C. J. Am. Chem. Soc. 1990, 112, 3803.
33. Skaf, M. S. J. Chem. Phys. 1997, 107, 7996.
34. Skaf, M. S. Mol. Phys. 1997, 90, 25.
35. Liu, H.; Mueller-Plathe, F.; van Gunsteren, W. F. J. Am. Chem. Soc.
1995, 117, 4363.
36. Thomas, R.; Shoemaker, C. B.; Eriks, K. Acta Cryst. 1966, 21, 12.
37. Itoh, S.; Ohtaki, H. Z. Naturforsch., A: Phys. Sci. 1987, 42, 858.
38. Onthong, U.; Bako, I.; Radnai, T.; Hermansson, K.; Probst, M. Int. J.
Mass Spectrom. 2003, 223-224, 263.
39. Onthong, U.; Megyes, T.; Bako, I.; Radnai, T.; Grosz, T.; Hermansson,
K.; Probst, M. Phys. Chem. Chem. Phys. 2004, 6, 2136.
40. Gabrielian, L. S.; Markarian, S. A. J. Mol. Liq. 2004, 112, 137.
41. Markarian, S. A.; Gabrielian, L. S.; Bonora, S.; Fagnano, C. Spectrochim.
Acta, Part A 2003, 59, 575.
2. DMSO and Water/DMSO Mixtures
59
42. Vaisman, I. I.; Berkowitz, M. L. J. Am. Chem. Soc. 1992, 114, 7889.
43. Kirchner, B.; Reiher, M. J. Am. Chem. Soc. 2002, 124, 6206.
44. Luzar, A.; Chandler, D. J. Chem. Phys. 1993, 98, 8160.
45. Laria, D.; Skaf, M. S. J. Chem. Phys. 1999, 111, 300.
46. Skaf, M. S. J. Phys. Chem. A 1999, 103, 10719.
47. Borin, I. A.; Skaf, M. S. J. Chem. Phys. 1999, 110, 6412.
48. Vechi, S. M.; Skaf, M. S. J. Chem. Phys. 2005, 123, 154507/1.
49. Das, A. K.; Madhusoodanan, M.; Tembe, B. L. J. Phys. Chem. A 1997,
101, 2862.
50. Madhusoodanan, M.; Tembe, B. L. J. Phys. Chem. 1995, 99, 44.
51. Madhusoodanan, M.; Tembe, B. L. J. Phys. Chem. 1994, 98, 7090.
52. Mierke, D. F.; Kessler, H. J. Am. Chem. Soc. 1991, 113, 9466.
53. Zheng, Y.-J.; Ornstein, R. L. J. Am. Chem. Soc. 1996, 118, 4175.
54. Markarian, S. A.; Gabrielyan, L. S.; Grigoryan, K. R. J. Solution Chem.
2004, 33, 1005.
55. Markarian, S. A.; Gabrielian, L. S.; Zatikyan, A. L.; Bonora, S.;
Trinchero, A. Vib. Spectrosc. 2005, 39, 220.
56. Gabrielyan, L. S.; Markaryan, S. A. Zh. Fiz. Khim. 2001, 75, 1792.
57. Alia, J. M.; Edwards, H. G. M. Vib. Spectrosc. 2000, 24, 185.
58. Markarian, S. A.; Stockhausen, M. Z. Phys. Chem. 2000, 214, 139.
59. Wang, Z.; Huang, B.; Wang, S.; Xue, R.; Huang, X.; Chen, L.
Electrochim. Acta 1997, 42, 2611.
60. Kloss, A. A.; Ronald Fawcett, W. J. Chem. Soc., Faraday Trans. 1998,
94, 1587.
61. Das, A. K.; Tembe, B. L. J. Chem. Phys. 1998, 108, 2930.
62. Onthong, U.; Megyes, T.; Bako, I.; Radnai, T.; Hermansson, K.; Probst,
M. Chem. Phys. Lett. 2005, 401, 217.
63. Megyes, T.; Bako, I.; Radnai, T.; Grosz, T.; Kosztolanyi, T.; Mroz, B.;
Probst, M. Chem. Phys. 2006, 321, 100.
64. Burk, P.; Koppel, I. A.; Koppel, I.; Kurg, R.; Gal, J.-F.; Maria, P.-C.;
Herreros, M.; Notario, R.; Abboud, J.-L. M.; Anvia, F.; Taft, R. W. J.
Phys. Chem. A 2000, 104, 2824.
2. DMSO and Water/DMSO Mixtures
60
65. Westphal, E.; Pliego Josefredo, R., Jr. J. Chem. Phys. 2005, 123,
074508.
66. Goldenberg, M. S.; Kruus, P.; Luk, S. K. F. Can. J. Chem. 1975, 53, 1007.
67. Kruus, P.; Poppe, B. E. Can. J. Chem. 1979, 57, 538.
68. Frolov, Y. L.; Guchik, I. V.; Shagun, V. A.; Vaschenko, A. V.; Trofimov, B.
A. J. Struct. Chem. 2003, 44, 927.
69. Kalugin, O. N.; Adya, A. K.; Volobuev, M. N.; Kolesnik, Y. V. Phys. Chem.
Chem. Phys. 2003, 5, 1536.
70. Das, A. K.; Tembe, B. L. J. Chem. Phys. 1999, 111, 7526.
71. Zamir, T.; Khan, A. J. Chem. Soc. Pak. 2005, 27, 130.
72. Kriz, J.; Abbrent, S.; Dybal, J.; Kurkova, D.; Lindgren, J.; Tegenfeldt, J.;
Wendsjo, A. J. Phys. Chem. A 1999, 103, 8505.
73. Emons, H. H.; Birkeneder, F.; Pollmer, K. Z. Anorg. Allg. Chem. 1986,
534, 175.
74. Alia, J. M.; Edwards, H. G. M. Vib. Spectrosc. 2004, 34, 225.
75. Ali, A.; Nain, A. K. Phys. Chem. Liq. 1997, 34, 25.
76. Chauhan, M. S.; Sharma, K. C.; Kumar, G.; Chauhan, S. Collect. Czech.
Chem. Commun. 2002, 67, 1141.
77. Maciel, G. E.; Hancock, J. K.; Lafferty, L. F.; Mueller, P. A.; Musker, W.
K. Inorg. Chem. 1966, 5, 554.
78. Cahen, Y. M.; Handy, P. R.; Roach, E. T.; Popov, A. I. J. Phys. Chem.
1975, 79, 80.
79. Wuepper, J. L.; Popov, A. I. J. Am. Chem. Soc. 1970, 92, 1493.
80. Mandal, B.; Sooksimuang, T.; Griffin, B.; Padhi, A.; Filler, R. Solid State
Ionics 2004, 175, 267.
81. Abe, T.; Kawabata, N.; Mizutani, Y.; Inaba, M.; Ogumi, Z. J.
Electrochem. Soc. 2003, 150, A257.
82. Kumar, G.; Janakiraman; Namboodiri, N.; Gangadharan; Saminathan;
Kanagaraj; Sharief, N. J. Chem. Eng. Data 1991, 36, 467.
83. Barthel, J.; Gores, H. J.; Neueder, R.; Schmid, A. Pure Appl. Chem.
1999, 71, 1705.
84. Becke, A. D. J. Chem. Phys. 1993, 98, 5648.
2. DMSO and Water/DMSO Mixtures
61
85. Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B: Condens. Matter Mater.
Phys. 1988, 37, 785.
86. Stephens, P. J.; Devlin, F. J.; Chabalowski, C. F.; Frisch, M. J. J. Phys.
Chem. 1994, 98, 11623.
87. Dunning, T. H.; Jr.; Hay, P. J. Modern Theoretical Chemistry; Plenum:
New York, 1976; Vol. 3, p 1.
88. Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 270.
89. Wadt, W. R.; Hay, P. J. J. Chem. Phys. 1985, 82, 284.
90. Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 299.
91. Huzinaga, S.; Andzelm, J.; Klobukowski, M.; Radzio-Andzelm, E.; Sakai,
Y.; Tatewaki, H. Gaussian basis sets for molecular calculations;
Elsevier: Amsterdam, 1984.
92. Koch, W.; Holthausen, M. C. A Chemist's Guide to Density Functional
Theory, 2nd Edition; John Wiley & Sons, Inc.: Chichester, UK, 2001; p
528.
93. Davidson, E. R.; Feller, D. Chem. Rev. 1986, 86, 681.
94. Barone, V.; Cossi, M. J. Phys. Chem. A 1998, 102, 1995.
95. Cossi, M.; Rega, N.; Scalmani, G.; Barone, V. J. Comput. Chem. 2003,
24, 669.
96. Cammi, R.; Mennucci, B.; Tomasi, J. J. Phys. Chem. A 2000, 104, 5631.
97. Foresman, J. B.; Keith, T. A.; Wiberg, K. B.; Snoonian, J.; Frisch, M. J. J.
Phys. Chem. 1996, 100, 16098.
98. Illner, P.; Zahl, A.; Puchta, R.; van Eikema Hommes, N.; Wasserscheid,
P.; van Eldik, R. J. Organomet. Chem. 2005, 690, 3567.
99. Puchta, R.; Clark, T.; Bauer, W. J. Mol. Model. 2006, 12, 739.
100. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.;
Cheeseman, J. R.; Montgomery, J. A.; Jr.; Vreven, T.; Kudin, K. N.;
Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.;
Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.;
Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa,
J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.;
Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Bakken, V.; Adamo, C.;
2. DMSO and Water/DMSO Mixtures
62
Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.;
Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.;
Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich,
S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.;
Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.;
Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.;
Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham,
M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.;
Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian
03, Revision B.03, Gaussian Inc.; Wallingford, CT, 2004.
101. Maxey, B. W.; Popov, A. I. J. Am. Chem. Soc. 1968, 90, 4470.
102. Marcus, Y. Chem. Soc. Rev. 1993, 22, 409.
103. Richens, D. T. The Chemistry of Aqua Ions; Wiley&Sons: Chichester,
1997.
104. Helm, L.; Merbach, A. E. Chem. Rev. 2005, 105, 1923.
105. Dunand, F. A.; Helm, L.; Merbach, A. E. Adv. Inorg. Chem. 2003, 54, 1.
3. Acetonitrile and Hydrogen Cyanide
63
3. Ligand Exchange Processes on Solvated
Lithium Cations. Acetonitrile and Hydrogen
Cyanide
3.1. General remark
The work presented in this chapter was conducted by several authors within a
joint project and has been published as:
Pasgreta, E.; Puchta, R.; Zahl, A.; van Eldik, R. European Journal of Inorganic
Chemistry 2007, 1815-1822.
3.2. Abstract
Solutions of LiClO4 in solvent mixtures of acetonitrile and water, or
acetonitrile and nitromethane, were studied by 7Li NMR spectroscopy.
Measured chemical shifts indicate that the Li+ cation is coordinated by four
acetonitrile molecules. In the binary water/acetonitrile mixture, water
coordinates more strongly to Li+ than acetonitrile such that addition of water
immediately leads to the formation of [Li(H2O)4]+. The solvent-exchange
mechanism of [Li(L)4]+ (L = CH3CN and HCN) was studied using DFT
calculations (RB3LYP/6-311+G**). This process was found to follow a limiting
associative mechanism involving the formation of relatively stable five-
coordinate intermediates. The suggested mechanisms are discussed with
reference to available experimental and theoretical data.
3. Acetonitrile and Hydrogen Cyanide
64
3.3. Introduction
The ongoing discussion on sustainability in chemistry is partly focused on
the role of solvents and especially on alternatives for common organic solvents.
Such alternatives could be, for example, ionic liquids1-4 or supercritical carbon
dioxide.5-8 A detailed knowledge of the properties and structural motifs, as well
as of elementary reactions such as solvent-exchange processes in different
solvents, is an important prerequisite for selecting, optimizing and tailoring
the ideal solvent.
Acetonitrile is a common solvent in chemical laboratories, for example it is
used in reactions9, 10 or in the investigation of complexation phenomena and
complex-formation constants.11-25 Use of its hydrogen derivative HCN is mostly
prevented by its high toxicity and its boiling point of 25 °C. Experimental
studies dating back to the middle of the 20th century show that HCN is a
water-like solvent that is well suited for dissolving organic and inorganic
compounds. Salts such as LiCl, LiBr or LiClO4 are highly soluble in hydrogen
cyanide, most likely because of its high dielectric constant of 123 (at 15.6 °C).26-
30 However, HCN serves as an excellent simplified model of CH3CN and thus
can be used instead in quantum chemical studies of CH3CN.
The behaviour of pure acetonitrile and acetonitrile mixtures was studied
experimentally31, 32 and computationally.31, 33 Since an ion in solution strongly
disturbs the local solvent structure, a detailed knowledge of the environment is
an essential prerequisite for understanding reactions in the neighbourhood of
the ion. Several studies investigated solvated Li+ ions in acetonitrile both
experimentally with various spectroscopic techniques34-44 and computationally
with different classical molecular dynamics (MD) methods45-52 and Monte
Carlo simulations on solvated alkali and halide ions in CH3CN.53 Over the past
decade, DFT and ab initio calculations on [Li(CH3CN)n]+ and related clusters
were also performed by different groups.37, 54-57
There has been a special interest in the properties of Li salts, such as
conductivity,58-64 transport ability,65-67 viscosity,68 and osmotic behaviour,69 in
3. Acetonitrile and Hydrogen Cyanide
65
solvents and solvent mixtures because of the use of Li+ ions in batteries of high
energy density.70 Such studies are not restricted to liquid systems. Li+ mobility
in potential solid electrolytes was also investigated intensively both
experimentally and theoretically.71-73
To the best of our knowledge, however, investigations of solvent-exchange
reactions in acetonitrile and acetonitrile/water mixtures or in HCN and
HCN/water mixtures have not been performed. Whereas experimental
investigations study reactions under real conditions in a whole ensemble of
molecules,74, 75 quantum chemical calculations focus on one molecule and a
small number of solvent molecules, for example, the first solvent shell.76-82 This
approach enables a more detailed mechanistic insight.
In this contribution, we extend our combined quantum chemical and
experimental investigations on ligand coordination and ligand exchange on Li+
in acetonitrile and HCN, as well as in CH3CN/H2O and HCN/H2O mixtures.
The calculations mainly focus on HCN for two reasons: (1) to gain a deeper
insight into this solvent, which is not used often, and (2) HCN is a good
working model for CH3-CN as its use prevents well known problems associated
with rotating CH3 groups and to facilitate studies with solvent models.
3.4. Experimental Section
3.4.1. General Procedures
All transformations were carried out under an Ar or N2 atmosphere by
using standard Schlenk techniques, under Ar atmosphere in a glove box, or
under vacuum. LiClO4 (Aldrich, battery grade) was vacuum dried at 120 °C for
48 h and was then stored under dry nitrogen. Acetonitrile (Acros, extra dry,
water < 10 ppm) was used as received. Nitromethane (Roth) was purified and
dried by a previously reported method.83 Solutions of LiClO4 were prepared
under dry nitrogen. For 7Li NMR spectroscopic studies, these solutions were
sealed under an Ar atmosphere in 5-mm NMR tubes.
3. Acetonitrile and Hydrogen Cyanide
66
3.4.2. 7Li NMR Spectroscopy
Fourier transform 7Li NMR spectra were recorded at a frequency of 155
MHz on a Bruker Avance DRX 400WB spectrometer equipped with a
superconducting BC-94/89 magnet system. Spinning 5-mm tubes were used
with a 1-mm o.d. melting point capillary inserted coaxially in the spinning tube
and filled with an external reference solution (usually 1 M LiClO4 in DMF). The
experiments were conducted at room temperature and at ambient pressure.
For a typical kinetic run, 0.8 mL of solution was transferred under Ar to a
Wilmad screw cup NMR tube equipped with a poly(tetrafluoroethylene)
septum. All samples were prepared in the same way.
3.4.3. Quantum-Chemical Calculations
All structures were fully optimized at the B3LYP/6-311+G**84-87 level of
theory and characterized as minima or transition-state structures by
computation of vibrational frequencies (for minima, all frequencies are
positive, NIMAG = 0; for transition-state structures, exactly one imaginary
frequency is present, NIMAG = 1). Single-point energy calculations were
performed at the MP2(full)/6-311+G**//B3LYP/6-311+G** level;88 the
influence of th bulk solvent was probed by using the IPCM89 and CPCM90,91
formalism and water or acetonitrile as solvent, i.e. B3LYP(IPCM)/6-
311+G**//B3LYP/6-311+G** and B3LYP(CPCM)/6-311+G**//B3LYP/6-
311+G**.92 The Gaussian 03 suite of programs was used throughout.93
3.5. Results and Discussion
3.5.1. Coordination Number of Li+
The structures of Li+-acetonitrile solvates are in general less well known
than the structures of Li+-water solvates. The Li+ ion was found to be four
coordinate by NMR spectroscopic studies in which acetonitrile was gradually
replaced by water when water was added to an acetonitrile solution of LiClO4
(1.6 M),94 and on the basis of IR spectroscopic intensities measured for the
acetonitrile CN stretching vibrations.95, 96 Even compounds with mixed
3. Acetonitrile and Hydrogen Cyanide
67
coordination of a counterion and acetonitrile were reported to be four-
coordinate, namely [Li(CH3CN)3Br] formed from 0.58 M LiBr in CH3CN.42
Extensive concentration-dependent ion-pairing was also found for weak
coordinating counterions, such as in LiBF437 and LiClO4,95 by studying the CN
vibration.97 These findings for the coordination number are in good agreement
with published X-ray structures.98-100
However, Sajeevkumar and Singh found the average coordination number
to be 4.7,101 whereas Megyes et al. found up to six acetonitrile molecules
surrounding the Li+ ion in the gas phase by means of mass spectrometry.102
Therefore, we reinvestigated the maximum coordination number of
[Li(CH3CN)n]+ both experimentally and computationally, and purely quantum
chemically for [Li(NCH)n]+.
In order to determine the coordination number of Li+ by using NMR
techniques, the concentration of LiClO4 was kept constant and the CH3CN
concentration was varied over a wide range. To use this method, the cation
must be dissolved in solvent mixtures consisting of one coordinating and one
non-coordinating component. The measured chemical shifts are then plotted
against the mole ratio of cation to coordinating solvent. When such a plot
shows a marked discontinuity, the appropriate solvent/cation ratio can be
taken as the coordination number of the cation. There are, however, a few
conditions that must be obeyed. An inert non-coordinating solvent must be
available that will dissolve all solution components, a measurable parameter
must exist that is a function of the solvent composition, and the formation
constant of the solvate must be sufficiently large so that a limiting value of the
parameter can be reached. In this case, the selected parameter was the Li+
NMR signal; its position in the NMR spectrum depends on the composition of
the coordination sphere of the studied ion. Typical alkali metal salts are not
appreciably soluble in truly inert solvents. Therefore, it is necessary to select a
solvent that minimally interacts with CH3CN and the alkali metal ion.103
Nitromethane (Gutmann donor number DN = 2.7)104 represents a good
compromise for the various requirements. It appreciably dissolves alkali metal
3. Acetonitrile and Hydrogen Cyanide
68
salts and its interaction with Li+ is much weaker than that between CH3CN
(Gutmann donor number DN = 14.1) and a metal cation. The mole ratio data
obtained confirms the “inertness” of this solvent and indicates that it is in fact
sufficiently unreactive for the purpose of this study.
Figure 3.1 and Figure 3.2 illustrate the results obtained from the 7Li NMR
spectroscopic study. Figure 3.2 shows a plot of the chemical shift as a function
of the mole ratio of the CH3CN/LiClO4 system; the LiClO4 concentration was
kept at 0.05 M in nitromethane. A clear break can be observed at a
[CH3CN]/[Li+] ratio of 4:1, which indicates that Li+ is coordinated by 4
acetonitrile molecules under these conditions. It appears, therefore, that NMR
studies of alkali metal ion coordination can be carried out even in solvents that
are not totally inert toward the reacting metal ion provided that the interaction
between the metal ion and the stronger donor solvent dominates.
Figure 3.1. 7Li NMR spectra recorded as a function of CH3CN/LiClO4 mole ratio (0.05 M LiClO4 in the CH3CN/nitromethane mixture) at 25 ºC.
6.06.26.46.66.87.07.27.47.67.88.08.2(ppm)
6.6
5.7
4.7
3.8
3.3
2.4
1.9
1
8.5
mole ratio CH3CN/LiClO4
ref. 0.1 M LiClO4 in DMF
06.06.26.46.66.87.07.27.47.67.88.08.2
(ppm)6.06.26.46.66.87.07.27.47.67.88.08.2
(ppm)
6.6
5.7
4.7
3.8
3.3
2.4
1.9
1
8.5
mole ratio CH3CN/LiClO4
ref. 0.1 M LiClO4 in DMF
0
3. Acetonitrile and Hydrogen Cyanide
69
0 2 4 6 8 10
6.95
7.00
7.05
7.10
7.15
7.20
7.25
chem
ical
shi
ft, p
pm
mole ratio, CH3CN/LiClO4
Figure 3.2. 7Li NMR shift measured for 0.05 M LiClO4 in CH3CN/nitromethane solvent
mixtures.
