University of Huddersfield Repository

Ji, Pengju

Kinetics and Mechanisms of Organic Reactions in Liquid Ammonia

Original Citation

Ji, Pengju (2011) Kinetics and Mechanisms of Organic Reactions in Liquid Ammonia. Doctoral
thesis, University of Huddersfield.

This version is available at http://eprints.hud.ac.uk/10033/

The University Repository is a digital collection of the research output of the

University, available on Open Access. Copyright and Moral Rights for the items
on this site are retained by the individual author and/or other copyright owners.
Users may access full items free of charge; copies of full text items generally
can be reproduced, displayed or performed and given to third parties in any
format or medium for personal research or study, educational or not-for-profit
purposes without prior permission or charge, provided:

• The authors, title and full bibliographic details is credited in any copy;
• A hyperlink and/or URL is included for the original metadata page; and
• The content is not changed in any way.

For more information, including our policy and submission procedure, please
contact the Repository Team at: [email protected].

http://eprints.hud.ac.uk/
KINETICS AND MECHANISMS OF ORGANIC REACTIONS
IN LIQUID AMMONIA

Pengju JI

January 2011

A thesis submitted to the University of Huddersfield in partial fulfilment of the

requirements for the degree of Doctor of Philosophy

The Department of Chemical and Biological Sciences

The University of Huddersfield
Queensgate
Huddersfield
HD1 3DH
United Kingdom
 Acknowledgements

Acknowledgements

I would like to sincerely thank Prof. Michael I Page and Prof. John H Atherton for their
supervision of my PhD during the past 3 years, particularly their knowledge, patience and the
rigorous attitude towards the scientific research.

I am very grateful to the IPOS (Innovative Physical Organic Solutions) group of the
Huddersfield University for the financial support during the past 3 years, in particular, Dr.
Matt Stirling and Dr. Nick Powles, for their helpful discussions and advice, constant warm
encouragements, I am deeply in debt to them.

Special thanks to Dr. Bernard Gill, Dr. Michael Crampton and Dr. Andrew Laws for their
suggestive advice, also Dr. Neil McLay for the help of NMR studies.

Thanks are due to the University of Huddersfield for providing a postgraduate studentship
during my entire PhD.

Finally, I would like to give a huge thanks to my wife, Yang, who encourages me to pursue
my PhD, without your encouragement and support, this never would have been achieved.

i
 Copyright Statement

Copyright Statement

i. The author of this thesis (including any appendices and/or schedules to this thesis)
own any copyright in it (the “Copyright”) and s/he has given The University of
Huddersfield the right to use such Copyright for any administrative, promotional,
educational and/or teaching purposes.

ii. Copies of this thesis, either in full or in extracts, may be made only in accordance
with the regulations of the University Library. Details of these regulations may be
obtained from the Librarian. This page must form part of any such copies made.

iii. The ownership of any patents, designs, trademarks and any and all other
intellectual property rights except for the Copyright (the “Intellectual Property
Rights”) and any reproductions of copyright works, for example graphs and tables
(“Reproductions”), which may be described in this thesis, may not be owned by the
author and may be owned by third parties. Such Intellectual Property Rights and
Reproductions cannot and must not be made available for use without the prior
written permission of the owner(s) of the relevant Intellectual Property Rights
and/or Reproductions.

ii
 Abstract

Abstract

The rate constants for the reactions of a variety of nucleophiles reacting with substituted
benzyl chlorides in liquid ammonia (LNH3) have been determined. To fully interpret the
associated linear free-energy relationships, the ionisation constants of phenols ions in liquid
ammonia were obtained using UV spectra. These equilibrium constants are the product of
those for ion-pair formation and dissociation to the free ions, which can be separated by
evaluating the effect of added ammonium ions. There is a linear relationship between the pK a
of phenols in liquid ammonia and those in water of slope 1.68. Aminium ions exist in their
unprotonated free base form in liquid ammonia and their ionisation constants could not be
determined by NMR. The rates of solvolysis of substituted benzyl chlorides in liquid ammonia
at 25 oC show a Hammett ρ of zero, having little or no dependence upon ring substituents,
which is in stark contrast with the hydrolysis rates of substituted benzyl halides in water,
which vary 107 fold. The rate of substitution of benzyl chloride by substituted phenoxide ions
is first order in the concentration of the nucleophile indicative of a S N2 process, and the
dependence of the rate constants on the pKa of the phenol in liquid ammonia generates a
Brønsted βnuc = 0.40. Contrary to the solvolysis reaction, the reaction of phenoxide ion with 4-
substituted benzyl chlorides gives a Hammett ρ = 1.1, excluding the 4-methoxy derivative,
which shows the normal positive deviation. The second order rate constants for the
substitution of benzyl chlorides by neutral and anionic amines show a single Brønsted β nuc =
0.21 (based on the aqueous pKa of amine), but their dependence on the substituent in
substituted benzyl chlorides varies with a Hammett ρ of 0 for neutral amines, similar to that
seen for solvolysis, whereas that for amine anions is 0.93, similar to that seen for phenoxide
ion.

The rates of aromatic nucleophilic substitution reactions in liquid ammonia are much faster
than those in protic solvents indicating that liquid ammonia behaves like a typical dipolar
aprotic solvent in its solvent effects on organic reactions. Nitrofluorobenzenes (NFBs) readily
undergo solvolysis in liquid ammonia and 2-NFB is about 30 times more reactive than the 4-
substituted isomer. Oxygen nucleophiles, such as alkoxide and phenoxide ions, readily
displace fluorine of 4-NFB in liquid ammonia to give the corresponding substitution product
with little or no competing solvolysis product. Using the pKa of the substituted phenols in

iii
 Abstract

liquid ammonia, the Brønsted β nuc for the reaction of 4-NFB with para-substituted phenoxides
is 0.91, indicative of the removal of most of the negative charge on the oxygen anion and
complete bond formation in the transition state and therefore suggests that the decomposition
of the Meisenheimer σ-intermediate is rate limiting. The aminolysis of 4-NFB occurs without
general base catalysis by the amine and the second order rate constants generate a Brønsted
βnuc of 0.36 using either the pKa of aminium ion in acetonitrile or in water, which is also
interpreted in terms of rate limiting breakdown of Meisenheimer σ-intermediate. Nitrobenzene
and diazene are formed as unusual products from the reaction between sodium azide and 4-
NFB which may be due to the initially formed 4-nitroazidobenzene decomposing to give a
nitrene intermediate, which may dimerise and be trapped by ammonia to give the unstable
hydrazine which then yields nitrobenzene.

We have developed a method for the amination of aryl halides in liquid ammonia using copper
(I) catalysis which enables direct synthesis of a number of primary amines with excellent
yields. This method does not require strong base and ligands as additives and the amination in
liquid ammonia has exclusive selectivity for the formation of primary amines, even under
relative higher temperature. The amount of catalyst required for the reaction is relatively lower
than that generally used, and the convenience of products separation with liquid ammonia as
reaction medium indicate its potential industrial application. The preliminary mechanistic
investigation indicates that the rate of the amination is first order dependence on the
concentration of copper (I) catalyst, and the formation of triamminecopper (I)-aryl ring
intermediate is probably the rate limiting step in liquid ammonia. Due to strong coordination
of solvent molecules to the copper (I) ion, the kinetics of the reaction are generally insensitive
to the addition of other conventional ligands in liquid ammonia.

The copper (I) catalysed 1,3-Huisgen cycloaddition reaction of azide and alkynes (CuIAAC) in
liquid ammonia requires less catalyst than those in conventionally used solvents. The excellent
yield, exclusive selectivity, and most importantly, the ease of separation of the product
indicate the potential advantages of using liquid ammonia as the solvent for this reaction. The
preliminary mechanistic investigation suggests that CuIAAC reaction in liquid ammonia is a
stepwise process with the initial formation of copper (I)-acetylide ion complex, followed by its
combination with copper (I) coordinated azide.

iv
 Abbreviations

A list of commonly used symbols and abbreviations in this thesis

LNH3 liquid ammonia

HBD hydrogen bond donor
HBA hydrogen bond acceptor
εr dielectric constant
λmax maximum absorbance wavelength (nm)
εmax molar extinction coefficient at λmax
DMSO dimethyl sulphoxide
DCM dichloromethane
THF tetrahydrofuran
DMF N,N-dimethyl formaldehyde
HMPT hexamethylphosphoramide
AN acetonitrile
DMAc N,N-dimethylacetamide
NMP N-methyl-2-pyrrolidone
kobs observed pseudo-first-order rate constant
ksol solvolysis rate constant (s-1)
k2 second order rate constant (M-1s-1)
t1/2 half-life of reaction(s-1 or hr-1)
GC gas chromatography
HPLC high performance liquid chromatography
DSC differential scanning calorimetry
DN donor number, qualitative measure of Lewis
basicity (kcal mol-1)
I ionic strength (M)
Ki equilibrium constant for ion-pair formation
Kd equilibrium constant for ion-pair dissociation
K equilibrium constant
DEPT distortionless enhancement by polarisation
transfer
MDN malonodinitrile

v
 Abbreviations

NFBs nitrofluorobenzenes
4-NFB 4-nitrofluorobenzene
2-NFB 2-nitrofluorobenzene
2,4-DFNB 2,4-difluoronitrobenzene
4-NAB 4-nitroazidobenzene
2-NAB 2-nitroazidobenzene
ρ Hammett reaction constant
σ Hammett substituent constant
βnuc Brønsted constant for nucleophilic substitution
reaction
SNAr nucleophilic aromatic substitution with
addition-elimination mechanism

vi
 Contents

Contents Page

Abstract iii

Abbreviations v

Introduction 1

1. Solvent effects on organic reactions 2

1.1 The classification of solvents 3

1.2 Solvent effects on solubility 4

1.3 Solvent effects on chemical equiliribia 6

1.4 Solvent effects on the rates of organic reactions 12

1.5 Solvents effects on UV spectra 20

1.6 Empirical parameters of solvent polarity 21

2. Physical and chemical properties of liquid ammonia 24

3. Organic reactions and mechanisms in liquid ammonia 27

4. The project 41

Experimental 43

1. Materials 44

2. Pressure equipments 60

3. Instruments 65

4. General procedures 66

Results and Discussion 75

1. Acidity and spectral studies of compounds in liquid ammonia 75

vii
 Contents

1.1 UV-Vis studies 76

1.1.1 UV-Vis Spectra of aromatic nitro compounds 77

1.1.2 Ionisation of phenols 78

1.2 NMR studies 89

1.2.1 Ionisation of aminium salts 90

1.2.2 Ionisation of carbon acids 92

2. Solvolysis in liquid ammonia 106

2.1 Solvolysis of alkyl halides 107

2.2 Solvolysis of aromatic compounds 117

2.3 Solvolysis of epoxides, esters and sulfonyl chloride 124

2.4 Solvolysis of ketones with ammonium salt as catalyst 126

3. Aliphatic nucleophilic substitutions in liquid ammonia 128

3.1 Nucleophilic substitution with oxygen-nucleophiles 129

3.2 Nucleophilic substitution with nitrogen-nucleophiles 138

3.3 Nucleophilic substitution with sulfur-nucleophiles 145

3.4 Nucleophilic substitution with carbon-nucleophiles 145

4. Aromatic nucleophilic substitutions in liquid ammonia 152

4.1 Nucleophilic substitution with oxygen-nucleophiles 153

4.2 Nucleophilic substitution with nitrogen-nucleophiles 159

4.3 Nucleophilic substitution with sulfur-nucleophiles 168

4.4 Nucleophilic substitution with carbon-nucleophiles 170

viii
 Contents

5. Metal catalysed organic reactions in liquid ammonia 171

5.1 Copper catalysed amination of aryl halides 172

5.1.1 Copper (I) catalysed amination of aryl iodides 173

5.1.2 Copper (I) catalysed amination of aryl bromides and chlorides 185

5.2 Copper (I) catalysed azide-alkyne cycloaddition (CuIAAC) 188

Appendix 198

Appendix A: Tables and Figures 198

Appendix B: Derivations 234

Appendix C: NMR Spectra 236

Appendix D: Safety Protocols 252

References 255

viiii
 Introduction

Introduction Page

1. Solvent effects on organic reactions 2

2. Physical and chemical properties of liquid ammonia 24

3. Organic reactions and mechanisms in liquid ammonia 27

4. The project 41

1
 Introduction

1. Solvent effects on organic reactions

Solvents are ubiquitous, and in most circumstances, are indispensable in chemistry and, in a
sense, the extent of understanding of solutions reflects the development in chemistry. The
history of utilising and developing solvents can be traced back to the times of when Greek
Alchemists were searching for a universal solvent so-called “Alkahest”.1 Solvents are widely
used in nowadays, not only in academic research, but most importantly, in industry, in fact the
global solvents market is estimated to reach 19.9 million tons by 2015, according to a report
by Global Industry Analysts, Inc.2 Therefore, knowledge of solution properties and
investigations of solvents effects on organic reactions benefit academic chemical research as
well as industry, and the choice of the solvents has significant impact on the efficiency and
even nature of organic reactions. Although there have been many achievements on
understanding solvent properties and their influences on the organic reaction in recent years,1,3
due to the complex nature of solution and solvent effects on the reactions, our understanding
of solutions are still imperfect and the progress of research has been relatively slow compared
with the breakthrough in other chemical research areas, such as catalysis, synthesis and
materials. In addition, most modern research in this area remains academic and, therefore can
only benefit industry in a limited way. In recent years, as environmental issues become more
and more crucial, some conventionally used solvents in industry are listed as „unclean‟, and
potentially may harm the ecosystem. Thus it is imperative to discover „greener‟ solvents which
could replace some of the currently used ones but not at the expense of losing their beneficial
solvent effects.4

1.1 The classification of solvents

The nature of solvents can be quite different. Solvents can be classified in terms of chemical
constitution, physical properties, and acid-base behaviours.1 Despite the many criteria for the
classification of solvents, the solvents are often broadly divided into two different categories,
nonpolar and polar. The dielectric constants (εr) of solvents can provide a rough indication of a
solvent‟s polarity. Generally speaking, solvents with a dielectric constant of less than 15 are
considered to be nonpolar. Theoretically, the dielectric constant measures the solvent's ability
to decrease the field strength of the electric field surrounding a charged species immersed in it.
This reduction is then compared to the field strength of the charged species in vacuum.5

2
 Introduction

Although the dielectric constant gives a general indicator to group the solvents, considering
the enormous number of solvents, such a single parameter based classification is far from
satisfactory. For example, methanol (εr = 33) has roughly the same dielectric constant as DMF
(εr = 36.7) at 25 oC, but their other solvent properties including the solubility of solutes are
quite different, these differences become more pronounced when considering solvent effects
on the rates of nucleophilic substitution reactions in these two solvents. Obviously, the
classification of solvents needs to consider more parameters to fully explain and understand
solvent effects. Current classifications of solvents combine dielectric constant, dipole moment
(D), hydrogen bond donor (HBD) and acceptor (HBA) abilities of the solvent. Generally
speaking, solvents fall into one of three main categories:6

1. Protic: refer to solvents that possess a proton donating function, normally the
solvent molecule contains an -OH or -NH- group. Typical examples are
alcohols, amines, carboxylic acids and water. These have a large dipole
moment and a capacity for hydrogen bonding.
2. Dipolar aprotic: solvents of this category have no acidic proton, but possess a
large dipole moment. Representative examples are DMSO, DMF, acetonitrile,
nitromethane and ketones.
3. Non-polar aprotic: these solvents, which have only a small dipole moment, no
acidic protons, and also have a very weak or no ability to donate or accept
electrons. The intermolecular forces between solvent molecules are very weak.
Hydrocarbons, halocarbons and ethers are among the typical examples.

It is worth noting that due to the physical and chemical differences between solvents, it is
sometimes difficult to classify them neatly into one of the above classes and some are within
borderline. Some solvents that belong to protic category can be considered as amphiprotic and
act both as HBD and HBA. Table 1 shows the physical properties and classification of some
conventionally used solvents.

3
 Introduction

Table 1 Physical properties of some commonly used solvents at 25 oC7

solvent boiling point ( oC) dielectric constant dipole moment (D) classification

n-hexane 69 1.88 0 non-polar

benzene 80 2.3 0 non-polar

1,4-dioxane 101 2.3 0.45 non-polar

toluene 81 2.4 0.36 non-polar

diethyl ether 35 4.3 1.15 non-polar

chloroform 61 4.8 1.04 non-polar

ethyl acetate 77 6.0 1.78 polar aprotic

acetic acid 118 6.2 1.74 protic, amphiprotic

THF 66 7.5 1.75 polar aprotic

DCM 40 9.1 1.60 polar aprotic

Ammoniaa -33 16.7 1.42 dipolar aprotic

isopropanol 82 18 1.63 protic, amphiprotic

acetone 56 21 2.88 polar aprotic

ethanol 79 30 1.69 protic, amphiprotic

methanol 65 33 1.70 protic, amphiprotic

acetonitrile 82 37.5 3.92 dipolar aprotic

DMF 153 38 3.82 dipolar aprotic

DMSO 189 46.7 3.96 dipolar aprotic

water 100 80 1.85 protic, amphiprotic

a
Reference 8.

1.2 Solvent effects on solubility

The solubility of chemicals in solvents is one of the fundamental and direct factors that affect
the efficiency of organic reactions. The state of reactants dispersed in the solvent dictates
whether the reaction is under homogeneous and heterogeneous conditions, which is of great
importance for the kinetics and mechanisms of the reactions involved, more often than not,
this is also true for the selectivity and product yields of reactions. The classical view of
solvation process includes the following steps:9

4
 Introduction

1. A cavity must be created within the solvent to hold the solute molecule, this requires
the solvent molecules to overcome the intermolecular forces, such as hydrogen
bonding and dipole-dipole interaction between solvent molecules, so it is an
enthalpically unfavourable process. The creation of cavity also causes the order of the
solvent to increase, thus it is also an entropically unfavourable and so the total
contribution of this process to the overall free energy of solvation process is positive.
2. The solute must be separated from the bulk solute (solid or liquid) and requires the
conquest of solute-solute interactions to enable solute molecules to become free from,
for example, the restriction of crystal lattice energies in solid solutes. This process is
enthalpically unfavourable but generally entropically favourable. The net contribution
to the overall free energy of solvation is usually also positive.
3. The solute molecule inserts into the solvent cavity, and interacts and mixes with the
bulk solvent molecules. This usually results in enthalpically favourable solute-solvent
interactions, and due to the mixing, the mixture may become more disordered.
Therefore, this step is the key step for the dissolution of a solute, it contributes
negatively to the overall free energy of solvation.

The total free energy of solvation (ΔGsol) is the sum of the free energy changes from each
individual step. If dissolution of solute is to occur spontaneously, it must be accompanied by a
reduction in free energy. Empirically, dissolution occurs when solvent-solvent interactions are
similar to the interactions between solvent and solute, which is the principle behind the well-
known term “similia similibus solvuntur (like dissolves like).” There are several key
interactions that can affect the free energy of solvation, such as ion-dipole, dipole-dipole,
dipole-induced dipole, hydrogen bonding, electron-pair donor/electron-pair acceptor
interactions and last but not least, hydrophobic forces.

Kion-pair Kdissoc.
[B-A]solv [B A ]solv [B ]solv [ A ]solv
Ionisation Dissociation

Scheme 1

5
 Introduction

Solutions of compounds that can undergo ionisation and dissociation in solvents to produce
ion-pairs and free-ions involve further free energy changes (Scheme 1). The associated
equilibrium constants are: Kion-pair = [B A ] / [B-A]; Kdissoc. = [B ] • [A ] / [B A ].

A clear distinction must be made for the ionisation and dissociation steps, the former produces
solvent caged ion pairs by heterolysis of a covalent bond, or reduces the electrostatic forces of
the ionic bond by forming the solvent intervened contact ion-pairs (but still in a single solvent
cage), while the dissociation produces free-ions which are solvated and separated completely
by solvent molecules. The solvent can influence these two steps very differently, according to
the theoretical model (Equation 1), the potential energy of an ion-ion interaction of the
solvent separated ion pair is influenced by dielectric constant (εr), charge on the ion (Z • e) and
the distance between two charges (r).

1 z • z • e2
Uion-ion = •
4πε0 εr • r

Equation 1

Therefore, those solvents which have a sufficiently high dielectric constant are more likely to
cause the ion-pairs to become fully dissociated and solvated ions.

1.3 Solvent effects on chemical equilibria

Since the dissolution of chemicals in solvent to make a homogeneous solution involves the
interactions between solute and solvent molecules, and these interactions are key to the
solvation energy, it is not difficult to imagine that different solvents may exert a specific
solvation preference towards different solutes, which generates a differential stabilisation of
the reactant or product which is the origin of solvents effects on the chemical equilibrium. In
general, the equilibrium will be shifted by a change in solvent so as to favour the side most
stabilised by solvation.

6
 Introduction

1.3.1 Acid/Base equilibria

Of all the equilibria that can be affected by solvent effects, acid/base equilibria are probably
the most important and have been widely investigated.10 The ionisation equilibrium of an acid
or a base (Scheme 2), is often strongly influenced by a solvent change. Obviously, the basicity
or acidity of the solvent plays an important role in the ionisation of acids or protonation of
bases, respectively. For example, the ionisation of acids is essentially a proton transfer
reaction from the acid to the solvent. However, the dielectric constant and the ability of the
solvent to preferentially solvate the various species are also of crucial importance. The
dielectric constant of a solvent can, in some cases, significantly affect the acidity or basicity of
K
HAn+1 SH SH2 An
K'
An SH HAn+1 S

Scheme 2

compounds. For example, acetic acid has a pKa of 4.76 in water (εr = 78.4, 25oC) at 25oC,
while that in methanol11 (εr = 32.7, 25oC)12 is 9.63. Although water is only 10-20 times more
basic than methanol, there is a large decrease in the acidity constant. These observations are
consistent with the prediction from the theoretical model that the pKa of acid is inversely
proportional to the dielectric constant of the solvent. The effect of the solvent dielectric
constant on the ionisation constants depends very much on the charges of the species involved.
If the acid has a charge of +1, for example, NH4+, the dissociation constant of the acid has
very limited sensitivity towards the change of solvent dielectric constant, because there is a
positive charge on each side of the equilibrium. On the other hand, the dissociation constants
of those acids which are neutral or have negative charge are greatly affected by the solvent
dielectric constant, because of the change in charge upon ionisation. For neutral and charged
acids, the general trend is for the acidity to increase with increasing dielectric constant of the
solvent.

The rules above are sometimes not in very good agreement with experimental observations,
for example, the acidity of picric acid (2,4,6-trinitrophenol) at 25oC is only 3000 times greater
in water (pKa = 0.43) than in methanol (pKa = 3.90),13 less than the difference observed for

7
 Introduction

acetic acid, moreover, it is only about 5 times less acidic than in DMSO (pKa = -0.30),
although the dielectric constant of DMSO is about half of that for water. The reason for the
acidity of picric acid being less sensitive to the change of solvent is presumably due to the
charge on the oxygen anion being highly delocalised over the large benzene ring and nitro
groups, unlike the case for the carboxylate anion, where the charge is only delocalised over
two oxygen atoms. The example above show that in addition to the basicity/acidity of the
solvent and its dielectric constant, there are other interaction forces, such as ion-dipole, dipole-
dipole, hydrogen bonding, etc., which may result in a specific solvent effect on the acid/base
equilibria.

The ionising ability of a solvent increases with solvent ionic strength (I), especially for solutes
which are susceptible to a strong electric field effect, in other words, have a greater
polarisability. Adding salt can affect ionisation and dissociation constants differently, but in

4.8

4.7
pKa(water)

4.6

4.5

4.4
0 0.5 1 1.5 2 2.5
Ionic strength(NaNO3)/M

Figure 1 Aqueous pKa of acetic acid decreases with the increase of ionic strength (NaNO 3) at
25 oC

8
 Introduction

general, without the presence of common ion effect and at low salt concentration, the
ionisation and dissociation constant increases with increasing ionic strength. For example, the
pKa of acetic acid decreases from 4.76 in pure water to 4.45 when ionic strength (I) increases
to 1.5M (NaNO3) (Figure 1).

1.3.2 Tautomeric equilibria

The preferential solvation of different species in tautomeric equilibria, such as keto/enol,

imine/enamine, azo/hydrazone, ring/chain equilibria, etc. can also give rise to solvent effects.
Solvent effects on these equilibria are often key to the selectivity of organic reactions when
ambident nucleophilic species are involved. Among all of the tautomeric equilibria, keto/enol
equilibria have been most widely studied.14 For example, in a symmetrical 1,3-diketone, there
is a equilibrium between two possible tautomers (enol, and 1,3-diketo form) (Scheme 3):

enol 1,3-diketo

Scheme 3

The keto/enol equilibrium constant Ktauto can be expressed as: Ktauto = [enol]/[diketo] and the
values for acetylacetone and dimedone (5,5-dimethylcyclohexane-1,3-dione) in different
solvent are shown in Table 2.

In solution, the 1,3-diketo form is the dominant species for the open chained acetylacetone,
solvent effects on the equilibrium constant are relatively small in typical dipolar aprotic and
dipolar protic solvents. The enol form only becomes significant in some nonpolar aprotic
solvents, but there is still more than 70% of the 1,3-diketo form (in THF). Compared with the
1,3-diketo form, the enol is overall less polar due to an intramolecular hydrogen bond which
also reduces its susceptibility to stabilisation by H-bonding from the solvent, whereas the 1,3-
diketo form could be stabilised in this way. Thus, a change of solvent from a dipolar protic or
dipolar aprotic to a nonpolar solvent favours the formation of intramolecular hydrogen
bonding in the enol at the expense of the 1,3-diketo form. In contrast to the open chain

9
 Introduction

Table 2 Equilibrium constants of enol tautomers of acetylacetone and dimedone in different solvents at 20 oC1

Solvent (deuterated) Ktauto (acetylacetone) Ktauto (dimedone)

gas phasea 0.74 ___

tetrahydrofuran 0.40 ___

toluene 0.39 0.08

tetrachloromethane 0.29 ___

1,4-dioxane 0.13 2.8

acetone 0.13 4.2

chloroform 0.09 0.05

methanol 0.07 148

water 0.07 19

dimethyl sulfoxide 0.05 94

a
40 oC value.

acetylacetone, the cyclic dimedone has a significant amount of enol form but the variation
with solvent is not as clear as with the open chained compound. The enol form of dimedone is
favoured much more in dipolar solvents than in nonpolar solvents, presumably because the
1,3-diketo form is locked in the cis- arrangement with consequentially a high dipole moment
and the enol form is unable to easily form an intramolecular H-bond and so is much more
susceptible to H-bonding stabilisation from the solvent.15

An interesting example shows how intramolecular hydrogen bonding affects the enol/keto
equilibrium in various solvents with different polarity. For β-ketonitriles, due to the linear
structure of cyano group, the six-membered cyclic intramolecular hydrogen bond is unable to

C C

Scheme 4

form (Scheme 4). Without this stabilisation, therefore the intermolecular hydrogen bonding
between enolic OH and solvent molecules are more likely to form in protic polar solvents than

10
 Introduction

nonpolar ones. In addition, the conjugation between the double bond and the cyano group
further stabilises the enol form which also leads to some charge separation. Thus, the enol
content increases as the solvent changes from nonpolar to polar.16

Similar solvent effects are also in lactim/lactam equilibria. One of the most classic examples is
solvent effects on hydroxypyridine/pyridone equilibria (Scheme 5) in which hydroxypyridines
dominate in the gas phase. However, the reverse situation is found in solutions, and increasing
the solvent polarity significantly shifts the equilibrium towards the pyridone form. The
difference in equilibrium constant in water and the gas phase is greater than 104 for 4-
hydroxypyridine at 25 oC.17 Due to resonance, both 2- and 4- pyridone is a charge separated
species and so the keto form is more polar than the corresponding hydroxypyridine. Polar
solvents significantly stabilise the charge distribution of the pyridone form.

..
..

4-Hydroxylpridine 4-pyridone

.
.. O

2-Hydroxylpridine 2-pyridone

Scheme 5

Besides the examples described above, solvent effects can also be observed in Brønsted
acid/base, Lewis acid/base, conformational, and E/Z isomerisation equilibria, etc., all of which
reveal the influences of solvation on relative molecular stabilities.

11
 Introduction

1.4 Solvent effects on the rates of organic reactions

The rate and order of chemical reactions can be considerably changed by a change of solvent,
and in some extreme cases, rate accelerations as high as up to 109 can be achieved by a solvent
change.18 Studies of solvent effects on the rate of chemical reactions have a long and
enlightening history, for example, as early as 1890, Menschutkin demonstrated that the rate of
quaternisation of triethylamine by ethyl iodide was found to be very dependent on the reaction
medium. Compared with the rate in n-hexane, the rate of the Menschutkin reaction is 4 times
faster in ether, 280 times faster in methanol, and about 750 times faster in benzyl alcohol.19
The choice of solvents is especially important for the chemical industry, where an appropriate
choice of reaction medium can not only significantly reduce the cost of a product, but also can
be eco-friendly to the environment. Therefore the establishment of theories and rules for
solvent effects on reactions can facilitate the design and selection of a solvent for organic
synthesis.

In principle, the way in which solvents can affect reaction rates can be rationalised in terms of
transition-state theory3,20 which describes the differential solvation of reactants and activated
complexes leading to a change in the free energy of activation. Transition state theory is
essential for a qualitative understanding of solvent effects on reactions, although it has
limitation in some extreme cases, for example, when the rate of solvent reorganisation
becomes rate-limiting. The latter has only been given attention recently due to rate
measurement techniques becoming more advanced. It concerned with the non-equilibrium
solvation of the activated complex that may occur in some fast or ultra-fast reactions, which is
generally ignored in conventional transition state theory. 21 Under such circumstances, the rate
of reaction will depend on the solvent dynamics or friction and so with the density, viscosity,
internal pressure, etc.

1.4.1 Transition state theory3,20

A B [A-B]‡ C D
reactants activitate complex products

Scheme 6

12
 Introduction

Consider a reaction between reactants A and B passing through an activated complex [A-B]‡
to give products C and D (Scheme 6), in which the reactants are quasi-equilibrated with
activated complex[A-B]‡. The equilibrium constant can be given as follow:
K‡ = [A-B]‡ / [A][B], in which [A], [B] and [A-B]‡ corresponding to the concentration of
reactants and activated complex respectively. Transition state theory presumes that the
formation of products, C and D, does not affect the equilibrium between reactants and
activated complex and the free energy of the activated complex occupies the highest point
during the reaction process and is defined as transition state, the activated complex is the
corresponding chemical entity, and the structure of the activated complex is defined as
transition state structure (Figure 2).

E
[A-B]‡

transition state

ΔG‡

A+B

reactants
o
ΔGrnx
C+D
products

reaction coordinate

Figure 2

Figure 2 Single-dimensional reaction energy profile of the reaction (Scheme 6). ΔG‡ : Gibbs
energy of activation; ΔGrnxo: Gibbs energy of the reaction.

The decomposition of activated complex to give the products proceeds with a fixed rate
constant kT/h and so effectively rates of reactions vary as the concentration of the activated
complex vary. Assuming that the reactants and activated complex are in thermal equilibrium
with the solvent, a change in solvent will lead to a modification of the height of the reaction
energy barrier by differential solvation of the activated complex and reactants. A change of

13
 Introduction

solvent that reduces the energy barrier for reaction can be realised by either decreasing the
energy of the activated complex though solvation stabilisation or by increasing the energy of
reactant state though solvation destabilisation. It is the net differential change in the free
energies of solvation that determines whether a change in solvent results in an increase or
decrease in the rate of reaction. The solvation effects on the activated complex and reactants
could be in the same or opposite direction (Figure 3).

E activated complex activated complex

E

ΔG3‡
ΔG2‡
ΔG1‡ ΔG1‡

ΔG2‡
ΔG3‡

reactants reactants

products products

reaction coordinate reaction coordinate

(a) (b)

Figure 3

Figure 3 The reduction of the reaction energy barrier by differential solvation of (a) the
activated complex, (b) the reactants. ΔG1‡: Gibbs free energy of activation of the reaction in
solvent 1; ΔG2‡: Gibbs free energy of activation of the reaction in solvent 2; ΔG3‡: standard
Gibbs transfer free energy for the reactants (ΔGR1→2) or activated complex (ΔG‡1→2).

From the measurement of rate constants in different solvents and different temperatures, the
activation parameters can be acquired and compared, thus enabling analysis of solvent effects
on reaction rates.

14
 Introduction

1.4.2 Hughes-Ingold solvation ‘rules’

The structure of the activated complex in the transition state may be modified by changes in
solvent. According to the Hammond postulate,22 reactions can be described as proceeding
through a reactant-like or a product-like transition state structure. These structures can vary in
terms of bond formation/breaking and charge distribution. Like „normal‟ molecules, the
activated complexes are subjected to the interactions with solvent molecules and these may be
stabilising or destabilising, which, in some cases, may even lead to a change in reaction
mechanism. Generally, reactions proceed through activated complexes which can be grouped
roughly into polar, isopolar and free-radical types. Polar types of activated complexes have a
considerable charge separation which differs from the reactant state and normally exhibits the
largest solvent effects. Hughes and Ingold23 investigated solvent effects on a range of aliphatic
nucleophilic substitution reactions, but only the pure electrostatic interactions between ions or
dipolar molecules and solvent molecules in reactant and transition states were considered in
their qualitative solvation model. Based on some simple assumptions and observations they
concluded that the creation or destruction or dispersal of charge is expected to be subjected to
an increase and decrease of solvation, respectively. The overall effect of the solvent on the rate
of reactions of different charge types can be summarised as follows:

1. An increase in charge density ongoing from the reactants to activated complex will
lead to an increase in rate in a more polar solvent.
2. A decrease in charge density ongoing from the reactants to activated complex will
cause a decrease in rate in a more polar solvent.
3. A dispersion of charge ongoing from the reactants to activated complex results in a
small or negligible decrease in rate with a change of solvent polarity.

Some examples6 can be found that agree with these Hughes-Ingold rules (Scheme 7) and a
more generalised summary of the solvent effects on aliphatic substitution reaction is given in
Table 3.24

A further extension of the Hughes-Ingold rules was made by Westaway25 regarding the
solvent effects on the structure of activated complexes in SN2 reactions. This solvation rule
argued that the solvent effects on the transition state structure is primarily determined by the

15
 Introduction

interactions of solvent molecules with the incoming nucleophile and leaving group, and that
the activated complex, penta-valent central carbon, is unlikely to have a strong interaction

k H O / kEtOH
2

 -
‡
[ ] 1500

charge separation


-  ‡
HO NR3 [ HO C NR3 ] 0.001

charge neutralisation

- - ‡

HO Br [ HO C Br ] 0.2

charge dispersal

Scheme 7

Table 3 Predicted solvent effects on rates of nucleophilic substitution reactions

Reaction Initial Activated complex Charge alteration during Effect of increased

type reactants activation solvent polarity on rate

SN 1 R-X Rδ+…..X δ- Separation of unlike charges Large increase

SN 1 R-X+ Rδ+…..X δ+ Dispersal of charge Small increase

SN 2 Y + R-X Y δ+…..R…..X δ- Separation of unlike charges Large increase

SN 2 Y- + R-X Y δ-…..R…..X δ- Dispersal of charge Small increase

SN 2 Y + R-X+ Y δ+…..R…..X δ+ Dispersal of charge Small increase

SN 2 Y- + R-X+ Y δ-…..R…..X δ+ Destruction of charge Large decrease

with solvent. The transition state of a reaction in which the nucleophile and leaving group
have the same charge will not significantly be influenced by a change of solvent (Type I SN2
reaction). On the other hand, a change in solvent will lead to a change of transition state
energy for a reaction in which the nucleophile and leaving group bear different charges (Type

16
 Introduction

II SN2 reaction). In essence, the Westaway solvation rule is a refinement of Hughes-Ingold

rules, both of them established a good guide for predicting solvent effects on chemical
reactions in a qualitative way.

Despite numerous examples that have been found to be consistent with Hughes-Ingold rules, it
has its limitations. The model of Hughes-Ingold rules was built upon the assumption and
simplification that only pure static interactions are considered, and they are not capable of
evaluating solvent effects on reactions that have an isopolar activated complex. Moreover, the
rules are based on a continuous solvation model, assuming the solvent molecules equilibrate
thermodynamically with reactants and activated complexes, but this is not appropriate for
some very fast reactions in which the rate limiting step is the solvent reorganisation. A very
interesting and counterintuitive example challenges the validity of the rules. A theoretical
study shows that the Gibbs free energy of activation for S N1 ionisation of t-butyl halides in
solution decreasing with increasing solvent polarity, which is in line with the rules. However,
the study also shows a decrease in the stabilisation of the activated complexes with increasing
solvent polarity. This indicates that the activated complex becomes less charged with
increasing solvent polarity, which is not easily rationalised with the Hughes-Ingold model.26
However, if the charged transition state structure is stabilised by a more polar solvent, then,
according to the Hammond Postulate this should lead to a more reactant-like transition state
i.e. less bond fission and less charge development in the transition state. In addition, Hughes-
Ingold rules do not directly consider the contribution from the entropy of activation to the
overall Gibbs free energy of activation, and regard the enthalpy of activation as dominant.
However, solvent restriction will result from the interaction of solvent molecules with a
charged activated complex, giving rise to a large entropy loss. For example, the activation
parameters for SN1 reaction in other solvents and water indicate that changes in the highly
ordered water structure make the activation process entropy controlled.

1.4.3 Specific solvation effects on reaction rates

Besides pure electronic interactions, such as inductive and dispersion forces which exist
between solvent molecules and reactants or activated complexes, there are some other specific
intermolecular forces which are responsible for the solvent effects on the kinetics and
mechanisms of reactions in solution. Hydrogen bonding forces are among one of the most

17
 Introduction

important specific interactions. In nucleophilic substitution reactions in which the attacking

nucleophiles or leaving groups are ionic, hydrogen bonding forces are especially important for
the understanding of solvent effects on the kinetics and mechanisms of these reactions. In this
sense, protic solvents have a great advantage by forming hydrogen bonds with reactants and
leaving group. This is especially the case when the reactants and leaving groups are negatively
charged anions. Specific solvation of leaving anions by hydrogen bonding facilitates the
decomposition of activated complexes to form products, thus increasing the rate of reaction.
On the other hand, specific solvation of attacking anions by hydrogen bonding leads to a
decrease in the relative nucleophilicity of nucleophiles. For example, the rate of neighbouring
group assisted solvolysis of 4-methoxyneophyl tosylate (I) is tremendously varied by about

106-fold on changing solvent from diethyl ether to formic acid due to the difficulty of
solvating the leaving tosylate anion by ether.27 Counter-ions can also affect the nucleophilicity
of anionic nucleophiles, especially in non-polar solvents.

Dipolar aprotic solvents, such as DMSO, acetonitrile and DMF, normally have large dipole
moments and relatively high dielectric constants (Table 1). These solvents have no, or very
limited, ability to act as hydrogen bond donors and are often called dipolar non-HBD solvents.
Anions are poorly solvated in these solvents and hence are better nucleophiles than in, say,
water leading to large rate enhancements in these solvents. As dipolar aprotic solvents are not
good hydrogen bond donors, anionic nucleophiles are more „free‟ than in protic solvents in
which these nucleophiles are solvated by the hydrogen bonding and must be at least partially
desolvated for reaction to occur. For example, the rate of S NAr reaction between sodium azide
and 4-nitrofluorobenzne in a typical dipolar aprotic solvent is about 104-107 times faster than
in protic solvents and large rate enhancements are also seen for S N2 reactions involving
anions.28 In contrast, the rate of SNAr reaction between neutral secondary amines and 4-

18
 Introduction

fluorobenzene is not affected very much when the solvent is changed from protic to dipolar
aprotic (Table 4).

Table 4 Relative rates of representative SNAr, SN2 reactions in protic and dipolar aprotic solvents at 25 oC29

log (k2,solvent / k2,MeOH)

solvents
N3 N3

MeOH 0 0 0

H2O 0.8 N. A. N. A.

HCONH2 1.1 0.8 N. A.

acetonitrile 3.7 3.9 0.9

DMSO 3.1 3.9 2.3

DMF 3.4 4.5 1.8

acetone 3.6 4.9 0.4

HMPT 5.3 7.3 N. A.

It is worth noting that dipolar aprotic solvents also could influence the stability of activated
complexes. Theoretical studies show that the intermediate Meisenheimer complex for the
reaction of azide anion with 4-NFB is more stable relative to reactants in dipolar aprotic than
protic solvents (Figure 4).30

19
 Introduction

Figure 4 The QM/MM calculation of potential energy for the reaction of azide anion with 4-
NFB in different solvents. TS1: transition state 1; TS2: transition state 2; IC: intermediate
complex.

1.5 Solvent effects on UV spectra

In solution, the UV spectra of organic compounds may differ from one solvent to another in
their wavelength of maximum absorption and extinction coefficient. These changes can be
interpreted in terms of the changes of interaction between solute and solvent, which can alter
the energy gap(s) between ground and excited state of solute molecule. In many cases,
ionisation can play an important role on the UV absorption spectra of organic compounds and
these spectral changes can allow the determination of dissociation constants.31 Normally,
solvent effects on the UV absorption spectrum are pronounced when the energy gap between
the ground and excited state of the molecule is relatively large, whereas it is relatively small in
benzene, polyenes, etc.

20
 Introduction

1.6 Empirical parameters of solvent polarity

It is not easy to measure solvent polarity but simple physical properties, such as dielectric
constant or refraction index, are often used to correlate rate constants of reactions. For
example, it was found that logarithms of relative rate for Menschutkin reaction shows good
linear relationship with the Kirkwood function32 [(εr-1)/(2εr+1)] of the solvent (Figure 5).33
Unfortunately, in most cases it has been found that no correlation between dielectric constant
and logarithms of rate constants of solvent sensitive reactions. Obviously, the establishment of
solvent scale need to combine all possible detailed intermolecular interaction forces between
solvent and solute molecules, such as dipole-dipole, ion-dipole interactions, electrostatic
forces, hydrogen bonding and EPD/EPA interactions, etc. No single macroscopic physical
parameters could reflect the intermolecular forces on microscopic level.

Figure 5 The correlation between lg(k/ko) and the Kirkwood function [(εr-1)/(2εr+1)] for the
o
Menschutkin reaction between triethylamine and iodoethane at 40 C in binary
acetone/benzene and acetone/1,4-dioxane mixtures.

21
 Introduction

1.6.1 Grunwald-Winstein Equation34

Despite the difficulty and complexity to establish a satisfactory solvent polarity scale,
Grunwald and Winstein made one of most ambitious attempts to correlate rate constant with
empirical solvent polarity parameters. Based on the solvolysis rate of t-butyl chloride, they

Y = log k t-BuCl - log k0 t-BuCl

log (k/k0) = m • Y + c

Equation 2

(Equation 2, Grunwald-Winstein equation, where k0 t-BuCl : the solvolysis rate of t-butyl chloride in 80% (v/v)
aqueous ethanol at 25 oC (Y = 0); k t-BuCl : the solvolysis rate of t-butyl chloride in the solvent of interest at 25 oC).

found that polar solvents significantly accelerate the reaction and arbitrarily defined a Y
parameter that describes the ionising power of solvent, and a substrate parameter m that senses
the rate change to the ionisation power of solvent (Equation 2).

The correlation of Y and m is a type of linear free energy relationship (LFER), and is
reasonably successful in correlating the rates of the S N1 solvolytic reactions of various tertiary
alkyl halides, secondary alkyl sulfonates. However, an unsatisfactory correlation is found for
some solvolysis reactions with borderline mechanism between S N1 and SN2, such as the
solvolysis of secondary haloalkanes. The main concern with the Grunwald-Winstein equation
is whether the solvent intervenes in SN1 solvolysis processes. Schleyer later showed that the
solvolysis of t-butyl and 1- or 2-adamantyl chlorides and adamantyl tosylates occur without
nucleophilic assistance from the solvent and can be best regarded as „true‟ SN1 process,
although protic solvents can facilitate the expulsion of the leaving anion by strong

log (k/k0)RX = m • Y + l • N + c

Equation 3

(Equation 3, modified Grunwald-Winstein equation to account for nucleophilic solvent assistance, where l:
sensitivity of substrate to the solvent nucleophilic assistance; N: solvent nucleophilicity, in contrast to Y, ionising
power of solvent).

22
 Introduction

hydrogen bonding.35 However, studies also found that the solvolysis rates for t-butyl chloride
and adamantyl chloride are very different in some nucleophilic solvents, which suggests
partial nucleophilic assistance from some solvents. 36 In order to account for possible
nucleophilic solvent assistance, the Grunwald-Winstein equation was later modified to a
multiple parameter one (Equation 3).37

The Grunwald-Winstein solvent scale is applied mainly to SN1 reactions, and the SN2
Menschutkin reaction of tri-n-propylamine and iodomethane has been used to measure solvent
polarity.38

1.6.2 Other spectroscopically obtained empirical parameters

Spectroscopic methods such as UV-Vis, IR, and NMR have been used to establish a solvent
polarity scale. Several solvent scales were successfully established by using solvatochromic
dyes as indicator.39 Kosower set up a Z parameter as the molar transition energy based on 1-
ethyl-4-methoxycarbonyl)pyridium iodide as model, a higher Z value corresponding to a more
polar solvent. So far, about 60 commonly used solvents are included in Kosower Z scale.40
Dimroth and Reichardt proposed an ET (30) parameter based on the UV absorption of N-
phenolate betaine dyes.41 Due to the extraordinarily large range of solvatochromic absorption
of betaine dyes, a much wider solvent polarity spectrum can be established, and so far, 360
more ET (30) values have been measured. Based on solvent hydrogen bond acceptor basicity
and hydrogen bond donor acidity, a method called solvatochromic comparison was introduced
by Taft and Kamlet, using HMPT and methanol as a reference solvent for constructing
hydrogen bond acceptor (β) and donor (α) ability, respectively, a dual parameters model for
solvent relative HBD/HBA tendency was established.42 Also a parameter π* was introduced
for the description of solvent dipolarity and polarisability, which was based on the solvent
effects on π → π* electron transitions of several nitroarmatic compounds. Those parameters
reflect the solvent properties in different ways, and a combination of them sometimes can give
good correlations in multiple parameter models.43 However, it must be pointed out that there is
no single method so far that can give an absolute accurate solvent scale, unsatisfactory
correlations or limitations for each method are unavoidable which reflects the intriguing nature
of solvent effects on chemical reactions.

23
 Introduction

2. Physical and chemical properties of liquid ammonia 44

Ammonia has three covalent N-H bonds and a lone pair of electrons, and adopts a trigonal
pyramidal shape as the valence shell electron pair repulsion theory predicted (Figure 6). In the


o
1.01 A

106.5o

Figure 6 Structure of ammonia molecule

gas phase, the bond angle of ammonia (HNH) is 106.5o, smaller than that of methane
(109.5o) and greater than that of water (105o), and the N-H bond length of ammonia is 1.01 Å,
which is also in between of that for methane (1.10 Å) and water (0.96 Å). The ammonia
molecule has the form of a low pyramid of height 0.360 Å,45 and this configuration gives rise
to the possibility of the nitrogen atom passing from its equilibrium position on one side of the
plane of the hydrogen atoms through the plane to an equally stable position on the other side.
The energy barrier for this nitrogen inversion is 24.7 kJ mol-1 at room temperature.46 Because
ammonia has three possible hydrogen bond donors but only one lone pair for H-bond
acceptance, its arrangement in the liquid and solid phases is interesting. In the solid phase, X-
ray diffraction indicates that each ammonia molecule is surrounded by six nearest neighbours
though asymmetrical and nonlinear hydrogen bonds and each electron pair on a nitrogen atom
can accommodate three hydrogen bonds.47 While, in the liquid phase, an ammonia molecule is
surrounded by an average of 11 molecules close to its boiling point.48 However, interestingly,
infra red spectroscopy of liquid ammonia at room temperature shows that less than 3 hydrogen
bonds are formed,49 in stark contrast with the X-ray results, which suggests that temperature
and pressure effects are very important factor for the properties of liquid ammonia.

Table 5 lists some physical properties of liquid ammonia, which has a vapour pressure of 10
bar at room temperature. Its enthalpy of vapourisation (ΔHvap) of 23.25 kJ mol-1 is
intermediate between those of methane (8.19 kJ mol-1) and water (40.65 kJ mol-1), suggesting
a moderate degree of association.

24
 Introduction

Table 5 Some physical properties of liquid ammonia 50

critical temperature 132.4 oC

critical pressure 112.3 atm

vapour pressure of liquid log10 Pmm = 9.95028 - (1473.17)/(T - 3.8603) 10-3 T

enthalpy of formation 5.64 kJ mol-1

enthalpy of vapourisation 23.25 kJ mol-1

refractive index 1.325

electrical conductivity 1 × 10211 ohm-1 cm-1

dielectric constant 22.70 (-50 oC), 18.94 (5 oC), 17.82 (15 oC ), 16.70 (25
o
C)

The lone pair of ammonia is of great importance in determining the properties of ammonia. It
makes liquid ammonia one of the most basic molecular liquids and it enables ammonia to be a
good electron donor and hydrogen bond acceptor. As ammonia has only one lone pair and the
electronegativity of nitrogen (3.04) is smaller than that of oxygen (3.44), the properties of
ammonia are also rather different from those of water. The most obvious difference between
ammonia and water is boiling point. At one atmosphere, ammonia boils at -33.5 oC, compared
with 100 oC for water. Despite the difference in properties between liquid ammonia and water,
they are both classified as protic/amphiphilic solvent in a classical review1 and are often listed
together for comparison (Table 6).

At 25 oC the dielectric constant of water is about 4.7 times greater than that of liquid
ammonia, which will make an obvious difference on the ionisation and dissociation of
compounds in these two solvents. Specifically, the high dielectric constant of water facilitates
the dissociation of ion-pairs while low dielectric constant ammonia favours the formation of
ion-pairs but not their dissociation into free ions.51 The autoprotolysis (self-ionisation)
constant of ammonia is much lower than that of water, corresponding to a pK = 27.7. In
addition, the internal energy of liquid ammonia is about -21 kJ mol-1, which is around half the
value for water.52As stated earlier, ammonia has only one lone pair of electrons and yet with
three potential NH hydrogen bond donors, it cannot form a highly organised, web-like solvent

25
 Introduction

structure as in water. It is not surprising that liquid ammonia is highly soluble in water, to give
a basic solution.

Table 6 Some physical properties of liquid ammonia and water

Physical property Liquid ammonia Water

Melting point (oC) -77.7 0

Boiling point (oC) -33.35 100

Dielectric constant (25 oC) 16.7 80

Dipole moment (D) (25 oC) 1.42 1.85

Density (g/ml) (25 oC) 0.606 0.997

Polarisability (Å3) 2.21 1.84

Autoprotolysis constant (25 oC) 2.0 ×10-28 1.0 ×10-14

Empirical polarity (ET30 /kcal mol-1) 51.7a 63.1

a
Reference 53.

Liquid ammonia has a very strong tendency to donate lone pair electrons as indicated by its
very high donor number (DNLNH3 = 59 kcal mol-1), which is much greater than that for water
(DNwater = 18 kcal mol-1) and even HMPT (DNHMPT = 38.8 kcal mol-1),54 and, as a
23
consequence, it strongly solvates cations through a electron donation, as evidenced by Na
chemical shifts.55 However, unlike water, it is not a good hydrogen bond donor 56 and does not
significantly solvate anions, as shown by the high single ion transfer energies from water.57
Many synthetically useful salts are highly soluble, particularly ammonium salts, e.g. NH4N3,
67.3g/100g at -36 oC, 58
and even a ca. 30 M NH4NO3 is possible in liquid ammonia at 25
o 44a
C. Some other salts, such as fluorides, and those with multiple negative charged anions, are
difficult to dissolve in liquid ammonia at ambient temperature. For organic solutes, three
factors may affect the solubility of a molecular substance in liquid ammonia: the magnitude of
the dispersion forces, the polarity of the ammonia molecule, and the ability of solute to form
hydrogen bonds. Generally speaking, most common organic compounds are more soluble in
liquid ammonia then in water. Hydrocarbons are insoluble, but alkyl ethers, alcohols and
amines are miscible with ammonia, and most importantly, the majority of aromatic
compounds have a good or moderate solubility at ambient temperature. Solubility is obviously
important for the potential application of liquid ammonia as an organic reaction medium.

26
 Introduction

The chemical properties of ammonia are closely associated with its physical properties. The
lone pair of ammonia makes it a Lewis base and ammonia can accommodate a proton to form
a tetrahedral ammonium cation. Due to electron donation from the lone pair, metal ions are
coordinated by ammonia molecules. For examples, copper (II) cation forms
+
tetraamminediaquacopper (II), [Cu(NH3)4(H2O)2]2 complex in aqueous ammonia, silver (I)
forms diamminesilver (I), [Ag(NH3)2]+. Much attention has been given to the inorganic
chemistry in liquid ammonia by Lagowski and others.8,44c,59 One of the most extraordinary
phenomenon is that alkali metals can be dissolved and ionised in liquid ammonia to form a
solution of the metal cation and solvated electrons. Solutions of sodium or potassium in liquid
ammonia are widely used as reducing agents in the organic synthesis (vide infra).60

3. Organic reactions and mechanisms in liquid ammonia

Since the commercial availability of liquid ammonia at the beginning of 19th century,
chemical research in this unique solvent became popular. The early research work in this area
was performed by Cady, Franklin and Kraus,61 mainly covering on the physical and chemical
properties of liquid ammonia, and especially through Kraus' and his many years of dedication
to the chemistry of ammonia solutions, some fundamental data on liquid ammonia became
available, such as the solubility of organic and inorganic chemicals, the conductivities of
ammonia solutions, ionisation of metals in liquid ammonia,62 etc., Audrieth,,63 Watt64 and later
Lagowski et al, investigated some fundamental organic and inorganic reactions in liquid
ammonia. So far the largest body of literature related to liquid ammonia is on the reduction of
organic compounds in alkali metal/ammonia solution, and the application of sodium/lithium
amide in liquid ammonia as a strong base in synthesis. The last major review of the literature
on organic reaction in liquid ammonia was edited by Smith in 1963.65 Generally, the uses of
liquid ammonia in organic reaction are divided into the following areas:

1. As a supporting medium that dissolves alkali metals for reduction purposes.

2. As a medium that dissolves alkali metals for the production of very strong bases.
3. As a reagent (solvolysis or amination) or medium for organic reactions.

27
 Introduction

It should be pointed out that probably due to the limitation of equipment and the awkwardness
of handling liquid ammonia at ambient temperature or above, most synthetic work has been
done below or near the boiling point of ammonia.

3.1 As a solvent for reduction

The reduction of organic compounds with metals in liquid ammonia is a well-established

synthetic technique. When an alkali metal is dissolved in liquid ammonia, a deep blue solution
is formed. The colour is a characteristic of the solvated electron that is given up by metal in
the medium, and these electrons can be donated to substrates to form radical anions and
dianions, and the reduced product is generated after the protonation of the anion.

3.1.1 Cycloalkanes

Generally speaking, alkanes are inert to the reducing conditions, but the α-keto-cyclopropanes
can be reduced to the corresponding straight-chain ketone (Scheme 8).66,67 The reaction
probably proceeds through formation of the radical anion of the ketone followed by the
interaction of radical with the ring and subsequent reduction and protonation of the carbanion.

O O
Li/NH3 H+
R R

Scheme 8

3.1.2 Conjugated alkenes and alkynes

Non-conjugated alkenes are difficult to reduce by alkali metals in liquid ammonia and, in fact,
they are often the end product of the reduction of alkynes or conjugated alkenes and aromatic
rings. Conjugated alkenes can be reduced easily to mono-alkenes, however, the reactive
intermediate radical anion is active enough to undergo polymerisation or other radical
reactions, and the stereochemistry could not be predicted due to the conformational instability
of the allylic radical. For example, butadiene is reduced to a mixture of cis- and trans-2-butene
with lithium in liquid ammonia, the ratio of the two isomers depends on the reaction
temperature (Scheme 9).68

28
 Introduction

Li / NH3
+

Scheme 9

Alkynes are reduced predominantly to trans-alkenes, the stereochemistry of the reaction is

dictated by the configuration of the intermediate radical anion or dianion, which prefers the
trans form in order to minimise the interactions between the two newly-created negatively
charged sp2 centres. Thus, 2-heptyne is reduced via the dianion to give exclusively trans-2-
heptene (Scheme 10).69

+
Na/ NH3 H
C4H9 C4H9

Scheme 10

3.1.3 Carbonyl groups

All carbonyl groups, with the exception of the salts of carboxylic acids, are reduced by alkali
metals to hydroxyl groups in liquid ammonia. Due to the coupling of the reactive radical
intermediate, the reduction of esters and aldehydes has very little synthetic use, however, if a
proton donor, such as alcohol, is added to capture the reactive intermediate, the unwanted
coupling reaction can be avoided, and the reduction can then be run successfully. The
following are some typical examples (Scheme 11).70,71

Na/ NH3 H
O
t-Butanol OH
H

98%
Scheme 11

29
 Introduction

3.1.4 Aromatic Compounds

Possibly the best known of all alkali metal/ammonia reductions is the Birch reduction, the
partial reduction of aromatic systems to 1, 4-dihydrocyclohexadienes.72,73,74 The regio-
selectivity of the Birch reduction depends on whether the substituent is electro-withdrawing or
electro-donating. The intermediate radical anion tends to locate itself on the carbon ortho to an
electro-withdrawing group and on the carbon meta to the electro-donating group in order to
minimise the interactions.75,76 This is best illustrated by the two examples below (Scheme
12).77,78
OMe OMe

Na/NH3

CN CN

Na/NH3

Scheme 12

3.1.5 Halogen-containing compounds

Most of the work in this area has involved the reduction of aromatic halides, and there are few
examples of the reduction of aliphatic halides, although the following is an example of the
reduction of 2, 3-dibromobutane in liquid ammonia (Scheme 13).79

Na e-

LNH3 Na/LNH3 LNH3

Scheme 13

30
 Introduction

3.1.6 Nitro, nitrile groups and epoxides

Aromatic nitro groups are reduced to anilines, however, the yield is poor compared with
conventional reduction methods, and is probably due to the reactive intermediate radical or
radical anion involved.80 Surprisingly, aromatic nitriles are not easily reduced to the
corresponding amine, but, instead the aromatic ring is reduced. The intermediate radical anion

_
Na/LNH3 RX

H+

Scheme 14

attacks other substrates to give the nucleophilic substitution products which is synthetically
useful (Scheme 14).81 Alkali metal/ammonia reactions with epoxides open the ring and form
alcohols, the stereochemistry of the process depends on the stability of the anion or dianion
generated through the reduction. Interestingly, when styrene oxide undergoes the reduction
with metal/liquid ammonia, 2-phenylethanol is the sole product (Scheme 15).82

O
OH
Na/NH3

Scheme 15

3.2 Liquid ammonia as a solvent for the production and reaction of strong bases

Generally speaking, alkali metal amides in liquid ammonia react as bases rather than as
nucleophiles. There are many examples of carbanion formation using metal amides, the
protons are abstracted in sequence of their pKa value, and once a carbanion or dianion is
formed, it can undergo inter- or intra-molecular nucleophilic substitution, or elimination
reactions. For example, phenylacetonitrile in the presence of bromocyclohexane is
deprotonated by sodium amide in liquid ammonia to form the carbanion, which then attacks

31
 Introduction

bromocyclohexane to give the alkylated product,83 or the carbanion may attack a carbonyl
group of ester in the same way as a Grignard reagent (Scheme 16).84
CN

CN NaNH2 CN Br
NH3

CN

R
CN NaNH2 CN RCOOR'
NH3 O

Scheme 16

Terminal alkynes also can be deprotonated by metal amides in liquid ammonia to form the
corresponding anions, which can react with suitable electrophiles such as alkyl halides,
ketones, or even carbon dioxide (Scheme 17).85,86

HO HO
NaNH2/NH3
H R
RX

NaNH2 CO2
C4H9 H C4H9 COOH
NH3 H+

Scheme 17

Benzyne formation is a unique example of strong base-promoted elimination, which produces

a highly reactive localized “acetylene” analogue within the benzene ring. Nucleophiles can
add to this system to afford an anion intermediate which is then protonated to give substituted
aromatic compounds (Scheme 18).87

Br NH2

NaNH2 NH3
NH3

Scheme 18

32
 Introduction

The regio-selectivity of the elimination-addition process is determined by the nature of the

substituent group in the halobenzene. An electron-withdrawing group leads to para-addition of
the nucleophile to the benzyne intermediate, whereas electron-donating group give a mixture
of para and meta-substitution products for the reaction of 4-substituted bromobenzene with
NaNH2 at -33 oC (Scheme 19 and Table 7).88

X X X X

NaNH2 NH3
+
NH3
NH2
Br NH2

Scheme 19

Table 7 The product ratio of the reaction Scheme 19

X products ratio

para (%) meta (%)

CN 95-100 0-5

OMe 50-55 45-50

A range of condensation reactions in the presence of sodium amide or potassium amide in

liquid ammonia are known. Hauser et al. extensively investigated the Claisen condensation
reaction in liquid ammonia.89 For example, addition of 2-phenylethyl acetate to 0.9
equivalents of potassium amide in liquid ammonia at -33 oC gives 2-phenylethyl acetoacetate
in a 39% yield, but with 2 equivalents excess of potassium amide, the major product is
styrene, resulting from base catalyzed elimination (Scheme 20).90

O O

H Ph
KN 2
O eq. O
0.9 NH 3
Ph 2 eq
. KN
O NH H2
3

Scheme 20

33
 Introduction

Crossed Claisen condensation reaction between different esters in liquid ammonia is also
possible, lithium salts are normally used to avoid the enolate exchange (Scheme 21).91

OH O
O -33oC
O + O
NH3
O- +
Li

Scheme 21

Dieckmann condensation of diethyl adipate in liquid ammonia with one equivalent of sodium
amide is completed in minutes (Scheme 22).92
Na metal, benzene, 12 hr reflux

- Na+

NaNH2, NH3, -33oC, 5 minutes

Scheme 22

3.3 As a solvent for solvolysis

Solvolysis of substrates in liquid ammonia, involve ammonia as a reactant which leads to the
replacement of an activated atom or group, for example, halogen, or alkoxy, by an amino
group. Numerous solvolysis reactions in liquid ammonia are known, and some are widely used
in preparative organic synthesis. However, the possible solvolysis of reactants must also be
first considered as unwanted reaction for those reactions which use liquid ammonia as just a
reaction medium, such as nucleophilic substitution or condensation, etc.

3.3.1 Alkyl and aryl halides

Alkyl halides undergo solvolysis in liquid ammonia to give primary, secondary, and tertiary
amines, and in some cases, quaternary ammonium salts. For a given aliphatic group, the ease
of solvolysis for alkyl halides generally lies in the order: iodides > bromides >> chlorides.93

34
 Introduction

The solvolysis of sterically hindered alkyl halides is very slow, in stark contrast with that in
water. For example, the solvolysis of t-butyl chloride in liquid ammonia at 25 oC has a half life
of about 150 days94 versus 21 seconds in water.34a In addition, although ammonia is a basic
solvent, tertiary alkyl halides and activated halides do not undergo elimination to give alkenes
in liquid ammonia. Shatenshtein94 investigated the kinetics of solvolysis of some saturated
aliphatic halides in liquid ammonia at 25 oC (Table 8).

The data shows that the solvolysis rate of primary alkyl halides hardly changes with increasing
size of the alkyl group beyond ethyl, but decreases markedly on passing from a primary to a
branched alkyl chain halide, which suggests the solvolysis of alkyl halides in liquid ammonia
occurs by an SN2 mechanism.

Table 8 First order rate constants for the solvolysis of alkyl halides in liquid ammonia at 25 oC94

104 kobs (min-1)

halides X=Cl X=Br X=I

CH3CH2X 7.42 736 5020

CH3(CH2)2X 3.98 578 1870

CH3(CH2)3X 3.15 576 1550

CH3(CH2)4X 2.49 414 —

CH3(CH2)5X 3.74 496 —

CH3(CH2)6X 3.05 529 —

CH3(CH2)7X 1.53 275 1420

(CH3)2CH(CH2)2X 2.21 248 750

(CH3)2CHX 0.119 9.7 182

CH3CH2CHXCH3 0.077 — —

(CH3)3CX 0.033 — —

C6H5CH2Cl 544 — —

35
 Introduction

For a given halide, increasing the size of alkyl group in the alkyl halide tends to favour
increased formation of primary over secondary amine (Table 9),95 the latter product being the
result of the first formed primary amine, in preference to ammonia, reacting subsequently with
the alkyl halide. This is also indicates an SN2 mechanism, as the amine becomes bulkier, its
substitution reaction to form a secondary amine decreases. Inactivated aromatic halides,
without the presence of a catalyst, generally do not react with liquid ammonia at room
temperature or even higher,94 but some heteroaryl halides do slowly solvolyse, for example, 2-
chlorobenzthiazole gives 2-aminobenzthiazole in liquid ammonia at 20 oC.96

Table 9 The solvolysis of alkyl halides with different size of alkyl group in liquid ammonia at 25 oC after 24
hours

solvolysis product primary amine (%) secondary amine (%)

halides
n-Amyl chloride 10 80
n-Octyl chloride 45 43
n-Dodecyl chloride 90 trace

3.3.2 Epoxides

The opening of epoxides to give β-hydroxyamines occurs in aqueous ammonia and from metal
amide attack in liquid ammonia. For example, cyclohexene oxide is converted into trans-2-
amino-cyclohexanol by potassium amide in liquid ammonia (Scheme 23).97

O
KNH2
OH
NH3
NH2

Scheme 23

Due to the resistance of epoxides towards the solvolysis in just liquid ammonia, the solvolysis
rates can be very slow, and some researchers claimed that there was no sign of solvolysis of
styrene oxide in liquid ammonia at room temperature.98

36
 Introduction

3.3.3 Aldehydes and ketones

NH2 NH
NH3 NH3
O
-33 oC -33 oC
N

Scheme 24

The solvolysis of aldehydes in aqueous ammonia is well known. The solvolysis of

propionaldehyde and n-heptaldehyde in liquid ammonia at its boiling point, surprisingly gives
pyridine derivatives as products. Probably the reaction proceeds via an imine intermediate in
an aldol-type condensation (Scheme 24).99

In the presence of calcium chloride and ammonium chloride, acetone undergoes a series of
condensation reactions at 0 oC in liquid ammonia.100 Acetophenone gives a very low yield
(3%) of acetophenone imine on heating with ammonia at 180 oC for 4 hours, but the yield can
be improved to 30% by the addition of ammonium chloride as a catalyst.101 Benzophenone

O NH

NH3

130oC,120bar

Scheme 25

fails to react with liquid ammonia at room temperature over several weeks, whereas
fluorenone under the same conditions gives a quantitative yield of fluorenone imine. 102 Under
forcing conditions, benzophenone can be converted to the corresponding imine with high yield
(Scheme 25).103(vide infra)

3.3.4 Esters

Many carboxylic esters undergo solvolysis in liquid ammonia to give the corresponding
amide. The reaction is slow with the alkyl esters of simple aliphatic and aromatic carboxylic
acids, but the solvolysis may be accelerated by increasing the temperature or by adding the

37
 Introduction

ammonium salts. In liquid ammonia, four trends are clearly observed for the solvolysis of
ester:104

1. The reaction rate is accelerated by the presence of electro-withdrawing α-substituents

which increase the electrophilicity of the carbon atom of the ester carbonyl group.
2. The reaction rate is decreased by the presence of α, β-double bond which decreases the
electrophilicity of the carbonyl group by conjugation;
3. Aryl esters undergo the solvolysis reaction more rapidly than their aliphatic analogues.
4. The solvolysis rate is increased greatly by added the metal amide.

All of the above suggest that the ammonolysis of esters occurs by an addition-elimination
mechanism (Scheme 26).