3.5.2. Selective Coordination of Li+
Selective coordination occurs when the solvate prefers one of the solvent
components over the other. The exchange of ions between different
environments is usually rapid with respect to the NMR time scale, and this
results in only one resonance signal at an average frequency determined by the
magnetic shielding and lifetime of the nucleus in each of the sites. Variation of
parameters such as concentration, counterions and solvent produces changes
in the relative proportion and type of environment, which may be reflected by
the NMR spectra in terms of chemical shift, line shape, and/or line width of
the observed resonance.105
We performed the measurements by keeping the concentration of LiClO4
constant (0.1 M) and by changing the composition of the H2O/CH3CN
mixtures. The concentration of H2O (molar ratio) was increased from 0 to 100
%. For each sample, the 7Li NMR spectrum was recorded. The chemical shift
changed as the amount of water in the sample was varied.
3. Acetonitrile and Hydrogen Cyanide
70
It was observed that the position of the NMR signal moved significantly
(from 4.5 to 6.4 ppm) even when the amount of added water was very small
(only 10% H2O). The chemical shift value then increased slightly and remained
almost constant when the amount of water was changed from 10 to 100% (see
Figures 3.3 and 3.4). This observation confirms that water coordinates much
more strongly to the Li+ ion than acetonitrile. Thus, addition of water
immediately leads to the formation of the aquated cation.
Figure 3.3. 7Li NMR spectra of 0.1 M LiClO4 in H2O/CH3CN mixtures.
3.03.54.04.55.05.56.06.57.07.58.08.59.0(ppm)
percentage of H2O
30
20
10
8
6
4
2
0
90
80
70
60
50
40
100
ref. 0.1 M LiClO4 in DMF
3.03.54.04.55.05.56.06.57.07.58.08.59.0(ppm)
3.03.54.04.55.05.56.06.57.07.58.08.59.0(ppm)
percentage of H2O
30
20
10
8
6
4
2
0
90
80
70
60
50
40
100
ref. 0.1 M LiClO4 in DMF
3. Acetonitrile and Hydrogen Cyanide
71
0 20 40 60 80 100
4.5
5.0
5.5
6.0
6.5
7.0
chem
ical
shi
ft, p
pm
% mol H2O
Figure 3.4. 7Li NMR shifts for 0.1 M LiClO4 as a function of the H2O/CH3CN composition.
3.5.3. Ligand Exchange on Solvated Li+
It was not possible to follow the exchange process for CH3CN on Li+
experimentally. As reported before, the rate constant of water exchange on Li+
is extremely fast (approximately 109 s-1 at 25 ºC) although water is a relatively
strong donor (DN = 33.0). As CH3CN coordinates much more weakly (DN =
14.1), its exchange on Li+ is even faster, which makes it impossible to study
experimentally. Therefore, theoretical calculations are essential and may be
very useful in such investigations.
3.5.4. Quantum-Chemical Calculations
We studied the ligand exchange process in more detail and provide
computational evidence for a limiting associative exchange mechanism on the
HCN- and CH3CN-solvated lithium cation. As mentioned before, HCN was
employed as a simplified working model to perform quantum chemical
calculations on CH3CN. Our calculations corroborated the results of the
experimental study in solution (vide supra)35 and in the solid state,100 which
3. Acetonitrile and Hydrogen Cyanide
72
showed that the first coordination sphere around Li+ consists of four tightly
bound CH3CN and HCN molecules. The five-coordinate intermediates are 3.7
kcal/mol and 4.6 kcal/mol (3.2 kcal/mol B3LYP(IPCM: acetonitrile)/6-
311+G**//B3LYP/6-311+G**) higher in energy for acetonitrile and HCN,
respectively. The computed Li-N distance for Li(NCH)4 and Li(NCCH3)4 are
2.06 and 2.05 Å, respectively. Additional solvent molecules do not coordinate
to Li+, but instead form a second coordination sphere, mainly through
electrostatic interactions. The gas-phase binding energy for the fifth
acetonitrile molecule in [Li(NCCH3)4(NCCH3)]+ is 6.1 kcal/mol, with a Li-N
distance of 6.65 Å. The fifth molecule is stabilized only by a weak HCH3-NC-CH3
interaction of 2.32 Å. The fifth HCN molecule is bound through a weak
hydrogen bond (2.04 Å) in the second coordination sphere. This bond
stabilizes the system by 8.9 kcal/mol (see Figure 3.5).
Figure 3.5. Energy profile (RB3LYP/6-311+G**) for the exchange of HCN on [Li(NCH)4]+.
The HCN exchange itself proceeds through a trigonal-bipyramidal
intermediate [Li(NCH)5]+ that is reached via a late transition state. The
3. Acetonitrile and Hydrogen Cyanide
73
entering HCN molecule approaches the lithium cation directly and pushes
three coordinated solvent molecules away toward the equatorial positions. In
line with the experimental observation of a very fast exchange process (vide
supra), the computed activation barrier is only 5.5 kcal/mol, while the five-
coordinate intermediate is 4.6 kcal/mol less stable than the precursor complex.
The very low barrier toward dissociation of only 0.9 kcal/mol means that the
five-coordinate intermediate is very short-lived. The evaluation of the energy
at the MP2(full)/6-311+G**//B3LYP/6-311+G** and B3LYP/6-
311+G**//B3LYP/6-311+G** levels shows only minor changes in the computed
energies, except in those for the intermediate ([Li(L)4(L)]+) (see Table 3.1). We
attribute this to an incorrectly balanced description of the metal-ligand and
hydrogen-bond strengths by DFT methods as observed earlier by Rotzinger
and others in studies on water exchange reactions.106,107 Inclusion of solvent
effects at the B3LYP(CPCM)/6-311+G**//B3LYP/6-311+G** and
B3LYP(IPCM)/6-311+G**//B3LYP/6-311+G** level clearly favours a four-
coordinate structure with NCH in the second coordination sphere or
uncoordinated. Therefore, the transition state is also stabilized since the
entering NCH molecule can best be addressed as uncoordinated (see Table 3.1
and Figure 3.5). Unfortunately, the transition state is modelled as being too
stable in the B3LYP(IPCM)/6-311+G**//B3LYP/6-311+G** approximation,
irrespective of whether acetonitrile or water is used in the modelling by the
IPCM formalism.
In trigonal-bipyramidal [Li(NCH)5]+, the axial Li-N distances are 0.11 Å
longer than the equatorial distances, which may be taken as a clear indication
that one of the axial ligands will be expelled. The energy profile for the
exchange of acetonitrile on [Li(NCCH3)4]+ shows the same topology; only the
energy differences are somewhat smaller. We attribute this difference to
stabilization effects caused by the methyl group and to the absence of hydrogen
bonding. A detailed investigation was prevented since a local minimum was
not reached with [Li(NCCH3)4(NCCH3)]+ because of the rotation of methyl
groups, and IPCM calculations show the well-documented problems in
reaching an energy convergence.3,108
3. Acetonitrile and Hydrogen Cyanide
74
Table 3.1. Calculated relative energies (in kcal/mol) for the exchange of L at [Li(L)4]+ (L: HCN, CH3CN).[a]
L: HCN [Li(L)4]+ + L [Li(L)4…(L)]+ ts [Li(L)4(L)]+
B3LYP 8.9 0.0 5.5 4.6
MP2(full) 9.5 0.0 3.2 0.5
IPCM -2.8 0.0 -2.6 3.2
CPCM -0.4 0.0 1.8 1.4
L: H3CCN
B3LYP 6.1 0.0 4.1 3.7
[a] B3LYP: B3LYP/6-311+G**//B3LYP/6-311+G**; MP2(full): MP2(full)/6-311+G**//B3LYP/6-311+G**; IPCM: B3LYP/(IPCM: Acetonitrile)/6-311+G**//B3LYP/6-311+G**; IPCM: B3LYP/(CPCM: water)/6-311+G**//B3LYP/6-311+G**
These findings fully corroborate the results expected for small ions
coordinated by small ligands. The dissociation of a molecule of HCN from
[Li(NCH)4]+ results in the intermediate [Li(NCH)3]+ with a D3h symmetry,
which is 14.3 kcal/mol higher in energy and less favourable. In addition, there
is no visible steric crowding, which is necessary for a dissociative mechanism.
Interchange mechanisms are known for Li+, e.g. in the case of ammonia
exchange on Li+,82 but requires a five-coordinate transition state that would
contradict the five-coordinate intermediate in our mechanism. Analogous
arguments can also be applied for the second reaction under investigation.
Under actual experimental conditions water can never be excluded.
Therefore, and because of the importance of solvent mixtures in technical and
laboratory applications, we modelled the HCN/H2O exchange process
exemplarily for [Li(NCH)4]+ to [Li(NCH)3(H2O)]+ (Table 3.2). In this case,
NCH can again be considered as a model for acetonitrile.
From experiments (vide supra) and from the Gutmann donor number for
acetonitrile, which is only 50% of that for water, it is known that addition of
water to solutions of lithium ions in acetonitrile leads to the formation of a
3. Acetonitrile and Hydrogen Cyanide
75
water-coordinated Li+ complex. This can be reproduced for HCN as model by
quantum chemical calculations (RB3LYP/6-311+G**).
Table 3.2. Calculated energies for the subsequent reactions of [Li(NCH)4]+ with water.
B3LYP(IPCM)/6-311+G**//B3LYP/6-311+G**
[kcal/mol]
Gas
phase
IPCM:
acetonitrile
IPCM:
water
[Li(NCH)4]+ + H2O [Li(H2O)(H2O)3]+ + NCH -0.5 +1.7 +1.6
[Li(H2O)( NCH)3]+ + H2O [Li(H2O)2(NCH)2]+ +
NCH -0.5 -1.4 -1.6
[Li(H2O)2(NCH)2]+ + H2O [Li(H2O)3(NCH)]+ +
NCH -0.3 -3.6 -3.9
[Li(H2O)3(NCH)]+ + H2O [Li(H2O)4]+ + NCH +0.1 -0,9 -1.2
∑ -1.3 -4.2 -5.1
By including solvent effects at the B3LYP(IPCM)/6-311+G**//B3LYP/6-
311+G** level, the reaction energies are significantly increased, regardless of
whether water or acetonitrile is selected as solvent.
Figure 3.6. Energy profile (RB3LYP/6-311+G**) for the HCN/water exchange around
[Li(NCH)4]+.
3. Acetonitrile and Hydrogen Cyanide
76
We investigated the mechanism for the first reaction step in detail (see
Figure 3.6). As shown before,35,100 the first coordination sphere around Li+
consists of four tightly bound acetonitrile molecules, here NCH molecules. An
additional solvent molecule does not coordinate to the Li+ ion; a second
coordination sphere is formed in which water is bound by one linear hydrogen
bond (1.88 Å). The gas-phase hydrogen-bond energy for the H2O molecule in
[Li(NCH)4…(OH2)]+ is 9.5 kcal/mol. Again, the entering solvent, here water,
approaches the Li+ ion directly through a face of the [Li(NCH)4]+ tetrahedron
and pushes three coordinated HCN molecules away. According to the
movement of the water molecule in the transition state (7.1 kcal/mol higher in
energy), one would expect an intermediate with three HCN molecules in
equatorial positions and the fourth together with H2O in the axial position.
During the optimization steps, the structure rearranges to form an
intermediate 3.7 kcal/mol lower in energy, with two HCN molecules in the
axial position and the H2O molecule together with the two remaining HCN
molecules in the equatorial positions. The bonds for the axial molecules are
0.16 Å longer than those for the equatorial HCN ligands - it can be concluded
that one axial ligand will be expelled to reach a four-coordinate product. The
activation barrier for the release of an axial HCN molecule is rather low, (0.1
kcal/mol), which means that the five-coordinate intermediate will be very
short-lived. On the assumption that the expelled HCN will form a hydrogen
bond with the nearest hydrogen atom, one would expect the product
[Li(NCH)3(H2O)…(NCH)]+ to form at an energy that is 4.2 kcal/mol lower than
that of the second transition state. The energy required to destroy this
hydrogen bond is 9.7 kcal/mol (Table 3.3).
3. Acetonitrile and Hydrogen Cyanide
77
Table 3.3. Calculated relative energies for the exchange of H2O/NCH at [Li(NCH)4]+ (L = HCN; W = H2O).[a]
L:HCN; W: H2O [Li(L)4…W]+ ts [Li(L)2{W(L)2}]+ ts [Li(L)3(W)…L]+
B3LYP 0.7 7.8 4.1 4.2 0.0
MP2(full) 0.9 6.1 0.6 1.2 0.0
IPCM 0.1 2.8 1.3 1.8 0.0
CPCM 0.1 5.6 1.0 1.5 0.0
[a] B3LYP: B3LYP/6-311+G**// B3LYP/6-311+G**; MP2(full): MP2(full)/6-
311+G**//B3LYP/6-311+G**; IPCM: B3LYP/(IPCM: Acetonitrile)/6-311+G**//B3LYP/6-
311+G**
Whereas MP2(full)/6-311+G**//B3LYP/6-311+G** calculations show
typical discrepancies, the application of the IPCM- and CPCM-solvent models
dramatically lowers the energy for the intermediate and the transition states.
The barriers of 2.8 and 1.8 kcal/mol corroborate the experimental findings for
an efficient formation of a water coordinated Li+ complex.
3.6. Conclusions
It was confirmed both experimentally and theoretically that Li+ is
coordinated by four acetonitrile molecules. In the binary mixture of water and
acetonitrile, water is coordinated more strongly to the Li+ ion than acetonitrile.
Thus, addition of water immediately leads to the formation of the aquated
cation. Ligand exchange reactions on Li+ most probably follow an associative
mechanism as indicated by the formation of stable five-coordinate
intermediates.
3. Acetonitrile and Hydrogen Cyanide
78
3.7. References
1. Wasserscheid, P.; Welton, T. Ionic Liquids in Synthesis; Wiley-VCH
Verlag GmbH & Co. KGaA: Weinheim, Germany, 2003; p 364.
2. Wasserscheid, P. Chem. unserer Zeit 2003, 37, 52.
3. Illner, P.; Zahl, A.; Puchta, R.; van Eikema Hommes, N.; Wasserscheid,
P.; van Eldik, R. J. Organomet. Chem. 2005, 690, 3567.
4. Weber, C. F.; Puchta, R.; van Eikema Hommes, N. J. R.; Wasserscheid,
P.; van Eldik, R. Angew. Chem., Int. Ed. 2005, 44, 6033.
5. Clifford, A.; Clifford, T. Fundamentals of Supercritical Fluids; Oxford
University Press: Oxford, 1999; p 228.
6. Erkey, C. J. Supercrit. Fluids 2000, 17, 259.
7. Oakes, R. S.; Clifford, A. A.; Rayner, C. M. J. Chem. Soc., Perkin Trans. 1
2001, 9, 917.
8. Jessop, P. G.; Ikariya, T.; Noyori, R. Science 1995, 269, 1065.
9. Kinart, W. J.; Sniec, E.; Tylak, I.; Kinart, C. M. Phys. Chem. Liq. 2000,
38, 193.
10. Babu, B. S.; Balasubramanian, K. K. Synlett 1999, 8, 1261.
11. D'Aprano, A.; Salomon, M.; Mauro, V. J. Solution Chem. 1995, 24, 685.
12. Shamsipur, M.; Madrakian, T. Main Group Met. Chem. 2001, 24, 239.
13. Karkhaneei, E.; Zebarjadian, M. H.; Shamsipur, M. J. Solution Chem.
2001, 30, 323.
14. Shamsipur, M.; Madrakian, T. J. Coord. Chem. 2000, 52, 139.
15. Hojo, M.; Hisatsune, I.; Tsurui, H.; Minami, S.-I. Anal. Sci. 2000, 16,
1277.
16. Shamsipur, M.; Madrakian, T. Polyhedron 2000, 19, 1681.
17. Shamsipur, M.; Karkhaneei, E.; Afkhami, A. J. Coord. Chem. 1998, 44,
23.
18. Danil de Namor, A. F.; Pulcha Salazar, L. E.; Llosa Tanco, M. A.;
Kowalska, D.; Villanueva Salas, J.; Schulz, R. A. J. Chem. Soc., Faraday
Trans. 1998, 94, 3111.
3. Acetonitrile and Hydrogen Cyanide
79
19. Karkhaneei, E.; Zolgharnein, J.; Afkhami, A.; Shamsipur, M. J. Coord.
Chem. 1998, 46, 1.
20. de Namor, A. F. D.; Zapata-Ormachea, M. L.; Jafou, O.; Al Rawi, N. J.
Phys. Chem. B 1997, 101, 6772.
21. D'Aprano, A.; Vicens, J.; Asfari, Z.; Salomon, M.; Iammarino, M. J.
Solution Chem. 1996, 25, 955.
22. de Namor, A. F. D.; Ng, J. C. Y.; Tanco, M. A. L.; Salomon, M. J. Phys.
Chem. 1996, 100, 14485.
23. Fakhari, A. R.; Shamsipur, M. J. Inclusion Phenom. 1996, 26, 243.
24. Karkhaneei, E.; Afkhami, A.; Shamsipur, M. J. Coord. Chem. 1996, 39,
33.
25. de Namor, A. F. D.; Gil, E.; Tanco, M. A. L.; Tanaka, D. A. P.; Salazar, L.
E. P.; Schulz, R. A.; Wang, J. J. Phys. Chem. 1995, 99, 16776.
26. Jander, G.; Klemm, W. Die Chemie in wasserähnlichen Lösungsmitteln;
Springer-Verlag: Berlin, 1949; p 367.
27. Jander, J.; Lafrenz, C. Wasserähnliche Lösungsmittel; Springer-Verlag:
Berlin, 1968; Vol. 3, p 212.
28. Jander, J.; Tuerk, G. Chem. Ber. 1962, 95, 881.
29. Jander, J.; Tuerk, G. Chem. Ber. 1962, 95, 2314.
30. Jander, J.; Tuerk, G. Chem. Ber. 1965, 98, 894.
31. Bako, I.; Megyes, T.; Palinkas, G. Chem. Phys. 2005, 316, 235.
32. Lu, H.; Wang, J.; Zhao, Y.; Xuan, X.; Zhuo, K. J. Chem. Eng. Data 2001,
46, 631.
33. Rivelino, R.; Cabral, B. J. C.; Coutinho, K.; Canuto, S. Chem. Phys. Lett.
2005, 407, 13.
34. Mollner, A. K.; Brooksby, P. A.; Loring, J. S.; Bako, I.; Palinkas, G.;
Fawcett, W. R. J. Phys. Chem. A 2004, 108, 3344.
35. Barthel, J.; Buchner, R.; Wismeth, E. J. Solution Chem. 2000, 29, 937.
36. Brouillette, D.; Irish, D. E.; Taylor, N. J.; Perron, G.; Odziemkowski, M.;
Desnoyers, J. E. Phys. Chem. Chem. Phys. 2002, 4, 6063.
37. Xuan, X.; Zhang, H.; Wang, J.; Wang, H. J. Phys. Chem. A 2004, 108,
7513.
3. Acetonitrile and Hydrogen Cyanide
80
38. Fawcett, W. R.; Brooksby, P.; Verbovy, D.; Bako, I.; Palinkas, G. J. Mol.
Liq. 2005, 118, 171.
39. Il'in, K. K.; Demakhin, A. G. Russ. J. Gen. Chem. 1999, 69, 703.
40. Loring, J. S.; Fawcett, W. R. J. Phys. Chem. A 1999, 103, 3608.
41. Matsubara, K.; Kaneuchi, R.; Maekita, N. J. Chem. Soc., Faraday Trans.
1998, 94, 3601.
42. Cartailler, T.; Kunz, W.; Turq, P.; Bellisent-Funel, M. C. J. Phys.:
Condens. Matter 1991, 3, 9511.
43. Alia, J. M.; Diaz de Mera, Y.; Edwards, H. G. M.; Garcia, F. J.; Lawson,
E. E. J. Mol. Struct. 1997, 408-409, 439.
44. Alia, J. M.; Diaz De Mera, Y.; Edwards, H. G. M.; Garcia, F. J.; Lawson,
E. E. Z. Phys. Chem. 1996, 196, 209.
45. Maroncelli, M. J. Chem. Phys. 1991, 94, 2084.
46. Troxler, L.; Wipff, G. J. Am. Chem. Soc. 1994, 116, 1468.
47. Richardi, J.; Fries, P. H.; Krienke, H. J. Chem. Phys. 1998, 108, 4079.
48. Cabaleiro-Lago, E. M.; Rios, M. A. Chem. Phys. 1998, 236, 235.
49. Guardia, E.; Pinzon, R. J. Mol. Liq. 2000, 85, 33.
50. Gregoire, G.; Brenner, V.; Millie, P. J. Phys. Chem. A 2000, 104, 5204.
51. Spångberg, D.; Hermansson, K. Chem. Phys. 2004, 300, 165.
52. Kolesnik, Y. V.; Kalugin, O. N. Russ. J. Electrochem. 2003, 39, 427.
53. Fischer, R.; Richardi, J.; Fries, P. H.; Krienke, H. J. Chem. Phys. 2002,
117, 8467.
54. Dimitrova, Y. THEOCHEM 1995, 334, 215.
55. Cabaleiro-Lago, E. M.; Rios, M. A. Chem. Phys. 2000, 254, 11.
56. Pejov, L. Int. J. Quantum Chem. 2002, 86, 356.
57. Bako, I.; Megyes, T.; Radnai, T.; Palinkas, G.; Probst, M.; Fawcett, W. R.
Z. Phys. Chem. 2004, 218, 643.
58. Hojo, M.; Hasegawa, H.; Hiura, N. J. Phys. Chem. 1996, 100, 891.
59. D'Aprano, A.; Iammarino, M.; Mauro, V.; Sesta, B. J. Electroanal. Chem.
1995, 392, 27.