O O- O
NH2 -
NH2 + -
OR
OR NH2
OR

Scheme 26

The catalytic effect of ammonium salts is probably due to either the partial protonation of the
carbonyl oxygen by ammonium cation, which increase the electrophilicity of carbonyl carbon,
or protonation of the leaving alcohol presumably analogous to that in the acid-catalysed
hydrolysis of esters. Table 10 gives the quantitative outcome of the solvolysis of esters in
liquid ammonia in the absence or presence the ammonium salt.105

Obviously, esters having electron-withdrawing groups as α-substituents solvolyse more

quickly and added NH4Cl salt has very little effect on the rate of solvolysis, whereas less
reactive esters are subject to a pronounced catalytic effect with NH4Cl salt.

38
 Introduction

Table 10 The solvolysis of esters at 0 oC in liquid ammonia with (B) and without (A) ammonium chloride

yield of amide (%)a

esters after 24 hours after 48 hours

A B A B

NCCH2CO2C2H5 97 96 99 100

H2NCOCH2CO2C2H5 67 91 93 98

C2H5OCOCH2CO2C2H5 9 79 63 95

C6H5CHOHCO2C2H5 44 63 62 79

C2H5OCH2CO2C2H5 5 53 38 77

C6H5CH2CO2C2H5 0.6 2 1.2 4.7

CH3CO2C2H5 0 1 0 3
a
Data in columns A and B correspond to yields in the absence and presence respectively of ammonium chloride.
(0.025mole ester, 0.2g ammonium chloride was added as catalyst)

3.4 As a solvent for nucleophilic substitution reactions

The research in this area is mainly concerned with aromatic nucleophilic substitution (SNAr).
Shteingarts et al. have investigated the kinetics and regioselectivity in liquid ammonia.106,107
For example, the reaction of sodium methoxide with 4-NFB was carried out in liquid ammonia
at -70oC. The extrapolated reaction rate was claimed to be nine orders of magnitude faster in
liquid ammonia than in methanol (Scheme 27).101,102

NaOMe
-70°C, NH3

Scheme 27

The selectivity of aromatic nucleophilic substitution in liquid ammonia is also claimed to be

better than in conventional solvents, for example, the reaction of sodium methoxide with 5,7-
difluoroquinoline at -53 oC in liquid ammonia yields a mixture of 5 and 7 substituted product

39
 Introduction

in a ratio of 8:1, the same reaction in DMSO at 25 oC gives a ratio of 6:1(Scheme 28).
Interestingly, the selectivity of the reaction seems to decrease with increasing reaction
temperature.108

F OMe F

NaOMe
+
LNH3, -53oC
F N F N OMe N

5-substituted 7-substituted

Scheme 28

Vicarious nucleophilic substitution reaction (VNS) and oxidative nucleophilic substitution of

hydrogen (ONSH) also can be performed in liquid ammonia to give some interesting and
useful functionalised heterocycles (Scheme 29).109,110

tBuOK

LNH3

98%

NO2 NO2 NO2

NH2 NH2
LNH3 KMnO4
H [O]
N N -40oC N N N N

S S S

Scheme 29

Some unactivated aryl halides can react with nucleophiles to form C-C, C-N, C-P and C-S
bonds under photo irradiation or using alkali metals as radical initiators in liquid ammonia, the
reactions follow an SNR1 mechanism.111 For example, 1-naphthyl halides react with
triphenylphosphine in liquid ammonia under ultrasound or sodium metal to give
naphthyldiphenylphosphine oxide in moderate yields after the reaction mixture treated with
40% H2O2 (Scheme 30).112

40
 Introduction

Na(3%) H2O2
Ph2P X = Cl, Br
LNH3

60-70%

Scheme 30

4. The project

The choice of solvent is still a significant problem in industrial organic synthesis. Ever
increasing health and environmental concerns have resulted in some previously common
solvents, for example chloroform, being proscribed, whilst others, still commonly used in
research syntheses, are generally avoided at the manufacturing scale. Dipolar aprotic solvents
(DMSO, DMF, DMAc, and NMP) are used in around 10% of cases. They are expensive and
there are toxicity concerns with some of them. They are all difficult to recycle due to their
water miscibility, and are frequently disposed of by incineration. No single alternative solvent
is likely to solve these problems. Liquid ammonia is a promising candidate to replace dipolar
aprotic solvents in a number of applications. Although liquid ammonia is among the least
expensive bulk chemicals113 and is a promising candidate to replace dipolar aprotic solvents in
a number of industrial processes, its application as a common solvent is relatively unusual.
Historically it has been used only where its use was essential, because of handling difficulties.
It can be quantitatively recovered from water by distillation under pressure. There is an
extensive literature on the physical and chemical properties of liquid ammonia,64,114 on the
reduction of organic compounds in alkali metal/ammonia solution115 and the application of
alkali metal amides in liquid ammonia as strong bases in synthesis. 116 However, there is little
about the detailed kinetics and mechanisms of aromatic and aliphatic nucleophilic substitution
in liquid ammonia, and the description of liquid ammonia solvent effects on organic reactions
are also rare. To the best of our knowledge, there are no systematic studies on the mechanisms
of organic reactions in liquid ammonia. Present research in liquid ammonia is much associated
with theoretical studies on the solvent effects. Therefore, it is imperative to establish some

41
 Introduction

fundamental facts and concepts about the effect of liquid ammonia solvent on the behaviour of
organic reactions.

The purpose of this project is to investigate the solvent effects of liquid ammonia on the
kinetics and mechanisms of some fundamental reactions and to examine the scope of this
solvent in synthesis, and so provide some of the physical organic chemistry required to
support synthetic programmes.

42
 Experimental

Experimental Page

1. Materials 44

2. Pressure equipments 60

3. Instruments 65

4. General procedures 66

43
 Experimental

1. Materials

1.1 General

Liquid ammonia was purchased from BOC Ltd., has 99.98% purity with minimal levels of
moisture (<200 ppm) and other impurities (<5ppm oil).117 Ammonia was distilled once from
the cylinder to a burette, and no further purification procedure was made before using as
reaction solvent.

All organic and inorganic were purchased from commercial providers, and were used directly
without further purification unless otherwise noted. (1S,2S)-(+)-N-(4-toluenesulfonyl)-1,2-
diphenylethylenediamine(S,S-TsDPEN, 99.5%) was from Johnson Matthey Catalysis & Chiral
Technologies. The general solvents were from commercial providers and were Reagent or
HPLC grade and were used without further purification.

1.2 Synthesis and purification of organic compounds for starting materials and products
characterisation

1
H NMR and proton decoupled 13C NMR spectra were acquired with a Bruker Avance 400
(400 MHz, 100.1 MHz, respectively). Proton and carbon chemical shifts are reported on the δ
scale relative to tetramethylsilane (δΗ = 0.00 ppm) and CDCl3 (δC = 77.00 ppm) respectively as
internal standards. Mass spectra were obtained with Agilent 6890 GC and 5973 MS detector.
Melting points were measured using a Mettler Toledo STAR SW 9.01 DSC instrument.

1.2.1 Preparation of sodium phenoxides118 and triazolate salts119

Sodium phenoxides: phenol (2.4 g, 25 mmol) was dissolved in diethyl ether or THF (20ml),
under argon. Sodium metal (0.46 g, 20 mmol) was cut into small pieces and added into the
above solution in several portions at room temperature. After all the sodium was added, the
reaction was refluxed overnight. Solvent was removed under vacuum, and the residual solid
was repeatedly washed with petroleum ether (40-60) until no phenol was detected by GC in
the washing eluent. The solid was dried under vacuum and used without further purification.
Phenoxide salts are hygroscopic and susceptible to oxidation, therefore all the salts were kept
in well-sealed bottle and stored under argon. Other phenolates were prepared by the similar
method.

44
 Experimental

Sodium triazolate: NaOH (0.6 g, 15 mmol) and 1,2,4-triazole (1.0 g, 15 mmol) were dissolved
in water (10 ml). The solution was maintained at room temperature for 4 h, and then cooled to
0 oC for 2 hours. The crystalline precipitate was filtered and repeatedly washed with ether, and
dried under vacuum at 100 oC to give 1.0 g white solid (72% yield).

Sodium benzotriazolate: similar to the method for the preparation of sodium phenoxides.

1.2.2 Preparation of nitroazidobenzenes (Scheme 31)

NaN3
n = 1,2

(NO2)n (NO2)n

Scheme 31

CAUTION: As organic azides are potentially explosive, it must be handled with care. All aryl
azides have been stored in the freezer in the dark.

4-NAB: NaN3 (0.39 g, 6 mmol) was added in portions to 4-NFB (0.71 g, 5mmol) in DMF
(15ml) with stirring, the reaction temperature was maintained at 65 oC overnight, and the
reaction mixture then poured into ice-water (50ml). The yellow precipitate was extracted with
DCM (3 × 15ml), and the organic layer was dried over anhydrous Na 2SO4. After solvent was
removed under vacuum, a yellow solid (0.71 g, 87% yield) was obtained. The solid was
recrystallised twice from water-ethanol (2:1) to give 0.58 g (71% yield) yellow solid with a
purity of 98.5% (GC). Melting point: 74-75 oC120; 1H NMR (400 MHz, CDCl3): δ = 7.36 (d,
2H), 8.25 (d, 2H); 13C NMR (100 MHz, CDCl3): δ = 119.1, 126.5, 143.9, 147.1;121 MS (EI, 70
eV): m/z (%) = 164.1(M+, 53.8), 136.1 (51.5), 90.1 (63.0), 63.1 (100), 50.0 (17.6).

2-NAB: NaN3 (0.39 g, 6 mmol) was added in portions to 2-NFB (0.71 g, 5mmol) in DMF
(15ml) with stirring, then heated to 60 oC and held there overnight. The reaction mixture was
then poured into ice-water (50ml). The yellow precipitate was extracted with DCM (3 ×
15ml), and the organic layer was dried over anhydrous Na 2SO4. After solvent was removed
under vacuum, a yellow solid (0.79 g, 96% yield) was obtained. The solid was recrystallised

45
 Experimental

twice from water-ethanol (2:1) to give 0.67g (82% yield) of yellow crystals with a purity of
99% (GC). Melting point: 63-65 oC;122 1H NMR (400 MHz, CDCl3): 7.28-7.39 (m, 2H,), 7.66
(m, 1H), 8.17 (q, 1H); 123 13C NMR (100 MHz, CDCl3): δ = 121.5, 125.2, 126.3, 134.3, 134.8,
140.9;124 MS (EI, 70 eV): m/z (%) = 164.1(M+, 43.8), 136.1 (61.5), 105.1 (100), 50.1 (52.5).

2,4-Dinitroazidobenzene: NaN3 (0.20 g, 3 mmol) was added in portions to 2,4-

dinitrofluorobenzene (0.37 g, 2mmol) in DMF (10ml) solution with stirring, held at 40 oC
overnight, and poured into ice-water (20ml). The yellow precipitation was observed, the
mixture was extracted with DCM (2 × 10ml), and the organic layer was dried over anhydrous
Na2SO4. After solvent was removed under vacuum, a deep orange solid was obtained. The
solid was recrystallised from water-ethanol (2:1) to give 0.23 g (56% yield) orange powder
with a purity of 97% (GC). Melting point: 67-68 oC;125 1H NMR (400 MHz, CDCl3): δ = 7.50
(d, 1H), 8.48 (m, 1H), 8.81(d, 1H); 126 13C NMR (100 MHz, CDCl3): δ = 122.5, 127.4, 135.8,
147.2, 148.9; MS (EI, 70 eV): m/z (%) = 209.1 (M+, 7.1), 181.1 (96.5), 51.1 (100).

1.2.3 Preparation of 1-(4-nitrophenyl)pyrrolidine, piperidine and morpholine (Scheme

32)

R1R2NH
R1R2NH =

Scheme 32

1-(4-nitrophenyl)pyrrolidine: pyrrolidine (1.8 g, 25 mmol) and 4-NFB (2.8 g, 20 mmol) were

dissolved in DMF (25ml) under argon, the reaction temperature was maintained at 70 oC
overnight. The reaction mixture then poured into 50ml ice-water, the yellow precipitation was
filtered and washed with water, then dissolved in DCM (10ml). The filtrates was extracted
with DCM (2 × 10ml) and organic layer was combined and washed with water until no DMF
was detected in the organic layer by GC. The solvent was removed under vacuum to afford
2.5g of yellow solid (64% yield). Melting point: 176-177 oC;127 1H NMR (400 MHz, CDCl3):

46
 Experimental

δ = 2.05-2.20 (m, 4H, CH2), 3.38-3.47 (m, 4H, CH2N), 6.48 (d, 2H, aromatic), 8.13 (d, 2H,
aromatic); 13C NMR (100 MHz, CDCl3): δ = 25.5, 47.9, 110.4, 126.4, 136.5, 151.8.128 MS (EI,
70 eV): m/z (%) = 192.1 (M+, 100), 191.1 (92.9), 136.1 (23.9), 120.0 (7.1), 104.0 (7.1), 90
(9.7), 77.0 (10.6).

1-(4-nitrophenyl)piperidine: pyrrolidine (2.1 g, 25 mmol) and 4-NFB (2.8 g, 20 mmol) were

dissolved in DMF (25ml) under argon and kept at 70 oC overnight. The reaction mixture then
poured into ice-water (50ml), the yellow precipitation was filtered off and washed with water,
then dissolved in DCM (10ml). The filtrates was extracted with DCM (2 × 10ml) and organic
layer was combined and washed with water until no DMF was detected in organic layer by
GC. The solvent was removed under vacuum to afford 2.8 g yellow solid (68% yield). Melting
point: 104-105 oC 129 ; 1H NMR (400 MHz, CDCl3): δ = 1.67 (m, 6H, CH2), 3.45 (m, 4H,
13
CH2N), 6.76-6.80 (d, 2H, aromatic), 8.06-8.11 (d, 2H, aromatic); C NMR (100 MHz,
CDCl3): δ = 24.3, 25.3, 48.3, 112.2, 126.2, 137.2, 154.9.128 MS (EI, 70 eV): m/z (%) = 206.1
(M+, 89.4), 205.1 (100), 176.1 (8.0), 165.1 (15.9), 159.1(23.9), 150.0 (18.6), 120.0 (13.3), 77.0
(11.5).

1-(4-nitrophenyl)morpholine: morpholine (2.1 g, 25 mmol) and 4-NFB (2.8 g, 20 mmol) were

dissolved in DMF (25ml) under argon and kept at 70 oC overnight. The reaction mixture then
poured into 50ml ice-water, the yellow precipitation was filtered off and washed with water,
then dissolved in DCM (10ml). The filtrates was extracted with DCM (2 × 10ml) and organic
layer was combined and washed with water until no DMF was detected in organic layer by
GC. The organic layer was dried over anhydrous Na 2SO4, and solvent was removed under
vacuum to afford 1.8 g yellow solid (44% yield). Melting point: 152-154 oC130; 1H NMR (400
MHz, CDCl3): δ = 3.32-3.40 (m, 4H, CH2), 3.83-3.89 (m, 4H, CH2N), 6.80-6.87 (d, 2H,
13
aromatic), 8.10-8.17 (d, 2H, aromatic); C NMR (100 MHz, CDCl3): δ = 25.2, 47.9, 110.8,
126.6, 137.1, 151.7.128 MS (EI, 70 eV): m/z (%) = 208.0 (M+, 86.7), 150.0 (100), 120.0 (40.8),
104.0 (12.4), 77.0 (21.2).

47
 Experimental

1.2.4 Preparation of 1-(4-nitrophenyl)-1H-1,2,4-triazole, 4-(4-nitrophenyl)-4H-1,2,4-

triazole and 1-(4-nitrophenyl)-1H-imidazole (Schemes 33 and 34)

NaH

Scheme 33

K2CO3

Scheme 34

1,2,4-Triazole (2.4 g, 35 mmol) was dissolved in 40 ml DMF, NaH (1.2 g, 50 mmol ) were
added and mixture was stirred for 1 hour. 4-nitrofluorobenzne (4.9 g, 35 mmol) was added
dropwise at ambient temperature for 4 hours. The reaction mixture was poured into 100ml ice-
water and the precipitate was isolated by filtration. The residue was dissolved in DCM, the
insoluble solid was filtered off and collected, recrystallised from water-ethanol (1:1) to give
1.2g of pale white powder as 4-(4-nitrophenyl)-4H-1,2,4-triazole. The filtrate, DCM layer was
washed with water until no DMF was observed in washing eluent by GC analysis. The organic
layer was dried over anhydrous Na2SO4, and solvent was removed under vacuum to afford
2.8g of pale yellow solid (60% total yield) as 1-(4-nitrophenyl)-1H-1,2,4-triazole.

1-(4-nitrophenyl)-1H-1,2,4-triazole: Melting point: 194-195 oC; 131 1H NMR (400 MHz,

CDCl3): δ = 7.93-7.97 (m, 2H, Ph), 8.18 (s, 1H, triazole), 8.41-8.49 (m, 2H, Ph), 8.74 (s, 1H,

48
 Experimental

triazole);132 13C NMR (100 MHz, CDCl3): δ = 119.9, 125.7, 141.3, 146.1, 153.5; MS (EI, 70
eV): m/z (%) = 190.1 (M+, 100), 163.1 (30.4), 136.0 (37.6), 90.1 (54.0), 80.1 (15.5), 63.1
(44.2), 50.1 (14.6).

4-(4-nitrophenyl)-4H-1,2,4-triazole: Melting point: >300 oC (decomp.) 1H NMR (400 MHz,

13
DMSO-d6/CDCl3): δ = 8.14 (m, 2H, Ph), 8.54 (m, 2H, Ph), 8.86 (s, 2H, triazole); C NMR
(100 MHz, DMSO-d6/CDCl3): δ = 122.8, 126.6, 142.0, 141.8, 147.3. MS (EI, 70 eV): m/z (%)
= 190.1(M+, 100), 160.1 (33.6), 136.1 (40.2), 90 (45.6), 63.1 (28.8), 30.0 (10.6).

1-(4-nitrophenyl)-1H-imidazole: a mixture of imidazole (0.34 g, 5 mmol) and 4-NFB (0.71 g,

5 mmol) in 15ml DMF was added K2CO3 (1 g, 7.5 mmol) and stirred at 75 oC for overnight.
The reaction mixture was poured into 50ml ice-water and the precipitate was collected and
washed with water. The solid then recrystalled from ethanol water (2:1) to give 0.81 g (86%
yield) pale white solid. Melting point: 202-204 oC;133 1H NMR (400 MHz, CDCl3): δ = 7.27
(s, 1H, imidazole), 7.38 (s, 1H, imidazole), 7.58 (d, 2H, Ph), 7.98 (s, 1H, imidazole), 8.42 (d,
2H, Ph); 13C NMR (100 MHz, CDCl3): δ = 117.6, 121.4, 126.2, 134.9, 131.9, 142.2, 146.7.134

1.2.5 Preparation of 4-substituted phenyl 4-nitrophenyl ether (Scheme 35)

X = CN, tBu

Scheme 35

4-substituted phenols (5 mmol), 4-NFB (5 mmol) and K2CO3 (1 g, 7.5 mmol) in DMF (15ml)
was stirred at 85oC for 5 hours. The reaction mixture was poured into ice-water (50ml) and
extracted with DCM, then the organic layer was washed with water and dried over anhydrous
Na2SO4. The solvent was removed under vacuum and the solid was recrystallised twice from
water-ethanol (1:1).

4-cyanophenyl 4-nitrophenyl ether: 1.1 g light yellow solid (91% yield). Melting point: 161-
162 oC;135 1H NMR (400 MHz, CDCl3): δ = 7.13-7.19 (m, 4H), 7.73 (d, 2H), 8.29 (d, 2H); 13C

49
 Experimental

NMR (100 MHz, CDCl3): δ = 108.4, 118.3, 119.0, 126.2, 134.6, 143.9, 159.0, 160.9.136 MS
(EI, 70 eV): m/z (%) = 240.0 (M+, 100), 210.0 (30.1), 166.0 (16.8), 140.0 (27.4), 102.0 (16.8),
75.0 (10.6), 50.0 (10.6).

4-tert-butyl phenyl 4-nitrophenyl ether: 0.91 g pale yellow solid (67% yield). Melting point:
58-59 oC; 1H NMR (400 MHz, CDCl3): δ = 1.35 (s, 9H), 7.00 (m, 4H), 7.46 (d, 2H), 8.19 (d,
2H);137 13C NMR (100 MHz, CDCl3): δ = 31.6, 35.6, 117.8, 120.7, 125.4, 127.9, 130.3, 142.5,
154.8, 164.1.

1.2.6 Preparation of 1,2-bis(4-nitrophenyl)diazene (Scheme 36)138

NaBH4
MeOH/NaOH

Scheme 36

1,4-dinitrobenzene (0.34 g, 2mmol) was dissolved together with NaOH (1.0 g, 2.5 mmol) in
methanol (10ml) at 60 oC, the solution was stirred and a 5 ml methanol solution of NaBH4 (0.1
g, 2.5 mmol) was added dropwise, the progress of the reaction being monitored by GC until no
1,4-dinitrobenzene was observed. Solvent was removed under vacuum and the residue syrup
was washed with water and extracted with toluene, the organic layer was dried over anhydrous
Na2SO4. The solvent was removed under vacuum, and the deep brown solid was recrystallised
from toluene/ethanol (5:1) twice to give 0.22 g (41% yield) of dark orange crystals. Melting
point: 225-227 oC;139 1H NMR (400 MHz, CDCl3): δ = 8.14 (d, 4H), 8.46 (d, 4H); 13C NMR
(100 MHz, CDCl3): δ = 124.1, 124.9, 150.2, 156.0;138 MS (EI, 70 eV): m/z (%) = 272.0 (M+,
36.1), 242.1 (8.4), 150.0 (55.9), 122.0 (100), 92.0 (44.9), 75 (42.3), 74 (40.9), 64 (10.6), 51
(4.4).

50
 Experimental

1.2.7 Preparation of benzyl azide (Scheme 37)

NaN3

Scheme 37

NaN3( 2.5g, 38.5mmol) was added to benzyl chloride (3.78 g, 30 mmol) in DMF (25ml) with
stirring at room temperature, then the temperature was raised to 70 oC and maintained
overnight. The reaction mixture was poured into ice-water (100ml) and extracted with DCM
(3 ×20ml), and then organic layer was washed with water until no DMF was detected by GC
analysis. The organic layer was dried over anhydrous Na 2SO4 and solvent removed under
vacuum. The oily residue was distilled (58-60 oC / 3mmHg) to give a colourless liquid (2.6g,
yield 65%). 1H NMR (400 MHz, CDCl3): δ = 4.38 (s, 2H, CH2N3), 7.36 (m, 5H, aromatic);
13
C NMR (100 MHz, CDCl3): δ = 55.6, 128.5, 128.5, 128.7.140

1.2.8 Preparation of N-benzylpyrrolidine, N-benzylpiperidine and N-benzylmorpholine

(Scheme 38)

R1R2NH
R1R2NH =

Scheme 38

Secondary cyclic amine (15 mmol) and benzyl chloride (15 mmol) in acetonitrile (30ml) were
added together with Na2CO3 (2.1 g, 20 mmol), reaction temperature was 85 oC for 5 hours, the
reaction mixture was poured into 50ml ice-water, and extracted with DCM (3×10ml).
Combined organic layers were dried over anhydrous Na2SO4. The solvent was removed under
vacuum. The crude product purified by flash column (hexane/ethyl acetate = 5:1).

N-benzylpyrrolidine: 2.1 g colourless oil (yield 87%), 1H NMR (400 MHz, CDCl3): δ = 1.71-
1.83 (m, 4 H, CH2), 2.42-2.52 (m, 4 H, NCH2), 3.51(s, 2H, CH2Ph), 7.31 (m, 5H, aromatic);

51
 Experimental
13
C NMR (100 MHz, CDCl3): δ = 23.4, 54.3, 60.8, 127.2, 128.2, 129.2, 139.5;141 MS (EI, 70
eV): m/z (%) = 161.2(M+, 54.0), 160.2 (77.0), 132.1 (7.1), 91.1 (100), 84.1 (56.6), 70.1 (39.8),
65.1 (20.8).

N-benzylpiperidine: 1.9 g colourless oil (yield 72%), 1H NMR (400 MHz, CDCl3): δ = 1.44-
1.48 (m, 2 H, CH2), 1.60 (m, 4H, CH2), 2.41 (m, 4H, NCH2), 3.52 (s, 2H, CH2Ph), 7.32 (m,
5H, aromatic); 13C NMR (100 MHz, CDCl3): δ = 24.4, 25.7, 54.4, 63.6, 127.1, 128.2, 128.9,
138.5.142

N-benzylmorpholine: 2.3 g light yellow oil (yield 88%), 1H NMR (400 MHz, CDCl3): δ = 2.46
(t, 4 H, NCH2), 3.51 (s, 2H, CH2), 3.72 (t, 4H, OCH2), 3.51 (s, 2H, CH2Ph), 7.34 (m, 5H,
aromatic); 13C NMR (100 MHz, CDCl3): δ = 53.6, 63.4, 66.9, 127.2, 128.3, 128.9, 138.9.143

1.2.9 Preparation of 1-Benzylimidazole (Scheme 39)

K2CO3

Scheme 39

Imidazole (1.4 g, 20mmol) and benzyl chloride (2.5 g, 20mmol) added together with K2CO3
(3.5 g, 25mmol) in THF (25ml), the reaction mixture was stirred under reflux overnight. After
filtration of solid, solvent was removed under vacuum to afford a yellow solid as crude
product. The solid was dissolved in DCM (30ml), washed with brine (3×15ml), and dried over
anhydrous Na2SO4. After removing the solvent under vacuum, yellow crystals (1.5 g, 47%
yield) with a melting point 74-75 oC were obtained. 1H NMR (400 MHz, CDCl3): δ = 5.15 (s,
2 H, PhCH2), 6.88 (s, 1H, imidazole), 7.07 (s, 1H, imidazole), 7.15 (d, 2H, benzene), 7.31-
7.33(m, 3H, benzene), 7.54 (s, 1H, imidazole); 13C NMR (100 MHz, CDCl3): δ = 50.8, 127.2,
128.3, 128.9, 129.8, 136.2, 137.4.144

52
 Experimental

1.2.10 Preparation of 1-benzyl-1,2,4-triazole and 4-benzyl-1,2,4-triazole (Scheme 40)

Scheme 40

Sodium 1,2,4-triazolate salt was prepared according to the literature,119 and used directly
without further purification. Sodium 1,2,4-triazolate salt (2.3 g, 25mmol) and benzyl chloride
(3.2 g, 25mmol) was added to acetonitrile (25ml), the reaction was stirred vigorously under
heterogeneous conditions at 85 oC and the progress of the reaction was monitored by GC
analysis. After 5 hours, the solvent was removed under vacuum to give yellow solid (2.97 g)
as crude products (the molar ratio between 1-benzylated and 4-benzylated triazole was 6.9:1
based on GC analysis). The solid was washed with hot ethyl acetate (3×15ml), and filtrates
were evaporated in vacuo to give a light yellow solid. The solid residue was purified using
flash column with DCM/hexane = 6:1 as eluent to give 2.2 g 1-benzyl-1,2,4-triazole as white
solid with a melting point 55-56 oC and 0.25 g 4-benzyl-1,2,4-triazole as white powder with a
melting point 107-108 oC (62% total yield).

1-benzyl-1,2,4-triazole: 1H NMR (400 MHz, CDCl3): δ = 5.34 (s, 2 H, PhCH2), 7.25-7.28 (m,
13
2H, Ph), 7.33-7.40 (m, 2H, Ph), 7.91 (s, 1H, triazole), 8.06 (s, 1H, triazole); C NMR (100
MHz, CDCl3): δ = 53.6, 128.0, 128.8, 128.9, 129.1, 134.6, 143.1, 152.2. MS (EI, 70 eV): m/z
(%) = 159.1(M+, 39.2), 158.1 (51.1), 132.1 (46.8), 105.1 (11.5), 91.1 (100), 65 (17.7).145

4-benzyl-1,2,4-triazole: 1H NMR (400 MHz, CDCl3): δ = 5.20 (s, 2H, PhCH2), 7.19-7.23 (m,
1H, Ph), 7.36-7.43 (m, 1H, Ph), 8.19 (s, 1H, triazole); 13C NMR (100 MHz, CDCl3): δ = 49.1,
127.6, 129.0, 129.4, 134.2, 142.9. MS (EI, 70 eV): m/z (%) = 159.1 (M+, 44.3), 158.1 (7.0),
132.1 (6.4), 91.1 (100), 65 (12.2).145

53
 Experimental

1.2.11 Preparation of (S)-α-methyl benzyl alcohol (Scheme 41)

S,S- TsDPEN, TEAF

[Ru(p-cymene)Cl2]2

Scheme 41

A 25ml vessel was pre-charged with [Ru(p-cymene)Cl2]2 (27.5 mg, 0.0375 mmol) and of
S,S-TsDPEN[(1S,2S)-(-)-N-(4-toluenesulfonyl)-1,2-diphenylethylenediamine] (23 mg,
0.075mmol), TEAF (7.5ml, a mixture of formic acid and triethylamine in molar ratio of 5:2)
was added slowly in several portions until the catalysts were fully dissolved, then the
remainder was quickly added followed by acetophenone (1.81 g, 15mmol). The solution was
maintained at 35 oC overnight. The reaction mixture was washed with 1M NaOH (10 ml)
solution, then extracted with DCM, the organic layer was washed with 1M HCl (2×10ml), and
then with water several times, and dried over anhydrous Na2SO4. The solvent was remove
under vacuum, and the crude product was purified over silica gel column [ethyl acetate/n-
hexane=1:2.5 (v/v)] to give 1.8 g (S)-α-methyl benzyl alcohol as colourless oil (yield 98.5%,
99%ee). 1H NMR (400 MHz, CDCl3):  1.44 (3H, d, CH3), 2.04 (1H, br. s), 4.85 (1H, q, CH),
7.26-7.38 (5H, m); 13C NMR (100 MHz, CDCl3): δ = 25.4, 70.5, 125.5, 127.8, 128.6, 145.7.146

GC condition for the analysis of (R, S)-α-methyl benzyl alcohol:

Inlet temperature: 150 oC; detector temperature: 150 oC;
Column: Varian Chrompack CP-Chiral-Dex CB (25.0m×250μm×0.25 μm);
Oven temperature: Isothermal, 120 oC for 35mins;
Inlet pressure: 14 psig;
Retention time of alcohol enatiomers:
(S)-α-methyl benzyl alcohol: 20.6mins;
(R)-α-methyl benzyl alcohol: 23.6mins.

54
 Experimental

1.2.12 Preparation of (S)-α-methyl benzyl chloride (Scheme 42)

SOCl2

DCM or hexane

Scheme 42

A 25 ml vessel with ice-bath was charged with thionyl chloride (1.2 g, 10 mmol) and 10ml
DCM or n-hexane, (S)-α-methyl benzyl alcohol (0.61g, 5 mmol, 99%ee) in 5 ml DCM was
added dropwise with caution so that the reaction temperature remained below 0 oC and the
reaction mixture agitated vigorously and was under argon during whole process. After adding
of alcohol, the reaction temperature was allowed to reach the ambient temperature. The
reaction completed after the reaction mixture agitated vigorously at room temperature for an
hour. The solvent was removed under vacuum to give a colourless liquid (0.57, 81% yield,
40.1%ee) as crude product. The crude product was used directly without further
purification.147 The enantiomeric excess of chloride was independent on the reaction solvent
used. The stability of the chloride was monitored by GC, the results shows that the chloride is
stable at room temperature, no decomposition or racemisation was observed after days at room
temperature. 1H NMR (400 MHz, CDCl3):  1.92 (3H, d, CH), 5.16 (1H, q, CH3), 7.28-7.51
(5H, m); 13C NMR (100 MHz, CDCl3): δ = 26.6, 58.9, 126.6, 128.3, 128.6, 142.9; MS (EI, 70
eV): m/z (%) = 140 (M+, 48.7), 125 (25.2), 105 (100), 77 (39.8), 63, (9.7), 51 (22.1).

GC conditions for the analysis of (R, S)-α-methyl benzyl chloride:

Inlet temperature: 120 oC; detector temperature: 120 oC;
Column: Varian Chrompack CP-Chiral-Dex CB (25.0m×250μm×0.25 μm);
Oven Temperature: Isothermal temperature fixed at 100 oC for 20 mins;
Inlet pressure: 14 psig;
Retention time of the chloride enantiomers:
(S)- α-methyl benzyl chloride: 12.1mins;
(R)- α-methyl benzyl chloride: 11.5mins.

55
 Experimental

GC condition for the analysis of (R, S)-α-methyl benzylamine:

Inlet temperature: 100 oC; detector temperature: 100 oC;
Column: Varian Chrompack CP-Chiral-Dex CB (25.0m×250μm×0.25 μm);
Oven Temperature: Isothermal temperature fixed at 100 oC for 35 mins;
Inlet pressure: 10 psig;
Retention time of the amine enantiomers:
(S)- α-methyl benzylamine: 29.16mins;
(R)- α-methyl benzylamine: 30.02mins.
The structures of amines were confirmed by authenticated samples and GC-MS.

1.2.13 Preparation of (S)-α-methylbenzyl acetate (Scheme 43)

Ac2O/NaOAc

chloroform

Scheme 43

To a 50ml 3-neck round bottom flask was charged chloroform (25 ml) and (S)-α-methylbenzyl
alcohol (1.25 g 10 mmol), acetic anhydride (1.53 g, 15 mmol) and ammonium acetate 1 mmol.
The reaction was held at 75 oC for about 40 hours, until all the alcohol was consumed. (GC
was used for the monitoring the progressing of the reaction). The reaction mixture then treated
with saturated NaHCO3, extracted with DCM. Organic layer was died over anhydrous
Na2SO4, and solvent was removed under vacuum to give 1.2g colourless liquid (73% yield).
H NMR (400 MHz, CDCl3):  1.56 (d, 3H), 2.10 (s, 3H), 5.90 (q, 1H), 7.28(s, 1H), 7.29-7.36
1

(m, 1H), 7.38 (d, 3H);148 13C NMR (100 MHz, CDCl3): δ = 21.4, 22.3, 72.3, 126.1, 127.9,
128.5, 141.7, 170.4;149 MS (EI, 70 eV): m/z (%) = 164.0 (M+, 27.9), 122.0 (100), 105.0 (75.2),
104.0 (96.5), 77.0 (34.1), 43.0 (35.4).

56
 Experimental

1.2.14 Preparation of 2-benzylmalononitrile (Scheme 44)

NaH

Scheme 44

Malononitrile (0.87 g, 13 mmol) and benzyl chloride (1.1 g, 8.7 mmol) were dissolved in 20ml
THF with stirring at room temperature, NaH (0.53 g, 13 mmol, 60%w/w in mineral oil) was
added in portions. The temperature was increased to 65 oC and held for 4 hours. After washing
with water, then extracted with DCM (3 × 20ml), the organic layer was dried over anhydrous
Na2SO4, and solvent was removed under vacuum. The residue was purified using a flash
column (DCM/hexane = 6:1) to give 0.91g (45% yield) as colourless prisms. Melting point:
89-90 oC; 1H NMR (400 MHz, DMSO-d6): 3.38(d, 2H), 5.15(t, 1H), 7.36-7.46(m, 5H); 13
C
NMR (100 MHz, DMSO-d6): δ = 24.8, 35.0, 114.6, 128.4, 129.3, 129.8, 135.2.150 MS (EI, 70
eV): m/z (%) = 156.0 (M+, 15.0), 91.1 (100), 77.1 (4.0), 65.1 (13.3), 77.0 (3.5), 5 1.1 (6.2).

1.2.15 Preparation of 4-substituted phenyl benzyl ether (Scheme 45)

K2CO3
DMF

X = Cl, OMe, Me, CN, NO 2, COOMe, tBu

Scheme 45

Benzyl chloride (10 mmol) was dissolved in DMF (25 ml) with stirring, and the 4-substittuted
phenoxide (10 mmol) and K2CO3 (2.0 g, 14 mmol) were added, the mixture was stirred at 80
o
C for 12 hours. The reaction mixture was poured into ice-water (50 ml), the precipitate were
filtered and washed with water. The filtrate was extracted with DCM (3×15ml) and the
organic layer was combined and washed with water until no DMF could be detected by GC.

57
 Experimental

The organic layer was dried over anhydrous Na 2SO4, solvent was removed under vacuum, and
the solid residue recrystallised from ethanol/water (1:1) twice before used as standards.

4-methylphenyl benzyl ether: 1.7 g white solid (86% yield), melting point: 41-42 oC; 151 1H
NMR (400 MHz, CDCl3): 2.29(s, 3H), 5.05 (s, 2H), 6.87(d, 2H), 7.09(d, 2H), 7.31-7.44 (m,
5H); 13C NMR (100 MHz, CDCl3): δ = 20.5, 70.1, 114.8, 127.4, 127.9, 128.7, 129.3, 129.8,
130.2, 137.3, 157.0.151

4-chlorophenyl benzyl ether: 1.8 g white solid (83% yield), melting point: 72-73 oC;152 1H
NMR (400 MHz, CDCl3): 5.04 (s, 2H), 6.89 (d, 2H), 7.23(d, 2H), 7.31-7.42 (m, 5H); 13C
NMR (100 MHz, CDCl3): δ = 70.1, 116.2, 125.7, 127.4, 128.0, 128.7, 129.3, 136.5, 157.9.151

4-methoxyphenyl benzyl ether: 2.0 g white solid (92% yield), melting point: 68-70 oC;153 1H
NMR (400 MHz, CDCl3): 3.77 (s, 3H), 5.01 (s, 2H), 6.75-6.93 (m, 4H), 7.31-7.44 (m, 5H);
13
C NMR (100 MHz, CDCl3): δ = 55.7, 70.6, 114.6, 115.8, 127.6, 127.8, 128.6, 137.6, 152.9,
153.1, 154.151

4-cyanophenyl benzyl ether: 1.7 g white crystal (72% yield), melting point: 92-92.5 oC (very
sharp);154 1H NMR (400 MHz, CDCl3): 5.14 (s, 2H), 7.03-7.06 (d, 2H), 7.36-7.48 (m, 5H),
7.59-7.62 (d, 2H); 155 13C NMR (100 MHz, CDCl3): δ = 70.3, 104.2, 115.6, 119.3, 127.5,
128.5, 128.8 ,134.1, 135.7,162.0.156

4-tert-butylphenyl benzyl ether: 1.8 g white powder (76% yield), melting point: 62-63 oC;157
1
H NMR (400 MHz, CDCl3): 1.29(s, 9H), 5.04 (s, 2H), 6.87 (d, 2H), 7.27-7.34 (m, 5H),
7.41(d, 2H);158 13C NMR (100 MHz, CDCl3): δ = 22.6, 37.6, 69.6, 114.8, 127.3, 127.7, 128.3,
129.0, 129.6, 130.1, 137.5, 156.5.

4-carbomethoxyphenyl benzyl ether: 1.7 g white crystal (71% yield), melting point: 99.5-100
o
C (sharp);159 1H NMR (400 MHz, CDCl3): 3.86(s, 3H), 5.10 (s, 2H), 7.00 (d, 2H), 7.31-7.46
(m, 5H), 7.98(d, 2H); 13C NMR (100 MHz, CDCl3): δ = 52.3, 71.1, 114.6, 122.7, 127.5, 129.1,
131.6, 136.3, 162.5, 166.9.160

58
 Experimental

4-nitrophenyl benzyl ether: 1.6 g yellowish solid (70% yield), melting point: 106-107 oC;161
1
H NMR (400 MHz, CDCl3): 5.18(s, 2H), 7.04 (d, 2H), 7.38-7.42 (m, 5H), 8.23(d, 2H); 13C
NMR

(100 MHz, CDCl3): δ = 70.6, 114.8, 125.9, 127.5, 128.5, 128.9, 135.5, 141.7, 164.1.156

1.2.16 Preparation of (iodoethynyl)benzene (Scheme 46)162

NaH, THF

-78°C

Scheme 46

Phenylacetylene (520 mg, 5 mmol) was dissolved in THF (5 ml) at -10 oC with a cooling bath,
and, with vigorous stirring, NaH (240 mg, 60%wt in mineral oil, 6 mmol) was added and the
reaction temperature was further lowered down to -78 oC. A solution of iodine (1.0 g, 8 mmol)
in THF (15 ml) was added dropwise at -78 oC, and then the temperature was allowed to rise to
ambient and the mixture left overnight for 14 hours. The reaction mixture was poured into
saturated Na2S2O3 solution (35ml), extracted with diethyl ether (3×15ml) and the organic layer
dried over anhydrous Na2SO4. Solvent was removed under vacuum, and the residue was
purified on a flash column (DCM/n-hexane = 4:1) to give 430mg (38% yield) yellow oil. 1H
13
NMR (400 MHz, CDCl3): 7.25-7.31 (m, 3H), 7.38-7.42 (m, 2H); C NMR (100 MHz,
163
CDCl3): δ = 6.6, 94.6, 123.8, 129.1, 132.7. MS (EI, 70 eV): m/z (%) = 228.0 (M+, 100),
175.9 (3.5), 126.9 (5.3), 101.1 (24.8), 75.1 (26.5), 5 1.1 (8.8).

59
 Experimental

2. Pressure equipments

Note: Use of these pressure equipments with liquid ammonia must follow the safety
protocols which are described in the Appendix D: Safety Protocols.

2.1 Glassware design

2.1.1 Ammonia tank (B) and burette (C)

standard Omnifit necks

standard Omnifit neck

110 mm
160 mm

standard Omnifit neck

18 mm

18 mm

Ammonia Tank (B) Ammonia Burette (C)

Figure 7 The design of ammonia tank (B) and burette (C)

The ammonia tank (B) and burette (C) were made of glass by a professional glassware
company (HGL Ltd., Southampton, UK). The ammonia tank was uncalibrated has a volume
capacity of about 40ml, while that for burette is about 30 ml. The burette was calibrated and
had a minimum volume unit of 0.5 ml. All the necks had a standard Omnifit® which allows

60
 Experimental

connection to standard Omnifit® valves, connectors and septa. Both B and C were pressure
tested up to 35 bar (Figure 7).

2.1.2 Reaction vessel (D)

The reaction vessel (D) also was made of glass and has a volume capacity of 15 ml. It has 2
Omnifit® necks and a standard GL14 neck which can be fitted with a GL14 lid (the GL14 lid
has a PTFE coated silicone rubber septa which is inert to the attack from ammonia).

GL14 Neck
Unit: mm

15 Thermo couple
Omnifit neck
GL14 Neck

15

85
Thermo fluid

Thermo fluid
Omnifit neck

25
25

Figure 8 The design of reaction vessel (D).

The reaction vessel also has a jacket which allows connecting with thermo regulators. The
vessel was pressure tested up to 35 bar and the maximum working temperature of the reactions
in the vessel was 45 oC. The Omnifit® necks allow the reaction vessel (D) connects to

61
 Experimental

Omnifit® valves, septum and tubing, Swagelok connections, ammonia burette (C), pressure
UV-cell, IR-cell, and pressure NMR tube (Figure 8).

2.2 Pressure glassware set-up

As shown in Figure 9, ammonia cylinder (A) is connected to ammonia tank (B) through
Omnifit® connectors and PTFE tubing (1/16 inch O.D.), which condenses ammonia gas from
A into B which is cooled by dry-ice or liquid nitrogen. Then liquid ammonia in B is warmed
to room temperature and is transferred from B to ammonia burette (C) by opening and closing
the Omnifit® valves. The burette (C) is connected to the reactor (D) through several Omnifit®
3-way and 2-way valves in order to keep the pressure balanced between the reactor and the
burette during the liquid ammonia transfer from (C) to (D), thus the required amount of liquid
ammonia then can be accurately dispensed from (C) to (D).

A: Ammonia Cylinder
B: Ammonia Tank
A C C: Burrette
D: Reactor(with jacket)

Omnifit 3-way valve

Omnifit 2-way valve

D Thermo Fluid

62
 Experimental

Figure 9 Diagram of the set-up of pressure equipment for the studies in liquid ammonia. B, C
and D are fixed into a 60  40 cm wooden board by clamps. B and C are placed inside a
wooden protection box with a Perspex sheet as front window and a protection sheet is placed
in front of D when A is charged with liquid ammonia. The maximum working temperature
allowed for the system is 45 oC (18 bar).

2.3 Pressure UV cells and NMR tubes

2.3.1 Pressure UV cells

Pressure UV cells were based on a design by Gill and were on loan from Syngenta. (Figure

Figure 10 A picture of pressure UV cell

63
 Experimental

10). 164 The body of the pressure UV cell is made of PTFE, and with an inlet and outlet
controlled by Kel-F valves, the windows of the UV cell are made from CaF2, the path length
between two windows is 10mm. The top of the UV cell has a standard Swagelok which can be
connected to the Omnifit® valves, thus the cell can be connected with D (Figure 8) and allows
the liquid ammonia solution to be transferred from D to the cell.

2.3.2 Pressure NMR tubes

Pressure NMR tube rated to 200 psi (14 bar) was purchased from Wilmad-Lab Glass® with an
O.D. of 5mm, a tube length of 7 inches and was the thin-walled (0.38mm) model which was
specially designed for 500Hz NMR instruments (Model 522-PV-7).165 The top inlet valve of
the pressure NMR tube has a standard Swagelok connection which allows the connection to
Omnifit® valve, thus ammonia solution can be transferred from the pressure vessel (D) to the
tube (Figure 11).

Figure 11 A diagram of pressure NMR tube

2.3.3 Pressure syringe

Pressure syringe was from SGE which has a total volume capacity of 1ml, the minimum
volume unit of the syringe is 0.01 ml. The top of the pressure syringe has a standard Swagelok
fitting and a gas tight PTFE on/off valve that enables to connect with various apparatus, such
as pressure reaction vessels (D), UV cells and NMR tubes. Frequently it was used to inject one

Figure 12 A picture of high pressure syringe

64
 Experimental

of ether solutions of the reactant into the pressure vessel (D) through an Omnifit® septum,
therefore it also equipped with a special side-hole needle which has a length of 71 mm and an
O.D. of 1.07mm. The pressure syringe was pressure rated to 500 psi (33 bar) (Figure 12).

2.4 Pressure tube reactors

Several pressure tube reactors were made from standard 1/4 inch O. D. Swagelok stainless
steel tubing (316L seamless, ammonia resistant), the length of these tube reactors was varied
from 10-20cm. Both ends of these reactors had standard Swagelok fittings which can be sealed
with Swagelok caps. These tube reactors had been tested to be able to hold pressure up to
1200 bar by the provider. Due to requirements form the Swagelok Fittings Protocols, the
individual reactor only can be used under pressure for 10-15 times. The working temperature
for these reactors was from 50-110 oC. Due to liquid ammonia has a very high coefficient of
expansion with temperature, great care must be taken to avoid overfilling these pressure tube
reactors. Ammonia expands by 20% on heating from 25 oC to 100 oC (volumetric coefficient
of expansion for liquid ammonia is 2.45 × 10-3 K-1, the largest among some common liquids).

3. Instruments

3.1 Thermo regulator

Thermo regulator was a Huber-Unistat Tango Nuevo which enables the temperature control of
pressure vessel (D) accurately (±0.01 oC) in a range of -40-200 oC. The instrument has a Pt-
100 thermo sensor which can be used as a temperature probe to measure the actual
temperature difference between thermo fluid and reaction vessel. The temperature control also
can be monitored by the computer. Normally the temperature that required for the
investigation kinetics in liquid ammonia is range between -10- 45 oC.

3.2 Analytical instruments

GC instrument was an Agilent 7980, the column for general kinetic study was an Agilent J&W
Scientific 19091J-433 HP5 (30 m × 0.25 mm × 0.25 µm); the column for the chiral product
analysis was a Varian Chrompack CP-Chiral-Dex CB (25 m × 250μm × 0.25 μm). The general
GC analysis conditions were:

65
 Experimental

Injection and detector temperature were 250 oC. Carrier gas was helium at constant pressure
14 psi, GC oven started at 50 oC for 1.5mins then ramped with a temperature increasing rate of
20 oC/min to 280 oC and held the temperature for 2 mins.

GC-MS was an Agilent 6980 with a 5973 MS selective detector (EI, 70 eV), the column was
an Agilent J&W Scientific DB-1701(30m × 0.25mm × 0.25µm), the general GC-MS analysis
condition:

Injection and detector temperature were 250 oC. Carrying gas was helium at constant pressure
8 psi, GC oven started at 50 oC for 1.5mins then ramped with a temperature increasing rate of
20 oC/min to 280 oC and held the temperature for 2 mins.

HPLC instrument was an Agilent 1200 series with a fraction collector for purification and
products of interest. The column used for general kinetic investigation was an Agilent Zorbax
Extend C-18 (4.6 mm × 150 mm × 5µm). The general HPLC condition:

Mobile phase: 75% methanol/25% water, flow rate was 1ml/min, the column temperature was
set at constant 30 oC, and the UV detector was normally set at 254 nm and 365 nm.

UV-Vis instrument was a Varian Cara 300 series and the pressure cells were fitted inside by
modification of the chamber.

Melting points were obtained from a Mettler Toledo DSC (STAR SW 9.01) instrument.

NMR Instruments were Bruker Avance DPX 400 and 500 MHz NMR spectrometers.

4. General procedures

4.1 Kinetics

4.1.1 Nucleophilic substitution

Generally, the kinetic measurement for nucleophilic substitution reactions in liquid ammonia
were carried out under pseudo first order conditions. The concentration of nucleophile was at
least 10 times greater than those for substrates. Ether solutions of substrates, for example,
benzyl chloride, 4-nitrofluorobenzne, etc., were prepared firstly. Normally, the concentration
of these ether solutions was 0.5 M or 1 M and 0.1-0.2 ml of these solutions was injected into

66
 Experimental

10 ml liquid ammonia, therefore all the kinetic measurements were made in liquid ammonia
containing 1-2% (v/v) diethyl ether. Normally nucleophiles were pre-charged in the reaction
vessel (D), together with internal standard, for example, biphenyl or ethyl phenyl ether, etc.
Generally, 10ml liquid ammonia was released from burette (C) to reaction vessel (D), and the
concentration of nucleophiles was varied from 0.1M to 1M. With the temperature control from
thermo regulator and vigorous stirring, liquid ammonia solution was allowed to equilibrate
with system temperature for 1 hour before the diethyl ether solution of substrate was injected
through the pressure syringe. The sample (0.5 ml) was released into a 3 ml sample vial at the
required time interval by opening and closing two way Omnifit® valves, and the reaction was
stopped with quenching agent, such as saturated NH4Cl solution, 1M HCl or 1M NaOH,
depending on the nature of the reaction. The aqueous solution was extracted with 1-2ml DCM
or toluene, the organic layer was separated, dried over anhydrous Na 2SO4, and analysed by
GC, or HPLC. The peak area of starting material and product acquired from GC or HPLC
instrument were normalised against the peak area of internal standard (Equation 4). The
reaction profiles were obtained by plotting the normalised area against time by Microsoft
Excel.

Peak area of reactants or products

Normalised area =
Peak area of internal standard

Equation 4

The data then transferred to the commercial data fitting software, such as Berkeley Madonna
or Scientist 3.0,166 and the pseudo first order rate (kobs) was obtained by following both starting
material and product through data fitting. Table 11, Figures 13 and 14a, b give a typical
example of data processing and fitting.

67
 Experimental

Table 11 The GC area and normalised area for the reaction between 0.01 M 4-NFB and 0.1 M sodium phenoxide
in liquid ammonia at 25 oC (I = 0.2M NaNO3)a

time (mins) area 4-NFB area product area IS NA 4-NFB NA product

0 2238.4 0 2373.1 0.943 0

0.33 133.5 20.8 159.7 0.836 0.130

0.67 149.6 42.5 190.9 0.784 0.223

1.0 129.7 58.1 179.0 0.725 0.325

1.5 106.7 77.0 163.7 0.652 0.470

2.0 116.7 121.3 203.5 0.573 0.596

2.5 103.4 149.6 207.1 0.499 0.722

3.0 125.8 224.6 271.5 0.463 0.827

5.0 95.9 345.8 305.9 0.314 1.130

7.0 79.4 503.3 365.8 0.217 1.376

10.0 33.6 481.2 305.0 0.110 1.578

15.0 9.9 459.2 265.1 0.037 1.732

20.0 4.0 558.6 312.6 0.013 1.787

31.0 0 591.2 329.6 0 1.794

a
IS = internal standard; NA = normalised area.

68
 Experimental

1.8

1.6

1.4
Normalised area
1.2

0.8

0.6

0.4

0.2
0
0 5 10 15 20 25 30 35
Time (mins)

Figure 13 Reaction profile for the reaction between 0.01 M 4-NFB and 0.1 M sodium
phenoxide in liquid ammonia at 25 oC (I = 0.2 M NaNO3)

Figure 14a Scientist 3.0 data fittings for the reaction between 0.01 M 4-NFB and 0.1 M
sodium phenoxide in liquid ammonia at 25 oC (I = 0.2M NaNO3), following the decreasing of
4-NFB.

69
 Experimental

Figure 14b Scientist 3.0 data fittings for the reaction between 0.01 M 4-NFB and 0.1 M
sodium phenoxide in liquid ammonia at 25 oC (I = 0.2M NaNO3), following the increasing of
4-nitrophenyl phenyl ether.

Sub-NH2
ksol
NH3
Sub-X
k2
Nuc
Sub-Nuc

Scheme 47

-d[Sub]/dt = kobs[Sub]

-d[sub]/dt = ksol [NH3][Sub] + k2[Nuc][Sub]

kobs = ksol [NH3] + k2[Nuc]

kobs _ ksol [NH3]

k2 =
[Nuc]

Equation 5

70
 Experimental

Due to the background solvolysis of substrate (Scheme 47), the second order rate constant of
nucleophilic substitution reaction (k2) was obtained from the slope of pseudo first order (kobs)
against concentration of nucleophile ([Nuc]), or calculated based on pseudo first order (k obs),
the concentration of nucleophile ([Nuc]) and background solvolysis rate (k sol) according to
Equation 5.

Because the rates of some reactions were too fast to be accurately measured by sampling
method, for example, pseudo first order reaction between thiophenoxide and benzyl chloride,
or the reaction samples which could not be satisfactorily quenched, for example, the reactions
between 4-nitrofluorobezene, benzyl chloride and secondary cyclic amines, the kinetic of
these reactions were acquired by a competition method, so the rate constants were obtained

Mole of nucleophilic product ksol [NH3]

=
Mole of solvolysis product k2[Nuc]

Equation 6

from the molar ratio of products through the product analysis (Equation 6). If the background
solvolysis rate is so slow compared with nucleophilic substitution rates, for example,
solvolysis rate of 4-NFB is negligible compared with nucleophilic substitution

Mole of nucleophilic product 1 k2,Nuc1[Nuc1]

=
Mole of nucleophilic product 2 k2,Nuc2[Nuc2]

Equation 7

reaction rate of 4-NFB with phenoxides, alkoxides, therefore, the rates of some very fast
reactions, such as thiophenoxide with 4-NFB, can be measured by referencing to the known
nucleophilic substitution reaction rates according to Equation 7. Rates measured by the
competition method were the average of at least 3 individual runs.

71
 Experimental

4.1.2 Solvolysis

General procedure for the solvolysis was similar to that described for nucleophilic substitution.
The substrates were prepared as standard diethyl ether solutions and injected into reaction
vessel (D) which was pre-charged with 10 ml of liquid ammonia and internal standard. The
samples were quenched with 1M NaOH and extracted with DCM or toluene, the organic layer
was dried over anhydrous Na2SO4 and analysed by general GC or HPLC methods. The
substrate concentration was varied from 0.01M to 0.02 M, and the solvolysis rate is
independent of substrate concentration due to constant concentration of ammonia.

The kinetics of some relatively slow reactions, for example, solvolysis of 4-NFB, were
measured by both sampling and UV methods. The concentrations of substrates in UV kinetic
study were range from 2.5 × 10-5 M to 5 × 10-4M depending on the molar extinction
coefficient of the substrate. The initial rate of solvolysis reaction was obtained according
Equation 8, where A: absorbance of reactant or product; max: the molar extinction coefficient
at maximum absorbance wavelength; C0: the initial concentration of the substrate.

ksol = A/(maxt • C0)

Equation 8

4.2 Ionisation of phenols by UV-Vis study

Standard solution of phenols (0.05M or 0.1M) were prepared in ether, then 50-100 μL of this
solution was injected into the pressure vessel (D) using a SGE syringe. If adjustment of the
ionic strength was required, the salt was also pre-charged in the vessel. With stirring, 10 ml of
liquid ammonia was released from burette (C) and the liquid ammonia solution was left to
equilibrate at room temperature for 1 hour. Then the solution was transferred into the pressure
UV cell. The UV spectra of phenols in acidified, basified water and ether were also acquired
and compared with these in liquid ammonia. The concentration range of phenol was from 10-5
M to 10-3M depending on the absorbance intensity of the phenol. The UV spectra were
recorded at ambient temperature with the dual beam method and the absorbance of phenol was
relative to the solvent blank.

72
 Experimental

4.3 Ionisation of amines, carbon acids by NMR study

A liquid ammonia solution of amine or carbon acid was prepared as described above, the
concentration of substrate being in the range 0.1M to 1M. A small amount of deuterated
standard (2.5%) such as DMSO-d6, toluene-d8, or benzene-d6 was also added to the solution to
provide a the deuterium lock, the solution was transferred into a pressure NMR tube, and the
NMR spectrum was recorded in a Bruker 500 MHz NMR (125 MHz for carbon) instrument.
For the study of ionisation of carbon acids, in order to confirm the deprotonation, a known
amount of acetonitrile, THF or benzene was added in the liquid ammonia solution as the
internal reference. The NMR spectra of amines and carbon acids were also recorded in normal
deuterium solvent, normally D2O, CDCl3 and DMSO-d6, and compared with these in liquid
ammonia. In some cases, strong base was required to deprotonate the carbon acid, so the NMR
solvent was HMPT with benzene-d6 as deuterium lock. In the case of weak quaternary carbon
signal, a small amount of shiftless relaxation reagent Cr(acac)3, was added in order to shorten
the spin-lattice relaxation times (T1) of quaternary carbon nuclei. 167 Proton and carbon
chemical shifts were on the δ scale relative to DMSO (δΗ = 2.50 ppm, δC = 39.50 ppm) or
19
benzene (δΗ = 7.36 ppm, δC = 128.30 ppm) as internal standards. F NMR chemical shift was
relative to CFCl3 (δF = 0.00 ppm) as reference. The NMR spectra in normal solvents were
acquired from a Bruker 400 MHz NMR instrument.

4.4 Copper (I) catalysed amination of aryl halides and 1,3-dipolar cycloaddition

A thick walled GC sample vial (2 ml total volume capacity, from Agilent) was pre-charged
with copper salt and the required amount of reactants and additives. The vial was pre-cooled in
a cold bath (ethanol slurry, -35 to -50 oC), then 1ml liquid ammonia was released into the vial
carefully, and the vial was quickly sealed with an aluminum cap which has a strong and inert
PTFE septa. Without stirring, the vial was allows to rise to ambient temperature and kept in a
protection jar. After the required time interval, the vial was cooled with liquid nitrogen and
decapped, upon the vapourisation of ammonia, the reaction mixture was treated with 1M
NaOH then extracted with DCM or toluene, the organic layer was dried over anhydrous
Na2SO4, and analysed by general GC or HPLC methods. The high temperature reactions, for
example, copper catalysed amination of aryl bromides and chlorides at 100 oC or higher, were
performed in the pressure tube reactors, and the procedure was similar to those introduced

73
 Experimental

above. The heating apparatus was a GC oven (Agilent 5890) which can accurately control the
temperature for those reactions that require high temperature conditions (normally range from
50-120 oC) in liquid ammonia.

74
 Results and Discussion

1. Acidity and spectral studies of compounds in liquid ammonia Page

1.1 UV-Vis studies 76

1.1.1 UV-Vis spectra of aromatic nitro compounds 77

1.1.2 Ionisation of phenols 78

1.2 NMR studies 89

1.2.1 Ionisation of aminium salts 90

1.2.2 Ionisation of carbon acids 92

75
 Results and Discussion

1.1 UV-Vis studies

Due to the difficulties of easily handling liquid ammonia without the necessary pressure
equipment at room temperature, the studies on the UV-Vis spectra of organic compounds in
liquid ammonia above its boiling point are rare. However, UV investigations of liquid
ammonia solutions of alkali metals below ammonia boiling point have been widely studied.168
Being a good electron donor but with limited ability to form hydrogen bonds, liquid ammonia
affects UV-Vis spectra of organic compounds showing very different shapes, wavelengths of
maximum absorption and intensities from other solvents. For example, the UV spectrum of 4-
nitrophenol in liquid ammonia shows a large bathochromic shift of max from its ether solution
due to ionisation (Figure 1.1).

Figure 1.1 UV absorbance of 4-nitrophenol in different solvents at room temperature, pH = 1

for acidified water, pH = 13 for basified water.

76
 Results and Discussion

1.1.1 UV-Vis spectra of aromatic nitro compounds

The dielectric constant of liquid ammonia is quite different from that for water and the
extinction coefficients of compounds are generally greater in liquid ammonia compared with
those in water and also ether under the similar conditions, although the maximum wavelength
of absorption of some nitrobenzene derivatives changes little on going from water to liquid
ammonia (Table 1.1).

Table 1.1 Molar extinction coefficients (max) and wavelength of maximum absorbance (max) of nitrobenzene
compounds in LNH3 compared with ether at room temperature

max(nm) max(M-1cm-1)

compound ether LNH3 ether LNH3

2-NFB 282 295 3000 5150
4-BFB 260 267 6950 8350
2-nitroaniline 358 410 3900 4100
4-nitroaniline 348 386 13700 14200
2-NAB 393 413 5850 5900
4-NAB 304 383 10350 3500
2,4-dinitroaniline 378 537 4600 6100
1,3-dinitrobenzene 300 555 1350 3950

The compounds in Table 1.1 are either stable in liquid ammonia or their solvolysis rates in
liquid ammonia at room temperature are relatively slow. The wavelength of absorption
maximum of 4-NFB, 4-nitroaniline and 4-NAB is significantly shorter than the corresponding
ortho analogue and may indicate a stronger resonance effect from the ortho nitro group, which
decreases the energy gaps between the ground and excited states. Conversely the extinction
coefficients are greater for the para-isomers. The absorption spectra of 1,3-dinitrobenzene and
2,4-dinitroaniline in liquid ammonia show several absorptions (257, 353, 387, 537 nm), which
are more complicated than those in ether and water (220 and 300 nm). One possible
interpretation is that 2,4-dinitroaniline is attacked by a solvent molecule to form stable a
Meisenheimer σ-complexes (Scheme 1.1), which are written as anions rather than zwitterions,
as we have shown that the aminium ions exist as free bases in liquid ammonia.219b
The formation of Meisenheimer σ-complexes between highly activated benzenes and a
nucleophile is well known, and has been shown spectroscopically, 169 including NMR. 170

77
 Results and Discussion

Generally speaking, the UV-Vis absorption of a typical Meisenheimer σ-complex is in a range

between 450-600 nm,171 but sometimes fall in the range of the visible spectrum (380-750 nm)

NH3

NH4+ - -

Scheme 1.1

and therefore often colour is observed during the formation of the adducts. 10-5 to 10-4M
Liquid ammonia solution of 1,3-dinitrobenzene appears a pink red colour, while that of 2,4-
dinitroaniline is dark green at room temperature. The absorption of 2,4-dinitroaniline at 537
nm thus can be assigned to one or both of the intermediates (Scheme 1.1). The similar
absorption at 555 nm for 1,3-dinitrobenzene also can be regarded as due to a Meisenheimer σ-
complex. There is no new product formed after vapourisation of liquid ammonia solution of
2,4-dinitroaniline and 1,3-dinitrobenzene and it appears that the Meisenheimer σ-complex is
formed rapidly but reversibly and removal of the ammonia solvent results in its reversion to
starting material. The stable Meisenheimer σ-complexes of electron deficient arenes have been
previously demonstrated in liquid ammonia,172 however, in the presence of a strong base,173
i.e., NaNH2, or a strong oxidation agent,174 i.e., KMnO4, the decomposition of intermediates
occurs to yield the amination product in liquid ammonia.

1.1.2 Ionisation of phenols

K

Equation 1.1

Liquid ammonia is a basic solvent with a very low self-ionisation constant (pK = 27.6 at 25
o
C) and the ionisation of acids in this solvent generates equivalent amounts of the conjugate
base and the ammonium ion (Equation 1.1). The low dielectric constant of liquid ammonia
indicates that most ionic species will be strongly associated in this solvent and conductivity

78
 Results and Discussion

data shows that ion-pairing occurs even at low concentrations and probably larger aggregates
form at higher concentrations. 175 There have been several methods used to determine
ionisation and dissociation constants of acids in liquid ammonia including spectroscopic,
conductivity and NMR, 176 however, to our knowledge, there has been no systematic
evaluation of substituent effects on any one class of acids. We are interested in the relationship
between ionisation constants in liquid ammonia compared with other solvents and their
variation with substituents to aid the interpretation of linear free energy relationships in liquid
ammonia (vide infra).

1.1.2.1 Ionisation without salt effect

The max and max of some phenols in liquid ammonia at room temperature are compared with
those in ether, basic and acidic water, (Table 1.2) and the results show the expected
bathochromic shift and intensified absorption upon ionisation.

Table 1.2 Molar extinction coefficients (max) and wavelengths of maximum absorbance (max) of phenols in
LNH3 at room temperature

 max (nm) max (M-1∙cm-1)

pKaa
phenol Et2O H2 O H2 O LNH3 Et2O H2 O H2 O LNH3
pH=1 pH=13 pH=1 pH=13
4-methoxyphenol 10.21 290 287 306 296 2650 2800 2780 3530
phenol 9.99 274 270 287 275 1980 1370 2600 2340
4-chlorophenol 9.20 283 285 294 287 1833 1890 2150 1885
3-chlorophenol 9.02 276 274 292 279 2215 1850 2835 2440
4-carbomethoxyphenol 8.47 251 255 295 319 25050 20360 28010 28200
3-nitrophenol 8.35 346 340 398 440 2320 2210 3930 4210
4-cyanophenol 7.95 244 243 274 298 16450 17220 22710 35200
2,4-dibromophenol 7.79 287 285 307 331 4240 3360 5670 4260
2,4-dichlorophenol 7.65 286 283 305 325 4880 3120 4850 6240
3,5-dichlorophenol 7.51 282 279 297 320 3265 2295 4425 5470
Dichlorophenol
5-methyl-2-nitrophenol 7.41 282 295 417 441 8765 4275 6004 9640
2-nitrophenol 7.23 272 278 418 444 7410 5985 4620 9140
4-nitrophenol 7.14 302 316 399 434 13165 9155 15990 39060
a
Aqueous pKa, ref.177.

The max of phenol and 3-chlorophenol are similar to those in acidified water, while those for
4-cyano, 3,5-dichloro, 2-nitro, 4-nitrophenol exhibit large bathochromic shifts along with
intensified absorption and are similar to the corresponding spectra in basified water. The

79
 Results and Discussion

difference in the wavelength of maximum absorption, Δmax, between that in acidic water and
liquid ammonia increases dramatically when the corresponding aqueous pKa of phenol drops
below 8 (Figure 1.2). This clearly indicates that phenols with aqueous pKa  7.0 are fully
ionised in liquid ammonia at room temperature, but not those with pK a > 8.5. It is also worth
noting that the molar extinction coefficients of the phenolate ions are, except for 2,4-
dibromophenol, significantly greater in liquid ammonia compared with those in water. For
examples, phenols with electron-withdrawing group at para position normally have greater

extinction coefficient, such as shown by 4-nitro, 4-cyano and 4-carbomethoxyphenol.