60. Brouillette, D.; Perron, G.; Desnoyers, J. E. J. Solution Chem. 1998, 27,
151.
3. Acetonitrile and Hydrogen Cyanide
81
61. Hojo, M.; Ueda, T.; Nishimura, M.; Hamada, H.; Matsui, M.; Umetani,
S. J. Phys. Chem. B 1999, 103, 8965.
62. Izonfuo, W.-A. L.; Obunwo, C. C. Indian J. Chem., Sect. A: Inorg., Bio-
inorg., Phys., Theor. Anal. Chem. 1999, 38A, 939.
63. Barthel, J.; Kleebauer, M.; Buchner, R. J. Solution Chem. 1995, 24, 1.
64. Croce, F.; D'Aprano, A.; Nanjundiah, C.; Koch, V. R.; Walker, C. W.;
Salomon, M. J. Electrochem. Soc. 1996, 143, 154.
65. D'Aprano, A.; Sesta, B.; Capalbi, A.; Iammarino, M.; Mauro, V. J.
Electroanal. Chem. 1996, 403, 257.
66. Mauro, V.; D'Aprano, A.; Croce, F.; Salomon, M. J. Power Sources
2005, 141, 167.
67. Faverio, C. L.; Mussini, P. R.; Mussini, T. Anal. Chem. 1998, 70, 2589.
68. Brouillette, D.; Perron, G.; Desnoyers, J. E. Electrochim. Acta 1999, 44,
4721.
69. Barthel, J.; Neueder, R.; Poepke, H.; Wittmann, H. J. Solution Chem.
1999, 28, 489.
70. Mandal, B.; Sooksimuang, T.; Griffin, B.; Padhi, A.; Filler, R. Solid State
Ionics 2004, 175, 267.
71. Murugavel, S.; Roling, B. J. Phys. Chem. B 2004, 108, 2564.
72. Mariappan, C. R.; Govindaraj, G.; Roling, B. Solid State Ionics 2005,
176, 723.
73. Heuer, A.; Murugavel, S.; Roling, B. Phys. Rev. B: Condens. Matter
Mater. Phys. 2005, 72, 174304/1.
74. Helm, L.; Merbach, A. E. Chem. Rev. 2005, 105, 1923.
75. Helm, L.; Merbach, A. E. Coord. Chem. Rev. 1999, 187, 151.
76. Erras-Hanauer, H.; Clark, T.; van Eldik, R. Coord. Chem. Rev. 2003,
238-239, 233.
77. Hartmann, M.; Clark, T.; van Eldik, R. J. Mol. Model. 1996, 2, 354.
78. Hartmann, M.; Clark, T.; van Eldik, R. J. Am. Chem. Soc. 1997, 119,
7843.
79. Hartmann, M.; Clark, T.; Van Eldik, R. J. Phys. Chem. A 1999, 103,
9899.
3. Acetonitrile and Hydrogen Cyanide
82
80. Rotzinger, F. P. Chem. Rev. 2005, 105, 2003.
81. Puchta, R.; van Eikema Hommes, N.; van Eldik, R. Helv. Chim. Acta
2005, 88, 911.
82. Puchta, R.; Galle, M.; Van Eikema Hommes, N.; Pasgreta, E.; Van Eldik,
R. Inorg. Chem. 2004, 43, 8227.
83. Coetzee, J. F.; Chang, T. H. Pure Appl. Chem. 1986, 58, 1541.
84. Stephens, P. J.; Devlin, F. J.; Chabalowski, C. F.; Frisch, M. J. J. Phys.
Chem. 1994, 98, 11623.
85. Becke, A. D. J. Chem. Phys. 1993, 98, 5648.
86. Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B: Condens. Matter Mater.
Phys. 1988, 37, 785.
87. Koch, W.; Holthausen, M. C. A Chemist's Guide to Density Functional
Theory, 2nd Edition; John Wiley & Sons, Inc.: Chichester, UK, 2001; p
528.
88. Hehre, W. J.; Radom, L.; Schleyer, P. v. R.; Pople, J. A. Ab Initio
Molecular Orbital Theory; John Wiley & Sons: New York, 1986; p 548.
89. Foresman, J. B.; Keith, T. A.; Wiberg, K. B.; Snoonian, J.; Frisch, M. J. J.
Phys. Chem. 1996, 100, 16098.
90. Barone, V.; Cossi, M. J. Phys. Chem. A 1998, 102, 1995.
91. Cossi, M.; Rega, N.; Scalmani, G.; Barone, V. J. Comput. Chem. 2003,
24, 669.
92. In the CPCM formalism some molecules failed while using symmetry in
the molecular cavity generation. This problem was solved successfully by
the keyword “nosymm”.
93. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.;
Cheeseman, J. R.; Montgomery, J. A.; Jr.; Vreven, T.; Kudin, K. N.;
Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.;
Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.;
Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa,
J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.;
Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Bakken, V.; Adamo, C.;
Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.;
3. Acetonitrile and Hydrogen Cyanide
83
Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.;
Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich,
S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.;
Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.;
Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.;
Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham,
M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.;
Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian
03, Revision B.03, Gaussian Inc.; Wallingford, CT, 2004.
94. Keil, R. G.; Johnson, D. W.; Fryling, M. A.; O'Brien, J. F. Inorg. Chem.
1989, 28, 2764.
95. Fawcett, W. R.; Liu, G.; Faguy, P. W.; Foss, C. A., Jr.; Motheo, A. J. J.
Chem. Soc., Faraday Trans. 1993, 89, 811.
96. Barthel, J.; Deser, R. J. Solution Chem. 1994, 23, 1133.
97. Solovieva, L. A.; Akopyan, S. K.; Vilaseca, E. Spectrochim. Acta, Part A
1994, 50A, 683.
98. Brooks, N. R.; Henderson, W. A.; Smyrl, W. H. Acta Crystallogr., Sect.
E: Struct. Rep. Online 2002, E58, m176.
99. Chitsaz, S.; Pauls, J.; Neumuller, B. Z. Naturforsch., B: Chem. Sci. 2001,
56, 245.
100. Yokota, Y.; Young, V. G., Jr.; Verkade, J. G. Acta Crystallogr., Sect. C:
Cryst. Struct. Commun. 1999, C55, 196.
101. Sajeevkumar, V. A.; Singh, S. J. Mol. Struct. 1996, 382, 101.
102. Megyes, T.; Radnai, T.; Wakisaka, A. J. Mol. Liq. 2003, 103-104, 319.
103. Maxey, B. W.; Popov, A. I. J. Am. Chem. Soc. 1968, 90, 4470.
104. Gutmann, V. Electrochim. Acta 1976, 21, 661.
105. Cahen, Y. M.; Handy, P. R.; Roach, E. T.; Popov, A. I. J. Phys. Chem.
1975, 79, 80.
106. Rotzinger, F. P. J. Phys. Chem. B 2005, 109, 1510.
107. Hanauer, H.; Puchta, R.; Clark, T.; van Eldik, R. Inorg. Chem. 2007, 46,
1112.
108. Puchta, R.; Clark, T.; Bauer, W. J. Mol. Model. 2006, 12, 7
4. Complexation by Cryptands in Acetone as Solvent
84
4. Ligand Exchange Processes on Solvated
Lithium Cations. Complexation by
Cryptands in Acetone as Solvent
4.1. General remark
The work presented in this chapter was conducted by several authors within a
joint project and has been published as:
Pasgreta, E.; Puchta, R.; Zahl, A.; van Eldik, R. European Journal of Inorganic
Chemistry 2007, 3067-3076.
4.2. Abstract
Kinetic studies on Li+ exchange between the cryptands C222 and C221 and
acetone as solvent were performed as a function of ligand-to-metal ratio,
temperature and pressure using 7Li NMR spectroscopy. Temperature and
pressure dependence measurements were performed in the presence of an
excess of Li+. The rate and activation parameters for the exchange process are:
C222: k298 = (1.7 ± 0.5) x 103 s-1, ∆H# = 24 ± 1 kJ mol-1 and ∆S# = -105 ± 4 J K-1
mol-1; C221: k298 = 2.24 ± 0.09 s-1, ∆H# = 29 ± 2 kJ mol-1 and ∆S# = -141 ± 5 J
K-1 mol-1. The influence of pressure on the exchange rate is insignificant for
both ligands, such that the activation volume is around zero within the
experimental error limits. The reported activation parameters suggest that the
exchange of Li+ between solvated and chelated ions follows an associative
interchange mechanism.
4. Complexation by Cryptands in Acetone as Solvent
85
4.3. Introduction
Following the work of Pedersen on two-dimensional crowns, Lehn and
coworkers developed in the late 1960s a range of three-dimensional, polycyclic
ligand systems which they named cryptands (Greek: cryptos = cave).1, 2 Until
today this class of compounds has been intensively explored and extended in
different directions3, 4 in terms of cavity size,5, 6 donor atoms,7 and metalla-
topomers.8-11 Because cryptands readily form complexes with a wide range of
metal ions and NH4+,12 provided the ion involved is not too large to be
contained in the macrocyclic cavity,13 they find their application in various
areas like catalysis, medicine, biomimetic research,14-21 and as receptors22 and
electrides.23 Studies on the complexes (cryptates) formed between alkali-metal
ions and cryptands, have revealed a substantial understanding of the effects of
metal-ion size and cryptand-cavity size on the structure, stability and lability of
cryptates, and the mechanism of the cryptate complex-formation processes.24,
25 Such selective complexation has led to many biological and chemical studies.
Kinetic studies on macrocyclic complex-formation reactions with alkali
cations not only result in important information on the rates and mechanisms
of complex-formation reactions, but also lead to a better understanding of the
high selectivity of these ligands toward different cations. The broad impact of
such studies spans from the theory of complexation to cation transport
through membranes, and makes the ability to control cation selectivity via
tactical structural changes in the complexing agent of prime importance.18
Polyoxo macrobicyclic diamines, represented by the structure in Figure 4.1, are
able to form metal-ion complexes in which the metal ion is located in the cavity
of the macromolecule and two nitrogen donors along with the oxygen donors
participate in binding the metal cation. All complexes have a 1:1 stoichiometry
with the metal ion positioned at the center of the ligand cavity.24
4. Complexation by Cryptands in Acetone as Solvent
86
O
N
OO
O
N O O
m
m
n Figure 4.1. Chelate structures: C211: m = 0, n = 1; C221: m = 1, n = 0; C222: m = 1, n = 1.
In case of crown ethers and cryptands in non-aqueous solutions, the rates
of complex-formation (kf) with alkali-metal cations are in general diffusion-
controlled and, consequently, the complexation selectivities are governed by
the decomplexation rates (kd).26 These reactions have received significant
attention, initially from Lehn and co-workers and more recently from Popov et
al.27-29 and Lincoln et al..30-37 Many examples have been included in a series of
review articles.38-41 It has been reported that ligand structure and solvent
properties have considerable effects on the reaction rates, activation
parameters, and exchange mechanism of cations between the solvated and
complexed sites. We have now extended our earlier studies42 on Li+ exchange
reactions to acetone as solvent.
Non-aqueous solvents and their mixtures combined with Li+ are often used
in secondary lithium batteries.43, 44 Li+ in acetone is also known to catalyze
Diels-Alder-reactions45 and the ene reaction.46 The behavior of Li+-ions in
different solvents is therefore an important research area.47-57 A variety of
studies on solutions of Li+ in acetone have been mainly studied
experimentally52, 55, 58-63 and some theoretically,64, 65 as well as for
solvent/acetone combinations53 or chelator/acetone combinations.53, 66 In this
contribution, we focus on acetone as a solvent and the competition between
solvated and endohedral complexed Li+, report kinetic parameters for the
exchange of Li+ between solvated and complexed states in acetone under
ambient and elevated pressure conditions, and discuss the mechanistic
implications.
4. Complexation by Cryptands in Acetone as Solvent
87
4.4. Experimental Section
4.4.1. General Procedures
The preparation of test solutions was carried out under an Ar or N2 using
standard Schlenk techniques, under an Ar in a glove box, or under vacuum.
Lithium perchlorate (Aldrich, battery grade) was vacuum-dried at 120 °C for
48 h and then stored under dry nitrogen. The cryptands C222 (purchased from
Fluka), C221 and C211 (purchased from Aldrich) were dried under vacuum for
48 h and stored under dry nitrogen. Acetone (Acros, extra dry, water < 50
ppm) was used as received. Solutions of LiClO4, C222, C221 and C211 in
acetone were prepared under dry nitrogen. For the 7Li NMR measurements the
solutions were sealed in 5-mm NMR tubes under Ar. In order to study the
kinetics of the exchange of Li+ between C222 and C221 and acetone as solvent,
three solutions of different concentration ratio were prepared for each system.
The concentration of lithium perchlorate in each solution was kept constant at
0.04 M, whereas the concentration of ligand was varied between 0.01 M and
0.032 M. In this way the measurements were always performed in the
presence of an excess of Li+.
4.4.2. 7Li NMR Spectroscopy
Variable-temperature and variable-pressure Fourier transform 7Li NMR
spectra were recorded at a frequency of 155 MHz on a Bruker Avance DRX
400WB spectrometer equipped with a superconducting BC-94/89 magnet
system. Spinning tubes (5 mm) were used with a 1-mm o.d. melting point
capillary inserted coaxially in the spinning tube and filled with an external
reference solution (1 M LiClO4 in acetonitrile). The temperature dependence
measurements were performed over a wide as possible temperature range for
each system. We did not use a temperature calibration by insertion of a
capillary filled with methanol (low temperature calibration) or ethylene glycol
(high temperature calibration) into the sample tube, as the corresponding
signals of these solvents would disturb the NMR spectra in the region of
intrest. In such a case a temperature-calibrated NMR probe has to be used in
4. Complexation by Cryptands in Acetone as Solvent
88
order to determine the temperature of the sample accurately. Methanol was
used for the low-temperature calibration of the NMR probe, whereas ethylene
glycol was used for high-temperature calibration. The actual sample
temperature can be calculated from the chemical shift differences. If there is
any difference between the sample temperature and the temperature detected
by the thermocouple of the probe, the probe can be calibrated by using
standard Bruker Topspin software.
A homemade high-pressure probe described in the literature67 was used for
the variable-pressure experiments which were conducted at a selected
temperature and at ambient, 30, 60, 75, 90, 120 and 150 MPa pressure. A
standard 5-mm NMR tube cut to a length of 45 mm was used for the sample
solution. The high-pressure NMR tube was sealed with a MACOR plug and an
O-ring made of VITON. The plug and O-ring were shown to be stable in the
employed solvent. The pressure was transmitted by the movable MACOR
piston to the sample, and the temperature of the sample was controlled with a
Pt-100 resistor.67 The probe itself was externally heated with a thermostatted
circulating fluid that was pumped through a double helix that was cut into the
outer surface of the autoclave. Because the sample is embedded in the high-
pressure transmitting fluid that serves as a thermostating bath, a reliable
temperature setting is guaranteed.
For a typical kinetic run, 0.8 ml of solution was transferred under Ar to a
Wilmad screw-cap NMR tube equipped with a poly(tetrafluoroethylene)
septum. All test samples were prepared in the same way.
4.4.3. Programs
The line widths of the NMR signals were obtained by fitting Lorentzian
functions to the experimental spectra using the NMRICMA 2.7 program68 for
Matlab.69 The adjustable parameters were the resonance frequency, intensity,
line width and baseline. Complete line-shape analysis based on the Kubo-Sack
formalism using modified Bloch equations70 was also performed with
NMRICMA 2.7 to extract rate constants from experimental spectra. The
temperature and pressure dependent 7Li line widths and chemical shifts
4. Complexation by Cryptands in Acetone as Solvent
89
employed in the line shape analysis were extrapolated from low temperatures
where no exchange-induced modification occurred.
4.4.4. Quantum-Chemical Calculations
All structures were optimized with the B3LYP71-73 and the basis set
D95Vp,74, 75 or with 6-311+G** with no constrains other than symmetry, and
confirmed to be minima by computation of vibrational frequencies. Relative
energies were corrected for zero point vibrational energy differences. The
D95V basis set, augmented with polarisation functions (D95Vp), was used.76 If
not otherwise mentioned, calculations were done without a solvent model. The
influence of bulk solvent was probed using the CPCM formalism and water as
solvent, i.e., B3LYP(CPCM)/D95Vp//B3LYP/D95Vp. The Gaussian 03 suite of
programs was used throughout.77
4.5. Results and Discussion
4.5.1. Spectroscopic and Kinetic Observations
In preliminary experiments we investigated the Li+ exchange in different
solvents. In water and methanol the exchange process was too fast to be
studied by 7Li NMR (only a single signal was observed). The process became
slower in acetonitrile, and in acetone two completely separated signals were
observed at low temperature. We, therefore, investigated Li+ exchange between
the selected cryptands and acetone as solvent. A similar system was also
investigated before by Popov et al..28 Although they reported a significantly
negative value for the activation entropy, they suggested the operation of a
dissociative exchange mechanism. We realized that it would be worth to
perform a reinvestigation of this system in acetone in order to clarify this
apparent discrepancy, in particular by means of high pressure 7Li NMR
spectroscopy.
Initially, a series of 7Li NMR measurements for the Li+-cryptand system in
acetone were performed. For each of the cryptands C222, C221 and C211,
molar ratio dependent measurements were performed by variation of both the
4. Complexation by Cryptands in Acetone as Solvent
90
cryptand/Li+ and Li+/cryptand molar ratios. The results enabled us to observe
the distribution of particular species on variation of the concentration ratio.
Basically, two signals of varying intensity were observed in the 7Li NMR
spectra that refer to solvated Li+ and Li+ bound inside the cryptand cavity. In
case of the smallest chelate C211, however, an additional signal appeared
which can be ascribed to Li+ bound outside the cryptand and being partially
solvated. In the case of Li(C211)+, the complex is very stable and almost no
exchange takes place. A detailed study of the observed rate constant, kobs, for
the exchange of Li+ as a function of temperature and pressure was therefore
undertaken only for C221 and C222. The lithium exchange reaction between
cryptand and acetone was studied for three solutions with different ligand-to-
metal ratios in the presence of an excess of Li+.
Three signals can be observed in Figure 4.2; viz. one coming from the
reference (δ = 5.8 ppm) and the other two from the exchanging sites (δ = 9.0
ppm for solvated Li+ and 6.6 ppm for complexed Li+). At low temperature the
signals are sharp and completely separated due to very slow or no exchange.
Because the formation constant for Li(C221)+ is very high,24 the concentration
of bound lithium equals the amount of cryptand added. At ca. 283 K, the
signals start to broaden and the exchange process takes place. The linear
temperature dependence for the temperature range in which exchange takes
place can be seen in Figure 4.5. In the case of C222 (Figure 4.3), the exchange
process is faster due to the larger cavity of the cryptand. The signals that refer
to solvated and complexed Li cation (8.7 and 6.8 ppm, respectively at 223 K)
coalescence at ca. 263 K and at high temperature (>313 K) only one sharp
signal at 7.7 ppm was observed as a consequence of a very fast exchange
process. An accurate temperature dependence for C222 over a large
temperature range can be observed in Figure 4.6.
4. Complexation by Cryptands in Acetone as Solvent
91
Figure 4.2. 7Li NMR spectra recorded as a function of temperature for the mole ratio
C221/LiClO4 = 0.5.
Figure 4.3. 7Li NMR spectra recorded as a function of temperature for the mole ratio
C222/LiClO4 = 0.5.
The rate of the exchange reaction increased not only with increasing cavity
size of the cryptand but also with increasing concentration of the chelate, as
4. Complexation by Cryptands in Acetone as Solvent
92
can be recognized from the shape of the signals at 223 K in Figure 4.4 and
from the values of kobs in Table 4.1 (see Figure 4.5 and Figure 4.6).
Figure 4.4. 7Li NMR spectra of C222 and LiClO4 in acetone. Temperature dependence for
the sample ratios C222/LiClO4 = 0.25, 0.5 and 0.75, respectively.
Figure 4.5. Eyring plots for kobs as a function of temperature for C221 in the temperature range in which exchange was observed for the concentration ratios C221/LiClO4 = 0.5, 0.65 and 0.8, respectively.
Figure 4.6. Eyring plots for kobs as a function of temperature for C222 for the concentration ratios C222/LiClO4 = 0.25, 0.5 and 0.75, respectively.
0.0031 0.0032 0.0033 0.0034-4.5
-4.2
-3.9
-3.6
-3.3
ln (k/T)
1/T, K-10.0030 0.0031 0.0032 0.0033 0.0034
-3.0
-2.7
-2.4
-2.1
-1.8
ln (k/T)
1/T, K-10.0031 0.0032 0.0033
-4.8
-4.4
-4.0
-3.6
ln (k/T)
1/T, K-10.0031 0.0032 0.0033 0.0034
-4.5
-4.2
-3.9
-3.6
-3.3
ln (k/T)
1/T, K-10.0030 0.0031 0.0032 0.0033 0.0034
-3.0
-2.7
-2.4
-2.1
-1.8
ln (k/T)
1/T, K-10.0031 0.0032 0.0033
-4.8
-4.4
-4.0
-3.6
ln (k/T)
1/T, K-1
0.0032 0.0036 0.0040 0.0044-4
-3
-2
-1
0
1
2
ln (k
/T)
1/T, K-10.0032 0.0036 0.0040 0.0044
-1
0
1
2
3
4
ln (k
/T)
1/T, K-1
0.0032 0.0036 0.0040 0.0044-3
-2
-1
0
1
2
3
ln (k
/T)
1/T, K-10.0032 0.0036 0.0040 0.0044
-4
-3
-2
-1
0
1
2
ln (k
/T)
1/T, K-10.0032 0.0036 0.0040 0.0044
-1
0
1
2
3
4
ln (k
/T)
1/T, K-1
0.0032 0.0036 0.0040 0.0044-3
-2
-1
0
1
2
3
ln (k
/T)
1/T, K-1
4. Complexation by Cryptands in Acetone as Solvent
93
Table 4.1. Values of the observed rate constants (kobs) and activation parameters calculated from the line-shape analysis for the exchange of Li+ between solvated and chelated sites at 25 °C.