Interestingly, although both ortho, meta and para-nitrophenol are fully ionised in liquid
ammonia, the molar extinction coefficients are very different in the order para >> ortho >
meta.

150

100
Δλmax/nm

50

0
5 6 7 8 9 10 11
pKa (aq.)

Figure 1.2 A plot of the bathochromic shift (Δmax) of phenols in acidified water (pH = 1) and
liquid ammonia against the corresponding aqueous pKa of the phenol. (Δmax = max, LNH3 _
max, acidified water).

80
 Results and Discussion

1.1.2.2 Ionisation of phenols with added salts

1.1.2.2a. Measurement of the apparent pKa of phenols

Salt effects on the ionisation of acids in solution are well known, as salts can influence the
activity coefficients of the solutes thus changing the ionisation and dissociation process.178
Normally the acidity of acids increases with the ionic strength of the medium. In liquid
ammonia, the UV spectra of phenols change with the ionic strength with respect to the
positions of the absorption bands, the peak shapes and intensities. For example, the absorption
of 4-chlorophenol at max (326 nm) increases gradually with the increasing concentration of
added KClO4 (0 to 0.6M) in liquid ammonia, and finally reaches a maximum which does not
increase by further added salt (Table 1.3, Figures 1.3 and 1.4). This phenomenon suggests
that under sufficient ionic strength, the phenol can be forced to fully ionise in liquid ammonia,
and so the molar extinction coefficient of fully ionised species can be determined.

Table 1.3 The absorbance of 4-chlorophenol (10-4 M) at max (326 nm) with various salt concentrations (I = 0 to
0.6M, KClO4) in liquid ammonia at 25 oC

I (KClO4)/M absorbance (326nm)

0 0.0561

0.05 0.155

0.1 0.222

0.2 0.281

0.3 0.304

0.5 0.305

0.6 0.309

81
 Results and Discussion

Figure 1.3 The absorbance of 4-chlorophenol (10-4M) increases at max (326 nm) with
increasing concentrations of added salt (KClO4, I = 0 to 0.6M) in liquid ammonia at room
temperature and reaches a maximum.

82
 Results and Discussion

0.3
Absorbance at 326 nm

0.2

0.1

0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
I(KClO4)/M

Figure 1.4 The absorption of 4-chlorophenol at max (326 nm) as a function of added salt
(KClO4) in liquid ammonia at room temperature.

Furthermore, the saturation absorbance of phenol is independent of nature of salt, for example,
3-chlorophenol (10-4M) can be forced into fully ionised species with NaCl or KClO 4 in liquid
ammonia and reaches the same saturation absorbance (Appendix A: Tables A1 and A2,
Figures A1 to A4).

The sensitivities of the change in absorbance to ionic strength vary with the aqueous pKa of
the phenol. The salt effect on the UV absorbance of phenols becomes more pronounced when
the aqueous pKa of phenol is closer to 8.5. For example, the absorption of 4-nitrophenol
(aqueous pKa = 7.14) does not change with added salt, even under relative high salt
concentration (Appendix A: Figure A5a), while those for 3- and 4-chlorophenol (aqueous pKa
= 9.02 and 9.20 respectively) are very sensitive to the added salt in liquid ammonia (vide
supra). This is compatible with phenols of aqueous pKa < 8 being already fully ionised in the
absence of added salt.

83
 Results and Discussion

Using these observations and extinction coefficients of the substituted phenoxide ions, and the
absorbance at max in liquid ammonia, the apparent ionisation constant for phenol (pKa) under
different concentration of added salt can be calculated, taking the activity coefficients (γ+,-) at
low concentrations as unity (Equation 1.2).

_
pKa = log

Equation 1.2

Empirically, a reasonable linear relationship is found between these constants and the square
root of the ionic strength (I1/2) and this allows an estimate of the apparent pKa for substituted
phenols at zero ionic strength (Table 1.4, for details see: Appendix A: Tables A3 to A9;
Figures A5 to A11).

Table 1.4 Apparent pKa of some phenols in LNH3 at room temperature

phenol pKa in water a pKa in LNH3 (I = 0)

4-methoxyphenol 10.27 6.62

phenol 9.99 6.02

1-naphthol 9.37 4.97

4-chlorophenol 9.20 4.69

3-chlorophenol 9.02 4.50

4-carbomethoxy phenol 8.47 4.04

3-nitrophenol 8.36 3.61

4-nitrophenol 7.14 1.10

a
Aqueous pKa value is from reference 177.

Interestingly, there is a linear relationship between these apparent pK a values and the
corresponding aqueous ones with a slope of 1.68 (Figure 1.5). This compares with similar
plots of the acidity constants in other solvents, for example, those in dipolar aprotic solvents
acetonitrile and DMSO, against the corresponding values in water give slopes of 2.00 and
1.84,179 respectively (Figure 1.5, Appendix A: Tables A41 and A42), and these slopes are
very different from those in protic solvents such as methanol, which gives a slope of 1.15

84
 Results and Discussion

(Appendix A: Table A43 and Figure A25). 180 The greater dependence of the acidity of
phenols on substituents in liquid ammonia compared with water presumably results from the
poorer solvation of the phenoxide anions in the non-aqueous solvent so their stability is more
dependent on negative charge delocalization through the substituent. The similarity of slopes
for apparent pKa of phenols in liquid ammonia, DMSO and acetonitrile against that in water
indicate that liquid ammonia behaves like a dipolar aprotic solvent in its effects upon
ionisation of organic acids.

35

AN, slope = 2.00

pKa(dipolar aprotic solvent)

25

DMSO, slope = 1.84

15

LNH3, slope = 1.68

-5
5 6 7 8 9 10 11 12
pKa(aq.)

Figure 1.5 Plots of the pKa of phenols in liquid ammonia (LNH3), DMSO and acetonitrile
(AN) against the corresponding aqueous pKa

1.1.2.2b. Separation of ionisation and dissociation constant

The apparent pKa of phenols as described above are actually the product of the two constants
Ki, for ion pair formation, and Kd, for dissociation to the free ions (Scheme 1.2 and Equation
1.3). The degree of dissociation is dependent on the concentration of ammonium ions and the

85
 Results and Discussion

Ki _ Kd _
ArOH [ ArO ] ArO
ip

_
[ ArO ] ip [NH4 ] [ArO- ]
Ki = Kd = _
[ArOH] [ ArO ]
ip

Scheme 1.2

Apparent pKa = -log( Ki • Kd )

Equation 1.3

addition of the latter generally decreases the UV absorption of the phenoxide ion which then
levels out when only the ion-pair is present. At higher concentrations of ammonium ions, the
absorbance then increases again in line with that expected from the ionic strength effect

Ki (Kd + [NH4 ])
Abs. = . . [ArOH]t
[NH4 ] Ki (Kd + [NH4 ])

Equation 1.4

described earlier. The values of Ki and Kd can be approximated from Equation 1.4, where
[ArOH]t is the total amount of phenol present, and if it is assumed that the extinction
coefficients of the phenoxide ions are the same for the ion pair and the free ion. 181 For
example, the absorbance of 5.0×10-5M 4-carbomethoxyphenol, which is about 70% ionised in
liquid ammonia, decreases markedly with even small concentrations of NH4Cl (up to 0.05M)
and then levels out but the absorbance subsequently increases with further increasing NH4Cl
concentrations (up to 0.2M). Interestingly, the absorbance increases with higher NH4Cl
concentrations is not as marked as seen with other salts (Figure 1.6), similar observations
were found with 4-nitrophenol (Appendix A: Table A12, Figure A14). The rapid decrease in
absorbance at low concentration of ammonium ions is due to the suppression of the
dissociation step (Appendix A: Table A10, Figure A12), while later slower increase at higher

86
 Results and Discussion

NH4+ concentrations can be regarded as a general salt effect increasing ionisation but which is
counterbalanced by the specific ion effect on dissociation.

1.6

1.4
Absorbance

1.2

0.8

0.6
0 0.2 0.4 0.6 0.8 1 1.2
NH4Cl concentration/M

Figure 1.6 Change in absorbance at 315nm of 4-carbomethoxy phenol (5×10-5M) with NH4Cl
in liquid ammonia at room temperature

At the minimum of absorption, presumably there is no free phenoxide ion in solution and all
phenoxide ions are ion-paired with NH4+. Based upon this assumption and together with the
mass balance of the process and molar extinction coefficient of the ionised species, a model
can be established to separate the Ki and Kd by data fitting in Excel (for the detailed model
building and derivations, see Appendix B: Derivation 1). It is also necessary to assume that
the extinction coefficient is the same for the phenoxide ion whether it is in the ion-pair or
‘free’ and that it is independent to the nature of its counter ion. A similar titration of phenols
by NH4Cl but under constant ionic strength (I = 0.2M, NaCl) in liquid ammonia showed that
phenoxide absorbance decreased more slowly than with just ammonium chloride and varying
ionic strength, although the minimum absorbance is the same (Appendix A: Tables A10 to
A13, Figures A12 to A15).

87
 Results and Discussion

The dissociation constant Kd for the equilibrium between the ion-pair and free ions increases
with ionic strength, whereas the ionisation constant Ki is almost independent (Table 1.5) and
it appears that the effect is greater for the weaker acid. At low ionic strength (I<0.2M), Ki >>
Kd, which indicates that ionised phenols mainly exist as ion pairs in liquid ammonia at room
temperature, and is consistent with relatively low dielectric constant of liquid ammonia.
However, the dissociation constant of the ion-pair to the free ions, Kd, is very dependent on

Table 1.5 The equlibium constants Ki and Kd of some phenols in liquid ammonia at room temperature

Kd Ki

phenol Cphenol
(I=0.2M, (I=0.2M,
no ionic strength no ionic strength
NaCl) NaCl)
control control

4-cabomethoxyphenol 5×10-5M 1.2×10-4M 5.0×10-2M 0.65M-1 0.63M-1

4-nitrophenol 2.5×10-5M 1.5×10-2M 0.11M 5.6M-1 1.9M-1

the ionic strength of the medium, the difference between Ki and Kd decreases with increasing
salt concentration, so that a significant amount of phenoxide ion exists as the ‘free’ ion.

For 4-nitrophenol at I = 0.2M (NaCl), Ki = 1.9M-1 and Kd = 0.11M whereas at low ionic
strength (the concentration of ammonium ion added varying from 0 to 5×10-3M) Ki = 5.6M-1
and Kd = 1.5×10-2M. The latter data generates an apparent pKa for 4-nitrophenol from Ki and
Kd of 1.08 (Equation 1.3), in good agreement with the value of 1.10 obtained by extrapolation
to zero ionic strength (Table 1.5). The linear relationship between the apparent pKa of phenols
in liquid ammonia and those in water (Figure 1.5) probably reflects a good correlation
between Kd and the aqueous pKa.

The suppression of the concentration of free phenoxide ion by adding ammonium salts is also
reflected in the kinetics of their reactions as reported in the following sections. The half life of
the pseudo first order reaction between phenoxide and benzyl chloride is significantly
increased by adding small amount of NH4Cl in the reaction system, and no reaction at all is
observed if the NH4Cl concentration is very high.

88
 Results and Discussion

Attempts to measure the ionisation of aminium ions by examining the effect on the UV
absorbance by adding amines to phenols in liquid ammonia were unsuccessful. For example,
adding 0.1M triethylamine or piperidine to 3-chlorophenol (1mM) showed less than 5%
increase in absorption at the max of the phenoxide ion, indicating that the pKa of
triethylammonium ion in liquid ammonia is < -1 (Appendix B: Derivations 2).

1.2 NMR studies in liquid ammonia

Figure 1.7 1H NMR spectrum of ammonia blank at room temperature (500 MHz)

1
H NMR studies to investigate the ionisation of organic compounds in liquid ammonia are not
new, 182 but are rare, especially at room temperature, again probably due to equipment
availability. Another reason maybe the assumption is that the ammonia solvent peak would be
very large and broad, which would limit the scope of 1H NMR spectrum in liquid ammonia.
However, although the 1H NMR spectrum of liquid ammonia solvent at room temperature
(Figure 1.7) does show a large ammonia singlet peak at 0.65 ppm (referenced to DMSO-d6, δ
= 2.50 ppm), it is surprisingly sharp and the downfield region is very clean. However, due to
the large solvent peak, any compound with a chemical shift below 1.00 ppm may be difficult
to identify but 1H NMR spectra in liquid ammonia still can be useful for the majority of

89
 Results and Discussion

organic compounds. The 1H NMR spectra of 0.1M and 1M NH4Cl solution show that solvent
peak shifted downfield to 0.70 and 1.0 ppm (Figure 1.8), respectively, but still a sharp singlet
as expected.

LNH3 blank

DMSO peaks
as reference
0.1M NH4Cl
solution

Figure 1.8 The superimposition of the 1H NMR spectra of ammonia blank and 0.1M NH4Cl
liquid ammonia solution at 25 oC

1.2.1 Ionisation of aminium salts

The ionisation of aminium ions in liquid ammonia were investigated using 1H NMR at 25 oC.
The control spectra were the chemical shift differences of the protonated and free base forms
of the amine seen in other solvents. For example, 0.1M trifluoroethylamine hydrochloride
(aqueous pKa = 5.8) in liquid ammonia shows the same 1H NMR spectrum as the free base
indicating that it is fully deprotonated. Surprisingly, 0.1M benzylamine hydrochloride
(aqueous pKa = 9.33) and 1M piperidine hydrochloride (aqueous pKa = 11.27) also show the
same 1HNMR spectrum as their free bases indicating that they also are fully deprotonated in
liquid ammonia (Tables 1.6, 1.7 and 1.8, Appendix C: Figures N1 to N7). These observations
agree with the UV-Vis studies of mixtures of phenols and amines which showed that amines

90
 Results and Discussion

K
RNH3 NH3 RNH2 NH4

Equation 1.5

do not deprotonate phenols in liquid ammonia. The equilibrium for the dissociation of
aminium ions in liquid ammonia (Equation 1.5) must lie well over to the right, suggesting
that ammonia solvent stabilises the ammonium ion (NH4+) more than the aminium ions
(RNH3+), or put another way, in liquid ammonia, ammonia is a much stronger base than other
amines, which is different from the situation in water.

Table 1.6 1H NMR shift of trifluoroethylamine and trifluoroethylamine hydrochloride in DMSO-d6 and in liquid
ammonia at 25 oC

trifluoroethylamine trifluoroethylamine hydrochloride

DMSO-d6 δ 3.13(q, 2H) δ 3.17(q, 2H)

LNH3 δ 3.15(q, 2H) δ 3.18(q, 2H)

Table 1.7 1H NMR shift of benzylamine, benzylamine hydrochloride and triethylbenzylammonium chloride in
DMSO-d6 and in liquid ammonia at 25 oC

benzylamine benzylamine hydrochloride triethylbenzylammonium chloride

δ 3.71(s, 2H), δ 4.00(s, 2H), δ 1.30(t, 9H), 3.20(q, 6H), 4.59(s,

DMSO-d6
7.11-7.37(m, 5H) 7.32-7.53(m, 5H) 2H), 7.47-7.58(m, 5H)

δ 3.69(s, 2H), δ 3.69(s, 2H), δ 1.36(t, 9H), 3.20(q, 6H), 4.69(s,

LNH3
7.12-7.26(m, 5H) 7.14-7.29(m, 5H) 2H), 7.49-7.62(m, 5H)

Table 1.8 1H NMR shift of piperidine and piperidine hydrochloride in DMSO-d6 and in liquid ammonia at 25 oC

piperidine piperidine hydrochloride

DMSO-d6 δ 2.64(t, 4H), 1.36-1.54(m, 6H) δ 2.96(t, 4H), 1.52-1.71(m, 6H)

a a
LNH3 δ 2.61(t, 4H), 1.35-1.41(m, 6H) δ 2.59(t, 4H), 1.34-1.41(m, 6H)
b b
LNH3 δ 2.61(t, 4H), 1.35-1.43(m, 6H) δ 2.57(t, 4H), 1.31-1.38(m, 6H)
a
0.1M. b 1M.

91
 Results and Discussion

Although all amines exist effectively solely in their free base form in liquid ammonia, as will
be discussed later their nucleophilic reactivity still varies with their aqueous basicities (vide
infra).

1.2.2 Ionisation of carbon acids

Several conventional carbon acids with different aqueous pKa are studied in liquid ammonia
by 1H and 13
C NMR at 25 oC. The NMR spectra are compared with those in other solvents,
such as DMSO-d6 or CDCl3 in order to confirm the ionisation of these carbon acids in liquid
ammonia.

Dimedone (5,5-dimethylcyclohexane-1,3-dione) (1)

3 1

Table 1.9 1H NMR spectra of dimedone (1) in various solvents at 25 oC (Appendix C: Figure N8)

1
solvent H NMR shift (ppm)

CDCl3 1.09 (s, 3H); 1.11 (s, 3H); 2.31 (s, 2H); 2.54 (s, 2H); 3.35 (s, 1H); 5.51 (s, 1H)

DMSO-d6 0.99 (s, 6H); 2.21 (s, 4H); 5.20 (s, 1H)

LNH3 0.89 (s, 6H); 1.84 (s, 4H); 4.62 (s, 1H)

Table 1.10 13C NMR of dimedone (1) in various solvents at 25 oC (Appendix C: Figure N9)

13
solvent C NMR shift (ppm)

CDCl3 28.2; 30.9; 32.7; 54.1; 57.28; 103.0; 191.4; 203.8

DMSO-d6 28.4; 32.6; 102.8

LNH3 29.5; 31.5; 50.4; 99.0; 192.4

92
 Results and Discussion

LNH3 CDCl3

or DMSO-d6

1a 1 1b
deprotonation tautomerisation

Scheme 1.3

Dimedone (1) has an aqueous pKa of 5.25 and, as expected, the 1H NMR and 13
C NMR
spectra of dimedone (1) in CDCl3 and DMSO-d6 indicate that it is unionised in these two
solvents, although the ratio between enol (1b)/1,3-diketone (1) significantly changes with the
properties of the solvent (Scheme 1.3). 183 However, dimedone (1) is deprotonated to its
mono-anion form (1a) in liquid ammonia (Scheme 1.3). 1H NMR and DEPT135 spectra of 1
in liquid ammonia (Tables 1.9 and 1.10, Appendix C: Figure N10) show only one proton
attached to carbon 2 between the two carbonyl groups, and the carbon shift of the carbonyl
carbons (carbon 1 and 3) is about 10 ppm smaller than that for a normal carbonyl group which
comes about 200 ppm. These observations in liquid ammonia are not due to the enol form,
where intramolecular hydrogen bonding can cause the upfield shift of carbonyl groups as seen
for open chained 1,3-diketones, such as acetylacetone in CDCl3 which has a carbon shift of
192 ppm in its enol form.184 Intramolecular hydrogen bonding in the enol form of dimedone (1)
is not possible. The ionisation of 1 in liquid ammonia also supported by the synthetic
experiment, in which equal molar reaction of 1 and benzyl chloride in liquid ammonia affords
only mono-substituted product 1c, without the aid from strong bases.

1c

93
 Results and Discussion

1-Nitropropane (2)

Table 1.11 1H NMR of 1-nitropropane (2) in various solvents at 25 oC (Appendix C: Figure N11)

1
solvent H NMR shift (ppm)

CDCl3 1.03 (t, 3H); 2.03 (m, 2H); 4.36 (t, 2H)

DMSO-d6 0.91 (t, 3H); 1.91 (m, 2H); 4.51 (t, 2H)

LNH3 0.89 (t, 3H); 1.91 (m, 2H); 4.54 (s, 2H)

1-Nitropropane (2) has an aqueous pKa of 9 and a pKa of 17.2 in DMSO, which suggests that
there are significant solvation differences of the nitro group by protic and dipolar aprotic
solvents. 1-Nitropropane (2) is not ionised in liquid ammonia, or in DMSO, based on 1H NMR
data (Table 1.11). Previous studies on the pKa measurement of nitroalkanes by NMR in liquid
ammonia showed that the ionisation of nitromethane and nitroethane are very dependent on
the initial concentration of the nitro compound and the temperature, for example, higher
concentrations of nitromethane and lower temperature resulted in a greater degree of
deprotonation, e.g., 0.39M nitromethane at -10 oC is deprotonated by only 2% in liquid
ammonia, the authors rationalised these observations in terms of ion aggregation.185 In our
study, the concentration of 2 is 0.1M and the experiment is carried out at 25 oC, and no
ionisation is found at all. It is perhaps worth noting that phenols of aqueous pKa of 9 are also
not ionised in liquid ammonia, as described previously in this chapter.

Malonate diethyl ester (3)

Malonate diethyl ester (3) has an aqueous pKa of 13.3 and a pKa of 15.9 in DMSO, but,
perhaps surprisingly, it is not ionised in liquid ammonia (Table 1.12). A broad single peak of

94
 Results and Discussion

the protons on the central carbon of 3 with a chemical shift similar to that in DMSO-d6 is
observed in liquid ammonia, probably due to a fast exchange of these protons with solvent In
liquid ammonia, most of 3 exists in its 1,3-diketone form and there is no evidence of any of
the enol form.

Table 1.12 1H NMR of malonate diethyl ester (3) in various solvents at 25 oC (Appendix C: Figure N12)

1
solvent H NMR shift (ppm)

CDCl3 1.29 (t, 6H); 3.36 (s, 2H); 4.21 (q, 4H) (J=7.3)

DMSO-d6 1.20 (t, 6H); 3.47 (s, 2H); 4.11 (q, 4H) (J=8.4)

LNH3 1.17 (t, 6H); 3.46 (bs, 2H); 4.11 (q, 4H) (J=5.7)

Benzylmalonodinitrile (4)

Table 1.13 1H NMR of benzylmalonodinitrile (4) in various solvents at 25 oC (Appendix C: Figure N13)

1
solvent H NMR shift (ppm)

CDCl3 3.28 (d, 2H); 3.92 (m, 1H); 7.32-7.39 (m, 5H)

DMSO-d6 3.38 (d, 2H); 5.15 (t, 1H); 7.36-7.46 (m, 5H)

LNH3 3.31 (s,2H); 7.34-7.54 (m, 5H)

Table 1.14 13C NMR of benzylmalonodinitrile (4) in various solvents at 25 oC (Appendix C: Figure N14)

13
solvent C NMR shift (ppm)

CDCl3 24.9; 36.7; 112.1; 128.8; 129.1; 129.3; 132.9

DMSO-d6 24.8; 35.0; 114.6; 128.4; 129.3; 129.8; 135.2

LNH3 11.5; 35.0; 125.0; 128.1; 134.0; 137.7; 145.1

95
 Results and Discussion

LNH3 _

4 4a

Scheme 1.4

The 1H NMR spectrum of benzylmalonodinitrile (4) in liquid ammonia does not show a proton
attached to the methine carbon and 13C also supports this by showing, compared with CDCl3, a
downfield shift of the cyano groups and an upfield shift of the central carbon as expected from
an increased negative charge density on this carbon. (Tables 1.13 and 1.14) Ionisation of 4 in
liquid ammonia is further supported by DEPT 135 spectrum (Appendix C: Figure N15) which
show no coupling between the methine carbon and its attached proton in neutral (4). The large
downfield shift of the cyano groups (145.1 ppm), although perhaps surprising, is in agreement
with that reported for the cyano group of the lithium salt of benzyl cyanide anion in THF
which has a carbon shift of 144.3 ppm.186

Benzyl cyanide (5)

Benzyl cyanide (5) has an aqueous pKa of 22 and a pKa of 21.9 in DMSO and, as expected, is
not ionised in liquid ammonia, based on the comparison of 1H NMR and 13
C NMR data in
different solvents (Tables 1.15 and 1.16).

Table 1.15 1H NMR of benzyl cyanide (5) in various solvents at 25 oC (Appendix C: Figures N16-N18)

1
solvent H NMR shift (ppm)

CDCl3 3.75 (s, 2H); 7.34-7.38 (m, 5H)

DMSO-d6 4.02 (s, 2H); 7.32-7.43 (m, 5H)

LNH3a 4.06 (s,2H); 7.30-7.40 (m, 5H)

a
Proton relaxation time prolonged to 10s

96
 Results and Discussion

Table 1.16 13C NMR of benzyl cyanide (5) in various solvents at 25 oC (Appendix C: Figure N19)

13
solvent C NMR shift (ppm)

CDCl3 142.1; 129.1; 127.9; 127.8; 117.8; 23.5

DMSO-d6 131.7; 129.3; 128.8; 128.1; 119.6; 23.0

LNH3 131.5; 129.1; 128.0; 127.7; 119.4; 22.8

It is worth noting that the relaxation time of the methylene protons of 5 are extraordinarily fast
compared with the aromatic protons 187 and, consequently, it takes 10s relaxation time to
obtain the expected integration ratio between aromatic protons and methylene protons in liquid
ammonia (Appendix C: Figures N16-N18). This is not seen in other solvents. The reason for
this phenomenon in liquid ammonia is not clear, but it is likely due to the specific solute-
solvent interactions of liquid ammonia solutions.188

Acetylacetone (6) and 2-acetocyclohexanone (7)

NH3

NH3

Scheme 1.5

Acetylacetone (6) and 2-acetocyclohexanone (7) have an aqueous pKa of 8.95 and 10.1,
respectively, and a pKa of 13.3 and 14.1 in DMSO, respectively. In liquid ammonia,
acetylacetone and 2-acetocyclohexanone react rapidly with ammonia to give corresponding
enamines (Scheme 1.5), confirmed by GC-MS. Also equal molar of 6 and 7 react with benzyl
chloride give the corresponding mono-substituted products together with the solvolysis
products in the absence of strong bases, which suggest that both 6 and 7 are ionised in liquid

97
 Results and Discussion

ammonia at room temperature (Scheme 3.11, Section 3.4). Due to the instability of 6 and 7 in
liquid ammonia at room temperature, NMR investigation is not possible.

Malonodinitrile (8)

1
2

Malonodinitrile, MDN, (8) has an aqueous pKa of 11.2 and a pKa of 11.1 in DMSO. MDN (8)
appears to behave very unusually in liquid ammonia, its 1H NMR spectrum at 25 oC shows no
protons attached to the central methylene carbon (Appendix C: Figure N20), which is in stark
contrast to those observed in other solvents (Table 1.17), and suggesting that both protons
have been removed to form a carbon dianion. The lack of a proton signal is not due to H-D
exchange with the deuteriated dimethyl sulfoxide used to ‘lock’ the spectrometer as the same
result is observed with d8 toluene or d6 benzene as a lock. A similar spectrum is also seen
when one equivalent of acetonitrile is added using its three methyl hydrogens as an internal
standard, apparently indicating that there is less than 2% MDN with hydrogen still attached. It
is also not due to any rapid exchange mechanism leading to a very broad signal that is not
observable because there are also no H signals associated with the methylene carbon of MDN
in the 1H NMR spectrum at -40 oC (Appendix C: Figure N22).

Table 1.17 1H NMR of malonodinitrile (6) in various solvents at 25 oC

1
solvent H NMR shift (ppm)

CDCl3 3.62 (s,2H)

D2Oa 4.79 (s,HOD)

HMPTb 5.34, s; 3.86, s (ratio=4:1)

DMSO-d6 4.42, s; 3.37, s (ratio=4:1)

LNH3c not observed

a
Fast proton exchange between 8 and D2O generates the HOD signal at 4.79, all protons of 8 are replaced by
deuterium, which is confirmed by GC-MS analysis. b With benzene-d6 as deuterium lock, 1H NMR of HMPT: δH
(CDCl3, ppm): 2.65 (d, 18H), J = 9.2 Hz; HMPT: δH (D2O, ppm): 2.58 (d, 18H), J = 9.5Hz.

98
 Results and Discussion

Ionisation of MDN would transfer the protons to ammonia to form ammonium ions but the
addition of ammonium chloride does not affect the 1H NMR spectrum (Appendix C: Figures
N21 to N24).

Table 1.18 13C NMR and DEPT 135 of malonodinitrile (8) in various solvents at 25 oC

13
solvent C NMR shift (ppm) DEPT 135

CDCl3 8.77; 109.4 8.77(CH2)

D2O -2.94; 8.44; 111.4 not observed

HMPTa 7.08; 112.5 7.08(CH2); 112.51(CH)b

DMSO-d6 8.79; 112.1 8.79(CH2); 112.39(CH)c

LNH3d -3.12; 131.5 not observed

a
With benzene-d6 as deuterium lock, 13C NMR of HMPT: δC (CDCl3, ppm): 36.81; 36.83; HMPT: δC (D2O,
ppm): 35.93; 35.97. b CH peak is weak, the peak intensity ratio between the CH2 and CH carbon is 32:1. c CH
peak is also weak, the peak intensity ratio between the CH2 and CH carbon is 12:1. d 13C NMR is recorded with
DMSO-d6 or benzene-d6 as deuterium lock.

The 13C NMR spectrum of MDN in liquid ammonia at 25 oC shows a very high field carbon
signal at -3.12ppm (Table 1.18, Appendix C: Figure N21), consistent with the formation of a
negatively charged carbon,189 which can be compared with the methylene carbon signal of
MDN in dimethyl sulfoxide which occurs at 8.8ppm. There are only two quaternary carbons as
expected for the MDN carbon dianion, (CN)2C2– and the DEPT 135 spectrum of
malonodinitrile (8) in liquid ammonia shows no carbons attached to H. Interestingly, the 13C
NMR spectrum shows the cyano carbons shifted downfield in liquid ammonia to 131ppm
compared with 112.1ppm in the neutral MDN in dimethyl sulfoxide.

All of these observations are not due to a chemical reaction of MDN. If MDN is left in liquid
ammonia for two days at room temperature, followed by evaporation of the ammonia, acid
neutralisation and extraction yields unreacted MDN, indicating that MDN is stable in liquid
ammonia.

Malonodinitrile (8) undergoes fast H-D exchange in D2O to give the mono- and then the di-
deuterated compound, as shown from 1H NMR and 13
C NMR spectra in D2O (Table 1.17).
MDN is not significantly ionised in water as shown by 13C NMR spectra (Table 1.18), which

99
 Results and Discussion

is consistent with its low acidity in water (aqueous pKa = 11.2) however, the very small
negative carbon signal (-2.94 ppm) is probably due to the partial formation of the monoanion.
Although the evidence appears to support the formation of a stable carbon dianion formed by
the removal of two hydrogens from a single methylene in MDN, it seems difficult to believe.
The removal of one proton from MDN in water to form the carbon monoanion is associated
with pKa of 11.2, so the ability of liquid ammonia to increase the acidity of MDN to such an
extent that it forms the dianion is unexpected. For example, although the effect of the solvent
on the acidity of carbon acids is well known the pKa of dimedone is 5.3 in water but is
increased to11.2 in dimethyl sulfoxide. We can compare the data for MDN with that for a
monoalkylated MDN, benzyl malonodinitrile, (4), which contains only one ionisable hydrogen
and in DMSO (4) is unionised but in liquid ammonia it loses a proton to form the carbon
monoanion. There is no proton signal associated with the CH in the 1H NMR spectrum of (4)
13
in liquid ammonia and the C NMR spectrum shows this carbon to have moved upfield in
liquid ammonia to 11.5ppm from 24.8ppm for neutral (4) in dimethyl sulfoxide (Table 1.14)
which is consistent with the addition of a negative charge. The DEPT spectrum of (4) in liquid
ammonia shows that the methine carbon CH is no longer attached to H whereas the other
carbons show their expected environment. As with MDN, the cyano carbons of benzyl
malonodinitrile in liquid ammonia shift downfield to 145ppm compared with 114ppm in the
neutral compound in dimethyl sulfoxide.

There are also some reactions of MDN in liquid ammonia that are consistent with, but not
proof of, the formation of a carbon dianion. MDN reacts with one equivalent of benzyl
chloride C6H5CH2Cl in liquid ammonia to form the dialkylated product (C 6H5CH2)2C(CN)2
and with 1,2-dibromoethane (BrCH2CH2Br) to form the cyclic product 1,1-
dicyanocyclopropane in excellent yields.

LNH3 _

8 9

Scheme 1.6

100
 Results and Discussion

It still does not seem clear that MDN actually forms a carbon dianion in liquid ammonia or
simply forms a monoanion with unusual NMR properties. Consequently we decided to make
the mono-anion (9) independently (Scheme 1.6) and compare its NMR spectra with those in
other typical dipolar aprotic solvents, such as DMSO and HMPT.

The negative 13C NMR chemical shift of the central carbon in MDN (-3.12 ppm) certainly is
indicative that it does not exist in its neutral undissociated form in liquid ammonia. There is
also a significant cyano carbon downfield shift of MDN in liquid ammonia compared with
those in DMSO and HMPT (Tables 1.17 and 1.18). However, the 1H NMR data for the
monoanion 9 in other dipolar aprotic solvents also shows apparently the absence of any
hydrogens (Table 1.19), similar to that seen in liquid ammonia. The 13C NMR spectrum of the
MDN monoanion in HMPT shows an upfield shift to 2.6ppm for the central carbon from
7.1ppm in neutral MDN, but not as much as the -3.12ppm seen in liquid ammonia (Tables
1.18 and 1.20), which, as already stated, is significantly further upfield compared with that
(11.5ppm) of benzylmalonodinitrile anion (4a) in liquid ammonia (Table 1.14). It does appear
therefore that the carbon shift of MDN in liquid ammonia is unusually low, but this could still
be due to a poorly solvated monoanion with consequently a relatively larger negative charge
density on the central carbon in liquid ammonia compared with other solvents.

Table 1.19 1H NMR of malononitrile monoanion (9) in various solvents at 25 oC

1
solvent H NMR shift (ppm)

D2Oa 4.78 (s,HOD)

HMPTb not observed

DMSO-d6c not observed

LNH3d not observed

a
40% wt NaOD was used to generate the monoanion (9), the molar ratio between 8 and NaOD was 1:1. b NaH
was added into 8 in HMPT solution, the molar ratio between NaH and 8 is 1:1, benezene-d6 as deuterium lock,
benzene or biphenyl is also added as internal reference. It is worth noting that 1H NMR spectrum of NaH with
HMPT blank shows a broad single peak at 3.86 ppm, and chemical shift of protons of HMPT move upfield. 1H
NMR of HMPT with NaH: δH (HMPT, ppm): 2.46 (d, 18H), J = 8.4 Hz. c A broad single peak with chemical shift
of 3.40-3.50 ppm is observed for DMSO-d6 blank with NaH. d 2 eq. of NaNH2 to 8 is added into the ammonia
solution, with benzene-d6 as deuterium lock (Appendix C: Figure N25).