Molar ratio [L]:[Li+] kobs298, s-1 ΔH#, kJ mol-1 ΔS#, J K-1 mol-1
C221
0.5 2.10 ± 0.04 41 ± 3 -102 ± 12
0.65 4.5 ± 0.1 37 ± 1 -109 ± 3
0.8 19.4 ± 0.8 31 ± 1 -116 ± 4
C222
0.25 1001 ± 20 33.3 ± 0.5 -76 ± 1
0.5 3263 ± 98 34.1 ± 0.4 -64 ± 1
0.75 6346 ± 254 29 ± 1 -75 ± 2
4.5.2. Suggested Exchange Mechanism
Two mechanisms should be considered for the decomplexation processes of
cryptand-alkali-metal complexes: unimolecular (dissociative) and bimolecular
(associative) cation exchange. The solvent plays a major role in the
unimolecular mechanism and this mechanism is therefore favored in strong
donor solvents.40
Lincoln et al.30-37 investigated many similar systems where the Li+ cation
was exchanged between different cryptands or crown ethers and common
solvents. In case of crown ethers and cryptands in non-aqueous solutions, the
rates of formation (kc) for their interaction with alkali-metal cations are
generally diffusion controlled and, consequently, the complexation selectivities
are governed by the decomplexation rates (kd). They proposed the operation of
a monomolecular mechanism for the decomplexation of Li+ from the cryptand,
notwithstanding the fact that decomplexation is characterized by very negative
activation entropies.
4. Complexation by Cryptands in Acetone as Solvent
94
According to Popov et al.,28 there are two possible mechanisms that need to
be considered for the exchange of Li+ cation between the solvated and the
complexed sites, viz. bimolecular (associative) (I) and unimolecular
(dissociative) (back reaction in II) cation exchange processes, where S presents
solvent and L the employed chelate.28, 78 (In this description we have added the
terms ‘associative’ and ‘dissociative’ because this is implied by the mechanistic
interpretation offered by the authors).
At equilibrium
]Li(S)][Li(L)[*]Li(L)][Li(S)[* 11 −= kk
]S][Li(L)[]L][Li(S)[ 22 −= kk
Therefore see Equation (4.1) and Equation (4.2)
]S][Li(L)[]Li(L)][Li(S)[]Li(S)[]Li(S)[ 21 −+==− kkkdtd
obs (4.1)
and
]Li(S)[
]S][Li(L)[]Li(L)[ 2
1−+=
kkkobs (4.2)
An alternative form of Equation (4.2) is Equation (4.3), which was used to fit
the data in Figure 4.7 and Figure 4.9.
]Li(S)['
]Li(L)[1
21 −+= kkkobs (4.3)
k1
k-1
*Li(S)+ + Li(L)+ Li(S)+ + *Li(L)+ (I)
k2
k-2
Li(S)+ + L Li(L)+ + S (II)
k1
k-1
*Li(S)+ + Li(L)+ Li(S)+ + *Li(L)+ (I)
k2
k-2
Li(S)+ + L Li(L)+ + S (II)
4. Complexation by Cryptands in Acetone as Solvent
95
Plots of kobs/[Li(L)] vs. 1/[Li(S)] were found to be linear with varying slopes
(k-2’ = k-2[S]) and intercepts (k1) (see Figure 4.7 and Figure 4.9). The plots
indicate that both pathways compete with each other and the competition is
controlled by temperature and the nature of the cryptand. In the case of C222
(Figure 4.7), the plots at low temperature exhibit no meaning intercept,
whereas the plots at high temperature exhibit very large intercepts. Thus the
contribution from k1 increases significantly with temperature. In the case of
C221 (Figure 4.9), all the plots exhibit almost no intercept, such that the
exchange process is mainly controlled by k-2. The corresponding Eyring plots
are shown in Figure 4.8 and Figure 4.10, respectively, and the thermal
activation parameters are summarized in Table 4.2.
0 10 20 30 40 50 60 70 80 90 1000.0
2.0x104
4.0x104
6.0x104
1.6x105
2.4x105
3.2x105
4.0x105
k obs
/[Li(L
)], M
-1S
-1
1/[Li(S)], M-1
313 K 298 K 273 K 253 K 243 K 233 K
Figure 4.7. Plots of kobs/[Li(L)] vs. 1/[Li(S)] for C222 as a function of temperature.
4. Complexation by Cryptands in Acetone as Solvent
96
0.0032 0.0036 0.0040 0.0044-2
0
2
4
6 k1 k-2
ln (k
/T)
1/T, K-1
Figure 4.8. Eyring plots for k1 and k-2 as a function of temperature for C222.
0 50 100 150 200 2500
200
400
600
800
1000
1200 318 K 313 K 308 K 303 K 298 K
k obs
/[Li(L
)], M
-1s-1
1/[Li(S)], M-1
Figure 4.9. Plots of kobs/[Li(L)] vs. 1/[Li(S)] for C221 as a function of temperature.
4. Complexation by Cryptands in Acetone as Solvent
97
0.00315 0.00320 0.00325 0.00330 0.00335
-4.9
-4.8
-4.7
-4.6
-4.5
-4.4
-4.3
-4.2
-4.1
k-2
ln (k
/T)
1/T, K-1
Figure 4.10. Eyring plot for k-2 as a function of temperature for C221.
Table 4.2. Thermal activation parameters obtained from the values of k1 and k-2 as a function of temperature.
Chelate k1298, M-1 s-1 k-2298, s-1 ∆H#, kJ mol-1 ∆S#, J K-1 mol-1
C221 2.24 ± 0.09 29 ± 2 -141 ± 5
C222 (1.7 ± 0.5) x 103 24 ± 1 -105 ± 4
C222 (6 ± 3) x 104 54 ± 4 +29 ± 15
The activation parameters derived from the temperature dependence of k1
are rather inaccurate due to the large error limits of the intercepts and the
reported activation entropy has no mechanistic meaning. On the basis of the
values of k-2 and the significantly negative activation entropies obtained for
this pathway, the activation parameters support an associative type of
exchange process. k-2 for C222 is almost three orders of magnitude larger than
for C221. Furthermore, the activation parameters are very similar to those
obtained directly from the values of kobs as a function of temperature. This
4. Complexation by Cryptands in Acetone as Solvent
98
would be in a good agreement with that postulated by Lincoln, who assumed
that only the decomplexation process contributes to the observed reaction
under the given experimental conditions. The negative activation entropies
found for the decomplexation reactions of both C221 and C222 strongly
suggest that attack of the solvent on the complexed Li+ must play an important
role in the decomplexation process, i.e. decomplexation follows an associative
mechanism (see theoretical calculations).
The reported activation parameters are within the range of those reported
by Popov et al. for closely related systems.28 However, it should be pointed out
that they used a different rate law than the one proposed by Shchori et al.79
and the one used in the present study, which complicates a direct comparison
of the reported activation parameters. Nevertheless, although they proposed
another mechanism, the values of the activation parameters obtained from
their study are very similar to ours, as shown in Table 4.3.below.
Table 4.3. Comparison of activation parameters.
kb, s-1 ΔH#, kJ mol-1 ΔS#, J K-1 mol-1
Popov et al.28 794 ± 42 20 ± 1 -121 ± 3
This work 1724 ± 484 24 ± 1 -105 ± 4
It is reasonable to expect that solvents such as acetonitrile, acetone and γ-
butyrolactone, due to their similar properties as expected from their rather
similar Gutmann donor numbers (DN = 14.1, 17.0 and 18.0, respectively,80)
should coordinate to Li+ in a similar way and should have a similar influence
on the exchange mechanism. This observation corresponds to that reported by
Shamsipur et al.81 for the exchange of Li+ between C221 and
acetonitrile/nitromethane mixtures. In all solvent mixtures used, Li+ exchange
was found to occur through an associative mechanism, as expected for weakly
donor solvents (DN = 14.1 and 2.7 for acetonitrile and nitromethane,
respectively).
4. Complexation by Cryptands in Acetone as Solvent
99
4.5.3. High Pressure 7Li NMR Measurements
Additional mechanistic information on the intimate nature of the exchange
mechanism can in principle be obtained from the effect of pressure on the Li+
exchange process.82-85 A systematic pressure dependence study was performed
to determine the volume of activation for the Li+ exchange process for both
chelates. Pressure-dependent NMR spectra were recorded for each sample
over the pressure range of 0.1 to 150 MPa at 313.2 K. The higher temperature
had to be chosen, since in case of C221 the exchange process hardly occurs at
room temperature, whereas in case of C222 the signals are broad at room
temperature due to coalescence and it was difficult to observe line broadening
that resulted from increasing pressure. There are two signals to be seen in
Figure 4.11: one for the solvated Li at 8.8 ppm and the one for Li complexed by
C221 at 6.5 ppm. It can be observed that there is a very small difference in the
shape of the NMR spectra recorded at various pressures. The same effect was
observed for the second ligand, C222 (Figure 4.12), where only one signal
appears due to the fast exchange. Its shape almost does not change with the
varying pressure for any of the three samples. The rate constants calculated for
each pressure were very similar within the experimental error limits and
resulted in a practically zero value for ∆V#obs for both ligands, which points to a
concerted interchange type of exchange mechanism.86 In terms of an
interchange mechanism, bond formation and bond cleavage occur in a
concerted fashion without the formation of a distinct intermediate of higher or
lower coordination number. In addition, decomplexation will be accompanied
by charge creation due to the release of Li+, which in turn will lead to an
increase in electrostriction which will offset the intrinsic volume changes.
4. Complexation by Cryptands in Acetone as Solvent
100
Figure 4.11. 7Li NMR spectra recorded as a function of pressure at 313.2 K for the mole ratio C221/LiClO4 = 0.5.
Figure 4.12. 7Li NMR spectra of C222 and LiClO4 in acetone. Pressure dependence for the sample ratios C222/LiClO4 = 0.25, 0.5, and 0.75, respectively, at 313.2 K.
4.5.4. Theoretical Calculations
For the closely related γ-butyrolactone and Li+/γ-butyrolactone high level
ab initio studies were recently performed by Rey et al..87 For acetone and
Li+/acetone only Hartree-Fock calculations were performed.65 In our DFT
calculations we found that the Li+ cation coordinates to the carbonyl oxygen
with an almost co-linear configuration relative to the carbon-oxygen bond, as
for γ-butyrolactone. This configuration holds for clusters of up to four
4. Complexation by Cryptands in Acetone as Solvent
101
molecules and also in the solid phase, where a tetrahedral first solvation
(coordination) sphere was found (see Table 4.4).66
Table 4.4. Calculated bond length for Li+ solvated in actone.
Bond length [Å] PG B3LYP/D95Vp B3LYP/6-311+G**
[Li(acetone)]+ C2 1.77 1.75
[Li(acetone)2]+ D2 1.81 1.79
[Li(acetone)3]+ D3 1.88 1.86
[Li(acetone)4]+ C1 1.96 (averaged) 1.96 (averaged)
Experimentally, a decrease in the cavity size of the cryptands hampers the
exchange of Li+ between the cryptand and the solvent. We were able to observe
it in the case of γ-butyrolactone and again we can see it in the present case. The
improved binding of Li+ in a smaller cavity is visible from the structure and the
following model reaction [Equation (4.4)] that describes the energy required
for the dechelation process. Application of the CPCM solvent model
strengthens this trend significantly (see Table 4.5).
[Li ⊂ Cryptand]+ + 4 solvent [Li(solvent)4]+ + Cryptand (4.4)
In agreement with previous studies by Wipff et al.,5 we also found only a
limited flexibility for the investigated cryptands in our DFT calculations,
therefore all cryptands were calculated within the highest possible symmetry
(Table 4.5). We recently demonstrated that reaction (4.4) is a valuable tool to
determine ion selectivity as a function of the ionic radii88 and to examine cavity
size of the cryptands as a variable.42
4. Complexation by Cryptands in Acetone as Solvent
102
Table 4.5. Dechelation energy, Edechelation, for the reaction outlined in Equation (4.4).[a]
Acetone C222 C221 C211
Edechelation (DFT) [kcal/mol] -8.6 +2.0 +3.0
Edechelation (CPCM) [kcal/mol] +5.4 +13.6 +20.4
γ-Butyrolactone C222 C221 C211
Edechelation (DFT) [kcal/mol]42 -16.8 -11.6 -5.3
[a] DFT: B3LYP/D95Vp, CPCM: B3LYP(CPCM)/D95Vp//B3LYP/D95Vp.
Table 4.6. Calculated (RB3LYP/D95Vp) and X-ray bond lengths (in Å).
d [Å] C222 C221 C211 [Li(OH2)4]+ [Li(acetone)4]+ [Li(NH3)4]+
Li-O
2.14
2.06
-
2.08
2.10
2.14
2.11
1.95
1.96
-.--
Li-N 2.22 2.28 2.38 -.-- -.-- 2.13
PG C3 C1 C2 T C1 C1
X-ray 89 90 91 92 93
Li-O
-.--
1.99
2.06
2.10
2.17
2.08
1.96
1.94, 1.91, 1.93,
1.95
-.--
Li-N -.-- 2.40
2.44
2.29 -.-- -.-- 2.08
In agreement with the experimental kinetic data, the smaller cryptand
binds Li+ better and slows down its exchange rate (Table 4.2). On comparing
the calculated Li-O and Li-N bond lengths of the three structures with the
bond length in [Li(OH2)4]+ and [Li(NH3)4]+, the trend can easily be
understood (see Table 4.6 and Figure 4.11). In C222, Li+ is four-fold weakly
4. Complexation by Cryptands in Acetone as Solvent
103
coordinated in a trigonal-pyramidal shape by one Li-N and three Li-O
interactions all clearly elongated (0.2 and 0.1 Å, respectively). The Li+ cation in
C221 is five-fold coordinated (distorted trigonal-bipyramidal structure) by one
nitrogen and four oxygen bonds all extended by around 0.15 Å. The
coordinating bonds in [Li ⊂ C211]+ are even more extended, but the Li-O bonds
form a somewhat distorted tetrahedral coordination sphere for the lithium
monocation, while the stronger elongated Li-N bonds form a kind of second
coordination sphere. Therefore, the lithium ion can formally adopt a six-
coordinate state, which strongly supports the proposed associative character of
the decomplexation process that involves binding of external solvent molecules
as suggested by the very negative activation entropies. A comparison of the
calculated and X-ray bond lengths shows a good agreement. A comparison
between the calculated data for γ-butyrolactone and acetone shows the same
trend for both solvents. Li+ is better coordinated in the smaller cavities of the
smaller cryptands, as expected.
To draw a conclusion on solvation abilities from a direct comparison of the
dechelating energies in acetone and γ-butyrolactone, is risky. The calculated
dechelating energies can not be considered as a proof of the rule of Madrakian
et al., that there is an inverse relationship between the stability of the
complexes and the solvating ability of the solvent.66 The different values can be
attributed to the different number of non-hydrogen atoms in both solvents.
Acetone has only four non-hydrogen atoms, while γ-butyrolactone has six and
can therefore stabilize charge much better.
4. Complexation by Cryptands in Acetone as Solvent
104
Figure 4.13. Optimized structures for [Li ⊂ C222]+, [Li ⊂ C221]+, [Li ⊂ C211]+ and
[Li(Acetone)4]+.
4.6. Conclusions
It can be concluded from the results of the present study that for the Li+
exchange process between C222, C221 and acetone as solvent, two pathways,
viz. bimolecular (associative) and unimolecular (dissociative) compete with
each other. The pressure dependence of the exchange process suggests the
operation of a pure interchange mechanism, whereas the negative activation
entropy values favor an associative interchange process for the decomplexation
reaction. The basic idea of an interchange process involves concerted bond
4. Complexation by Cryptands in Acetone as Solvent
105
formation and bond cleavage, which is in agreement with the consideration
that acetone is a rather weak donor. The results are in good agreement with
that expected on the basis of a systematic comparison with data reported in the
literature for solvents of similar donor strength.
4. Complexation by Cryptands in Acetone as Solvent
106
4.7. References
1. Dietrich, B.; Lehn, J. M.; Sauvage, J. P. Tetrahedron Lett. 1969, 34,
2889.
2. Dietrich, B.; Lehn, J. M.; Sauvage, J. P. Tetrahedron Lett. 1969, 34,
2885.
3. Lehn, J. M. Acc. Chem. Res. 1978, 11, 49.
4. Lehn, J. M. Angew. Chem. 1988, 100, 91.
5. Wipff, G.; Wurtz, J. M. New J. Chem. 1989, 13, 807.
6. Zhang, X. X.; Izatt, R. M.; Krakowiak, K. E.; Bradshaw, J. S. Inorg.
Chim. Acta 1997, 254, 43.
7. Lehn, J. M.; Montavon, F. Helv. Chim. Acta 1976, 59, 1566.
8. Puchta, R.; Seitz, V.; van Eikema Hommes, N. J. R.; Saalfrank, R. W. J.
Mol. Model. 2000, 6, 126.
9. Saalfrank, R. W.; Dresel, A.; Seitz, V.; Trummer, S.; Hampel, F.;
Teichert, M.; Stalke, D.; Stadler, C.; Daub, J.; Schunemann, V.;
Trautwein, A. X. Chem.-Eur. J. 1997, 3, 2058.
10. Saalfrank, R. W.; Seitz, V.; Caulder, D. L.; Raymond, K. N.; Teichert, M.;
Stalke, D. Eur. J. Inorg. Chem. 1998, 9, 1313.
11. Saalfrank, R. W.; Seitz, V.; Heinemann, F. W.; Gobel, C.; Herbst-Irmer,
R. J. Chem. Soc., Dalton Trans. 2001, 5, 599.
12. Dietrich, B.; Kintzinger, J. P.; Lehn, J. M.; Metz, B.; Zahidi, A. J. Phys.
Chem. 1987, 91, 6600.
13. Lindoy, L. F. The Chemistry of Macrocyclic Ligand Complexes;
Cambridge University Press: Cambridge, 1992; p 269.
14. Farahbakhsh, M.; Schmidt, H.; Rehder, D. Chem. Ber. Recl. 1997, 130,
1123.
15. Housecroft, C. E. Cluster Compounds of Main Group Elements; VCH:
Weinheim, Germany, 1996; p 94.
16. Kirch, M.; Lehn, J. M. Angew. Chem. 1975, 87, 542.
17. Lehn, J. M.; Montavon, F. Helv. Chim. Acta 1978, 61, 67.
18. Lehn, J. M.; Sauvage, J. P. J. Chem. Soc., Chem. Commun. 1971, 9, 440.
4. Complexation by Cryptands in Acetone as Solvent
107
19. Morf, W. E.; Simon, W. Helv. Chim. Acta 1971, 54, 2683.
20. Pregel, M. J.; Buncel, E. J. Am. Chem. Soc. 1993, 115, 10.
21. Simon, W.; Morf, W. E.; Meier, P. C. Struct. Bond. 1973, 16, 113.
22. Severin, K. Coord. Chem. Rev. 2003, 245, 3.
23. Huang, R. H.; Wagner, M. J.; Gilbert, D. J.; Reidy-Cedergren, K. A.;
Ward, D. L.; Faber, M. K.; Dye, J. L. J. Am. Chem. Soc. 1997, 119, 3765.
24. Lehn, J. M.; Sauvage, J. P. J. Am. Chem. Soc. 1975, 97, 6700.
25. Wipff, G.; Kollman, P. Nouv. J. Chim. 1985, 9, 457.
26. Cox, B. G.; Garcia-Rosas, J.; Schneider, H. J. Am. Chem. Soc. 1981, 103,
1054.
27. Rhinebarger, R. R.; Popov, A. I. Polyhedron 1988, 7, 1341.
28. Shamsipur, M.; Popov, A. I. J. Phys. Chem. 1986, 90, 5997.
29. Smetana, A. J.; Popov, A. I. J. Solution Chem. 1980, 9, 183.
30. Abou-Hamdan, A.; Lincoln, S. F. Inorg. Chem. 1991, 30, 462.
31. Clarke, P.; Gulbis, J. M.; Lincoln, S. F.; Tiekink, E. R. T. Inorg. Chem.
1992, 31, 3398.
32. Clarke, P.; Lincoln, S. F.; Tiekink, E. R. T. Inorg. Chem. 1991, 30, 2747.
33. Lincoln, S. F.; Abou-Hamdan, A. Inorg. Chem. 1990, 29, 3584.
34. Lincoln, S. F.; Rodopoulos, T. Inorg. Chim. Acta 1991, 190, 223.
35. Lincoln, S. F.; Rodopoulos, T. Inorg. Chim. Acta 1993, 205, 23.
36. Lincoln, S. F.; Stephens, A. K. W. Inorg. Chem. 1991, 30, 3529.
37. Stephens, A. K. W.; Dhillon, R. S.; Madbak, S. E.; Whitbread, S. L.;
Lincoln, S. F. Inorg. Chem. 1996, 35, 2019.
38. Izatt, R. M.; Bradshaw, J. S.; Nielsen, S. A.; Lamb, J. D.; Christensen, J.
J.; Sen, D. Chem. Rev. 1985, 85, 271.
39. Izatt, R. M.; Bradshaw, J. S.; Pawlak, K.; Bruening, R. L.; Tarbet, B. J.
Chem. Rev. 1992, 92, 1261.
40. Izatt, R. M.; Pawlak, K.; Bradshaw, J. S.; Bruening, R. L. Chem. Rev.
1991, 91, 1721.
41. Izatt, R. M.; Pawlak, K.; Bradshaw, J. S.; Bruening, R. L. Chemical
Reviews 1995, 95, 2529.
4. Complexation by Cryptands in Acetone as Solvent
108
42. Pasgreta, E.; Puchta, R.; Galle, M.; van Eikema Hommes, N.; van Eldik,
R. J. Inclusion Phenom. Macrocyclic Chem. 2007, 58, 81.
43. Sonoda, K.; Deguchi, M.; Koshina, H.; Armand, M.; Michot, C.;
Gauthier, M. Proc. - Electrochem. Soc. 2003, 2002-26, 512.
44. Takami, N.; Sekino, M.; Ohsaki, T.; Kanda, M.; Yamamoto, M. J. Power
Sources 2001, 97-98, 677.
45. Handy, S. T.; Grieco, P. A.; Mineur, C.; Ghosez, L. Synlett 1995, Spec.
Issue, 565.
46. Kinart, W. J.; Sniec, E.; Tylak, I.; Kinart, C. M. Phys. Chem. Liq. 2000,
38, 193.
47. Barthel, J.; Buchner, R.; Wismeth, E. J. Solution Chem. 2000, 29, 937.
48. Barthel, J.; Gores, H. J.; Neueder, R.; Schmid, A. Pure Appl. Chem.
1999, 71, 1705.
49. Croce, F.; D'Aprano, A.; Nanjundiah, C.; Koch, V. R.; Walker, C. W.;
Salomon, M. J. Electrochem. Soc. 1996, 143, 154.
50. Matsubara, K.; Kaneuchi, R.; Maekita, N. J. Chem. Soc., Faraday Trans.
1998, 94, 3601.
51. Alia, J. M.; Edwards, H. G. M. Vib. Spectrosc. 2000, 24, 185.
52. Himeno, S.; Takamoto, M.; Ueda, T.; Santo, R.; Ichimura, A.
Electroanalysis 2004, 16, 656.
53. Karkhaneei, E.; Zebrajadian, M. H.; Shamsipur, M. J. Inclusion Phenom.
Macrocyclic Chem. 2001, 40, 309.
54. Kumar, G.; Janakiraman; Namboodiri, N.; Gangadharan; Saminathan;
Kanagaraj; Sharief, N. J. Chem. Eng. Data 1991, 36, 467.
55. Hojo, M.; Hasegawa, H.; Hiura, N. J. Phys. Chem. 1996, 100, 891.
56. Puchta, R.; Galle, M.; Van Eikema Hommes, N.; Pasgreta, E.; Van Eldik,
R. Inorg. Chem. 2004, 43, 8227.
57. Spångberg, D.; Hermansson, K. Chem. Phys. 2004, 300, 165.
58. Alia, J. M.; Diaz de Mera, Y.; Edwards, H. G. M.; Garcia, F. J.; Lawson,
E. E. J. Mol. Struct. 1997, 408-409, 439.
59. Chabanel, M.; Legoff, D.; Touaj, K. J. Chem. Soc., Faraday Trans. 1996,
92, 4199.
4. Complexation by Cryptands in Acetone as Solvent
109
60. Hojo, M.; Hasegawa, H.; Morimoto, Y. Anal. Sci. 1996, 12, 521.
61. Barthel, J.; Neueder, R.; Poepke, H.; Wittmann, H. J. Solution Chem.
1999, 28, 489.
62. Kiselev, V. D.; Kashaeva, E. A.; Luzanova, N. A.; Konovalov, A. I.
Thermochim. Acta 1997, 303, 225.
63. Kameda, Y.; Kudoh, N.; Suzuki, S.; Usuki, T.; Uemura, O. Bull. Chem.
Soc. Jpn. 2001, 74, 1009.
64. Blint, R. J. J. Electrochem. Soc. 1995, 142, 696.
65. Jarek, R. L.; Miles, T. D.; Trester, M. L.; Denson, S. C.; Shin, S. K. J.
Phys. Chem. A 2000, 104, 2230.
66. Madrakian, T.; Afkhami, A.; Ghasemi, J.; Shamsipur, M. Polyhedron
1996, 15, 3647.
67. Zahl, A.; Neubrand, A.; Aygen, S.; van Eldik, R. Rev. Sci. Instrum. 1994,
65, 882.
68. Helm, L.; Borel, A. NMRICMA 2.7, University of Lausanne, 2000.
69. Matlab, version 5.3.1.; Mathworks Inc.: 1999.
70. Pople, J. A.; Schneider, W. G.; Bernstein, H. J. High-Resolution Nuclear
Magnetic Resonance; McGrow-Hill Book Co.: New York, 1959; p 513.
71. Becke, A. D. J. Chem. Phys. 1993, 98, 5648.
72. Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B: Condens. Matter Mater.
Phys. 1988, 37, 785.
73. Stephens, P. J.; Devlin, F. J.; Chabalowski, C. F.; Frisch, M. J. J. Phys.
Chem. 1994, 98, 11623.
74. Dunning, T. H.; Jr.; Hay, P. J. Modern Theoretical Chemistry; Plenum:
New York, 1976; Vol. 3, p 1.
75. Huzinaga, S.; Andzelm, J.; Klobukowski, M.; Radzio-Andzelm, E.; Sakai,
Y.; Tatewaki, H. Gaussian basis sets for molecular calculations;
Elsevier: Amsterdam, 1984.
76. The performance of the computational level employed in this study is
well documented, see for example: Galle, M.; Puchta, R.; van Eikema
Hommes, N.; van Eldik, R. Z. Phys. Chem. 2006, 220, 511; Puchta, R.;
van Eikema Hommes, N. J. R.; Meier, R.; van Eldik, R. Dalton Trans.
4. Complexation by Cryptands in Acetone as Solvent
110
2006, 28, 3392; Klaus, S.; Neumann, H.; Jiao, H.; Jacobi von Wangelin,
A.; Goerdes, D.; Struebing, D.; Huebner, S.; Hateley, M.; Weckbecker,
C.; Huthmacher, K.; Riermeier, T.; Beller, M. J. Organomet. Chem.
2004, 689, 3685; Saalfrank, R. W.; Deutscher, C.; Maid, H.; Ako, A. M.;
Sperner, S.; Nakajima, T.; Bauer, W.; Hampel, F.; Hess, B. A.; van
Eikema Hommes, N. J. R.; Puchta, R.; Heinemann, F. W. Chem.-Eur. J.
2004, 10, 1899; Scheurer, A.; Maid, H.; Hampel, F.; Saalfrank, R. W.;
Toupet, L.; Mosset, P.; Puchta, R.; van Eikema Hommes, N. J. R. Eur. J.
Org. Chem. 2005, 12, 2566; Weber, C. F.; Puchta, R.; van Eikema
Hommes, N. J. R.; Wasserscheid, P.; van Eldik, R. Angew. Chem. 2005,
117, 6187; Weber, C. F.; Puchta, R.; van Eikema Hommes, N. J. R.;
Wasserscheid, P.; van Eldik, R. Angew. Chem. (Int. Ed. Eng.) 2005, 44,
6033.
77. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.;
Cheeseman, J. R.; Montgomery, J. A.; Jr.; Vreven, T.; Kudin, K. N.;
Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.;
Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.;
Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa,
J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.;
Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Bakken, V.; Adamo, C.;
Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.;
Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.;
Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich,
S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.;
Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.;
Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.;
Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham,
M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.;
Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian
03, Revision B.03, Gaussian Inc.; Wallingford, CT, 2004.
78. Shchori, E.; Jagur-Grodzinski, J.; Shporer, M. J. Am. Chem. Soc. 1973,
95, 3842.
4. Complexation by Cryptands in Acetone as Solvent
111
79. Shchori, E.; Jagur-Grodzinski, J.; Luz, Z.; Shporer, M. J. Am. Chem. Soc.
1971, 93, 7133.
80. Marcus, Y. Chem. Soc. Rev. 1993, 22, 409.
81. Shamsipur, M.; Karkhaneei, E.; Afkhami, A. J. Coord. Chem. 1998, 44,
23.
82. Davis, A. V.; Fiedler, D.; Seeber, G.; Zahl, A.; Van Eldik, R.; Raymond, K.
N. J. Am. Chem. Soc. 2006, 128, 1324.
83. van Eldik, R.; Hubbard, C. D. Chemistry at Extreme Conditions;
Elsevier: Amsterdam, 2005; p 109.
84. Wolak, M.; van Eldik, R. J. Am. Chem. Soc. 2005, 127, 13312.
85. Zahl, A.; Van Eldik, R.; Matsumoto, M.; Swaddle, T. W. Inorg. Chem.
2003, 42, 3718.
86. Helm, L.; Merbach, A. E. Chem. Rev. 2005, 105, 1923.
87. Masia, M.; Rey, R. J. Phys. Chem. B 2004, 108, 17992.
88. Galle, M.; Puchta, R.; van Eikema Hommes, N. J. R.; van Eldik, R. Z.
Phys. Chem. 2006, 220, 511.
89. Choua, S.; Sidorenkova, H.; Berclaz, T.; Geoffroy, M.; Rosa, P.;
Mezailles, N.; Ricard, L.; Mathey, F.; Le Floch, P. J. Am. Chem. Soc.
2000, 122, 12227.
90. Moras, D.; Weiss, R. Acta Crystallogr.,Sect. B: Struct. Crystallogr.
Cryst. Chem. 1973, 29, 400.
91. Meske, W.; Babel, D. Z. Anorg. Allg. Chem. 1998, 624, 1751.
92. Cotton, F. A.; Daniels, L. M.; Murillo, C. A.; Zhou, H. C. Inorg. Chim.
Acta 2000, 300-302, 319.
93. Scotti, N.; Zachwieja, U.; Jacobs, H. Z. Anorg. Allg. Chem. 1997, 623,
1503.
5. Complexation by Cryptands in γ-Butyrolactone as Solvent
112
5. Ligand exchange processes on solvated
lithium cations. Complexation by Cryptands
in γ-Butyrolactone as Solvent
5.1. General remark
The work presented in this chapter was conducted by several authors within a
joint project and has been published as:
Pasgreta, E.; Puchta, R.; Galle, M.; van Eikema-Hommes N.; Zahl A.; van Eldik
R. Journal of Inclusion Phenomena and Macrocyclic Chemistry 2007, 58, 81-
88.
5.2. Abstract
Kinetic studies on Li+ exchange between the cryptands C222 and C221 and
γ-butyrolactone as solvent were performed as a function of ligand-to-metal
ratio, temperature and pressure using 7Li NMR. The thermal rate and
activation parameters are: C222: k298 = (3.3 ± 0.8) x 104 M-1 s-1, ∆H# = 35 ± 1
kJ mol-1 and ∆S# = -41 ± 3 J K-1 mol-1; C221: k298 = 105 ± 32 M-1 s-1, ∆H# = 48 ±
1 kJ mol-1 and ∆S# = -45 ± 2 J K-1 mol-1. Temperature and pressure dependence
measurements were performed in the presence of an excess of Li+. The
influence of pressure on the exchange rate is insignificant for both ligands,
such that the value of activation volume is around zero within the experimental
error limits. The activation parameters obtained in this study indicate that the
exchange of Li+ between solvated and chelated Li+ ions follows an associative
interchange mechanism.
5. Complexation by Cryptands in γ-Butyrolactone as Solvent
113
5.3. Introduction
Following the work of Pedersen on two-dimensional crowns, Lehn and
coworkers developed in 1969 a range of three-dimensional, polycyclic ligand
systems which they named cryptands (Greek: cryptos = cave).1, 2 Until today
this class of compounds has been intensively explored and extended in
different directions3, 4 in terms of cavity size,5, 6 donor atoms,7 and metalla-
topomers.8-11 Since cryptands readily form complexes with a wide range of
metal ions and NH4+,12 provided the ion involved is not too large to be
contained in the macrocyclic cavity,13 they find their application in various
areas like catalysis, medicine, biomimetic research,14-21 and as receptors22 and
electrides.23 Studies on the complexes (cryptates) formed between alkali metal
ions and cryptands, have revealed a substantial understanding of the effects of
metal ion size and cryptand cavity size on the structure, stability and lability of
cryptates, and the mechanism of the cryptate complex-formation processes.24-
28 Such selective complexation has led to many biological and chemical studies.
Kinetic studies on macrocyclic complex-formation reactions with alkali
cations not only result in important information on the rates and mechanisms
of complex-formation reactions, but also lead to a better understanding of the
high selectivity of these ligands toward different cations. The broad impact of
such studies spans from the theory of complexation to cation transport
through membranes, and makes the ability to control cation selectivity via
tactical structural changes in the complexing agent of prime importance.18
Polyoxo macrobicyclic diamines, represented by the structure in Figure 5.1, are
able to form metal ion complexes in which the metal ion is located in the cavity
of the macromolecule and two nitrogen atoms along with the oxygen atoms
participate in binding the metal cation. All complexes have 1:1 stoichiometry
with the metal ion positioned in the center of the ligand cavity.24
5. Complexation by Cryptands in γ-Butyrolactone as Solvent
114
O
N
OO
O
N O O
m
m
n Figure 5.1. Chelate structures. C211: m = 0, n = 1. C221: m = 1, n = 0. C222: m = 1, n = 1.
In case of crown ethers and cryptands in non-aqueous solutions, the rates
of complex-formation (kf) for the interaction with alkali-metal cations are
generally diffusion-controlled and, consequently, the complexation
selectivities are governed by the decomplexation rates kd.29 These reactions
have received significant attention, initially from Lehn and co-workers and
more recently from Popov et al.30-32 and Lincoln et al..33-40 These and other
groups performed a series of detailed kinetic studies and found that two
mechanisms need to be considered for the decomplexation process of
cryptand-alkali metal cation complexes, viz., bimolecular associative (I) and
unimolecular dissociative (back reaction in II) cation exchange processes,
where S presents solvent and L the employed chelate.41, 42
The solvent plays a major role in the unimolecular mechanism II,43, 44
which is therefore favored in strong donor solvents. Many examples have been
included in a series of review articles.41, 45-47 It has been reported that ligand
structure and solvent properties have considerable effects on the reaction
rates, activation parameters, and exchange mechanism of cations between the
k1
k-1
*Li(S)+ + Li(L)+ Li(S)+ + *Li(L)+ (I)
k2
k-2
Li(S)+ + L Li(L)+ + S (II)
k1
k-1
*Li(S)+ + Li(L)+ Li(S)+ + *Li(L)+ (I)
k2
k-2
Li(S)+ + L Li(L)+ + S (II)
5. Complexation by Cryptands in γ-Butyrolactone as Solvent
115
solvated and complexed sites. We have now extended the study of Li+ exchange
to γ-butyrolactone (see Figure 5.2) as solvent.
O
O
Figure 5.2. Structure of γ-butyrolactone.
γ-Butyrolactone was chosen as solvent due to its relevance in lithium
batteries and capacitors.48, 49 It is a model cyclic ester used as a major chemical
compound with extensive application in pharmaceuticals, pesticides and
petrochemicals. The butyrolactone ring is an integral building block of many
natural products of biological activity,50, 51 like the sesquiterpene lactones,
flavor components, alkaloids, antileukemics, and pheromones. γ-Butyrolactone
is one of the important intermediates in fine chemical industrial practices, for
example in the synthesis of pyrrolidine, herbicides and rubber additives.52 It is
also used as a solvent for surface treatment of textiles and metal coated
plastics, as a paint remover, in photochemical etching, in vitamin and
pharmaceutical preparations.
Recently, computational studies on γ-butyrolactone and Li+/γ-
butyrolactone have been performed by Rey et al..53 It was found that the Li+
cation coordinates to the carbonyl oxygen with an almost co-linear
configuration relative to the carbon-oxygen bond but with a slight tilting
toward the lactone oxygen. This configuration holds for clusters of up to four
molecules and also in the liquid phase, where a tetrahedral first solvation
(coordination) sphere was found.
In this contribution, we report kinetic parameters for the exchange of Li+ in
γ-butyrolactone under ambient and elevated pressure conditions, and discuss
their mechanistic implications.
5. Complexation by Cryptands in γ-Butyrolactone as Solvent
116
5.4. Experimental Section
5.4.1. General Procedures
The preparation of test solutions was carried out under an Ar or N2
atmosphere using standard Schlenk techniques, under an Ar atmosphere in a
glove box, or under vacuum. Lithium perchlorate (Aldrich, battery grade) was
vacuum-dried at 120 °C for 48 h and then stored under dry nitrogen. The
cryptands C222 (purchased from Fluka), C221 and C211 (purchased from
Aldrich) were dried under vacuum for 48 h and stored under dry nitrogen. γ-
Butyrolactone (Aldrich) was dried over anhydrous CaSO4, then fractionally
destilled under vacuum and stored under nitrogen over Linde 3Å molecular
sieves. Solutions of LiClO4, C222, C221 and C211 in γ-butyrolactone were
prepared under dry nitrogen. For the 7Li NMR measurements the solutions
were sealed in 5 mm NMR tubes under an Ar atmosphere. In order to study the
kinetics of the exchange of Li+ between C222 and C221 and γ-butyrolactone as
solvent, four solutions of different concentration ratio were prepared for each
system. The concentration of lithium perchlorate in each solution was kept
constant at 0.05 M, whereas the concentration of ligand was varied between
0.0175 M and 0.04 M. In this way the measurements were always performed in
the presence of an excess of Li+.
5.4.2. 7Li NMR Spectroscopy
Variable-temperature and variable-pressure Fourier transform 7Li NMR
spectra were recorded at a frequency of 155 MHz on a Bruker Avance DRX
400WB spectrometer equipped with a superconducting BC-94/89 magnet
system. 5 mm spinning tubes were used with a 1 mm o.d. melting point
capillary inserted coaxially in the spinning tube and filled with an external
reference solution (usually 1 M LiClO4 in DMF). The temperature dependence
measurements were performed over a wide as possible temperature range for
each system. A homemade high-pressure probe described in the literature 54
was used for the variable-pressure experiments which were conducted at a
selected temperature and at ambient, 30, 60, 90, 120 and 150 MPa pressure. A
5. Complexation by Cryptands in γ-Butyrolactone as Solvent
117
standard 5 mm NMR tube cut to a length of 45 mm was used for the sample
solution. The high-pressure NMR tube was sealed with a MACOR plug and an
O-ring made of VITON. The plug and O-ring were shown to be stable in the
employed solvent. The pressure was transmitted by the movable MACOR
piston to the sample, and the temperature was controlled as described
elsewhere.54 For a typical kinetic run, 0.8 ml of solution was transferred under
an Ar atmosphere to a Wilmad screw-cap NMR tube equipped with a
poly(tetrafluoroethylene) septum. All test samples were prepared in the same
way.
5.4.3. Programs
The line widths of the NMR signals were obtained by fitting Lorentzian
functions to the experimental spectra using the NMRICMA 2.7 program55 for
Matlab.56 The adjustable parameters were the resonance frequency, intensity,
line width and baseline. Complete line-shape analysis based on the Kubo-Sack
formalism using modified Bloch equations57 was also performed with
NMRICMA 2.7 to extract rate constants from experimental spectra. The
temperature and pressure dependent 7Li line widths and chemical shifts
employed in the line shape analysis were extrapolated from low temperatures
where no exchange-induced modification occurred.
5.4.4. Quantum-Chemical Calculations
All structures were optimized at the B3LYP/D95Vp hybrid DFT level 58-60 61,
62 with no constrains other than symmetry and confirmed to be minima by
computation of vibrational frequencies. Relative energies were corrected for
zero point vibrational energy differences. The D95V basis set, augmented with
polarisation functions (D95Vp), was used.63 The Gaussian suite of programs
was used throughout.64
5. Complexation by Cryptands in γ-Butyrolactone as Solvent
118
5.5. Results and Discussion
5.5.1. Spectroscopic Observations
In preliminary experiments, a series of 7Li NMR measurements for the Li+-
cryptand system in γ-butyrolactone as solvent were performed. For each of the
cryptands C222, C221 and C211, mole ratio dependent measurements were
performed by variation of both the cryptand:Li+ and Li+:cryptand molar ratios.