101
 Results and Discussion

The carbon chemical shifts of MDN with 2 equivalents of NaH2 in liquid ammonia are similar
to those without the added base which does not distinguish between mono- and di-anion
formation in both cases (Appendix C: Figures N24 and N25).

Table 1.20 13C NMR of malonodinitrile monoanion (9) in various solvents at 25 oCa

13
solvent C NMR shift (ppm) DEPT 135

D2Ob -1.37; 117.7; 124.2; 126.6; 165.1 not observed

HMPTc 2.56; 122.75 not observed

DMSO-d6 not observed not observed

LNH3d -3.27; 131.37 not observed

a
Small amount of Cr(acac)3, is added in order to shorten the spin-lattice relaxation times (T1) of quaternary
carbon nuclei. 190 b 13C NMR of 8 in a molar ratio of 1:2 with NaOD in D2O: -1.58; 117.9; 124.2; 126.5; 134.2;
170.1. c NaH is added into 8 in HMPT solution, the molar ratio between NaH and 8 is 1:1, benezene-d6 as
deuterium lock, benzene or biphenyl is also added as internal reference. 13C NMR of HMPT with NaH: δC
(HMPT, ppm): 36.13; 36.46. d 2 eq. of NaNH2 to 8 is added into the ammonia solution, with benzene-d6 as
deuterium lock (Appendix C: Figure N26).

Very highfield carbon shifts are seen for carbon dianions or quasi-dianons in other solvents,
for example, carbon suboxide (10) has a very negative carbon shift of -14.6 in CDCl3 at -40
o
C,191 while iminopropadienones 11 and 12 have shifts, respectively, of -6.8 and -3.8 ppm in
CDCl3 at room temperature.192

129.7 111.6 132.5 107.6 132.6

O=C=C=C=O N=C=C=C=O N=C=C=C=O

-6.8 -3.8
-14.6

10 11 12

If MDN does exist as its dianion (13) in liquid ammonia it must be stabilised by delocalisation
of the charges formally on the central carbon to give 13a, in which the two negative charges
are separated to reduce electrostatic repulsion (Scheme 1.7). Such delocalisation would result
structures not too dissimilar from the allene type derivatives 10, 11 and 12.

102
 Results and Discussion

_ _ _
_ N=C=C=C=N

13 13a

Scheme 1.7

In conclusion, it does not seem possible at present for NMR studies to unambiguously rule out
the formation of the malonodinitrile dianion in liquid ammonia. However, it is clear that at
least the monoanion is formed spontaneously from MDN in liquid ammonia, but the NMR
studies of the monoanion in other aprotic polar solvents do seem to indicate that what we
observe is more likely to be due to the unusual NMR properties of the monoanion rather than
due to the formation of the malonodinitrile dianion.

The structure of monoanion 9 in liquid ammonia is itself of interest, because of the potential
equilibrium between the delocalised anion and its tautomer 9a (Scheme 1.8). In principle,
structure 9a should show C-H coupling which should be determined by DEPT 135, while 9b
has an imine motif with a characteristic NH but it could potentially rapidly exchange its proton
with ammonia solvent so that no proton would be seen in 1H NMR. To test the latter, the 1H
NMR of benzophenone imine in liquid ammonia shows a distinct sharp single peak for the the
imine proton (Appendix C: Figure N27),193 indicating slow exchange. This indicates that the
malonodinitrile monoanion does not exist as its tautomer 9b, because there is no signal
observed in its 1H NMR spectrum.

H
_ _
_
NC C=C=N HN=C=C=C=N
NC CN H
9 9a 9b

Scheme 1.8

Richard and Gao 194 investigated deprotonation rate of several cyanoalkanes in D2O, and
proposed that the significant resonance stabilisation of α-cyano carbanions is attributed to the
differential solvation of cyanoalkanes and cyanocarbanions. The theoretical calculation of the

103
 Results and Discussion

free energy change for the highly unfavorable tautomerisation of acetonitrile to ketenimine in
water is computed as high as ΔGT = 30.7 kcal/mol. Based on the experimental and calculation
results, they also concluded that the large instability of the ketenimine cumulative double bond
favours the valence bond resonance form of the α-cyanocarbanion in which there is a formal
carbon-nitrogen triple bond and the negative charge is localised at the α-carbon. This may
explain why there is such a highfield signal for MDN anion in liquid ammonia – there is a
large negative charge density on the central carbon.

Another interesting observation is that there are two singlet peaks in the 1H NMR spectrum of
MDN 8 in DMSO-d6 (H,DMSO4.42 and 3.37ppm) and HMPT (HHMPT5.34 and 3.86ppm),
the ratio between these two peaks is 4:1 in both solvents (Table 1.17). The 13C NMR of MDN
in HMPT shows the central carbon with a chemical shift at 112 ppm being a methylene CH2
structure, whereas that at 7 ppm has a CH structure (Table 1.17) as shown by DEPT 135
spectrum. This can be explained by the tautomerisation of 8 and 8c in those solvents as
described in Scheme 1.9 with an equilibrium constant, Keq., for the tautomerisation in these
solvents of 0.25 at 25 oC, corresponding to a ΔGo298K = 3.4 kJ mol-1. These observations
suggest that perhaps the difference calculated by Richard for the tautomerisation of
acetonitrile to ketenimine is overestimated (vide supra).194

Keq.= k1/k-1
k-1
NC C CN HN C C C N
k1

8 8c

Scheme 1.9

In summary, due to the basic properties of liquid ammonia, the ionisation of some carbon
acids occurs spontaneously to form the corresponding anions. It appears that carbon acids
with a pKa in DMSO of <15 are ionised in liquid ammonia (Table 1.21) but there is not a
simple correlation with their aqueous pKa. An obvious anomaly is nitropropane, which is also
shown by its deviation from a reasonably linear plot of the pKa of carbon acids in water and
DMSO (Figure 3.10) which will be discussed in the following section. Liquid ammonia

104
 Results and Discussion

appears to behave like a typical dipolar aprotic solvents in its effects on ionisation and
solvation of organic acids.

Table 1.21 The ionisation of carbon acids in liquid ammonia at 25 oC by NMR compared with pKa values in
water and DMSO

carbon acid pKa (aq.)a pKa (DMSO)b ionisation in LNH3

malononitrile 11.2 11.1 Yes

dimedone 5.25 11.2 Yes

HCN 9.21 12.9 Yes

acetylacetone 9.0 13.3 Yes

2-acetylcyclohexanone 10.1 14.1 Yes

diethyl malonate 12.9c 16.4c No

1-nitropropane 9.0 17.2 No

benzyl cyanide 21.9d 21.9 No

a b
Aqueous pKa is from reference 10c except otherwise noted. pKa in DMSO is from reference 195 except
otherwise noted. c reference 196. d pKa in DMSO.

105
 Results and Discussion

2. Solvolysis in liquid ammonia Page

2.1 Solvolysis of alkyl halides 107

2.2 Solvolysis of aromatic compounds 117

2.3 Solvolysis of epoxides, esters and sulfonyl chloride 124

2.4 Solvolysis of ketones with ammonium salt as catalyst 126

106
 Results and Discussion

2. Solvolysis in liquid ammonia

Solvolysis and hence stability of organic compounds in liquid ammonia needs to be studied
first before synthetic work and kinetic measurement of nucleophilic substitution reactions are
undertaken. Due to the strong electron donating character of ammonia and its constant high
concentration (36M/25 oC), when used as a solvent, ammonia can compete with nucleophiles
to react with substrates in nucleophilic substitution reactions in liquid ammonia, The rates of
these processes are especially important for the accurate determination of the rates of those
reactions in which weak nucleophiles are involved.

From a synthetic point of view, solvolysis in liquid ammonia also can be very useful for the
synthesis of amines, amides and imines, etc. and kinetic and product analysis enables the
evaluation of rate and selectivity of these amination reactions. Strictly speaking, solvolysis in
liquid ammonia is a ‘special’ case of nucleophilic substitution in which the nucleophile is
ammonia which may have an effective constant concentration if it is in vast excess of the
substrate concentration.

For solvolysis in liquid ammonia, the solvent, in addition to its role as a nucleophile, could act
as a general base in an SN3 type process,197 and influence the mechanism of substitution by its
effects on the stability of any intermediate and on its solvation of the leaving group. These
solvent effects are also present for substitutions by other nucleophiles, especially by solvation
of the nucleophile and leaving group in determining the extent of ‘push and pull.’

2.1 Solvolysis of alkyl halides

Previous investigations of the solvolysis of organic compounds in liquid ammonia have been
described earlier (Introduction 3.3). In order to determine the solvent effect on the transition
state of the aliphatic nucleophilic substitution reactions and how the solvent may modify
substituent effects, we investigated the rates of solvolysis of substituted benzyl chlorides in
liquid ammonia.

The solvolysis of substituted benzyl chlorides in liquid ammonia give the corresponding
benzylamines (Scheme 2.1) and the rates of appearance of the product and the disappearance
of the reactant, followed by GC analysis, change exponentially with time. The derived pseudo

107
 Results and Discussion

first order rate constants show little or no dependence of on the substituent (Table 2.1). This is
very different from hydrolysis in water where the hydrolysis rates increase by about 7 orders
of magnitude on going from 4-nitrobenzyl chloride to 4-methoxybenzyl chloride (Table 2.1).
The rates of solvolysis in liquid ammonia are generally faster than those in water, but the
difference decreases with electron-donating substituents, so that 4-methoxybenzyl chloride is
more reactive in water than in liquid ammonia.

25°C

Scheme 2.1

Table 2.1 Pseudo first order rate constants (k0) for the solvolysis of substituted benzyl chlorides in LNH3 and
water at 25 oC

substrate k0, LNH3(s-1) k0, water(s-1)

4-methylbenzyl chloride 7.85×10-4 a

2.91×10-4

benzyl chloride 8.89×10-4 a

1.33×10-5

4-chlorobenzyl chloride 9.81×10-4 a

7.56×10-6

4-carbomethoxybenzyl chloride 11.0×10-4 N.A.

4-cyanobenzyl chloride 13.3×10-4 N.A.

4-nitrobenzyl chloride 15.3×10-4 a

3.38×10-7

3-methoxybenzyl chloride 7.82×10-4 N. A.

4-methoxybenzyl chloride 19.1×10-4 3.70 b

α-methyl benzyl chloride 6.71×10-6 0.48 c

α-methyl 4-methoxybenzyl chloride 9.68×10-4 1.55 d

a
The solvolysis rates were extrapolated data from ref. 198. b ref. 199. c ref. 34c. d ref. 200, the data is for 80%
(v/v) ethanol in water.

108
 Results and Discussion

There is no clear consensus on the mechanism of hydrolysis of benzyl halides in water and
aqueous binary solvents, although that for 4-methoxybenzyl halides is described as an SN1
mechanism with the intermediate formation of a carbocation,201 and that for the 4-nitro
derivative is claimed to be SN2.202 The rate of solvolysis in liquid ammonia is retarded
dramatically when one of methylene hydrogen of benzyl chloride in the α-position is replaced
by a methyl group, α-methyl benzyl chloride is solvolysed about 130 times slower than benzyl
chloride. This is strong evidence for solvolysis occurring by an S N2 mechanism due to a more
sterically hindered transition state. It is also worth noting that the solvolysis rate of t-butyl
chloride in liquid ammonia at 25 oC is very slow, the half life of the reaction was reported as
long as 146 days,203 which is in stark contrast with that for the hydrolysis and is indicative of a
bimolecular process in liquid ammonia. Furthermore, α-methyl 4-methoxybenzyl chloride,
which undergoes solvolysis by the unimolecular mechanism in protic solvents, 204 is solvolysed
nearly at the same rate as benzyl chloride in liquid ammonia, but about 1600 and 400 times
slower than in 80% (v/v) ethanol/water and 100% methanol at 25 oC, respectively.200 This
indicates that, even with activated benzyl chloride derivatives, the solvolysis of these
compounds in liquid ammonia proceeds through a concerted bimolecular mechanism. The
activation parameters for the solvolysis of substituted benzyl halides in liquid ammonia are
significantly different from those in water (Table 2.2, Appendix A: Tables A14 to A18,
Figures A16 to A20).

Table 2.2 Activation parameters for the solvolysis of substituted benzyl chlorides in LNH3 and water at 25 oC

LNH3 Water a

substrate ΔH‡ ( kJ mol-1) ΔS‡ ( J K-1 mol-1) ΔH‡ (kJ mol-1) ΔS‡ (J K-1 mol-1)

benzyl chloride 39.9 -200 83.1 -38.0

4-chlorobenzyl chloride 40.2 -197 85.9 -37.2

4-nitrobenzyl chloride 37.8 -202 87.6 -50.1

4-methoxybenzyl chloride 41.3 -188 70.6 -5.10

α-methyl benzyl chloride 67.7 -147 71.0 -50.2

a
Activation parameters for 4-substituted benzyl chlorides were extrapolated from refs.198, 199 and ref.205; the
data for 4-methoxybenzyl chloride were for 20% (v/v) methanol in water; the data for α-methyl benzyl chloride
were from ref.34c.

109
 Results and Discussion

Very large negative entropies of activation (ΔS‡) are observed for the solvolysis of all
substituted benzyl chlorides in liquid ammonia, indicative of a restricted activated complex
relative to the reactant and a bimolecular concerted SN2 mechanism for all derivatives.
Bimolecular nucleophilic substitution processes are usually characterised by large negative
entropies of activation of -90 and -120 J K-1 mol-1 due to the loss of translational and rotational
entropy of the reactants and the development of charge in the transition state leading to solvent
restriction.206 By contrast the ΔS‡ for typical hydrolytic reactions which follow the SN1
mechanism are often positive, e.g. the ΔS‡ of the hydrolysis of t-butyl chloride34f is about 50 J
K-1 mol-1. The additional methyl group on the α-position of benzyl chloride increases ΔH‡ of
the reaction significantly in liquid ammonia (Table 2.2), as expected from a more sterically
hindered reaction in a bimolecular process. This is distinctly different from the hydrolysis in
water, where the additional methyl group causes a decrease in ΔH‡, presumably due to the
formation of a more stable carbocation.

The C-Cl bond of benzyl halide has a dipole moment of 1.85D in gas phase.207 Compared with
liquid ammonia, the C-Cl bond of benzyl chloride is presumably more polarised in water due
to the latter’s greater hydrogen bond donating ability and so the fission of the C-Cl bond may
be expected to be easier in water than in liquid ammonia. Although liquid ammonia is
generally described as a protic solvent,1 like water, with good hydrogen bond donor (HBD)
and acceptor (HBA) ability, there is very little evidence to support this assertion. It actually
has a very limited HBD ability, not only in the gas phase but also in the condensed phase, 56
and anions are poorly solvated in liquid ammonia. Expulsion of the chloride anion is thus
expected to be more difficult in liquid ammonia than in water and so the probability of a
unimolecular mechanism is less with the necessity of more ‘push’ required from the incoming
nucleophile. In addition, ammonia is a better nucleophile than water, all of which indicate a
more likely bimolecular process in liquid ammonia. Finally, based on a comparison of the
effect of solvents on the rates of reactions, and, contrary to commonly accepted views, it
appears that liquid ammonia acts more like a dipolar aprotic solvent in nucleophilic
substitution reactions.208

There is surprisingly little or no dependence of the rate of solvolysis of benzyl chlorides in

liquid ammonia upon ring substituents, again in contrast to that in water (Table 2.1).

110
 Results and Discussion

Consequently the Hammett plot for the solvolysis in liquid ammonia is very different from
that in water (Figure 2.1), the ρ value for the reaction in liquid ammonia is zero, while that in
water, although not ideally linear, shows ρ = -4.32. It is worth noting that the data for
solvolysis in liquid ammonia is effectively zero whether σp+ or σp is used. These different
values suggest that for the solvolysis of benzyl chlorides in liquid ammonia there is little or no
charge developed on the central carbon atom in the transition state with any charge developed
due to partial fission of the bond to the leaving group being counterbalanced by an equal
transfer of charge from the incoming nucleophile.

A typical SN1 solvolysis reaction generally has a high sensitivity to the solvent polarity and the
rate constants tend to be accelerated by increasing the ionising power of the solvent. However,
whether salts increase or decrease the rate of a solvolysis reaction depends on the substrate
structure. For example, addition of halide ions decreases the rate of hydrolysis for
diphenylmethyl in water but not of t-butyl halides.209

-0.5
log(kobs/s-1)

-4.5

-8.5
-1.3 -0.3 0.7
σp+

Figure 2.1 The Hammett plot for the solvolysis of 4-substituted benzyl chlorides in LNH3 (◊)
and water (×) at 25 oC.

111
 Results and Discussion

Normally, the more stable the intermediate formed in an SN1 process, the greater is the
selectivity towards nucleophilicity of the nucleophile in the product determining step, but the
similar is the sensitivity towards solvent polarity on the rate limiting step.210 On the other
hand, the salt effect normally has a positive effect on the rate of a SN2 process when a neutral
nucleophile attacks a neutral substrate,211 due to the generation of charge in the transition state.
Despite the difficulty of interpretation, salt effects on the rates of reaction have been used to
distinguish between SN1 and SN2 processes.212

The pseudo first order rate constants for the rate of the solvolysis of 4-methoxybenzyl chloride
in liquid ammonia increases linearly with increasing concentration of KClO 4 (Table 2.3,
Figure 2.1), but the positive effect is not as marked as that for a typical S N1 reaction, for
which the rate often increases exponentially. 213 The solvolysis rate of 4-methoxybenzyl
chloride is also increased both by adding NH4Cl and KClO4 in liquid ammonia (Table 2.4,
Figure 2.1), but not significantly more than that with potassium perchlorate. Even though
chloride ion is not a particularly good hydrogen bond acceptor, it may have been anticipated
that ammonium ions may facilitate chloride ion expulsion but its effect is simply that of a
simple salt effect. In addition, the solvolysis rate is also accelerated the same amount by
adding NaCl or ammonium chloride which excludes the possibility of specific aid from the
ammonium cation (Table 2.4). The smaller salt effect on the rate of solvolysis of 4-
methoxybenzyl chloride in liquid ammonia compared with observed for typical SN1 reactions
in aqueous or aqueous binary organic solvents is also indicative of a bimolecular SN2 process.

Table 2.3 Solvolysis rates of 4-methoxybenzyl chloride in presence of KClO4 in LNH3 at 25 oC

concentration of KClO4(M) 0 0.5 1.0 1.5

solvolysis rate, k0 (103s-1) 1.91 2.69 3.94 4.70

Table 2.4 Solvolysis rates of 4-methoxybenzyl chloride in presence of NH4Cl and NaCla in LNH3 at 25 oC

concentration of NH4Cl (M) 0 0.2 0.5 0.8 1.0

solvolysis rate,k0 (103s-1) 1.98 2.54 3.37 3.91 4.66

a
Solvolysis rate of 4-methoxybenzyl chloride in liquid ammonia at 25 C in presence of 0.2M NaCl is 2.54×10-3
o

s , identical to that for NH4Cl, however, the poor solubility of NaCl in liquid ammonia at 25 oC (3.02g/100g)44
-1

prevents the kinetic measurement at higher NaCl concentration.

112
 Results and Discussion

5.5

4.5
103k0/s-1

3.5

2.5

1.5
0 0.4 0.8 1.2 1.6 2
Concetration of NH4Cl or KClO4 / M

Figure 2.2 Changes in the solvolysis rate constants of 4-methoxybenzyl chloride with the
addition of various salts in LNH3 at 25 oC: (Δ) KClO4; (□) NH4Cl.

Due to the poorer solvation of the leaving group the chloride anion, in the solvolysis of benzyl
chlorides in liquid ammonia, compared with that in water, the unimolecular S N1 mechanism is
unfavourable. The expulsion of the leaving group chloride ion is facilitated by increasing its
solubility in liquid ammonia by increasing solvent ionising power, similar to the effect of
increasing ionic strength on increasing the ionisation constants for phenols, described in the
previous section.

Unlike the ammonium ion (NH4+), quaternary ammonium cations (R4N+) are permanently
charged, independent of the pH of the solution. The quaternary ammonium group can exert
strong electron withdrawing and field effect as a substituent, as shown by its large positive
Hammett sigma constant derived from its effect on the ionisation of benzoic acids.214 It is also
a good leaving group in elimination215 and nucleophilic substitution216 reactions. However,
despite all of this, there is no solvolysis of benzyltriethyl ammmonium chloride observed in
liquid ammonia at 25 oC for days (Scheme 2.2), and there is also no elimination product

113
 Results and Discussion

detected. This is probably due to the bulky triethyl ammonium group prevents the attack from
ammonia molecule on either the benzylic or the ethyl methylene.

Cl

LNH3, 25°C

Scheme 2.2

The stereochemical consequences of aliphatic nucleophilic substitutions are classic criteria for
mechanisms. Based on a racemised product, Ingold claimed that the solvolysis of α-methyl
benzyl chloride in liquid ammonia at room temperature occurs completely through a
unimolecular path.217 However, in our hands the synthesis of enantiomerically pure α-methyl
benzyl chloride proved to be difficult; α-methyl benzyl alcohol racemises during the
chlorination process and also during the silica gel purification step. 217,218 The best result we
achieved was only about 40% ee chloride and enantiomeric synthesis via tosylation,
mesylation, phosphorylation of α-methyl benzyl alcohol all proved to be unsuccessful. α-
Methylbenzyl acetate proved to be very inert towards nucleophilic attack by ammonia and no
reaction was observed after 2 hours at 45 oC in liquid ammonia, and the lack of formation of
acetamide is also consistent with our observation219 that alkyl esters solvolyse slowly in liquid
ammonia (vide infra).

Complete inversion of configuration of 40% ee S-α-methyl benzyl chloride was observed

during the solvolysis to give 40% ee R-enriched α-methyl benzylamine (Scheme 2.3) with
almost 100% yield after the reaction reached completion and no elimination product was
found. This is another strong indication that the solvolysis of primary and secondary aliphatic
halides in liquid ammonia follows an SN2 mechanism.

LNH3
25°C

Scheme 2.3

114
 Results and Discussion

It is of interest to determine the degree of inversion not only in liquid ammonia but also in
aqueous liquid ammonia solutions to see if any adventitious water would compromise the use
of liquid ammonia as a solvent for large scale enantiomeric synthesis. However, there is little
indication that significant racemisation occurs with increasing water content. The solvolysis of
41% ee S-α-methyl benzyl chloride gives a barely detectable reduction in the ee of R-α-methyl
benzylamine in 10%, 20% and 30% water-ammonia solution (Table 2.5). The increasing
water content could facilitate the expulsion of chloride due to better solvation from water and
the rate of solvolysis does increase exponentially with the increasing of water content the half
life decreases from 29 hours in pure liquid ammonia to 3.7 hours in 10% water, however, even

Table 2.5 Solvolysis of enantiomerically rich S-α-methyl benzyl chloride in aqueous ammonia at 25 oC

H2O volume fraction 0% H2O 10% H2O 20% H2O 30% H2O

ee% of reactant chloride(S) 37.8 40.8 40.8 40.8

ee% of product amine(R) 37.8 36.6 36.5 35.9

Table 2.6 The rate and product analysis for the solvolysis of α-methyl benzyl chloride in aqueous ammonia at 25
o
C.

H2O volume fraction 0% H2O 10% H2O 20% H2O 30% H2O

rate/s-1 6.71×10-6 1.93×10-5 5.2×10-5 1.40×10-4

______
molar ratio (amine/alcohol) 110 :1 42:1 12:1

krel. 1 2.9 7.7 21

115
 Results and Discussion

1.5

104kobs/s -1 1

0.5

0
0 0.1 0.2 0.3 0.4
Volume fraction of water(v/v)

Figure 2.3 The increase in the solvolysis rate of α-methyl benzyl chloride with increasing
volume fraction of water in water-ammonia solution at 25 oC.

in 10% water-ammonia there is little or no competing hydrolysis, with less than 1% of alcohol
produced (Table 2.6). The rate of solvolysis of α-methyl benzyl chloride dramatically
increases with increasing content of water in liquid ammonia (Figure 2.3), as is generally
observed in aqueous binary organic solvents,34c,220 presumably the driving force coming from
the differences in solvation of the leaving chloride anion, as shown by the large positive Gibbs
transfer energy of chloride anion from water to ammonia or DMSO.221

H H
δ δ-
N C Cl N Cl
N:
H H
H

X X X

Scheme 2.4

116
 Results and Discussion

Figure 2.4 Jencks-More O’Ferrall reaction coordinate-energy diagram for the solvolysis of
benzyl chloride derivatives in liquid ammonia at 25 oC

In summary, the solvolysis of alkyl halides in liquid ammonia is a concerted bimolecular

process which follows a distinct SN2 mechanism, as indicated by the substituent effects,
activation parameters, salt effects and stereochemistry for the solvolysis of benzyl halides. The
solvolysis of alkyl halides proceeds through a symmetrical transition state structure that has
little charge development on the incoming nucleophile and the departing nucleofuge and little
or no change in charge on the central benzylic carbon compared with that in the initial state
(Schemes 2.4 and Figure 2.4).

2.2 Solvolysis of aromatic compounds

Aryl halides and aromatic heterocyclic halides undergo solvolysis in liquid ammonia to give
the corresponding aromatic amines. The rates of these reactions were determined by
monitoring the exponential appearance of product and disappearance of reactant using GC or
HPLC normalised area as a function of time to give the first order rate constants for solvolysis
which are dependent on the nature of the leaving group and aromatic substituents and show
the expected trends (Table 2.7).

117
 Results and Discussion

As expected, the rates of solvolysis of aryl halides in liquid ammonia are slower than those for
aliphatic benzyl halides222 but the reactivity of halobenzenes can be dramatically activated
through the introduction of electron-withdrawing groups (Table 2.7). In liquid ammonia, there

Table 2.7 Solvolysis rate constants of some aromatic compounds in liquid ammonia

Substrate Temp (oC) 106k0(s-1) t1/2

2-NFB 25 2.15×102 54mins

3-nitrofluorobenzene a 100 no reaction ______

4-NFB 20 7.86 24.4hrs

2,4-DFNB 25 6.72×103 1.7mins

2,4-dinitrofluorobenzene 25 >1.4×105 <5s

______
4-nitrochlorobenzene 25 no reaction

2,4-dinitrochlorobenzene 25 6.18×103 1.9mins

3-nitroiodobenzene b 25 no reaction ______

2-NAB 20 5.81 33.1 hrs

4-NAB 25 5.11 37.7hrs

2,4-dinitroazidobenzene 25 >1.4×105 <5s

2-chloropyrimidine 20 14.2 13.3hrs

2-chlorobenzthiazole 20 5.33 36.1hrs

2-fluoropyridine a 25 no reaction ______

a
Stable for days at 100 oC, no reaction observed. b Stable for days at 25 oC.

is no reaction of unsubstituted halobenzenes and 3-nitrohalobenzenes at ambient temperature,

but as expected, the 2- and 4-nitro activated derivatives and the substituted aryl fluorides are
much more reactive.223 The introduction of additional fluoro or nitro groups increases the
solvolysis rates by more than 4 orders of magnitude. The solvolysis rate of 4-NAB, in the
absence of salts, are similar to those for 4-NFB, but that for 2-NFB is nearly two orders of
magnitude greater than that for 2-NAB.

Given the demonstration that nitro substituted aromatic compounds without a leaving group
reversibly form Meisenheimer complexes in liquid ammonia (Scheme 1.1), it seems
reasonable to postulate the complexes as intermediates in solvolysis and nucleophilic

118
 Results and Discussion

substitution reactions of analogous compounds that do contain a leaving group (Scheme 2.5).
The first formed intermediate (A) will have a charged ammonium ion but product formation
probably requires deprotonation to form the anionic intermediate (B) before the leaving group
can be expelled, especially as liquid ammonia is a poor solvent for anions. We have shown
that aminium ions in liquid ammonia are invariably deprotonated by the vast excess of basic
solvent and so exist in their free base form (chapter 1). It is therefore likely that the
zwitterionic intermediate (A) is rapidly converted to the thermodynamically more stable
anionic intermediate (B) by proton transfer to the solvent (k2 step in Scheme 2.5 where B =
NH3). In fact the intermediate (B) may be formed directly from the reactants by general base
catalysis by solvent ammonia. The rate-limiting step for solvolysis is therefore likely to be the
breakdown of the intermediate (B), step k3 (Scheme 2.5) which is also compatible with the
observations for other aromatic nucleophilic substitutions in liquid ammonia (vide infra).

k1 k2 (B) k3 -
Lg
k-1 k-2

- - - -
A B

Scheme 2.5

The solvolysis rates of 4-NFB and 2-NFB show relatively small salt effects (Figures 2.5 and
2.6), however, it worth noting that 2M salt increases the rate for the 4-isomer nearly 3-fold but
by only 28% for the 2- isomer (Appendix A: Table A19).

119
 Results and Discussion

3.5

2.5
105kobs/s-1

1.5

0.5
0 0.5 1 1.5 2 2.5 3 3.5

Salt concentration/M

Figure 2.5 The first order rate constants for the solvolysis of 4-NFB as a function of added
salt concentration in LNH3 at 25 oC. (Δ) NH4Cl; (□) NaNO3

120
 Results and Discussion

104kobs/s-1

2.5

2
0 0.5 1 1.5 2 2.5

NaNO3 concentration/M

Figure 2.6 The first order rate constants for the solvolysis of 2-NFB as a function of added
salt concentration in LNH3 at 25 oC

The solvolysis rate of 4-nitroazidobenzene (4-NAB), in the absence of salts, is similar to that
for 4-NFB, but that for 2-NFB is nearly two orders of magnitude greater than that for 2-NAB
(Table 2.7). As well as a significant difference in salt effects, 2-NFB is nearly 30 times more
reactive towards solvolysis than its 4-substituted isomer, whereas the reactivities of 2- and 4-
nitroazidobenzenes are similar. This ortho effect for 2-NFB, but not 2-NAB, is also seen in the
solvolysis of the more reactive di-substituted 2,4-DFNB which gives almost exclusively the σ-
substituted derivative (14) as product in liquid ammonia at 25 oC (Scheme 2.6). The product
ratio of ortho to para isomers is compatible with the rate differences observed between 2-NFB
and 4-NFB (Table 2.7). Interestingly, the product analysis of the solvolysis of 2,4-DFNB in
liquid ammonia shows that the solvolysis product ratio between 14 and 15 increases with,

121
 Results and Discussion

although not very pronounced, the decreasing of the reaction temperature (Table 2.8,
Appendix C: Figures N28-N29).

NH3
25°C

98% 2%
14 15

Scheme 2.6

Table 2.8 Product analysis (19F NMR) for the solvolysis of 2,4-DFNB in liquid ammonia under various
temperature

temperature(K) molar ratio 12/13

288.2 52:1

295.2 49:1

308.2 43:1

Preferential substitution ortho to the nitro group is sometimes observed in the reactions of
halonitrobenzenes with neutral nucleophile, and even with sterically hindered nucleophiles.224
Bunnett indicated that a halogen atom ortho to a nitro group could lead to the rotation of the
nitro group out of benzene plane which weakens the conjugation due to the secondary steric
effect, thus the reduced ratio between ortho and para would be expected. 225 On the contrary,
Miller stated that although the nitro group ortho to the halogen atom is not coplanar to the ring
in the initial state, it can be so in the transition state, so the reduction in activating power due
to absence of coplanarity has little or no importance in determining the para and ortho
ratios.226

The reason for the higher reactivity of ortho over para position for nitrofluorobenzenes in
nucleophilic substitution is unlikely to be attributable to the steric effect of fluorine, given the

122
 Results and Discussion

Van Der Waals radius of fluorine is relatively small (1.47 A).227 It is therefore likely that the
enhanced reactivity of 2-NFB over 4-NFB is due to a transition state effect. In support of this,
entropy of activation for the ortho-isomer is less negative than that for the para-isomer (Table
2.9, Appendix A: Tables A21 to A23 and Figures A21 to A23). Although this is compatible
with some interaction between the ortho substituents such as the formation of an
intramolecular hydrogen bond within the activated complex to stabilise the intermediate
(16),228 it would not explain why this favourable interaction cannot occur with 2-NAB. The
generation of negative charge on the nitro group oxygens in the σ-adduct formed from 4-NFB
requires solvation and restriction of solvent molecules giving rise to a slightly more negative
entropy of activation.