The results enabled us to observe the distribution of particular species on
increasing the amount of chelate or lithium salt. Basically, two signals of
varying intensity were observed in the 7Li NMR spectra that refer to solvated
Li+ and Li+ bound inside the cryptand cavity. In case of the smallest chelate
C211, however, an additional signal appeared which can be ascribed to Li+
bound outside the cryptand and being partially solvated. The distribution plot
for the various species in a solution of C211 and LiClO4 in γ-butyrolactone is
presented in Figure 5.3. The concentration of LiClO4 was kept constant at 0.01
M and the concentration of C211 was varied in the range 0 to 0.04 M. Initially
only the signal of free (solvated) Li+ is observed, and then with increasing
amount of C211, the signals referring to the complexed Li+ appear. Contrary to
the larger cryptands, there are two signals that refer to bound Li+; one for Li+
inside the cavity and another of low intensity (max. 15 %) for Li+ presumably
bound outside the cryptand and being partially solvated. The latter signal,
however, disappears once the C211/LiClO4 molar ratio gets to 1, and then all
the Li+ is bound as internal complex. Due to the facts that the exchange
process among the three species cannot be clearly defined, Li(C211)+ is very
stable and there is almost no exchange taking place, a detailed study of the
observed rate constant, kobs, for the exchange of Li+ as a function of
temperature and pressure was undertaken only for C221 and C222. The
lithium exchange reaction between cryptand and γ-butyrolactone was studied
for four solutions with different ligand-to-metal ratios in the presence of an
excess of Li+.
5. Complexation by Cryptands in γ-Butyrolactone as Solvent
119
0.00 0.01 0.02 0.03 0.04
0
20
40
60
80
100
free lithium internal complex external complex
%
[C211], M
Figure 5.3. Distribution plot for various species in a solution of C211 and LiClO4 in γ-
butyrolactone.
Three signals can be observed in Figure 5.4; one comes from the reference
(6.6 ppm) and the other two from the exchanging sites (6.0 ppm solvated Li+
and 4.8 ppm complexed Li+). At low temperature the signals are narrow and
completely separated due to very slow or no exchange. Since the formation
constant for Li(C221)+ is very high,24 the concentration of bound lithium is
equal to the amount of cryptand added. At ca. 358 K the signals start to
coalescence and at 388 K only one broad signal (at 5.3 ppm) can be observed
due to fast exchange. In the case of C222 (Figure 5.5), the exchange process is
faster due to the larger cavity of the cryptand. The signals that refer to solvated
and complexed Li cation (8.3 ppm and 6.8 ppm, respectively at 218 K)
coalescence at a lower temperature (ca. 268 K), and at high temperature (368
K) only one sharp signal at 7.7 ppm was observed as a consequence of a very
fast exchange process. An accurate temperature dependence for C222 over a
large temperature range can be observed in Figure 5.7.
5. Complexation by Cryptands in γ-Butyrolactone as Solvent
120
Figure 5.4. 7Li NMR spectra recorded as a function of temperature for the mole ratio C221:LiClO4 = 0.65.
Figure 5.5. 7Li NMR spectra recorded as a function of temperature for the mole ratio C222:LiClO4 = 0.65.
The rate of the exchange reaction increased not only with increasing cavity
size of the cryptand but also with increasing concentration of the chelate, as
can be recognized from the shape of the signals at 338 K in Figure 5.6 and from
the values of kobs in Table 5.1.
5. Complexation by Cryptands in γ-Butyrolactone as Solvent
121
Figure 5.6. 7Li NMR spectra of C222 and LiClO4 in GBL. Temperature dependence for the sample ratios C222:LiClO4 = 0.35, 0.45, 0.55 and 0.65.
Figure 5.7. Eyring plots for kobs as a function of temperature for C222 for the concentration ratios C222:LiClO4 = 0.35, 0.45, 0.55 and 0.65, respectively.
6.57.07.58.0(ppm)
6.57.07.58.08.5(ppm)
6.57.07.58.08.5(ppm)
6.57.07.58.08.5(ppm)
248 K
268 K
283 K
298 K
318 K
338 K
6.57.07.58.0(ppm)
6.57.07.58.0 6.57.07.58.0(ppm)
6.57.07.58.08.5(ppm)
6.57.07.58.08.5 6.57.07.58.08.5(ppm)
6.57.07.58.08.5(ppm)
6.57.07.58.08.5 6.57.07.58.08.5(ppm)
6.57.07.58.08.5(ppm)
6.57.07.58.08.5 6.57.07.58.08.5(ppm)
248 K
268 K
283 K
298 K
318 K
338 K
0,0026 0,0028 0,0030 0,0032 0,0034 0,0036 0,0038 0,0040 0,0042 0,0044-4
-2
0
2
4 Y = A + B * X
Parameter Value Error------------------------------------------------------------A 14,22877 0,21476B -4047,99144 63,1407-----------------------------------------------------------
ln k
/T
1/T0,0026 0,0028 0,0030 0,0032 0,0034 0,0036 0,0038 0,0040 0,0042
-2
0
2
4
Y = A + B * X
Parameter Value Error------------------------------------------------------------A 16,71606 0,11624B -4674,27696 34,83709----------------------------------------------------------
ln k
/T
1/T
0,0026 0,0028 0,0030 0,0032 0,0034 0,0036 0,0038 0,0040 0,0042
-2
0
2
4
Y = A + B * X
Parameter Value Error------------------------------------------------------------A 16,78361 0,19905B -4604,44033 59,65542------------------------------------------------------------
ln k
/T
1/T0,0026 0,0028 0,0030 0,0032 0,0034 0,0036 0,0038 0,0040 0,0042 0,0044
-2
0
2
4
Y = A + B * X
Parameter Value Error------------------------------------------------------------A 16,70946 0,21618B -4530,05129 63,5587------------------------------------------------------------
ln k
/T
1/T
0,0026 0,0028 0,0030 0,0032 0,0034 0,0036 0,0038 0,0040 0,0042 0,0044-4
-2
0
2
4 Y = A + B * X
Parameter Value Error------------------------------------------------------------A 14,22877 0,21476B -4047,99144 63,1407-----------------------------------------------------------
ln k
/T
1/T0,0026 0,0028 0,0030 0,0032 0,0034 0,0036 0,0038 0,0040 0,0042
-2
0
2
4
Y = A + B * X
Parameter Value Error------------------------------------------------------------A 16,71606 0,11624B -4674,27696 34,83709----------------------------------------------------------
ln k
/T
1/T
0,0026 0,0028 0,0030 0,0032 0,0034 0,0036 0,0038 0,0040 0,0042
-2
0
2
4
Y = A + B * X
Parameter Value Error------------------------------------------------------------A 16,78361 0,19905B -4604,44033 59,65542------------------------------------------------------------
ln k
/T
1/T0,0026 0,0028 0,0030 0,0032 0,0034 0,0036 0,0038 0,0040 0,0042 0,0044
-2
0
2
4
Y = A + B * X
Parameter Value Error------------------------------------------------------------A 16,70946 0,21618B -4530,05129 63,5587------------------------------------------------------------
ln k
/T
1/T
5. Complexation by Cryptands in γ-Butyrolactone as Solvent
122
Table 5.1. Values of the observed rate constants (kobs) calculated from the line shape analysis for the exchange of Li+ between solvated and chelated sites at 25°C.
Mole ratio C221:Li+ kobs, s-1 Mole ratio C222:Li+ kobs, s-1
0.4 2.68 ± 0.05 0.35 654 ± 13
0.5 4.3 ± 0.1 0.45 851 ± 25
0.65 6.9 ± 0.3 0.55 1282 ± 51
0.8 9.6 ± 0.5 0.65 1352 ± 68
5.5.2. Suggested Exchange Mechanism
As mentioned above, there are two possible mechanisms for the exchange
of the lithium ion between the solvated and the complexed sites, namely
bimolecular associative (I) and unimolecular dissociative (II) mechanisms.
At equilibrium
]Li(S)][Li(L)[*]Li(L)][Li(S)[* 11 −= kk
]S][Li(L)[]L][Li(S)[ 22 −= kk
Therefore
]S][Li(L)[]Li(L)][Li(S)[]Li(S)[]Li(S)[ 21 −+==− kkkdtd
obs (5.1)
and
]Li(S)[
]S][Li(L)[]Li(L)[ 2
1−+=
kkkobs (5.2)
An alternative form of Equation (5.2) is Equation (5.3), which was used to fit
the data in Figure 5.8 and Figure 5.10.
]Li(S)['
]Li(L)[1
21 −+= kkkobs (5.3)
5. Complexation by Cryptands in γ-Butyrolactone as Solvent
123
Plots of kobs/[Li(L)] vs. 1/[Li(S)] were found to be linear with negligible
slopes (k-2’ = k-2[S]) (especially at low temperature) and significant intercepts
(k1) (see Figures 5.8 and 5.10). This indicates that the exchange mechanism
has a more associative character for both C221 and C222. This led us to take
only the values of k1 into consideration, from which it follows that the exchange
process is much faster for C222 than for C221. The corresponding Eyring plots
for k1 are shown in Figures 5.9 and 5.11, respectively, and the thermal
activation parameters are summarized in Table 5.2.
30 35 40 45 50 550
5000
10000
15000
20000
25000
30000
35000
40000
45000
50000
k obs/[
Li(L
)], M
-1s-1
1/[Li(S)], M-1
248K 258K 268K 278K 283K 288K 298K
Figure 5.8. Plots of kobs/[Li(L)] vs 1/[Li(S)] for C222 as a function of temperature.
5. Complexation by Cryptands in γ-Butyrolactone as Solvent
124
0.0030 0.0032 0.0034 0.0036 0.0038 0.0040 0.0042
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0 Y = A + B * X
Parameter Value Error------------------------------------------------------------A 18.85629 0.38719B -4237.51545 108.61204------------------------------------------------------------
ln (k
1/T)
1/T, K-1
Figure 5.9. Eyring plot for k1 as a function of temperature for C222.
30 40 50 60 70 80 90 1000
500
1000
1500
2000
2500
3000
3500
4000
4500
k obs/[L
i(L)],
M-1s-1
1/[Li(S)], M-1
298K 308K 318K 328K 338K 348K 358K
Figure 5.10. Plots of kobs/[Li(L)] vs 1/[Li(S)] for C221 as a function of temperature.
5. Complexation by Cryptands in γ-Butyrolactone as Solvent
125
0.0027 0.0028 0.0029 0.0030 0.0031 0.0032 0.0033 0.0034-1.5
-1.0
-0.5
0.0
0.5
1.0
1.5
2.0
2.5 Y = A + B * X
Parameter Value Error------------------------------------------------------------A 18.36172 0.29829B -5766.54595 97.29293------------------------------------------------------------
ln (k
1/T)
1/T, K-1
Figure 5.11. Eyring plot for k1 as a function of temperature for C221.
Table 5.2. Thermal activation parameters obtained from the values of k1 as a function of temperature.
Chelate k1298, M-1 s-1 ∆H#, kJ mol-1 ∆S#, J K-1 mol-1
C221 105 ± 32 48 ± 1 -45 ± 2
C222 (3.3 ± 0.8) x 104 35 ± 1 -41 ± 3
Based on the significantly negative activation entropy, the activation
parameters support an associative interchange type of exchange process.
Furthermore, the reported activation parameters are within the range of those
reported by Popov et al. for closely related systems,31 viz. k298 = 680 ± 42 M-1 s-
1, ∆H# = 28 ± 1 kJ mol-1 and ∆S# = -101 ± 2 J K-1 mol-1 for Li+/C221 in
acetonitrile, and k298 = 892 ± 50 M-1 s-1, ∆H# = 26 ± 1 kJ mol-1 and ∆S# = -109 ±
2 J K-1 mol-1 for Li+/C221 in propylene carbonate, for which the authors also
postulated an associative character for the Li+ exchange process. In the case of
Li+ exchange between C222 and solvents such as acetonitrile, propylene
carbonate and acetone, these authors found the mechanism to have more of a
dissociative character. However, it should be pointed out that they used a
different rate law than the one proposed by Shchori et al.65 and that used in the
present study, which complicates a direct comparison of the reported
5. Complexation by Cryptands in γ-Butyrolactone as Solvent
126
activation parameters with ours. It is reasonable to expect that solvents such as
propylene carbonate, ethylene carbonate and γ-butyrolactone, together
considered as a class of solvents that play an important role in polymer-gel
lithium-ion baterries,66-70 due to their similar structure and properties as
expected from their rather similar Gutmann donor numbers (DN = 15.1, 16.4
and 18.0, respectively71), should coordinate to Li+ in a similar way and should
have a similar influence on the exchange mechanism. This observation
corresponds to that reported by Shamsipur et al.72 for the exchange of Li+
between C221 and acetonitrile-nitromethane mixtures. In all solvent mixtures
used, Li+ exchange was found to occur via an associative mechanism, as
expected for weakly donor solvents (DN = 14.1 and 2.7 for acetonitrile and
nitromethane, respectively). The same group reported another case, where Li+
exchange occurred between C221 and mixtures of nitromethane (low donor
ability) and dimethylformamide (high DN = 26.6), where it was found that the
associative mechanism predominates in nitromethane, whereas a dissociative
pathway prevails in dimethylformamide solution.
5.5.3. High Pressure 7Li NMR Measurements
Additional mechanistic information on the intimate nature of the exchange
mechanism can in principle be obtained from the effect of pressure on the Li+
exchange process73-76. A systematic pressure dependence study was performed
to determine the volume of activation for the Li+ exchange process for both
chelates. Pressure dependent NMR spectra were recorded for each sample over
the pressure range of 0.2 to 150 MPa at 316.5 K. The effect of pressure was
found to be not very significant for these systems. There was a very small
difference in the shape of the NMR spectra recorded at various pressures as
can be seen in Figure 5.12. The rate constants calculated for each pressure were
very similar within the experimental error limits and resulted in a practically
zero value for ∆V#obs, which points to a concerted interchange type of exchange
mechanism.77 In terms of an interchange mechanism, bond formation and
bond cleavage occur in a concerted fashion without the formation of a distinct
intermediate of higher or lower coordination number.
5. Complexation by Cryptands in γ-Butyrolactone as Solvent
127
Figure 5.12. 7Li NMR spectra recorded as a function of pressure at 316.5 K for the mole
ratio C221:LiClO4 = 0.65.
5.5.4. Influence of Water
The rate of the exchange process was also investigated in a mixture of γ-
butyrolactone and water in order to probe the impact of the solvent polarity
and donor ability on the rate of the exchange reaction. The exchange
measurements were repeated for one of the samples (C222:LiClO4 = 0.35) with
1μl H2O added to the 0.8 ml of γ-butyrolactone solution. It was observed that
the exchange rate increased significantly as compared to the value obtained for
the sample in the absence of water. The values of kobs increase from 27 to 75 s-1
at 248 K, from 654 to 1282 s-1 at 298 K, and from 5945 to 12895 s-1 at 358 K, on
addition of water. This can be accounted for by the higher donor strength of
water compared to γ-butyrolactone. It follows that the presence of water plays
an important role in the exchange rate of Li+ between cryptand and solvent.
5.5.5. Theoretical Calculations
Experimentally a decrease in the cavity size hampers the exchange of Li+
between the cryptand and the solvent. The improved binding of Li+ in a
smaller cavity should become visible from the structure and the following
model reaction that describes the energy required for the dechelation process
(Table 5.3).
5. Complexation by Cryptands in γ-Butyrolactone as Solvent
128
[Li⊂Cryptand]+ + 4γ-Butyrolactone [Li(γ-Butyrolactone)4]+ + Cryptand (5.4)
It is noteworthy that in agreement with previous studies by Wipff et al.,5 we
also found only a limited flexibility for the investigated cryptands in our DFT
calculations, justifying that all cryptands were calculated within the highest
possible symmetry. In earlier work we demonstrated that eq. (5.4) is a valuable
tool to determine ion selectivity as a function of the ionic radii.78 We therefore
examined the cavity size of the cryptands as a variable.
Table 5.3. Dechelation energy, Edechelation, for the reaction outlined in (5.4).
γ-Butyrolactone C222 C221 C211
Edechelation [kJ/mol] -70.3 -48.6 -22.2
Table 5.4. Calculated (RB3LYP/D95Vp) and x-ray bond lengths (in Å).
d [Å] C222 C221 C211 [Li(OH2)4]+ [Li(γ-butyrolactone)4]+ [Li(NH3)4]+
Li-O
2.14
2.06
-
2.08
2.10
2.14
2.11
1.95
1.94
-.--
Li-N 2.22 2.28 2.3
8
-.-- -.-- 2.13
PG C3 C1 C2 T C1 C1
X-ray 79 80 81 82
Li-O
-.--
1.99
2.06
2.10
2.17
2.0
8
1.96
-.--
-.--
Li-N -.-- 2.40
2.44
2.29 -.-- -.-- 2.08
In agreement with the experimental kinetic data, the smaller cryptand
binds Li+ better and slows down its exchange rate (Table 5.2). On comparing
5. Complexation by Cryptands in γ-Butyrolactone as Solvent
129
the calculated Li-O and Li-N bond lengths of the three structures with the
bond distance in [Li(OH2)4]+ and [Li(NH3)4]+, the trend can easily be
understood (see Table 5.4 and Figure 5.13). In C222, Li+ is four times weakly
coordinated in a trigonal pyramidal shape by one Li-N and three Li-O
interactions all clearly elongated (0.2 and 0.1 Å, respectively). The Li+ cation in
C221 is coordinated five times (distorted trigonal bipyramidal structure) by
one nitrogen and four oxygen bonds all extended by around 0.15 Å. The
coordinating bonds in [Li ⊂ C211]+ are even more extended, but the Li-O bonds
form a somewhat distorted tetrahedral coordination sphere for the lithium
mono cation, while the stronger elongated Li-N bonds form a kind of second
coordination sphere. Therefore, the lithium ion is formally six-coordinate. A
comparison of the calculated and x-ray bond lengths shows a good agreement.
Figure 5.13. Optimized structures for [Li ⊂ C222]+, [Li ⊂ C221]+, [Li ⊂ C211]+ and [Li(γ-Butyrolactone)4]+.
5. Complexation by Cryptands in γ-Butyrolactone as Solvent
130
5.6. Conclusions
It can be concluded from the results of the present study that for the Li+
exchange process between C222, C221 and γ-butyrolactone as solvent, two
pathways, viz. bimolecular associative and unimolecular dissociative, compete
with each other. The pressure dependence of the exchange process suggests the
operation of a pure interchange mechanism, whereas the negative activation
entropy values favor an associative interchange process. The basic idea of an
interchange process involves concerted bond-formation and bond-cleavage,
which is in agreement with the consideration that γ-butyrolactone is a rather
weak donor. The results are in good agreement with that expected on the basis
of a systematic comparison with data reported in the literature for solvents of
similar donor strength.
5. Complexation by Cryptands in γ-Butyrolactone as Solvent
131
5.7. References
1. Dietrich, B.; Lehn, J. M.; Sauvage, J. P. Tetrahedron Lett. 1969, 34,
2889.
2. Dietrich, B.; Lehn, J. M.; Sauvage, J. P. Tetrahedron Lett. 1969, 34,
2885.
3. Lehn, J. M. Acc. Chem. Res. 1978, 11, 49.
4. Lehn, J. M. Angew. Chem. 1988, 100, 91.
5. Wipff, G.; Wurtz, J. M. New J. Chem. 1989, 13, 807.
6. Zhang, X. X.; Izatt, R. M.; Krakowiak, K. E.; Bradshaw, J. S. Inorg.
Chim. Acta 1997, 254, 43.
7. Lehn, J. M.; Montavon, F. Helv. Chim. Acta 1976, 59, 1566.
8. Puchta, R.; Seitz, V.; van Eikema Hommes, N. J. R.; Saalfrank, R. W. J.
Mol. Model. 2000, 6, 126.
9. Saalfrank, R. W.; Dresel, A.; Seitz, V.; Trummer, S.; Hampel, F.;
Teichert, M.; Stalke, D.; Stadler, C.; Daub, J.; Schunemann, V.;
Trautwein, A. X. Chem.-Eur. J. 1997, 3, 2058.
10. Saalfrank, R. W.; Seitz, V.; Caulder, D. L.; Raymond, K. N.; Teichert, M.;
Stalke, D. Eur. J. Inorg. Chem. 1998, 9, 1313.
11. Saalfrank, R. W.; Seitz, V.; Heinemann, F. W.; Gobel, C.; Herbst-Irmer,
R. J. Chem. Soc., Dalton Trans. 2001, 5, 599.
12. Dietrich, B.; Kintzinger, J. P.; Lehn, J. M.; Metz, B.; Zahidi, A. J. Phys.
Chem. 1987, 91, 6600.
13. Lindoy, L. F. The Chemistry of Macrocyclic Ligand Complexes;
Cambridge University Press: Cambridge, 1992; p 269.
14. Farahbakhsh, M.; Schmidt, H.; Rehder, D. Chem. Ber. Recl. 1997, 130,
1123.
15. Housecroft, C. E. Cluster Compounds of Main Group Elements; VCH:
Weinheim, Germany, 1996; p 94.
16. Kirch, M.; Lehn, J. M. Angew. Chem. 1975, 87, 542.
17. Lehn, J. M.; Montavon, F. Helv. Chim. Acta 1978, 61, 67.
18. Lehn, J. M.; Sauvage, J. P. J. Chem. Soc., Chem. Commun. 1971, 9, 440.
5. Complexation by Cryptands in γ-Butyrolactone as Solvent
132
19. Morf, W. E.; Simon, W. Helv. Chim. Acta 1971, 54, 2683.
20. Pregel, M. J.; Buncel, E. J. Am. Chem. Soc. 1993, 115, 10.
21. Simon, W.; Morf, W. E.; Meier, P. C. Struct. Bond. 1973, 16, 113.
22. Severin, K. Coord. Chem. Rev. 2003, 245, 3.
23. Huang, R. H.; Wagner, M. J.; Gilbert, D. J.; Reidy-Cedergren, K. A.;
Ward, D. L.; Faber, M. K.; Dye, J. L. J. Am. Chem. Soc. 1997, 119, 3765.