Table 2.9 The rates and activation parameters of the solvolysis of 2-NFB, 4-NFB and 2,4-DFNB in LNH3 at 25
o
C

substrate 106kobs / s-1 krel. ΔG‡ / kJ mole-1 ΔH‡ / kJ mole-1 ΔS‡ / J K-1mole-1

4-NFB 7.86 1 101.1 53.0 -161.3

2-NFB 215 27.1 94.1 51.3 -143.4

2,4-DFNB 6717 852 87.7 40.0 -151.6

16

The additional fluorine in 2,4-DFNB compared with 2-NFB increases the solvolysis rate by
about 30-fold presumably due to the inductive effect of fluorine which normally shows an
additive effect on the reactivity of the polyfluorobenzenes. 229 Introduction of a second nitro
group causes the solvolysis of 2,4-DNFB in liquid ammonia to be too fast to measure with our
present equipment. The solvolysis of 4-nitrochlorobenzene does not occur at room
temperature, unless under forcing conditions,230 while the solvolysis of 2,4-

123
 Results and Discussion

dinitrochlorobenzene completes within minutes, but much slower than that for 2,4-DNFB
(Table 2.7)

2.3 Solvolysis of epoxides, esters and sulfonyl chloride

LNH3

25°C
90% 10%

Scheme 2.7

Table 2.10 Pseudo first order rate constants (k0) for the solvolysis of epoxides, esters and ketones in LNH3 at 25
o
C

substrate k0, LNH3(s-1) t1/2

styrene oxide 3.06×10-6 62.9hrs

styrene oxide/1M NH4Cl 6.89×10-6 27.9hrs

phenyl acetate >0.14 <5s

4-nitrophenyl acetate >0.14 <5s

4-nitrophenyl pivolate >0.14 <5s

phenyl benzoate 7.70×10-3 1.5mins

methyl-4-nitrobenzoate 1.44×10-5 13.4hrs

ethyl-4-nitrobenzoate 5.29×10-6 36.4hrs

______
methyl-3-hydroxybenzoate no reaction

triphenylphosphate 3.5×10-6 54.4hrs

______
triethylphosphate no reaction

benzenesulfonyl chloride >0.14 <5s

The solvolysis of styrene oxide is slow in liquid ammonia at 25 oC (Table 2.10), but smoothly
to give the corresponding β-hydroxyamine as the major product, together with a small amount
of bis-β-hydroxyamine as the minor product (Scheme 2.7). No oxygen-benzylic carbon bond

124
 Results and Discussion

fission of styrene oxide is found by GC or HPLC analysis, which would lead to the formation
of 17. This is in stark contrast with the hydrolysis of styrene oxide in water at 25 oC. Although
the rate of hydrolysis (4.18×10-6s-1) is similar to that in liquid ammonia, the diol product is
mainly formed by the attacking of the benzylic carbon from water molecule.231 The solvolysis
rate for styrene oxide in liquid ammonia does not significantly accelerated by added NH4Cl,
which is also in stark contrast with that for acid catalysed hydrolysis process of styrene oxide
(Table 2.10).231Obviously, the hydrolysis of styrene oxide involves a process that the O-C
bond dissociation ahead of N-C bond formation (DN+AN), on the contrary, the solvolysis in
liquid ammonia probably follows a concerted mechanism (ANDN). It worth noting that, with
1M NH4Cl as Lewis acid catalyst for the solvolysis of styrene oxide in liquid ammonia, the
major products are bi- and tri-β-hydroxyamine.

17

In liquid ammonia, esters with aryl groups solvolyse much faster than those with alkyl groups
as leaving group (Table 2.10). The solvolysis of esters gives the corresponding amide and
alcohol or phenol, and follows a classical addition-elimination mechanism (ANDE) in liquid
ammonia. The solvolysis of triphenyl phosphate is slow and gives only mono-substituted
amide (18) in liquid ammonia (Scheme 2.8), surprisingly, the less bulky triethyl phosphate
fails to react with ammonia at 25 oC.

NH3

25°C
18

Scheme 2.8

125
 Results and Discussion

The rate for the solvolysis of benzenesulfonyl chloride in liquid ammonia is so fast that the
accurate measurement is beyond our current experimental condition (Scheme 2.9).
Interestingly, the solvolysis rate of benzenesulfonyl chloride in liquid ammonia is much faster
than the hydrolysis rate at 25 oC,232 although the leaving chloride anion is better solvated by
water than liquid ammonia.

NH3
25°C

Scheme 2.9

2.4 Solvolysis of ketones with ammonium salt as catalyst

In the absence of ammonium salts or other Lewis acids, aromatic ketones, such as
acetophenone and benzophenone, are very stable in liquid ammonia at room temperature or
even higher. However, they can be smoothly converted into the corresponding ketimines in the
presence of ammonium salts at room temperature with good to excellent yields in liquid
ammonia (Scheme 2.10 and Table 2.11).

NH4Cl
LNH3

R1,R2 = aryl or alkyl

Scheme 2.10
Table 2.11 Ketimine yields for the solvolysis of ketones in the presence of various concentration of NH4Cl in
liquid ammonia at 25 oCa

ketimine yield b

ketone 0.1M NH4Cl 0.5M NH4Cl 1M NH4Cl

acetophenone 95.2% 97.4% 98.5%

benzophenone 82.2% 86.9% 92.7%

a
Reaction condition: 0.1M ketones with ammonium salt in 10ml liquid ammonia for 12hours. b GC yield.

126
 Results and Discussion

The yield of the solvolysis of ketone is dependent on the concentration of NH4Cl added, but
reaction temperature is much lower and the yields are higher than previously reported. 233
Catalytic ammonium salts may have two functions in the solvolysis process, firstly it acts as a
Lewis acid to activate the carbonyl group of ketones, which facilitates the attack from
ammonia molecule; secondly it also can remove the resulting water from the system, thus
significantly change the equilibrium towards the formation of ketimine (Scheme 2.11).

+ - +
NH4 NH4 + +
proton transfer -NH4OH -H
..

.. +
NH3

Scheme 2.11

The solvolysis of ketones also can be efficiently catalysed by other Lewis acids in liquid
ammonia, such as TiO2, and gives excellent yields in a high temperature flow system using
liquid ammonia as eluent.234 These ketimines are very useful intermediates for the synthesis of
primary amines though reduction in liquid ammonia.

127
 Results and Discussion

3. Aliphatic nucleophilic substitution reactions in liquid ammonia Page

3.1 Nucleophilic substitution with oxygen-nucleophiles 129

3.2 Nucleophilic substitution with nitrogen-nucleophiles 138

3.3 Nucleophilic substitution with sulfur-nucleophiles 145

3.4 Nucleophilic substitution with carbon-nucleophiles 145

128
 Results and Discussion

3 Aliphatic nucleophilic substitution reactions

Nucleophilic displacement reactions at saturated carbon centres occur either with simultaneous
breaking and forming of the involved bonds (SN2 or ANDN) or by a mechanism where
breaking the old bond precedes formation of the new bond (S N1 or DN+AN). It is well known
that the nature of the solvent used for these reactions can influence the mechanism adopted
and the transition state structure with, for example, a gradation of transition states between the
SN1 and SN2 extremes with varying degrees of participation by the solvent. 235 However, some
theoretical calculations indicate that changes in solvent should not lead to significant changes
in transition state structure.29 An additional complication, particularly in solvents of low
polarity, is the possible intervention of ion pairs. There are many examples of reactions of
various nucleophiles with benzyl derivatives that show a mixed rate law compatible with the
occurrence of simultaneous SN1 and SN2 mechanisms. There have also been attempts to
predict when the change from one mechanism to the other occurs by variation in the structure
of the reactants or solvent.

3.1 Nucleophilic substitution with oxygen-nucleophiles

RO

LNH3

Scheme 3.1

Generally, in nucleophilic substitution reactions, oxygen-nucleophiles act as hard bases having

a small polarisability. 236 In protic solvents, these hard nucleophiles and oxygen anions are
highly solvated to solvent molecules through hydrogen bonding, therefore the nucleophilicity
of oxygen-nucleophiles are greatly decreased as bond formation to oxygen will require at least
partial desolvation. On the contrary, due to the poor solvation of anions in dipolar aprotic
solvents, the nucleophilicity of these oxygen-nucleophiles are generally greater than in protic
solvents.237 Aliphatic nucleophilic substitution reactions between anionic oxygen-nucleophiles
and a neutral substrate are generally regarded as charge dispersion processes, and so the rates
of these reactions should have a little or no sensitivity to the change of solvents according to

129
 Results and Discussion

the Hughes-Ingold rules, However, the specific solvation of nucleophiles and nucleofuges
could be crucial for the kinetics and mechanisms of these reactions.

5
103kobs/s-1

1
0 0.1 0.2 0.3
Concentration of sodium phenoxide/M

Figure 3.1 The dependence of the pseudo first rate constant on the concentration of sodium
phenoxide for the reaction of benzyl chloride with phenoxide anion in LNH3 (I = 0.3 M,
KClO4) at 25 oC

A variety of anionic oxygen-nucleophiles react with benzyl chloride in liquid ammonia to give
the corresponding substitution products (Scheme 3.1), although solvolysis is sometimes
competitive with these reactions and a mixture of products is obtained. The kinetics of
nucleophilic substitution was determined under pseudo first order conditions with excess of
nucleophile over substrate concentration. A typical plot of the rate constants against the
concentration of the nucleophile shows an intercept corresponding to the rate constant for
solvolysis under constant ionic strength (Figure 3.1, Appendix A: Table A24). The slope of
these plots gives the corresponding second order rate constants for alkoxide ions reacting with
benzyl chloride in liquid ammonia. This first order dependence of the pseudo first order

130
 Results and Discussion

constants on the concentration of the nucleophile supports the conclusion from solvolysis
kinetic data that these reactions follow a bimolecular S N2 type mechanism.

There is a large rate enhancement of about 104-fold for the reaction of benzylalkoxide ion with
benzyl chloride in liquid ammonia compared with that in methanol (Table 3.1). The second
order rate constants for the nucleophilic substitution of benzyl chloride by phenoxide ion are
similar in liquid ammonia and DMF, a typical dipolar aprotic solvent, and are about 5000
times greater than that in methanol (Table 3.2).

Table 3.1 The second order rate constants of the reactions between alkoxides and benzyl chloride in different
solvents at 25 oC

nucleophile temp/oC solvent k2/M-1s-1 krel. reference

BnONa/BnOHa 25 LNH3 0.456 12500 this work

MeONa/MeOHa 25 LNH3 0.201 5500 this work

MeONa 25 MeOH 3.61×10-5 1 238

MeOLi 25 MeOH 2.41×10-5 0.68 239

a
The solubility of sodium alkoxides in liquid ammonia at room temperature is very poor. The rate for methoxide
and benzylalkoxide with benzyl chloride was measured by adding 2.5 equivalents of the corresponding alcohol
(the total alcohol volume added is less than 3.5%v/v) in order to give a homogeneous solution. There is no
reaction between the alcohols and benzyl chloride in liquid ammonia at 25 oC.

Table 3.2 The second order rate constants of the reactions between phenoxide and benzyl chloride in different
solvent

nucleophile temp/oC solvent k2/M-1s-1 krel. reference

PhONa 20 MeOH 8.02×10-6 1 240

PhONa 25 LNH3 2.01×10-2 2500 this work

PhONa 20 DMF 4.15×10-2 5100 240

These rates increase on going from protic to dipolar aprotic solvents are attributable to the
specific solvation though hydrogen bonding of anionic nucleophiles in protic solvents, which
decreases their activity as nucleophiles due to the large desolvation energy required on going
from initial state to the transition state.241 This is also shown by the large positive Gibbs
transfer energies of anions from protic solvents to non-polar and dipolar aprotic solvents.57 On

131
 Results and Discussion

the other hand, some dipolar aprotic solvents strongly favour the solvation of cations due to
the availability of electron donation from a solvent molecule’s lone pair, which can lead to the
destruction ion pairs and thus an increased nucleophilicity of the counter-anion. One of main
reasons for dipolar aprotic solvents favouring bimolecular concerted over a unimolecular
mechanisms lies in their poor solvation of the leaving anion.

There is a small decrease in the second order rate constant of the reaction of benzyl chloride
with phenoxide ion with increasing ionic strength by adding non-nucleophilic inorganic salts,
e.g., NaNO3, KClO4. This is in line with the Hughes-Ingold rules,211 that the increasing
polarity of the medium will cause a small decrease of rate if the transition state involves
charge dispersion between nucleophile, substrate and leaving group. Similarly, increasing
ionic strength in liquid ammonia decreases the pKa of the phenol leading to a weaker
nucleophilic phenoxide ion.

The second order rate constants for substituted phenoxide ions reacting with benzyl chloride in
liquid ammonia vary significantly with substituents (Table 3.3). The rate difference between
4-cyano and 4-methoxy phenoxide ion reactions is about 40-fold in liquid ammonia, whereas,
in methanol or alcoholic solvent, the rate is insensitive to the substituent. 242 The Brønsted plot
for the rate constants in liquid ammonia using the aqueous pKa of the phenol shows a very
good linear free energy relationship (Figure 3.2a) with a βnuc = 0.66.

Table 3.3 The second order rate constants of the reactions between 4-substituted phenoxide and alkoxide ions
with benzyl chloride in LNH3 at 25 oC

phenoxide/alkoxide ion pKa (aq.) a pKa (LNH3) 102k2,LNH3 (M-1s-1)

4-cyanophenoxide 7.95 2.71 0.077

4-carbomethoxyphenoxide 8.47 4.04 0.273

4-chlorophenoxide 9.20 4.69 0.95

phenoxide 9.99 6.02 2.01

4-methoxyphenoxide 10.27 6.62 3.97

4-tbutylphenoxide 10.31 6.67 3.19

a
Corresponding aqueous pKa of phenols and alcohols are from ref.243.

132
 Results and Discussion

-1

-1.5 βnuc= 0.66

log(k2/M-1s-1)

-2

-2.5

-3

-3.5
7.5 8 8.5 9 9.5 10 10.5
pKa(aq.)

Figure 3.2a The Brønsted plot for the reaction of para substituted phenoxides with benzyl
chloride in LNH3 at 25 oC using the aqueous pKa of the phenols

As shown in chapter 1, there is a linear relationship between the ionisation of substituted

phenols in liquid ammonia and those in water, the slope of a plot of the pK a of the former
against the corresponding aqueous pKa is 1.68 (Figure 1.5). Hence a more meaningful
Brønsted βnuc for the substituted phenoxides reacting with benzyl chloride in LNH3 is 0.40
(Figure 3.2b). The latter value indicates the transfer of some negative charge from the
attacking phenoxide anion to the benzyl group and the leaving chloride ion and partial bond
formation between the phenoxide oxygen and the benzylic carbon in the transition state. The
βnuc of the analogous reaction in methanol is 0.28,244 suggesting an earlier type of transition
state with a smaller fraction of charge transferred from oxygen to carbon in this solvent. The
solvation ability of a solvent is not only a function of its dielectric constant and dipole
moment, but also of its ability to donate protons or electrons. Although the dielectric constant
and dipole moment of liquid ammonia are much less than those for common dipolar aprotic
and protic solvents, the enhanced rate of reaction between anionic O- nucleophiles and alkyl

133
 Results and Discussion

halides in liquid ammonia compared with alcoholic solvents is probably due to the poor
solvation of anions in the former compared with the latter and good solvation of the anion’s
counter-cation from the ammonia lone pair, thus decreasing ion-pair formation.

βnuc= 0.40
-1.5
log(k2/M-1s-1)

-2.5

-3.5
2.5 3.5 4.5 5.5 6.5
pKa(LNH3)

Figure 3.2b The Brønsted plot for the reaction of para substituted phenoxides with benzyl
chloride in LNH3 at 25 oC using the pKa of the phenols in liquid ammonia

In order to obtain a more detailed picture of the transition state structure the rates of phenoxide
ion reacting with substituted benzyl chloride were measured. A plot of the second order rate
constants for this reaction against the Hammett constant for the substituent (Figure 3.3,
Appendix A: Table A25) is linear apart from the typical deviation for 4-methoxybenzyl
chloride.245 The latter is not due to a change in mechanism from SN2 to SN1 as the kinetics are
still clearly first order in phenoxide ion (Figure 3.4, Appendix A: Table A26). The most
likely explanations are either: (i) a change in the structure of the transition state for a single
mechanism but with a differing degree of bond formation and cleavage, so that the 4-methoxy
derivative causes a shift to a transition state with more positive charge on the central carbon

134
 Results and Discussion

atom245a or (ii) a single transition state structure but with the 4-methoxy substituent stabilising
the transition state with a different balance of polar and resonance effects. 245b,c Further
investigation shows that 3-methoxy derivative behaves normally, as expected, and fits the
linear relationship well.246 The ρ value of 1.11 for the reaction of phenoxide ion with 4-
substituted benzyl chlorides, excluding the 4-methoxy derivative (Figure 3.3), suggests that
appreciable charge has been transferred from phenoxide oxygen to benzylic carbon in the
transition state and which is more than that lost to the departing chloride ion. The ρ value of
1.11 contrasts markedly with that of zero for the solvolysis reaction (Figure 2.1). Overall the
transition state structure for phenoxide-ion substitution is negatively charged compared with a
neutral one for solvolysis so it is not surprising that the rate with phenoxide-ion is enhanced by
electron-withdrawing groups.

ρ = 1.11
-1.0
log(k2/M-1s-1)

-1.5

-2.0
-0.4 -0.15 0.1 0.35 0.6 0.85
σp

Figure 3.3 The Hammett plot for the reaction between 4-substituted benzyl chloride with
phenoxide anion in LNH3 at 25 oC, triangle ( ) shows the positive deviation of 4-
methoxybenzyl chloride, while round circle ( ) shows 3-methoxybenzyl chloride fits the linear
relationship.

135
 Results and Discussion

7.5

5.5
103kobs/s-1

3.5

1.5
0 0.025 0.05 0.075 0.1 0.125

[PhO-](I = 0.1M, KClO4)/M

Figure 3.4 The dependence of the pseudo first order rate constant on the concentration of
sodium phenoxide for the reaction of 4-methoxybenzyl chloride with phenoxide anion in
LNH3 (I = 0.1 M, KClO4) at 25 oC

H H

_
δ- _
PhO C Cl PhO Cl
PhO

Relative large negative charge on the benzylic carbon

Scheme 3.2

The Hammett and Brønsted plots indicate that most of the negative charge is distributed
between the attacking phenoxide oxygen and the benzylic methylene and aromatic ring with
little transferred to the leaving chloride anion, therefore, there is relatively little C-Cl bond

136
 Results and Discussion

fission in the transition state (Scheme 3.2). The changes in transition state structure can be
envisaged on a Jencks-More O’Ferrall reaction coordinate-energy diagram (Figure 3.5) with
separate axes for O-C and C-Cl bond formation and cleavage, respectively.

Figure 3.5 Jencks-More O’Ferrall reaction diagram for the reaction between phenoxide anion
and benzyl chlorides in LNH3 at 25 oC

Phenoxide ion is a well known ambident nucleophile247 and it can undergo both C- and O-
alkylation (Scheme 3.3) and which reaction dominates depends very much on the medium.
Under homogeneous conditions in liquid ammonia, with 0.3M sodium phenoxide and 0.1M

BnX
Solvents

X = Cl or Br 19 20

Scheme 3.3

137
 Results and Discussion

benzyl chloride there is less than 0.5% solvolysis product and no C-alkylated product (19)
formed, giving a selectivity for O-alkylation (20) of almost 100% within half an hour. In
diethyl ether as solvent, under heterogeneous conditions, the reaction between benzyl bromide
and phenoxide ion, after 3 days at 35 oC, the major product (75%) is the C-alkylated one.247
Presumably, the differing solvation of the phenoxide anion, including tight ion-pair formation,
affects the relative negative charge density on oxygen and the ring carbons as well as the
stability of the two transition states leading to C-alkylation in ether and protic solvents.

The nucleophilic substitution reaction between 0.01M styrene oxide and 0.1M phenoxide in
liquid ammonia is slow (t1/2 ≈ 35 hours) but faster than that for the solvolysis and gives 21 as
the major product (Scheme 3.4).

_ LNH3
PhO
25oC

80% 20%
21
Scheme 3.4

3.2 Nucleophilic substitution with nitrogen-nucleophiles

The background solvolysis reaction in liquid ammonia obviously involves nucleophilic attack
by a nitrogen nucleophile, but benzyl chloride also reacts with secondary amines in liquid
ammonia to give predominantly the substituted product (Scheme 3.5). The pseudo first order

LNH3

Scheme 3.5

138
 Results and Discussion

rate constants for the aminolysis of benzyl chloride and the reactions of benzyl chloride with
nitrogen anionic nucleophiles were obtained by the general sampling method and measuring
the formation of the product and the disappearance of the reactant by GC analysis (Table 3.4).
The reaction between benzyl chloride and neutral amines may give neutral or charged
products depending on the pKa of the product relative to the ammonia solvent; although, as
was shown in chapter 1, all amines exist in their free base form in liquid ammonia. The
reaction between two neutral reactants will have a transition state structure in which the
overall charge is neutral, but charge is created on the N-nucleophile during the formation of
the activated complex, assuming no general base catalysis by the solvent. The differences in
solvation of the reactant amines by aprotic and dipolar aprotic solvents is not as marked as for
anions and the rate differences between protic and dipolar aprotic solvents is not so
pronounced as that seen for anionic oxygen nucleophiles.

There is a first order dependence of the pseudo first order rate constant on the concentration of
the amine for the aminolysis of benzyl chloride by morpholine (Figure 3.6, Appendix A:
Table A27), which again confirms that these reactions follow a bimolecular SN2 type
mechanism. The rate constant for hydrazine was obtained by adding 1M hydrazine
dihydrochloride to liquid ammonia to generate the free base and so to correct for the effect of
2M NH4Cl produced from 1M N2H6Cl2, the rate of solvolysis of benzyl chloride with was
measured in the presence of 2M NH4Cl (kobs = 3.42 × 10-3s-1) and so the derived second order
rate constant (Table 3.4) is that for ionic strength 2.0M. There is no reaction of neutral
imidazole or triazole with benzyl chloride in liquid ammonia, but there is with their anionic
salts, indicating that these anions remain deprotonated in liquid ammonia.

139
 Results and Discussion

5.5

4.5

3.5
103kobs/s-1

2.5

1.5

0.5
0 0.2 0.4 0.6 0.8 1 1.2

Concentration of morpholine/M

Figure 3.6 The dependence of the pseudo first order rate constant on the concentration of
morpholine for the reaction of benzyl chloride with morpholine in LNH3 at 25 oC

The second order rate constants for the aminolysis of benzyl chloride with various amines
(Table 3.4) generate a reasonable Brønsted plot using the aqueous pKa of the amine to give a
βnuc of 0.21 (Figure 3.7). This plot includes both neutral and negatively charged amines and
yet both types give a reasonable single plot. Given the work described earlier which showed
that all amines exist in their free base unprotonated form in liquid ammonia it is not possible
to evaluate a Brønsted plot using the pKa of the aminium ions in this solvent. Nonetheless, the
apparent value of 0.21 using the aqueous pKa does indicate that there is some dependence on
the basicity of the amine nucleophile, but much less than that for oxygen nucleophiles
suggesting a transition state with little charge developed on the amine nitrogen in the transition
state or charge removal in the case of negatively charged amine anions.

140
 Results and Discussion

Table 3.4 The second order rate constants for the aminolysis of benzyl chloride with various N-nucleophiles in
LNH3 at 25 oC

nucleophile pKa (aq.) a 102k2(M-1s-1)

pyrrolidine 11.4 2.67

piperidine 10.4 1.70

morpholine 8.5 0.324

sodium azide 4.7 0.773

sodium triazolate 10.3 0.942

sodium benzotriazolate 8.37 0.261

sodium imidazolate 14.5 5.56

hydrazineb 8.1 0.514

a
Corresponding aqueous pKa of amines are from and ref. 248. b 1M hydrazine dihydrogen chloride was used.

-1 βnuc=0.21
log(k2/M-1s-1)

-3

-5
6.0 8.0 10.0 12.0 14.0 16.0
pKa (aq.)

Figure 3.7 The Brønsted plot for the substitution of benzyl chloride by various amines in
LNH3 at 25 °C, ammonia ( ) is negatively deviated from the line, but the positive deviation of
azide anion is not shown.

141
 Results and Discussion

The second order rate constant for ammonia shows a large negative deviation from the
Brønsted plot explaining why aminolysis by amines is easily observed in liquid ammonia. This
could result from the formation, in the transition state, of only one possible hydrogen bond
donor on the attacking nucleophile from secondary amines compared with three from
ammonia, resulting in either a considerable energetic cost in terms of solvent reorganisation or
a lack of hydrogen bond stabilisation/solvation, making ammonia less reactive. The rate
constant for azide ion reacting with benzyl chloride in liquid ammonia shows a positive
deviation of about 15 fold from the Brønsted plot (Figure 3.7 and Table 3.4). It is not known
if there is a similar deviation in the plot of the pKa of aminium ions in liquid ammonia against
their corresponding values in water and so it is not known if this deviation is a consequence of
a higher basicity than ‘expected’ of azide ion in liquid ammonia or an enhanced reactivity. The
other anionic nitrogen nucleophiles studied may be subject to more steric hindrance compared
to azide ion. The rate constant for azide ion reacting with benzyl chloride in liquid ammonia
is only about 150-fold241a greater than that in methanol and this relatively modest rate
difference is in stark contrast to the 2500-fold rate enhancement for phenoxide anion reacting
with benzyl chloride in liquid ammonia compared with that in methanol. Presumably, the
difference in solvation energies of azide anion in liquid ammonia and methanol is relatively
smaller than that of phenoxide ion.

BnX

_
Solvents

X = Cl or Br Bn Bn
22 23

Scheme 3.6

1,2,4-Triazolate and benzotriazolate anions are widely used in agriculture and pharmaceutical
industries249 and, as they are ambident nucleophiles,250 the regioselectivity of their nucleophilic
reactions is important of interest. In liquid ammonia, within a few hours, the major product of
equimolar reaction between benzyl chloride and sodium triazolate is 1-benzyl-1,2,4-triazole
(22) rather than 4-benzyl-1,2,4-triazole (23) in a ratio of 12 : 1 (Scheme 3.6). Previous studies
of this reaction in other solvents, often under heterogeneous conditions, also preferentially

142
 Results and Discussion

gives 1-substituted triazole products, but with much lower selectivity and longer reaction
times.251

BnCl
LNH3

75% 25%
24

Scheme 3.7

Also 1-benzyl-1,2,3-benzotriazole (24) is the main product of the reaction of 0.1M

benzotriazolate and benzyl chloride in liquid ammonia at 25 oC (Scheme 3.7), however, the
background solvolysis reaction strongly competes with that of benzotriazolate anion. The rate
of reaction between hydrazine and benzyl chloride with 2M NH4Cl does not show a
pronounced α-effect, in common with its general reactivity for nucleophilic attack at sp 3
carbon centres.252

In order to see if the difference in the susceptibility of the rate of nucleophilic substitution to
the substituent in substituted benzyl chlorides for solvolysis (Hammett ρ = 0) and with
phenoxide-ion (ρ = 1.11) is a function of the basicity, charge or nature of the attacking
nucleophilic element (O vs N), the rate constants for the reaction of piperidine and triazolate
with 4-substituted benzyl chloride were determined. A Hammett plot of the second order rate
constants (Figure 3.8, Appendix A: Tables A28 and A29) shows that the reaction with the
neutral amine piperidine is insensitive to the para-substitutent in benzyl chloride, similar to
that seen for solvolysis. By contrast, the ρ value for the triazolate anion with 4-substituted
benzyl chloride is 0.93, similar to that seen with phenoxide anion. It appears that the increased
sensitivity to aromatic ring substituents is due to the requirement to accommodate a negative
charge in the transition state. Interestingly, the rate constant for triazolate anion shows the
typical deviation for 4-methoxybenzyl chloride, similar to that seen with phenoxide ion.

143
 Results and Discussion

-1.3 ρ = 0.93
log(k2/M-1s-1)

-1.8
ρ = 0.06

-2.3
-0.4 -0.2 0 0.2 0.4 0.6 0.8 1
σp

Figure 3.8 The Hammett plot for the reaction of 4-substituted benzyl chloride with piperidine
( ) and triazolate anion ( ) in LNH3 at 25 oC.

There is no nucleophilic reaction observed between 0.1M aniline and 0.01M benzyl chloride
in liquid ammonia at 25 oC, except for the solvolysis reaction product benzylamine, and the
rate of the solvolysis is not significantly changed by the added aniline. The predicted second
order rate constant determined by extrapolation of the Brønsted plot for amines (Figure 3.7)
for an aminium ion of pKa 4.6 (i.e. that corresponding to anilinium ion) is 5.50 × 10-4 M-1s-1,
so 0.1M aniline would be predicted to have a pseudo first order rate constant of 5.50 × 10-5 s-
1
compared with 8.89 × 10-4 s-1 for the solvolysis.

High regioselectivity for the reaction between 0.1M 1,2,4-triazolate anion and 0.01M styrene
oxide is found in liquid ammonia (Scheme 3.8). The reaction was complete after 85 hours at
room temperature, and the total yield of the reaction is 95% according to GC and HPLC
analysis. The ratio between 25 and 26 is 30:1, which is much higher than those previously
reported in conventionally used solvents.253

144
 Results and Discussion

LNH3
25°C

25 26

Total yield 95%, 25:26 = 30 : 1

Scheme 3.8

3.3 Nucleophilic substitution with sulfur-nucleophiles

Thiophenoxide is a good nucleophile in many reactions with electrophilic centres largely due
to the ‘softness’ of the sulfur anion.254 The rate for the reaction between thiophenoxide and
benzyl chloride in liquid ammonia is again much faster than that in typical protic solvents
under the same conditions (Table 3.5).

Table 3.5 Second order rate constants for reaction of thiophenoxide with benzyl chloride in various solvents

nucleophile temp/oC solvent 102k2/M-1s-1 reference

PhSNa 25 MeOH 2.42 239

PhSNH4a 20 LNH3 3.15×103 this work

PhSNa 25 DMSO ~108 255

a
Thiophenol (aqueous pKa = 6.5) is fully ionised in liquid ammonia, see ref. 256 for details.

3.4 Nucleophilic substitution with carbon-nucleophiles

_
CR2CH
(PhCH2)2CR2
LNH3, 25°C

Scheme 3.9

The nucleophilic substitution reactions of benzyl chloride by carbanions in liquid ammonia are
complicated by self-ionisation, dibenzylisation and instability of these carbon nucleophiles
(Scheme 3.9). For examples, as shown earlier by NMR studies, carbon acids with a pKa
(DMSO) < 15, such as malonodinitrile, and 2-acetylhexanone, will be deprotonated in liquid

145
 Results and Discussion

ammonia at room temperature, hence the nucleophilic substitution reactions between these
carbon acids and benzyl chloride can occur without the aid of strong bases in liquid ammonia
as is often required in other solvents. The reaction between 0.2M malonodinitrile and 0.01M
benzyl chloride gives both mono-benzylated and dibenzylated products, but with the former
dominant (Scheme 3.10), and with 0.5M malonodinitrile as nucleophile, less than 1%
dibenzylated product is formed. The reaction between 0.2M 2-acetylhexanone with 0.01M
benzyl chloride in liquid ammonia at 25 oC forms the corresponding mono-benzylated
product, and also a small amount of 2-iminocyclohexyl ethanone, which is due to the
solvolysis of the 1,3-diketone in liquid ammonia (Scheme 3.11).

LNH3

25°C

90% 10%

Scheme 3.10

LNH3

25°C

Scheme 3.11

Malonate diethyl ester and benzyl cyanide remain in their neutral forms and are stable in
liquid ammonia at room temperature, however, their anionic conjugate bases can be generated
in situ by using sodium amide in liquid ammonia and these carbanions react readily with
benzyl chloride to give only mono-benzylated products.

The second order rate constants for cyanide ion reacting with benzyl chloride in various
solvents are given in Table 3.6 which shows that its reactivity is similar to that in DMF. The
second order rate constants for carbanions reacting with benzyl chloride in liquid ammonia are
given in Table 3.7. The rates of reactions which involve different types of carbanions as
nucleophiles are often found to be poorly correlated with the corresponding pKa of their

146
 Results and Discussion

conjugate acids in various solvents.196,257 Some specific factors for a certian types of
carbanion, such as solvent reorganisation and rehybridisation of nitroalkanes carbanions, 258
cyano carbon acids,194,259 must be considered in order to rationalise the abnormal reactivities
of these carbanions in nucleophilic substitution reactions.

Table 3.6 Second order rate constants for reaction of carbanions with benzyl chloride in various solvents

nucleophile temp/oC solvent 103k2/M-1s-1 reference

NaCN 25 LNH3 2.18 this work

Et4NCN 30 DMF/water=95:5 3.20 260

Et4NCN 30 DMF/water=80:20 0.57 260

KCN 30 DMF/water=95:5 5.3 260

KCN 25 [bmim][PF6] 0.022 261

Table 3.7 Second order rate constants for various carbanions reacting with benzyl chloride in liquid ammonia at
25 oC and their corresponding pKa in water and DMSO

carbon acid pKa(aq) pKa(DMSO) 103k2(M-1s-1)

malonodinitrilea 11.2 11.1 8.88

HCN 9.21 12.9 2.18

2-acetylhexanoneb 10.1 14.1 6.44

diethyl malonatec 12.9 16.4 2.27

acetophenoned 19.1 19.8 256

benzyl cyanided 21.9e 21.9 1400

a
0.5M malonitrile and 0.01M benzyl chloride. b 0.2M 2-acetylhexanone and 0.01M benzyl chloride. c Sodium
malonate anion was generated in situ by reacting diethyl malonate(1.25eq.) with NaNH 2 (1eq.) in liquid ammonia
until all the solid (NaNH2) disappeared. The reaction was studied under the pseudo first order conditions. The
product was confirmed by GC-MS. The tests showed that without adding NaNH2 diethyl malonate does not react
with benzyl chloride and diethyl malonate is stable in liquid ammonia at 25 oC for days.d 0.15M carbon acids and
0.1M NaNH2, then 0.01M benzyl chloride injected into the reaction vessel. The excess carbon acids do not react
with benzyl chloride. e pKa in DMSO, no aqueous pKa data available.

The Brønsted plot for the reactions between various common carbanions and benzyl chloride
in liquid ammonia at 25 oC gives an apparent βnuc = 0.30, using pKa of their corresponding
carbon acids in DMSO. However, the exact pKa of these carbon acids in liquid ammonia are

147
 Results and Discussion

unknown (Table 3.7 and Figure 3.9). The positive deviation of malonodinitrile anion from the
correlation line is observed in other systems, 262 and reason for this is probably due to the
negative charge mainly resting on the central carbon as charge delocalisation is accompanied
by an unfavourable structure,194,196 thus makes malonodinitrile a better nucleophile compared
with strongly delocalised carbanions.

1.0

0.0 βnuc = 0.30

log(k2/M-1s-1)

-1.0

-2.0

-3.0
10 14 18 22 26
pKa(DMSO)

Figure 3.9 Brønsted plot for the reactions of some carbanions with benzyl chloride in liquid
ammonia at 25 oC using the pKa of their conjugate acids in DMSO. ( ) Shows malonodinitrile
positively deviated from the line, and ( ) shows malonate diethyl ester negatively deviated
from the line.