24. Lehn, J. M.; Sauvage, J. P. J. Am. Chem. Soc. 1975, 97, 6700.
25. Wipff, G.; Kollman, P. Nouv. J. Chim. 1985, 9, 457.
26. Shamsipur, M.; Karkhaneei, E.; Afkhami, A. Polyhedron 1998, 17, 3809.
27. Chen, Q.; Cannell, K.; Nicoll, J.; Dearden, D. V. J. Am. Chem. Soc. 1996,
118, 6335.
28. Chantooni, M. K., Jr.; Kolthoff, I. M. J. Solution Chem. 1985, 14, 1.
29. Cox, B. G.; Garcia-Rosas, J.; Schneider, H. J. Am. Chem. Soc. 1981, 103,
1054.
30. Rhinebarger, R. R.; Popov, A. I. Polyhedron 1988, 7, 1341.
31. Shamsipur, M.; Popov, A. I. J. Phys. Chem. 1986, 90, 5997.
32. Smetana, A. J.; Popov, A. I. J. Solution Chem. 1980, 9, 183.
33. Abou-Hamdan, A.; Lincoln, S. F. Inorg. Chem. 1991, 30, 462.
34. Clarke, P.; Gulbis, J. M.; Lincoln, S. F.; Tiekink, E. R. T. Inorg. Chem.
1992, 31, 3398.
35. Clarke, P.; Lincoln, S. F.; Tiekink, E. R. T. Inorg. Chem. 1991, 30, 2747.
36. Lincoln, S. F.; Abou-Hamdan, A. Inorg. Chem. 1990, 29, 3584.
37. Lincoln, S. F.; Rodopoulos, T. Inorg. Chim. Acta 1991, 190, 223.
38. Lincoln, S. F.; Rodopoulos, T. Inorg. Chim. Acta 1993, 205, 23.
39. Lincoln, S. F.; Stephens, A. K. W. Inorg. Chem. 1991, 30, 3529.
40. Stephens, A. K. W.; Dhillon, R. S.; Madbak, S. E.; Whitbread, S. L.;
Lincoln, S. F. Inorg. Chem. 1996, 35, 2019.
41. Izatt, R. M.; Pawlak, K.; Bradshaw, J. S.; Bruening, R. L. Chem. Rev.
1991, 91, 1721.
42. Shchori, E.; Jagur-Grodzinski, J.; Shporer, M. J. Am. Chem. Soc. 1973,
95, 3842.
5. Complexation by Cryptands in γ-Butyrolactone as Solvent
133
43. Strasser, B. O.; Hallenga, K.; Popov, A. I. J. Am. Chem. Soc. 1985, 107,
789.
44. Strasser, B. O.; Popov, A. I. J. Am. Chem. Soc. 1985, 107, 7921.
45. Izatt, R. M.; Bradshaw, J. S.; Nielsen, S. A.; Lamb, J. D.; Christensen, J.
J.; Sen, D. Chem. Rev. 1985, 85, 271.
46. Izatt, R. M.; Bradshaw, J. S.; Pawlak, K.; Bruening, R. L.; Tarbet, B. J.
Chem. Rev. 1992, 92, 1261.
47. Izatt, R. M.; Pawlak, K.; Bradshaw, J. S.; Bruening, R. L. Chemical
Reviews 1995, 95, 2529.
48. Sonoda, K.; Deguchi, M.; Koshina, H.; Armand, M.; Michot, C.;
Gauthier, M. Proc. - Electrochem. Soc. 2003, 2002-26, 512.
49. Takami, N.; Sekino, M.; Ohsaki, T.; Kanda, M.; Yamamoto, M. J. Power
Sources 2001, 97-98, 677.
50. Hoffmann, H. M. R.; Rabe, J. Angew. Chem. 1985, 97, 96.
51. Mandal, S. K.; Amin, S. R.; Crowe, W. E. J. Am. Chem. Soc. 2001, 123,
6457.
52. Zhu, Y.-L.; Xiang, H.-W.; Wu, G.-S.; Bai, L.; Li, Y.-W. Chem. Commun.
2002, 3, 254.
53. Masia, M.; Rey, R. J. Phys. Chem. B 2004, 108, 17992.
54. Zahl, A.; Neubrand, A.; Aygen, S.; van Eldik, R. Rev. Sci. Instrum. 1994,
65, 882.
55. Helm, L.; Borel, A. NMRICMA 2.7, University of Lausanne, 2000.
56. Matlab, version 5.3.1.; Mathworks Inc.: 1999.
57. Pople, J. A.; Schneider, W. G.; Bernstein, H. J. High-Resolution Nuclear
Magnetic Resonance; McGrow-Hill Book Co.: New York, 1959; p 513.
58. Becke, A. D. J. Chem. Phys. 1993, 98, 5648.
59. Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B: Condens. Matter Mater.
Phys. 1988, 37, 785.
60. Stephens, P. J.; Devlin, F. J.; Chabalowski, C. F.; Frisch, M. J. J. Phys.
Chem. 1994, 98, 11623.
61. Dunning, T. H.; Jr.; Hay, P. J. Modern Theoretical Chemistry; Plenum:
New York, 1976; Vol. 3, p 1.
5. Complexation by Cryptands in γ-Butyrolactone as Solvent
134
62. Huzinaga, S.; Andzelm, J.; Klobukowski, M.; Radzio-Andzelm, E.; Sakai,
Y.; Tatewaki, H. Gaussian basis sets for molecular calculations;
Elsevier: Amsterdam, 1984.
63. The performance of the computational level employed in this study is
well documented, see for example; Galle, M.; Puchta, R.; van Eikema
Hommes, N.; van Eldik, R. Z. Phys. Chem. 2006, 220, 511; Puchta, R.;
van Eikema Hommes, N. J. R.; Meier, R.; van Eldik, R. Dalton Trans.
2006, 28, 3392; Klaus, S.; Neumann, H.; Jiao, H.; Jacobi von Wangelin,
A.; Goerdes, D.; Struebing, D.; Huebner, S.; Hateley, M.; Weckbecker,
C.; Huthmacher, K.; Riermeier, T.; Beller, M. J. Organomet. Chem.
2004, 689, 3685; Saalfrank, R. W.; Deutscher, C.; Maid, H.; Ako, A. M.;
Sperner, S.; Nakajima, T.; Bauer, W.; Hampel, F.; Hess, B. A.; van
Eikema Hommes, N. J. R.; Puchta, R.; Heinemann, F. W. Chem.-Eur. J.
2004, 10, 1899; Scheurer, A.; Maid, H.; Hampel, F.; Saalfrank, R. W.;
Toupet, L.; Mosset, P.; Puchta, R.; van Eikema Hommes, N. J. R. Eur. J.
Org. Chem. 2005, 12, 2566; Weber, C. F.; Puchta, R.; van Eikema
Hommes, N. J. R.; Wasserscheid, P.; van Eldik, R. Angew. Chem. 2005,
117, 6187; Weber, C. F.; Puchta, R.; van Eikema Hommes, N. J. R.;
Wasserscheid, P.; van Eldik, R. Angew. Chem. (Int. Ed. Eng.) 2005, 44,
6033.
64. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.;
Cheeseman, J. R.; Montgomery, J. A.; Jr.; Vreven, T.; Kudin, K. N.;
Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.;
Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.;
Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa,
J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.;
Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Bakken, V.; Adamo, C.;
Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.;
Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.;
Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich,
S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.;
Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.;
5. Complexation by Cryptands in γ-Butyrolactone as Solvent
135
Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.;
Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham,
M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.;
Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian
03, Revision B.03, Gaussian Inc.; Wallingford, CT, 2004.
65. Shchori, E.; Jagur-Grodzinski, J.; Luz, Z.; Shporer, M. J. Am. Chem. Soc.
1971, 93, 7133.
66. Song, J. Y.; Wang, Y. Y.; Wan, C. C. J. Power Sources 1999, 77, 183.
67. Matsuda, Y.; Fukushima, T.; Hashimoto, H.; Arakawa, R. J.
Electrochem. Soc. 2002, 149, A1045.
68. Adebahr, J.; Forsyth, M.; MacFarlane, D. R.; Gavelin, P.; Jacobsson, P. J.
Mater. Chem. 2003, 13, 814.
69. Ding, M. S.; Xu, K.; Zheng, J. P.; Jow, T. R. J. Power Sources 2004, 138,
340.
70. Ue, M. J. Electrochem. Soc. 1994, 141, 3336.
71. Marcus, Y. Chem. Soc. Rev. 1993, 22, 409.
72. Shamsipur, M.; Karkhaneei, E.; Afkhami, A. J. Coord. Chem. 1998, 44,
23.
73. Davis, A. V.; Fiedler, D.; Seeber, G.; Zahl, A.; Van Eldik, R.; Raymond, K.
N. J. Am. Chem. Soc. 2006, 128, 1324.
74. van Eldik, R.; Hubbard, C. D. Chemistry at Extreme Conditions;
Elsevier: Amsterdam, 2005; p 109.
75. Wolak, M.; van Eldik, R. J. Am. Chem. Soc. 2005, 127, 13312.
76. Zahl, A.; Van Eldik, R.; Matsumoto, M.; Swaddle, T. W. Inorg. Chem.
2003, 42, 3718.
77. Helm, L.; Merbach, A. E. Chem. Rev. 2005, 105, 1923.
78. Galle, M.; Puchta, R.; van Eikema Hommes, N. J. R.; van Eldik, R. Z.
Phys. Chem. 2006, 220, 511.
79. Choua, S.; Sidorenkova, H.; Berclaz, T.; Geoffroy, M.; Rosa, P.;
Mezailles, N.; Ricard, L.; Mathey, F.; Le Floch, P. J. Am. Chem. Soc.
2000, 122, 12227.
5. Complexation by Cryptands in γ-Butyrolactone as Solvent
136
80. Moras, D.; Weiss, R. Acta Crystallogr.,Sect. B: Struct. Crystallogr.
Cryst. Chem. 1973, 29, 400.
81. Meske, W.; Babel, D. Z. Anorg. Allg. Chem. 1998, 624, 1751.
82. Scotti, N.; Zachwieja, U.; Jacobs, H. Z. Anorg. Allg. Chem. 1997, 623,
1503.
6. Summary
137
6. Summary
In this work the solvation process of Li+ ion, as well as solvent and ligand
exchange reactions on Li+ ion were studied. Li+ ions possess interesting
properties and like other alkali metal ions are known to form complexes with
macrocyclic ligands called cryptands. In this summary, an overview over the
insights gained in the factors that control the reactivity of Li+ complexes with
respect to the solvent and cryptand properties is presented.
Three main questions were addressed:
• How does the nature of the solvent influence the first coordination
sphere around the Li+ ion?
• How is solvent exchanged between the first and second coordination
spheres of the Li+ ion?
• How is the Li+ ion complexed by different cryptands and what is the
nature of the Li+ exchange mechanism between cryptand and
solvent?
In this thesis these aims were approached from the microscopic point of
view, making use of both theoretical calculations and experimental techniques.
The first two topics mentioned above constitute the contents of Chapters 2
and 3, whereas the third objective was approached and the results are reported
in Chapters 4 and 5.
The experimental method used throughout was 7Li NMR. It was shown that
7Li NMR is a very powerful technique for the study of alkali complexes with
macrocyclic ligands, particularly in non-aqueous solutions. Particular attention
was given to high pressure 7Li NMR, which represents the first example
reported for this kind of research. It could be demonstrated that 7Li NMR is a
very convenient method for the estimation of the coordination number on Li+
ion as well as for investigation of the selective solvation of Li+ in a mixture of
solvents. This method was also used for mechanistic studies, specifically for Li+
exchange between two environments: solvated and chelated species.
6. Summary
138
Solvents of different properties and their influence on the solvation shell
were investigated. Questions concerning selective solvation around Li+ and the
solvation number of the Li+ cation were studied as a function of the solvent
properties. The Gutmann Donor Number (DN) was mostly considered as a
factor to describe the donor properties of the solvent. It reflects the ease of
donation of the solvent molecule’s lone electron pair.
In order to determine the coordination number of Li+, the cation was
dissolved in a solvent mixture that consisted of a weakly and strongly
coordinating component. It could be shown in Chapter 2 that the Li+ ion in
DMSO is coordinated by 4 solvent molecules. γ-Butyrolactone was found to be
a convenient diluent in this kind of study. Contrary to DMSO (DN = 29.8), it is
rather a weak donor (DN = 18.0) which appreciably dissolves alkali metal salts
and its interaction with DMSO is much weaker than that between DMSO and a
metal cation. By plotting the chemical shift of the obtained 7Li NMR signal vs.
the DMSO/LiClO4 mole ratio, a clear brake point could be observed at a
DMSO/Li+ ratio of 4:1, indicating that the solvation number of Li+ in DMSO is
4. Using the same method for the solvent mixture of acetonitrile (DN = 14.1)
and nitromethane (DN = 2.7), it was found that also in this case the Li+ ion is
coordinated by four acetonitrile molecules, as reported in Chapter 3.
Selective solvation of alkali metal ions in mixed solvents can be monitored
from the extent to which the chemical shift of the alkali metal nucleus varies
with solvent composition. In the binary solvent mixture of water and DMSO
reported in Chapter 2, no selective solvation was observed, indicating that on
increasing the water content of the solvent mixture, DMSO is gradually
displaced by water in the coordination sphere of Li+. We did not observe any
rapid change in the position of the NMR signal when the amount of water was
increased. DMSO is a solvent of high solvating ability which can strongly
coordinate to the Li+ ion and can compete with water in the first coordination
sphere of the Li+ ion. It could be concluded that the composition of the first
coordination sphere in mixtures of H2O/DMSO gradually changes from
[Li(DMSO)4]+ to [Li(H2O)4]+ on increasing the water content of the solvent.
6. Summary
139
This is a consequence of the rather similar donor strength of these solvents as
expressed by their donor numbers.
In the binary water/acetonitrile mixture, however, water is coordinated
stronger to Li+ than acetonitrile such that addition of water immediately leads
to the formation of [Li(H2O)4]+. It was observed that the position of the NMR
signal moved significantly (from 4.5 to 6.4 ppm) even if the amount of added
water is very small (only 10 % H2O). Then the chemical shift value increased
slightly and stayed almost constant while the amount of water was changed
from 10 to 100 %. This observation confirms that water coordinates much
stronger to the Li+ ion than acetonitrile (Chapter 3).
Once the solvation shell of the Li+ ion was known, we wanted to investigate
the exchange of certain solvent molecules around Li+. The Li+ ion, however, is
very labile due to the low surface charge density and the absence of ligand field
stabilization effects. Solvent exchange on Li+ is therefore extremely fast,
making it difficult to study experimentally. The ligand exchange mechanism of
Li+ ions solvated by DMSO and water/DMSO mixtures was studied using DFT
calculations. Ligand exchange on [Li(DMSO)4]+ was found to follow a limiting
associative mechanism, whereas the displacement of coordinated H2O by
DMSO in [Li(H2O)4]+ follows an associative interchange mechanism.
In Chapter 3, we have extended our combined quantum chemical and
experimental investigations of the ligand coordination and ligand exchange on
Li+ in acetonitrile (CH3CN) and HCN, as well as in CH3CN/H2O and HCN/H2O
mixtures.
We studied the ligand exchange process in more detail and report
computational evidence for a limiting associative exchange mechanism on the
HCN and CH3CN solvated lithium cation. HCN was employed here as a
simplified model to perform quantum chemical studies on CH3CN. Our
calculations corroborated the experimental work that the first coordination
sphere around Li+ consists of four tightly bound CH3CN and HCN molecules.
Ligand exchange reactions on Li+ most probably follow an associative
6. Summary
140
mechanism as indicated by the formation of stable five-coordinate
intermediates.
The class of ligands that were found to form stable complexes with alkali
metals are macrocyclic ligands called cryptands. They were designed and
shown to form stable alkali metal (i.e. lithium) cryptate complexes by inclusion
into the three-dimensional molecular cavity. Here three cryptands, viz. C211,
C221, and C222 were investigated.
In this work, we were particularly interested in the mechanistic aspects of
the Li+ exchange process between cryptands and solvent. In preliminary
experiments we investigated the Li+ exchange in different solvents. It was
shown that the stability of the cryptate varies with the nature of the solvent.
The stability tends to decrease as DN increases. In water and methanol the
exchange process was too fast to be studied by 7Li NMR (only a single signal
was observed). The process became slower in acetonitrile, and in acetone and
γ-butyrolactone two completely separated signals were observed at low
temperature. Therefore, these two solvents, viz. acetone and γ-butyrolactone,
were chosen for more detailed mechanistic investigations.
In Chapter 4 we focused on acetone as a solvent and the competition
between solvated and endohedral complexed Li+, reported kinetic parameters
for the exchange of Li+ between solvated and complexed states in acetone
under ambient and elevated pressure conditions, and discussed the
mechanistic implications.
Initially, a series of 7Li NMR measurements for the Li+-cryptand system in
acetone were performed. The results enabled us to observe the distribution of
particular species on variation of the concentration ratio. Basically, two signals
of varying intensity were observed in the 7Li NMR spectra that refer to solvated
Li+ and Li+ bound inside the cryptand cavity. In the case of the smallest chelate
(C211), however, an additional signal appeared which could be ascribed to Li+
bound outside the cryptand and being partially solvated. This complex is very
stable and almost no exchange took place. A detailed study of the observed rate
6. Summary
141
constant, kobs, for the exchange of Li+ as a function of temperature and
pressure was therefore undertaken only for C221 and C222.
Stabilities of the cryptates increase in the sequence Li(C222)+ < Li(C221)+ <
Li(C211)+, consistent with the six donor atoms of the C211 cavity and the
optimal fit of Li+ (r = 1.48 Å) into the C211 cavity (r = 1.6 Å) in inclusive
Li(C211)+, generating the largest stability. The increase in stability Li(C222)+ <
Li(C221)+ indicates that the effect of the decrease in electrostatic interaction
with Li+ resulting from the decrease in the number of donor atoms from 8 in
C222 to 7 in C221 is offset by the smaller cavity size of C221, with the
consequence that Li(C221)+ is the more stable cryptate. The more open and
flexible structure of C222 accounts for the larger kobs values that characterize
Li(C222)+ by comparison with those of the relatively rigid inclusive Li(C211)+.
These comparisons illustrate the interplay of the effects of the number of
donor atoms and cryptand flexibility which produce variations in cryptate
stability and lability.
Two mechanisms were considered for the decomplexation processes of
cryptand-alkali metal complexes: bimolecular and unimolecular cation
exchange.
The obtained results indicate that both pathways compete with each other
and the competition is controlled by temperature and the nature of the
cryptand. However, the unimolecular pathway seems to dominate, indicating
the nonparticipation of solvated Li+ in the rate-determining step, resulting in
the first order rate constant k-2. Based on the values of k-2 and the significantly
negative activation entropies obtained for this pathway, the activation
parameters support an associative type of exchange process. k-2 for C222 is
almost three orders of magnitude larger than for C221. The negative activation
entropies found for the decomplexation reactions of both C221 and C222
strongly suggest that attack of the solvent on the complexed Li+ must play an
important role in the decomplexation process, i.e. decomplexation follows an
associative mechanism.
6. Summary
142
Additional mechanistic information on the intimate nature of the exchange
mechanism can in principle be obtained from the effect of pressure on the Li+
exchange process. It could be observed that there was a very small difference in
the shape of the NMR spectra recorded at various pressures. The observed rate
constants calculated for each pressure were very similar within the
experimental error limits and resulted in a practically zero value for ∆V#obs for
both ligands, which points to a concerted interchange type of exchange
mechanism. In terms of an interchange mechanism, bond formation and bond
cleavage occur in a concerted fashion without the formation of a distinct
intermediate of higher or lower coordination number. In addition,
decomplexation will be accompanied by charge creation due to the release of
Li+, which in turn will lead to an increase in electrostriction which will offset
the intrinsic volume changes.
It is not invariably the case that the unimolecular mechanism operates in
cryptate systems. This is exemplified by the system in γ-butyrolactone, for
which the Li+ exchange process follows rather the bimolecular mechanism in
which also solvated Li+ is involved and therefore the concentration of Li+ is
taken into account in the rate law.
In Chapter 5, we report kinetic parameters for the exchange of Li+ in γ-
butyrolactone under ambient and elevated pressure conditions, and discuss
their mechanistic implications. γ-Butyrolactone was chosen as solvent due to
its relevance in lithium batteries where it is used as plasticizer.
It could also be seen in this system that both pathways, bimolecular and
unimolecular compete with each other. However, the bimolecular pathway
seems to predominate, indicating participation of solvated Li+ in the rate-
determining step, resulting in the second order rate constant k1. Involvement
of the solvent in the decomplexation process was found to be less pronounced
than it was in the case of acetone. This is probably due to the fact that although
both solvents, acetone and γ-butyrolactone should coordinate to the naked Li+
cation comparatively strongly, it cannot be observed in the process in which
Li+ must be released from the cryptand cavity. In this case the steric hindrance
6. Summary
143
should be taken into account, i.e. acetone as a sterically less demanding
molecule has probably better access to the Li+ complexed by the cryptand as γ-
butyrolactone does. Therefore, second order rate constants were obtained for
both ligands, as a consequence of participation of the solvated Li+ in the
decomplexation process.
Based on the values of k1 and the significantly negative activation entropies
obtained for this pathway, the activation parameters support an associative
interchange type of exchange process. The effect of pressure was found to be
not very significant for these systems and resulted in a practically zero value
for ∆V#obs, which points to a concerted interchange type of exchange
mechanism.
This work clearly demonstrated that the ligand structure and solvent
properties have considerable effects on the reaction rate and the exchange
mechanism of the cation between the solvated and the complexed sites. The
stability of the cryptate complexes depends to a large extent on the size
relationship between the three-dimensional cavity of the cryptand and the
diameter of the complexed ion. The relative stabilities of cryptates depend on
the electrostatic interactions between metal ions and cryptands, the solvation
energies of metal ions and cryptands, and the fit of the metal ion into the
cryptand cavity.
In general it can be concluded that the present thesis has shown in a
systematic manner that the combination of experimental (dynamic 7Li NMR)
and theoretical (DFT) methods represents a powerful tool that enabled us to
gain valuable qualitative and quantitative insight into mechanistic details of
the coordination behavior of lithium cations and cryptands in different
solvents. To the best of our knowledge, this was for the first time that
mechanistic investigations on complex-formation reactions by cryptands were
carried out using high pressure NMR techniques, in particular 7Li NMR. The
results presented here give a promising outlook in terms of the possibility to
investigate further lithium ion technologically relevant systems by the methods
used in this study.
7. Zusammenfassung
144
7. Zusammenfassung
In der vorliegenden Arbeit wurde der Solvatisierungsprozess des
Lithiumkations, sowie der Lösungsmittel- und Ligandenaustausch am
Lithiumion, untersucht. Das Lithiumkation bildet ebenso wie die anderen
Alkalimetallkationen, Komplexe mit Kryptanden. In dieser Arbeit lag der
Schwerpunkt auf C211, C221 und C222. Diese Zusammenfassung soll einen
Überblick über die Faktoren geben, die die Reaktivität der Lithium-(I)-
komplexe in Abhängigkeit von Lösungsmittel- und Kryptandeneigenschaften
steuern.
Folgende drei grundsätzliche Fragen wurden angegangen:
Wie wird die erste Koordinationssphäre um das Lithiumion von der Natur
des Lösungsmittels beeinflusst?
Wie wird das Lösungsmittel zwischen der ersten und der zweiten
Koordinationssphäre des Lithiumions ausgetauscht?
Wie wird das Lithiumion von verschiedenen Kryptanden komplexiert und
wie ist der Mechanismus des Lithiumionaustausches zwischen Kryptand und
Lösungsmittel.
Die gesteckten Ziele wurden aus der mikroskopischen Perspektive
angegangen. Anwendung fanden hier sowohl experimentelle Methoden (7Li
NMR) als auch quantenchemische Rechnungen. Die ersten beiden oben
genannten Fragestellungen bilden den Kern der Kapitel 2 und 3, während die
erhaltene Ergebnisse des dritten Themas in den Kapitel 4 und 5 beschrieben
sind.
Wie aus der Literatur bereits bekannt war, stellt 7Li NMR Spektroskopie
eine sehr leistungsstarke Methode zur Untersuchung von Alkalimetallionen in
makrozyklischen Liganden, besonders in nichtwässrigen Lösungen dar. In der
vorgelegten Arbeit wird hierbei besondere Aufmerksamkeit der Hochdruck 7Li
NMR Spektroskopie gewidmet. Sie wird hier zum ersten Mal für
Untersuchungen an Kryptaten benutzt. Es wurde weiterhin gezeigt, dass die 7Li
7. Zusammenfassung
145
NMR Spektroskopie eine sehr leistungsstarke Methode zur Bestimmung der
Koordinationszahl des Lithiumions als auch zur Untersuchung der selektiven
Solvatation in gemischten Lösungsmitteln ist. Diese Methode wurde auch für
mechanistische Untersuchungen benutzt, hier konkret für den
Lithiumionenaustausch zwischen solvatisierten und chelatisierten Spezies.
Lösungsmittel von verschiedenen Eigenschaften und ihr Einfluss auf die
Solvatationshülle wurden untersucht. Das Problem der selektiven Solvatation
um das Lithiumion und die Koordinationszahl des Lithiumions wurden als
Funktion der Lösungsmitteleigenschaften studiert. Die Donorzahl nach
Gutmann (DN) wurde meistens als der Faktor eingesetzt, der die Fähigkeit des
Lösungsmittels als Donor zu agieren charakterisiert.
Um die Koordinationszahl des Lithiumkations zu bestimmen, wurde eine
Lösung vorbereitet, die aus einem Gemisch von schwach und stark
koordinierenden Lösungsmitteln bestand. In Kapitel 2 wurde gezeigt, dass
das Lithiumion in DMSO von 4 Lösungsmittelmolekülen koordiniert wird.
Wir konnten zeigen, dass γ-Butyrolakton ein gut geeignetes
Verdünnungsmittel für derartige Untersuchungen ist. Im Gegensatz zu DMSO
(DN = 29.8), stellt γ-Butyrolakton einen eher schwachen Donor (DN = 18.0)
dar. Es löst die Salze der Alkalimetalle in merklicher Weise, wobei die
Wechselwirkung mit den DMSO-Molekülen wesentlich schwächer ist als die
zwischen DMSO und den Metallkationen. Durch graphische Auftragung der
chemischen Verschiebung des 7Li NMR Signals gegen das Molverhältniss von
DMSO/LiClO4 konnte ein deutlicher Bruch in der Kurve bei einem
Molverhältniss von 4:1 beobachtet werden. Dies deutet auf eine
Koordinationszahl von 4 für Li+ in DMSO hin. Die Anwendung dieser Methode
auf ein Lösungsmittelgemisch bestehend aus Acetonitril (DN = 14.1) und
Nitromethan (DN = 2.7) ergab ebenfalls eine vierfache Koordination des
Lithiumkations in Acetonitril Lösung, wie im Kapitel 3 berichtet wird.
Selektive Solvatation der Alkalimetallionen in Lösungsmittelgemischen
kann anhand der Veränderung der chemischen Verschiebung des
Alkalimetallkerns in Abhängigkeit mit der Zusammensetzung des
7. Zusammenfassung
146
Lösungsmittelgemisches beobachtet werden. Im binären
Lösungsmittelgemisch Wasser/DMSO, über das im Kapitel 2 berichtet wird,
wurde keine selektive Solvatation beobachtet. Dies deutet darauf hin, dass, mit
wachsendem Wassergehalt im Lösungsmittelgemisch, DMSO rein statistisch
sukzessiv durch Wasser in der Koordinationssphäre des Lithiumions ersetz
wird. Wir beobachteten keine sprunghaften Änderungen des NMR-Signals bei
Erhöhung des Wasseranteils.
DMSO ist ein Lösungsmittel mit großer Solvatationsfähigkeit, das das
Lithiumion stark koordinieren kann, und deswegen mit Wasser in der ersten
Koordinationssphäre des Lithiumions konkurriert. Es kann die
Schlussfolgerung gezogen werden, dass sich die Zusammensetzung der ersten
Koordinationssphäre in Gemischen von H2O/DMSO mit zunehmendem
Wassergehalt im Lösungsmittelgemisch schrittweise von [Li(DMSO)4]+ zu
[Li(H2O)4]+ ändert. Dies lässt auf eine ähnliche Donorstärke dieser
Lösungsmittel schließen, wie sie sich auch aus ihren Donorzahlen ergibt.
In binären Gemischen von Wasser und Acetonitril ist jedoch Wasser zum
Lithiumion stärker koordiniert als Acetonitril, so dass die Zugabe von Wasser
sofort zur Bildung von [Li(H2O)4]+ führt. Wir beobachteten, dass sich die
Position des NMR-Signals signifikant ändert (4.5 zu 6.4 ppm) sogar wenn die
Menge des zugegebenen Wasser sehr gering ist (nur 10 % H2O). Danach
erhöhte sich die chemische Verschiebung nur moderat und blieb konstant
während die Menge des Wassers von 10 % auf 100 % gesteigert wurde. Diese
Beobachtung bestätigt, dass Wasser viel stärker als Acetonitril an das
Lithiumion koordiniert (Kapitel 3).
Nachdem die Solvatationshülle des Lithiumions aufgeklärt wurde, wollten
wir den Austausch von bestimmten Lösungsmittelmolekülen um das
Lithiumion untersuchen. Das Lithiumkation ist jedoch wegen seiner geringen
Oberflächenladungsdichte und fehlenden Ligandenfeldstabilisierungseffekten
sehr labil. Der Lösungsmittelaustausch am Lithiumion ist deswegen extrem
schnell und experimentell schwierig zu untersuchen. Der
Ligandenaustauschmechanismus an dem von DMSO und Wasser/DMSO-
7. Zusammenfassung
147
Gemischen solvatisierten Lithiumion wurde mittels DFT-Rechnungen
untersucht. Es wurde gefunden, dass der Lösungsmittelaustausch an
[Li(DMSO)4]+ durch einen reinen assoziativen Mechanismus (A) erfolgt,
während die Substitution von koordiniertem Wasser durch DMSO in
[Li(H2O)4]+ durch einen assoziativen Interchange Mechanismus erfolgt.
Im Kapitel 3 erweiterten wir unsere kombinierten quantenchemischen
und experimentellen Untersuchungen der Ligandenkoordination und des
Ligandenaustausches auf das Lithiumion in CH3CN und HCN, wie auch auf
CH3CN/H2O und HCN/H2O Gemische.
Wir studierten die Ligandenaustauschprozesse detaillierter und konnten
quantenchemische Hinweise für einen reinen assoziativen
Austauschmechanismus an dem von HCN und CH3CN solvatisierten
Lithiumion finden. HCN wurde hier als ein vereinfachtes Modell für
quantenchemische Untersuchungen von CH3CN eingesetzt. Unsere
Rechnungen untermauerten zusätzlich die experimentellen Ergebnisse, dass
die erste Koordinationssphäre um das Lithiumion aus vier koordinierten
CH3CN- und HCN-Molekülen besteht. Ligandenaustauschreaktionen am
Lithiumion erfolgen am wahrscheinlichsten durch einen assoziativen
Mechanismus, worauf die Bildung von stabilen fünffach-koordinierten
Intermediaten hindeutet.
Wie bereits ausgeführt, bilden Kryptanden mit Alkalimetallionen stabile
Komplexe, sog. Kryptate. Wir haben drei Kryptanden untersucht: C211, C221
und C222.
In dieser Arbeit waren wir insbesondere an mechanistischen Aspekten des
Austauschprozesses am Lithiumion zwischen Kryptanden und Lösungsmittel
interessiert. In experimentellen Vorarbeiten untersuchten wir den
Lithiumionenaustausch in verschiedenen Lösungsmitteln. Es konnte gezeigt
werden, dass die Stabilität des Kryptats von der Natur des Lösungsmittels
abhängt. Die Stabilität nimmt mit zunehmendem DN ab. In Wasser und
Methanol war der Austauschprozess zu schnell, um mit 7Li NMR
Spektroskopie studiert werden zu können (nur ein Signal wurde beobachtet).
7. Zusammenfassung
148
In Acetonitril ist der Prozess langsamer geworden, und in Aceton und γ-
Butyrolakton konnten zwei vollständig getrennte Signale bei tiefen
Temperaturen beobachtet werden. Infolgedessen wurden diese zwei
Lösungsmittel (Aceton und γ-Butyrolakton) für weitere detaillierte
mechanistische Untersuchungen ausgewählt.
Im Kapitel 4 konzentrierten wir uns auf Aceton als Lösungsmittel und auf
die Konkurenz zwischen solvatisirtem und endohedral komplexiertem
Lithiumion. Darüber hinaus wird in diesem Kapitel über kinetische Parameter
für den Lithiumionenustausch zwischen solvatisiertem und komplexiertem
Zustand in Aceton unter Umgebungs- und Hochdruckbedingungen berichtet
und mechanistische Folgerungen werden gezogen.
Zunächst wurde eine Serie von 7Li NMR-Messungen für das System
Li+/Kryptand durchgeführt. Die erhaltenen Ergebnisse ermöglichten uns die
Verteilung konkreter Spezies in Abhängigkeit von Konzentrationsänderungen
zu beobachten. Es wurden zwei Signale von variirender Intensität in den 7Li
NMR-Spektren beobachtet, die auf das solvatisierte und das im
Kryptandenhohlraum gebundene Lithiumion zurückgeführt werden können.
Im Falle des kleinsten Kryptanden (C211) erschien aber auch ein zusätzliches
drittes Signal, das auf das von außen gebundene und damit nur teilweise
solvatisierte Lithiumion zurückgeführt werden könnte. Der Li(C211)-Komplex
ist sehr stabil und es fand fast kein Austausch statt. Eine detaillierte Studie der
beobachtenen Geschwindigkeitskonstante (kobs) für den
Lithiumionenaustausch als Funtion von Temperatur und Druck wurde
deswegen nur für C221 und C222 durchgeführt.
Die Stabilität der Kryptate nimmt in der Reihnfolge Li(C222)+ < Li(C221)+
< Li(C211)+ zu. Dies ist im Einklang mit den sechs Donoratomen des
Hohlraums von C211 und damit, dass das Lithiumion (r = 1.48 Å) optimal in
den Hohlraum von C211 (r = 1.6 Å) passt. Die zunehmende Stabilität Li(C222)+
< Li(C221)+ < Li(C211)+ weist darauf hin, dass die Abnahme der Anzahl der
Donoratome von 8 in C222 auf 7 in C221 durch den kleineren Hohlraum von
C221 ausgeglichen wird. Dementsprechend ist Li(C221)+ der stabilere Kryptat.
7. Zusammenfassung
149
Die flexiblere Struktur von C222 hat den größeren kobs-Wert zu Folge, der für
Li(C222)+ charakteristisch ist, im Vergleich zu den Werten für den relativ
starren Li(C211)+ Einschlußkomplex. Diese Vergleiche stellen die Folgen der
unterschiedlichen Zahl von Donoratomen in Kombination mit der Flexibilität
des Kryptanden gut dar. Sie haben eine unterschiedliche Stabilität bzw.
Labilität der Kryptate zur Folge.
Es wurden zwei Mechanismen für den Dekomplexierungsprozess von
Alkalimetal-Kryptatkomplexen in Aceton in Betracht gezogen: der
bimolekulare und der unimolekulare Kationenaustausch. Die Ergebnisse
deuten darauf hin, dass beide Prozesse miteinander konkurrieren und von der
Temperatur und der Natur des Ligandes gesteuert werden. Allerdings scheint
der unimolekulare Prozess dominant zu sein, was auf die Nichtbeteiligung des
solvatisierten Lithiumions im geschwindigkeitsbestimmenden Schritt
hindeutet und was in einer Geschwindigkeitskonstante erster Ordnung
resultiert. Basierend auf den für diesen Prozess erhaltenen
Geschwindigkeitskonstanten und signifikant negativen Aktivierungsentropien,
kann auf einen assoziativen Austauschprozess geschlossen werden. Die
Geschwindigkeitskonstante für C222 ist fast um drei Größenordnungen größer
als für C221. Die negativen Aktivierungsentropien, die für die
Dekomplexierungsreaktionen von C221 und C222 beobachtet wurden, deuten
stark darauf hin, dass der Angriff des Lösungsmittels auf das komplexierte
Lithiumion eine wichtige Rolle im Dekomplexirungsprozess spielt, d.h. die
Dekomplexierung erfolgt durch einen assoziativen Mechanismus.
Weitere detaillierte mechanistische Informationen über die Natur des
Austauschmechanismuses kann aus dem Einfluss von hohen Drücken auf den
Li+-Austauschprozess gewonnen werden. Es wurde beobachtet, dass nur sehr
kleine Änderungen in den bei verschidenem Druck aufgenommenen NMR-
Spektren auftreten. Die in Abhängigkeit vom Druck berechneten
Geschwindigkeitskonstanten waren im Rahmen der experimentellen
Fehlergrenzen sehr ähnlich und resultierten in ΔV#obs-Werten von Null für
C221 und C222. Dies spricht eher für einen Interchangemechanismus (I). Die
7. Zusammenfassung
150
Bildung und Spaltung der Li-Donor-Bindungen erfolgen konzertiert ohne
Bildung eines eindeutigen Intermediats mit höherer oder niedrigerer
Koordinationszahl. Durch die Dekomplexierung der Li+-Ionen erfolgt eine
Freisetzung von beweglichen Landungsträgen, was auf der anderen Seite zu
zunehmender Elektrostriktion führt, die die intrinsische Volumenänderungen
ausgleicht.
Im Gegensatz zum unimolekularen Lithiumionenaustauschprozess bei in
Aceton gelösten Li-Kryptatkomplexen finden wir im Fall von in γ-Butyrolacton
gelösten Krytptaten eher einen bimolekularen Mechanismus. Dieser wird auch
durch die Konzentration der solvatisierten Lithiumionen beeinflusst,
dementsprechend muß auch die Lithiumkonzentration in die
Geschwindigkeitsberechnung einbezogen werden.
Im Kapitel 5 berichten wir über kinetische Parameter für den
Lithiumionenaustausch an in γ-Butyrolacton gelösten Krytptaten unter
verschiedenen Drücken und diskutieren ihre Folgen für den Mechanismus. Die
Wahl von γ-Butyrolacton als Lösungsmittel ergab sich aus dessen Bedeutung
für Lithiumbatterien.
Hier wurde ebenfalls beobachtet, dass beide Prozesse, der bimolekulare
und der unimolekulare, miteinander konkurrieren. Allerdings, ist in diesem
Fall der bimolekulare Prozess dominant, was auf die Beteiligung des
solvatisierten Lithiumions am geschwindigkeitsbestimmenden Schritt
hindeutet und in einer Geschwindigkeitskonstante zweiter Ordnung resultiert.
Die Beteiligung des Lösungsmittels am Dekomplexierungsprozess war geringer
als im Falle des Acetons. Beide Lösungsmittel, Aceton und γ-Butyrolacton,
koordinieren nicht komplexierte Lithiumkationen gleich gut. In Konkurrenz
zwischen den hier untersuchten Kryptanden und den Lösungsmittelmolekülen
um die Komplexierung kann dies jedoch nicht beobachtet werden. Da γ-
Butyrolacton im Gegensatz zu Aceton sterisch anspruchsvoller ist, muß der
Angriff auf ein im Kryptanden endohedral komplexiertes Li+-Ion als sterisch
gehindert angesehen werden. Somit spielen solvatisierte Li+-Ionen bei den
Austauschprozessen eine größere Rolle als im Fall des Acetons. Dies zeigt die
7. Zusammenfassung
151
für beide Liganden, C221 und C222, erhaltene Geschwindigkeitskonstanten
zweiter Ordnung.
Die für diesen Prozess erhaltenen Geschwindigkeitskonstanten und die
signifikant negativen Entropien, untermauern den assoziativen Interchange
Charakter der untersuchten Reaktionen. Es konnte hierbei kein signifikanter
Druckeinfluss beobachtet werden. Die ΔV#obs-Werten sind nahezu Null und
legen zusätzlich einen Interchange-Mechanismus (I) nahe.
Diese Arbeit hat eindeutig gezeigt, dass die Ligandenstruktur und die
Lösungsmitteleigenschaften einen erheblichen Einfluss auf die
Reaktionsgeschwindigkeit und den Mechanismus des Kationenaustausches
zwischen den solvatisierten und komplexierten Spezies hat. Die Stabilität der
Kryptatkomplexe hängt weitgehend vom Größenverhältnis zwischen dem
dreidimensionalen Hohlraum des Kryptanden und dem Durchmesser des
komplexierten Ions ab. Zusätzlich wird die relative Stabilität der Kryptate
durch die Wechselwirkungen zwischen den Li+-Ionen und den Kryptanden
sowie deren Solvatatisierungsenergien bedingt.
Allgemein kann zusammengefasst werden, dass die vorliegende
Dissertation auf systematische Art und Weise zeigt, dass das Zusammenspiel
der experimentellen (dynamische 7Li NMR Spektroskopie) und
quantenchemischen (DFT-Rechnungen) Methoden wichtige qualitative und
quantitative Einblicke in mechanistische Details des Koordinationsverhaltens
von Lithiumionen und Kryptanden in verschiedenen Lösungsmitteln
ermöglicht. Dies sind nach unserem Wissen die ersten mechanistischen
Untersuchungen an Kryptaten mittels Hochdruck-NMR (7Li NMR). Die hier
vorgestellten Ergebnisse sind vielversprechend für weitere Untersuchungen an
anwendungsbezogenen Systemen mit Lithiumionen.
Lebenslauf
Persönliche Daten:
Name: Ewa Maria Pasgreta
Geburtsdatum: 11. März 1976
Geburtsort: Więcbork (Polen)
Eltern: Urszula und Stefan Pasgreta
Staatsangehörigkeit: polnisch
Familienstand: ledig
Schulbildung:
September 1983 - Juni 1991 Grundschule in Runowo Krajeńskie, Polen
September 1991 - Juni 1995 Gymnasium in Bydgoszcz, Polen (abgeschlossen
mit Abitur)
Studium:
Oktober 1995 - Juni 2001 Universität in Toruń, Polen
Chemiestudium mit dem Abschluss Dipl.-Chem.
November 2001 – December 2006 Friedrich-Alexander Universität Erlangen-
Nürnberg,
Anfertigung der Promotion bei Prof. Dr. Dr. h.c.
mult. Rudi van Eldik an der Institut für
Anorganische Chemie
März 2007 Abschluss mit Promotion zum Dr. rer. nat.