148
 Results and Discussion

35

Slope =1.05 for 1,3-dicarbonyl derivatives

30

25
pKa (DMSO)

20

15
Slope = 1.05 for cyano carbon acids
10

0
0 5 10 15 20 25 30 35
pKa (aq.)

Figure 3.10 pKa of carbon acid in DMSO against its aqueous pKa. ( ) nitroalkanes; ( ) 1,3-
dicarbonyl derivatives; ( ) cyano carbon acids.

Interestingly, the aqueous pKa of some common used carbon acids, except for nitroalkanes,
correlate well with their corresponding pKa in DMSO, and has a slope of 1.05 (Figure 3.10,
for details, see: Appendix A: Table A31). This is very different from the similar plots seen
previously for phenols (Figure 1.5), and this slope of 1.05 indicates that the pKa of carbon
acids are insensitive to the solvation effects of dipolar aprotic and protic solvents. Presumably
this is due to the majority of the stabilisation coming from the delocalised carbanion structures
which do not require substantial stabilisation from the solvent. The small βnuc = 0.30 observed
in the Brønsted plot, for carbanions reacting as nucleophiles in liquid ammonia, using their
corresponding pKa of carbon acids in DMSO, therefore becomes more meaningful, even
without the knowledge of the pKa of these carbon acids in liquid ammonia. Thus, similar to
that seen for nitrogen anionic nucleophiles, the reactions between carbanions and benzyl

149
 Results and Discussion

chloride occur with a transition state in which only a small amount of charge is removed from
the carbanion to the benzylic carbon and leaving group and transferred in liquid ammonia.

In order to see if there is much charge development on the benzylic carbon in the transition
state, the rates of nucleophilic substitution of malonate diethyl ester carbanion with substituted
benzyl chlorides were determined. The Hammett plot the second order rate constants for the
reaction of malonate diethyl ester carbanion with 4-substituted benzyl chlorides in liquid
ammonia at 25 oC (Figure 3.11) is scattered, but an apparent ρ = 0.87 is obtained excluding 4-
methyl and 4-methoxy benzyl chlorides. This ρ value, is smaller compared to those observed
for the reaction of substituted benzyl chlorides reacting with phenoxide anion (ρ = 1.11) and
nitrogen anionic nucleophiles (ρ = 0.93).

-1.5

ρ = 0.87

-2.0
log(k2/M-1s-1)

-2.5

-3.0
-0.4 0 0.4 0.8 1.2
σp

Figure 3.11 Hammett plot for the reaction of malonate diethyl ester carbanion with 4-
substituted benzyl chlorides in liquid ammonia at 25 oC. ( ) 4-methoxy benzyl chloride and (
) 4-methyl benzyl chloride show positive deviations from the line.

150
 Results and Discussion

Interestingly, both 4-methyl and 4-methoxy benzyl chloride derivatives show a positive
deviation from the correlation. The Hammett plot could be seen as two parts, showing the ‘V’
shape (Figure 3.11, Appendix A: Table A30) often seen in other correlations.245c This is again
probably not because of a change in the reaction mechanism from SN2 to SN1, as the reaction
rates all show a first order dependence on the concentration of malonate carbanion. This
behaviour probably reflects a single transition state structure but with the 4-methoxy
substituent stabilising the transition state with a different balance of polar and resonance
effects.

151
 Results and Discussion

4. Aromatic nucleophilic substitution reactions in liquid ammonia Page

4.1 Nucleophilic substitution with oxygen-nucleophiles 153

4.2 Nucleophilic substitution with nitrogen-nucleophiles 159

4.3 Nucleophilic substitution with sulfur-nucleophiles 168

4.4 Nucleophilic substitution with carbon-nucleophiles 170

152
 Results and Discussion

4. Aromatic nucleophilic substitution reactions

A significant proportion of reactions carried out by the pharmaceutical and agrochemical

industries involve aromatic nucleophilic substitution reactions. The nature of the solvent used
for these reactions influences both the kinetics and mechanisms of the process,1 and when
solvents are used in large quantities on an industrial scale, their efficiency, cost and
environmental impact are major factors involved in their selection.

The rates of some SNAr reactions are much faster in dipolar aprotic solvents than aprotic
solvents largely due to the special solvation effects on these reactions. Generally speaking, due
to the large dissociation energies for sp2 C-X bond of aryl halides,263 unsubstituted aryl halides
are often inactive towards nucleophiles in aromatic substitution reactions, therefore electron
withdrawing groups are often required for the activation of aryl ring. As described previously,
4-NFB solvolyses rather slowly but is active enough to react with various nucleophiles in
liquid ammonia. For this reason, 4-NFB is chosen as a representative substrate to study the
kinetics and mechanisms of SNAr reactions in liquid ammonia.

4.1 Nucleophilic substitution with oxygen-nucleophiles

RO

LNH3

Scheme 4.1

Oxygen nucleophiles, such as alkoxide and phenoxide ions, react readily with 4-NFB in liquid
ammonia to give the corresponding substitution product (Scheme 4.1). There is little
solvolysis product formed as the background rate of reaction of 4-NFB with ammonia is too
slow to compete with the rates of substitution by anionic O-nucleophiles. With excess
nucleophile, the rate of substitution shows pseudo first order kinetics and the associated rate
constants show a first order dependence on the concentration of the anion and the derived

153
 Results and Discussion

second order rate constants (Table 4.1, Appendix A: Table A32) were obtained from the slope
of these plots (Figure 4.1).

The rates of reactions of 4-NFB with O-nucleophiles in liquid ammonia are are similar to
those in DMSO and are 4-5 orders of magnitude faster than in methanol and. This large rate
enhancement is probably due to the differences in solvation of the nucleophilic anions in
dipolar aprotic and protic solvents, giving rise to enhanced nucleophilicity of anions in liquid
ammonia.

Table 4.1 The second order rate constants for the nucleophilic substitution of 4-NFB by oxygen anions in
different solvents at 25 oC

O-nucleophile solvent k2/M-1s-1 krel. reference

MeO- LNH3 >3.5a >20,000 this work

MeO- DMSO-MeOHb 3.77×10-1 2,106 264

MeO- MeOH 1.79×10-4 1 265

PhO- LNH3 0.0528 41,000 this work

PhO- DMSOc 0.52 400,000 266

PhO- MeOH 1.29×10-6 1 28

a
Rate is too fast to be measured accurately. b 80% (%v/v) DMSO-MeOH solution. c 20 oC

The second order rate constant for the reaction of phenoxide with 4-nitrochlorobenzene,
2.51×10-7 M-1s-1 at 25 oC, is 5 orders magnitude smaller than that of 4-NFB (Table 4.1), which
probably reflects the less favourable formation of the σ-complex. It is usually assumed that the
mechanism of SNAr reactions involves a charge delocalised Meisenheimer intermediate, the σ-
complex, (Scheme 2.5) in which negative charge of an incoming nucleophile is spread into the
aromatic ring and substituents through resonance, and so any stabilising influence of the
solvent is expected to be less in the transition state than in the relatively localised reactant
anion. Liquid ammonia, in common with dipolar aprotic solvents and unlike protic ones, 267,219
increases the rate of aromatic nucleophilic substitution by anions by several orders of
magnitude primarily due to the less solvated but more reactive nucleophile.

154
 Results and Discussion

There are some interesting differences in the activation parameters for oxygen anions reacting
with 4-NFB in liquid ammonia compared with those in methanol. The lower free energies of
activation in liquid ammonia appear as much lower enthalpies of activation which more than
compensate for unfavourable large negative entropies of activation compared with those in the
protic solvent methanol (Table 4.2, Appendix A: Table A32).

10.0

8.0

6.0
103kobs/s-1

4.0

2.0

0.0
0 0.1 0.2 0.3
Concentration of phenoxide/M

Figure 4.1 The dependence of the pseudo first order rate constants on the concentration of
phenoxide ion for the reaction between 4-NFB and sodium phenoxide in LNH3 at 25 oC (I =
0.3M, KClO4).

Table 4.2 Activation parameters for the nucleophilic substitution of 4-NFB by oxygen anions in LNH3 and
methanol

nucleophile solvent ΔH‡ / kJ mol-1 ΔS‡ / J K-1 mol-1 reference

PhO- LNH3 38.1 -141.3 this work

PhO- MeOH 102.9 -19.7 224d,e

MeO- MeOH 88.6 -27.6 224a,268

155
 Results and Discussion

This suggests that the oxygen anions are strongly hydrogen-bonded to the solvent molecules in
methanol so that the large negative entropy loss expected for the bimolecular process 269 is
partially compensated by the release of solvent molecules on going from the initial reactant
state to the transition state. By contrast, in liquid ammonia, the poor solvation of the oxygen
anions leads to a smaller contribution to the entropy of activation from the solvent and
consequently there is a large negative loss of entropy as a result of covalently linking two
molecules together.9 The good solvation of metal cations in liquid ammonia through electron
donation from ammonia presumably also increases the nucleophilicity and reactivity of the
counter anion in this solvent, which is reflected in the low enthalpy of activation compared
with that in methanol.

A rigorous interpretation of linear free-energy relationships for reactions in liquid ammonia

requires a knowledge of the effect of substituents on equilibria in this solvent. In chapter 1 the
ionisation constants for substituted phenols in liquid ammonia and that these show a linear
relationship with the corresponding values in water (Figure 1.5). The Brønsted plot for the
reaction of 4-NFB with para-substituted phenoxides in liquid ammonia using the aqueous pKa
for the phenols gives an apparent β nuc of 1.49, but a more meaningful βnuc of 0.91 is obtained
using the pKa of the substituted phenols in liquid ammonia (Figures 4.2a and 4.2b, Appendix
A: Table A34). These values are much larger than those for the similar reaction of 4-NFB
with phenoxides or thiophenoxides in protic solvents which are around 0.5.270 The large value
is indicative of significant, if not total, removal of the negative charge on the oxygen anion
and complete bond formation in the transition state and therefore suggests that the
decomposition of σ-complex is the rate limiting step. This is probably due to the difficulty of
expelling and solvating the leaving fluoride anion from the Meisenheimer σ-intermediate
(Scheme 2.5) in liquid ammonia.271 As stated earlier, the second order rate constant for the
reaction of phenoxide with 4-nitrochlorobenzene is 5 orders magnitude smaller than that of 4-
NFB,272 which probably reflects the less favourable formation of the σ-complex and expulsion
of chloride ion.

156
 Results and Discussion

0.0

βnuc=1.49
-1.0

log(k2/M-1s-1) -2.0

-3.0

-4.0

-5.0
7.5 8.5 9.5 10.5
pKa(aq.)

Figure 4.2a Brønsted type plot for the second order rate constants for the reaction between
para-substituted phenoxides and 4-NFB in LNH3 against the aqueous pKa of the phenol

0.0

βnuc= 0.91
-1.0

-2.0
log(k2/M-1s-1)

-3.0

-4.0

-5.0
2.5 3.5 4.5 5.5 6.5 7.5
pKa(LNH3)

Figure 4.2b Brønsted type plot for the second order rate constants of the reaction between
para-substituted phenoxides and 4-NFB in LNH3 against the pKa of the phenol in LNH3

157
 Results and Discussion

An alternative interpretation of the large Brønsted βnuc for nucleophilic substitution of 4-NFB
by phenoxide ion is a single electron transfer (SET) mechanism which would convert the
phenoxide ion to a radical and the 4-NFB to an aromatic radical anion.278

As described earlier (chapter 2), the solvolysis rates of 4-NFB shows a relatively small salt
effect with 2M salt increasing the rate by nearly 3-fold but, by contrast, the rate of substitution
of 4-NFB by phenoxide ion shows a decrease in rate with increasing salt concentration
(Figure 4.3). Presumably, this reflects a greater dispersion of the negative charge in the
transition state.

5.5

4.5
103kobs/s-1

3.5

2.5

1.5
0 0.5 1 1.5 2

concetration of sodium nitrate/M

Figure 4.3 Dependence of the pseudo first order rate constant for the on the reaction of 4-NFB
with phenoxide anion upon the concentration of added salt in LNH3 at 25 oC

Interestingly, the rate of the reaction between 2-NFB and phenoxide anion is only less than 2
times faster than that for 4-NFB in liquid ammonia at 25 oC,273 which is in stark contrast with
the 30-fold rate difference in the solvolysis of NFBs in liquid ammonia. It was postulated that
the difference in solvolysis rates was due to stabilisation of the Meisenheimer σ-intermediate

158
 Results and Discussion

by intramolecular H-bonding between the ortho nitro group and the ammonium ion, but this is
not possible with the reaction of 4-NFB and phenoxide (27). The small rate enhancement for

_ _

_ _

27

the reaction of 2-NFB with phenoxide anion in liquid ammonia, compared with that for 4-NFB
as the substrate, is probably due to the net result of a stronger inductive and unfavourable
steric effect of the neighbouring ortho nitro group.

The reaction between 0.01M 2,4-DFNB and 0.1M phenoxide gives exclusive di-substituted
derivative (28) within minutes (Scheme 4.2), with less than 1% solvolysis product (14)
observed.

-
PhO

LNH3

28

Scheme 4.2

4.2 Nucleophilic substitution with nitrogen nucleophiles

The kinetics and mechanisms of secondary amines reacting with activated aryl halides have
been well studied.274 The solvent effects on those reactions are often complicated by base
catalysis and the extent of which depends on the reaction medium and substrate structure.
Normally, general base catalysis occurs in non-polar aprotic solvents especially when proton
removal is required from the attacking nucleophile before the leaving group is expelled,

159
 Results and Discussion

although it may also occur coupled with the decomposition of the σ-complex.275 However, in
dipolar aprotic solvents the general base catalysis is generally not observed.274e-g

LNH3
R2NH NH4F

Scheme 4.3

Secondary amines react with 4-NFB in liquid ammonia to form the corresponding nitroaniline
and ammonium fluoride (Scheme 4.3). The reaction progress was determined by monitoring
the appearance of the product using GC analysis which followed an exponential change with
time, from which the pseudo first order rate constants were calculated.

3
104kobs/s-1

0
0 0.2 0.4 0.6 0.8

Morpholine concentration/M

Figure 4.4 The dependence of the pseudo order rate constant for the reaction between 4-NFB
and morpholine on the concentration of morpholine in LNH3 at 25 oC.

160
 Results and Discussion

These observed pseudo first order rate constants increase linearly with increasing amine
concentration (Figure 4.4, Appendix A: Table A36), indicating just a first order dependence
on amine concentration and the absence of general base catalysis by a second molecule of
amine. The corresponding second order rate constants for secondary amines and other nitrogen
nucleophiles are shown in Table 4.3.

Table 4.3 The second order rate constants for the substitution of 4-NFB by nitrogen nucleophiles in LNH3 at 25
o
C

N-nucleophile pKa (aq.) 103k2 (M-1s-1)

sodium azide 4.70 0.382a
morpholine 8.50 0.401
1,2,4-triazolate 10.3 0.560
piperidine 10.4 2.23
pyrrolidine 11.4 5.01
imidazolate 14.5 5.73
a
Reaction conditions: 0.1M 4-NFB with 0.01M sodium azide.

As discussed in the solvolysis section, aminium ions exist only in their free base unprotonated
form in liquid ammonia i.e. the latter is more basic than amines. The anionic Meisenheimer σ-
intermediate (D) (Scheme 4.4) therefore is thermodynamically more stable than its conjugate
acid (C) in liquid ammonia and other amines are unlikely to be able to compete with solvent
ammonia in converting (C) to (D) and so the absence of general base catalysis by amines is
not surprising.


k1 k2 (B) k3
Lg -
k-1 k-2

- - - -
C D

Scheme 4.4

There is no nucleophilic substitution of 4-NFB with aniline, with DABCO [(1,4-

diazabicyclo(2.2.2)octane] and triethylamine in liquid ammonia and only the solvolysis

161
 Results and Discussion

product, 4-nitroaniline, is formed. The rate of solvolysis is not significantly increased by these
amines, again indicating no general base catalysis by these amines. The rates of reaction of
sodium azide and piperidine with 4-NFB are similar to those in some typical dipolar aprotic
solvents such as acetonitrile, DMSO, DMF (Appendix A: Tables A37 and A38).276 The
reaction between 1,2,4-triazolate and 4-NFB in liquid ammonia gives only the 1-substituted
product 29.

29

As already stated all amines exist in their free base unprotonated form in liquid ammonia and
so it has not been possible to evaluate the pKa of aminium ions in this solvent. Nonetheless,
the second order rate constants (Table 4.3) do increase with increasing aqueous basicity of the
amine, and there is actually a reasonable correlation between the second order rate constants
for aminolysis of 4-NFB in liquid ammonia and aqueous pKa values of the amines which
generates an apparent Bronsted βnuc = 0.36 (Figure 4.5).

-2

βnuc=0.36
-2.5
log(k2/M-1s-1)

-3

-3.5

-4
8 9 10 11 12
pKa(aq.)

Figure 4.5 Brønsted type plot for the substitution of 4-NFB by nitrogen nucleophiles in LNH3
using the corresponding aqueous pKa of the aminium ion.

162
 Results and Discussion

A plot of the pKa values of aminium ions in acetonitrile against those in water is linear of
277
slope = 1.0 and so using the pKa values in this aprotic solvent would generate the same
Brønsted βnuc. It seems reasonable to conclude that a similar or even smaller value of Brønsted
βnuc would be seen if the pKa of aminium ions in liquid ammonia could be used. The small
Brønsted βnuc contrasts with the βnuc of 0.91 observed with phenoxide anion which was
obtained using pKa values for phenols determined in ammonia. Without the knowledge of the
relative pKa in liquid ammonia it is not possible to interpret this value with any certainty, but it
is indicative of only a small amount of positive charge development on the amine nitrogen
nucleophile in the transition state. This small value is compatible with rate limiting breakdown
of the σ-complex (D), following the deprotonation of the aminium ion in the Meisenheimer
intermediate (C) (Scheme 4.4). This proton transfer step to the solvent ammonia is probably
thermodynamically favourable given the effect of the adjacent fluorine in reducing the pKa of
the aminium ion and the fact that all aminium ions are deprotonated in liquid ammonia. This
suggestion is further supported by the lack of reactivity of tertiary amines, discussed earlier,
which may well be due to the lack of a removable proton. An alternative mechanism could
involve, proton transfer to solvent being coupled to expulsion of the fluoride ion in a concerted
breakdown of the σ-complex. The small Brønsted βnuc value is incompatible with a SET
mechanism in which an electron is transferred from the amine to the aromatic residue
generating a positive charge on the amine to give a radical cation and a full positive charge on
the amine nitrogen.278

In addition to the enhanced reactivity of azide-ion compared with its aqueous basicity (Table
4.3), there are some unusual observations with the reactions of this nucleophile with 4-NFB in
liquid ammonia. The reaction between sodium azide and 4-NFB in other solvents affords, as
expected, the corresponding 4-NAB. However, in liquid ammonia the reaction is very
complicated giving no 4-NAB after the 4-NFB has completely reacted. The final reaction
products are 4-nitroaniline, nitrobenzene (30), diazene (31) and nitrogen (Scheme 4.5). The
molar ratio of 30 and 31 in the products is independent of whether the reaction vessel is
covered in aluminium foil or not. In the absence of air, with control of ionic strength (I = 3M,
NaNO3) the rate of the disappearance of 4-NFB in liquid ammonia is proportional to the azide
concentration at 25 oC (Figure 4.6, Appendix A: Table A39).

163
 Results and Discussion

NaN3
LNH3

30 31

70% 15% <5%

Scheme 4.5

2.5

1.5
104kobs/s-1

0.5

0
0 0.5 1 1.5 2 2.5 3 3.5
Concentration of sodium azide/M

Figure 4.6 The dependence of the pseudo first order rate constant concentration for the
reaction between 4-NFB and sodium azide on sodium azide in LNH3 at 25 oC (I = 3M,
NaNO3).

A careful investigation, by GC and HPLC, of the reaction of azide ion with 4-NFB shows that
4-NAB is, in fact, a reactive intermediate and its concentration initially increases, reaches a
maximum and then decreases (Figure 4.7).

164
 Results and Discussion

0.8
0.5M azide+0.015M 4-NFB

0.6
Normalised area

0.4

0.2

0
0 1000 2000 3000
Time (mins)

Figure 4.7 The reaction profile for 4-NFB reacting with sodium azide in LNH3 at 25oC ( : 4-
NFB; : 4-nitroaniline; : 4-NAB; : nitrobenzene)

Consequently, we investigated, in separate experiments, the solvolysis of 4-NAB in liquid

ammonia. In the absence of salts, 4-NAB has a half life of 38 hours and yields the same
products as does 4-NFB with azide anion (Scheme 4.5). Furthermore, unlike the other
aromatic substitution reactions, the rate of decomposition of 4-NAB in liquid ammonia is very
dependent upon salt concentration. For example, it is 35 fold faster in the presence of 1.0 M
perchlorate and the observed pseudo first order rate constant for the decomposition of 4-NAB
is proportional to the concentration of potassium perchlorate (Figure 4.8, Appendix A: Table
A40). The rate of decomposition of 4-NAB in liquid ammonia is independent of the nature of
salt, whether sodium nitrate, sodium azide or potassium perchlorate. The nitrogen gas formed
originates from 4-NAB and not from ammonia solvent as shown by using 15N enriched liquid
ammonia for the solvolysis of 4-NAB.279

165
 Results and Discussion

3.0

2.0
104kobs/s-1

1.0

0.0
0 0.5 1 1.5 2
Concentration of potassium perchlorate/M

Figure 4.8 The dependence of the observed pseudo first order rate constant for the
decomposition of 4-NAB in LNH3 at 25 oC on potassium perchlorate concentration.

The unusual products, nitrobenzene (30) and diazene (31) formed from 4-NFB and azide ion
(Scheme 4.5) could be explained by the formation of an intermediate nitrene which is trapped
by ammonia to form 4-nitrophenyl hydrazine. Interestingly, the reaction of hydrazine with 4-
NFB in liquid ammonia gives, after work-up with sodium hydroxide and extraction with
dichloromethane, a mixture of nitrobenzene, 4-nitroaniline and aniline in a molar ratio of
12:5:1.280 The formation of nitrobenzene and aniline in this reaction suggests that 4-
nitrophenyl hydrazine could be an unstable intermediate formed in the reaction of 4-NFB with
azide ion in liquid ammonia.

So a possible origin of nitrobenzene (30) and diazene (31) from the reaction of azide ion with
4-NFB in liquid ammonia is the decomposition by the loss of nitrogen of the initially formed
azide, 4-NAB, to give 4-nitrophenyl nitrene which is trapped by ammonia to form 4-
nitrophenyl hydrazine. The generation of the nitrene from 4-NAB by the release of nitrogen
does occur under thermal and photolytic conditions and also electrochemically,281 and singlet

166
 Results and Discussion

aromatic nitrenes with electron-withdrawing groups are known to undergo insertion into the
N-H bonds of amines to give the corresponding hydrazines. 282 The major product of 4-
nitrophenyl nitrene trapped by diethylamine is 4-nitroaniline, with the minor product being 4-
nitrophenyl hydrazine under photo irradiation, but no diazene (31) is found whether the
reaction is under high or low power photo irradiation.282a,b

32 33

The formation of the diazene (31) is perhaps surprising given the expected low concentration
and stability of the nitrene and its formation may require the triplet state 4-nitrophenyl
nitrene.282b No possible ring enlargement products, such as 5-nitro-1,2-didehydroazepine (32)
or its amination derivative (33) were found from the solvolysis of 4-NAB in liquid ammonia,

:
31
_
N2

- -

30

Scheme 4.6

167
 Results and Discussion

and the molar ratio between 4-nitroaniline and nitrobenzene is about 5:1 in the product
mixture. The diazene compound (31) is quite stable in liquid ammonia at room temperature.
The possible routes for the decomposition of 4-NAB in liquid ammonia are therefore
summarised as in Scheme 4.6.

The nitrene could undergo insertion into the N-H bond of solvent ammonia to form 4-
nitrophenylhydrazine, which decomposes into nitrobenzene and a small amount of 4-
nitroaniline. The nitrene could also dimerise to form the diazene (31). Some of the 4-
nitroaniline formed may also come from the direct solvolysis of 4-NAB. Interestingly, the
similar results were observed when 4-NAB was dissolved in DMSO together with excess of
tetraethylammonium azide, followed by passing ammonia gas into the solution for several
hours, however, no reaction was observed in methanol.

The reaction between equal molar of 2,4-difluoronitrobenzene (2,4-DFNB) and secondary

amines in liquid ammonia at 25 oC forms ortho substituted derivative (34) as the predominant
product, together with samll amount of para substituted derivate (35) and solvolysis product
14 (Scheme 4.7).

LNH3

25°C

34 35 14
90% 5% 3%

Scheme 4.7

4.3 Nucleophilic substitution with sulfur nucleophiles

4-NFB reacts rapidly with thiophenoxide anion in liquid ammonia to give 4,4'-dinitrodiphenyl
disulfide and a trace amount of diphenyl thioether, probably due to the oxidation of
thiophenoxide by air during the sample processing stage (Scheme 4.8). The second order rate

168
 Results and Discussion

constant for the reaction between thiophenoxide and 4-NFB in liquid ammonia is significantly
greater than that in methanol, and is similar to that in DMSO (Table 4.4).

_
PhS
LNH3

Scheme 4.8
Table 4.4 The second order rate constants for the reaction of thiophenoxide ion with 4-NFB in various solvents at
25 oC

nucleophile solvent k2/M-1s-1 krel. reference

PhSNa DMSO 10.4 5×104 283

PhSNH4 LNH3 3.1 1.5×104 this work

PhSNa MeOH 2.1×10-4 1 284

4.4 Nucleophilic substitution with carbon nucleophiles

To investigate aromatic nucleophilic substitution in liquid ammonia, the reactions of some

typical carbanions, i.e., cyanide and diethyl malonate anions, with 4-NFB were studied at 25
o
C. However, only the solvolysis product of 4-NFB is observed. Interestingly, orange or red
colours are observed for the reactions of 4-NFB in liquid ammonia in the presence of these
carbanions, but, upon the vapourisation of ammonia, followed by the general work up
procedure, only 4-nitroaniline is found.

The colour may be due to the reversible formation of an anionic Meisenheimer σ-intermediate
however, possibly carbanions do not attack the carbon 1 of 4-NFB attached to the fluoride
atom, but they attack the α-position to the nitro group (Scheme 4.9).

169
 Results and Discussion

LNH3 -
-
CHR1R2
25°C

Scheme 4.9

In summary, the rates of SNAr reactions in liquid ammonia are much greater than those in
protic solvents and are similar to those in dipolar aprotic solvents. In many cases the
nucleophilic substitution reactions are sufficiently faster than the background solvolysis
reaction so that useful synthetic procedures are possible in liquid ammonia. The rates of S NAr
reactions with neutral amines in liquid ammonia are slower than those for anionic O- and S-
nucleophiles of similar aqueous basicity. Liquid ammonia can increase the regioselectivity of
some reactions compared with more conventional solvents. These results indicate that liquid
ammonia has potential as an easily recoverable solvent in many reactions usually carried out
in dipolar aprotic solvents by the chemical industry.

170
 Results and Discussion

5. Metal catalysed organic reactions in liquid ammonia Page

5.1 Copper catalysed amination of aryl halides 172

5.1.1 Copper (I) catalysed amination of aryl iodides 173

5.1.2 Copper (I) catalysed amination of aryl bromides and chlorides 185

5.2 Copper (I) catalysed azide-alkyne cycloaddition (CuIAAC) 188

171
 Results and Discussion

5.1 Copper (I) catalysed amination of aryl halides

Primary aromatic amines are widely used in the manufacture of pharmaceuticals,

agrochemicals, dyes, polymers and thus are very important intermediates in the chemical
industry. 285 In recent years, transition metal catalysed C-N bond formation has become a
powerful and reliable method for the synthesis of a variety of aromatic amines under mild and
convenient conditions. 286 However, these metal catalysed reactions are rarely carried out
directly with ammonia in industry or for synthetic purpose due to: 1) the difficulty of forming
the metal complexes without ammonia displacing the chelating ligands and thus causing the
deactivation of the catalyst.287 2) the amination products, primary aromatic amines, could be
more reactive than ammonia and potentially further react with starting aryl halides to form di-
and triaryl amines in conventionally used solvents. 288 Despite the disadvantages mentioned
above, the investigation of metal catalysed direct amination of aryl halides using ammonia
itself rather than ammonia surrogates286a,b,d, 289 has been investigated by several research
groups. 290 Recent results have shown that, although using palladium or copper as metal
catalyst with auxiliary chelating ligands for the amination of aryl halides with ammonia in
conventional solvents are promising, there are still many problems associated with their
application in industry. For example, the use of expensive palladium catalysts and toxic
ligands; the unavoidable formation of di- and triaryl amines as by-products; the need to
incorporate a deprotection step in the overall transformation for those reactions involving
ammonia surrogates, which is not an atom economic process, and the relatively high loading
of copper catalyst, etc. are all barriers yet to be overcome.

As reported elsewhere in this thesis, we have been trying to combine the unique solvent
properties of liquid ammonia and the potential benefit of using ammonia as a direct nitrogen
source, instead of using ammonia surrogates, to realise a number of amination reactions in
liquid ammonia. Ideally this would be under catalytic conditions and would be applicable to a
range of reactions such as the reductive amination of ketones, and the direct amination of aryl
halides.

172
 Results and Discussion

5.1.1 Copper (I) catalysed amination of aryl iodides

Aryl iodides are stable in liquid ammonia at room temperature for days, however, in the
presence of a catalytic amount (1mol%) copper (I) iodide and 1mol% of ascorbic acid,
iodobenzenes react with ammonia smoothly to give the corresponding anilines with excellent
yields (Scheme 5.1, Table 5.1). No further coupling reactions between starting iodobenzenes
and product anilines are found. Presumably, the product anilines are unable to compete with
ammonia in forming the necessary complexes with copper ion due to the high effective
concentration (35.5M at 25 oC) and good nucleophilic properties of liquid ammonia.

1mol% Cu(I)

LNH3,25°C
18hrs
Scheme 5.1

Table 5.1 Amination of a variety of aryl iodides with copper (I) iodide (1mol%) and ascorbic acid (1mol%) in
liquid ammonia at room temperaturea

entry substrate product Yield(%)b

1 97.3(95c)

2 94.5

3 97.2

4 99.0

5 97.8(96c)

6 98.0

7 90.5

8 97.6

173
 Results and Discussion

entry substrate product Yield(%)b

9 96.8 (94c)

10 96.5

11 96.0

12 98.6

13 77.4 (93.6e)

14 57.2

15 99.3

16 98.2

a
General conditions except otherwise noted: 1mmol iodobenzenes, 1mol% CuI, 1eq. ascorbic acid to the catalyst
and 1ml liquid ammonia at 25o C for 18 hours. b GC or HPLC yield unless otherwise noted. c Isolated yields. e 36
hours.

Upon optimisation of the reaction conditions, we were pleased to find that the copper (I)
catalysed amination of iodobenzene in liquid ammonia at room temperature requires only 1
mol% copper catalyst, which is much lower than that generally reported.291 Also the reaction
does not require the presence of a strong base, such as tert-butoxide, which is widely used in
similar reactions in other conventional solvents. Furthermore, the product separation is much
easier with liquid ammonia as reaction medium, as the product can be simply collected by the
evaporation of ammonia.

It appears that electron-withdrawing groups increase the yields of the amination of aryl iodides
in liquid ammonia (Table 5.1, entries 4, 5, 6, 8 and 10), although given the high yields and
fixed reaction time, this data is not a clear indicator of the effect of substituents. The reaction
yields also seem to be independent of the position of the substituent to the iodo group, for
example, 3-nitro and 4-nitroiodobenzne give similar yields under same reaction conditions

174
 Results and Discussion

(Table 5.1, entries 4 and 5). Some heterocyclic iodides can also be converted to corresponding
amines with excellent yields (Table 5.1, entries 9 and 15). Furthermore, under the present
conditions, amination occurs preferentially with iodide as the leaving group compared with
other aryl halides (Table 5.1, entries 3, 12 and 16). Lower yields are observed when the ring
substituents of aryl iodides are hydroxyl (Table 5.1, entry 14) or amino group, although the
latter gives an excellent yield after a prolonged reaction time (Table 5.1, entry 13). Currently,
the palladium catalysed amination of aryl halides that contain free N-H or O-H bonds still
remain problematic.300b To our knowledge, there has been no previous report on the metal
catalysed amination of iodophenols, although Buchwald has reported a yield of 62% for the
copper (I) catalysed (5mol%) amination of 4-iodoaniline by benzylamine in isopropanol, after
38 hours at 90 oC.295a

Table 5.2 Amination of iodobenzene with various copper salts as catalyst in liquid ammonia at 20 oCa

entry copper (I) salt yields(%)b

1 CuI (99.999%) 97.3

2 CuCl (99%) 97.3

3 Cu(OAc) (97%) 97.6

4 Cu(CH3CN)4BF4 (97%) 98.5

entry copper (II) salt yields(%)b

5 Cu(OAc)2 (98%)c trace

6 Cu(OAc)2 (98%)d 95.2

entry copper (0) yields(%)b

7 copper powder 88.4

a
Reaction conditions: 1mmol iodobenzene, 1 mol% copper catalyst, 1eq. ascorbic acid to the catalyst and 1ml
liquid ammonia, 18 hours. b GC yields. c In the absence of ascorbic acid. d 2eq. Ascorbic acid to the catalyst.

Different copper (I) salts can be used to catalyse the amination of iodobenzene in liquid
ammonia, and the yields of the reaction appear to be independent of the source of copper (I)
(Table 5.2, entries 1 to 4). Copper (II) shows no catalytic activity in liquid ammonia, but can
be active in the presence of 2 equivalents of ascorbic acid (Table 5.2, entries 5 and 6).
Presumably, copper (II) is reduced to an active copper (I) species by the ascorbic acid.292 The
stability of Cu (I) relative to Cu(II) is very dependent on solvent and so, for example, Cu(I) is

175
 Results and Discussion

more stable than Cu(II) in acetonitrile and the tetrahedral ion [Cu(MeCN) 4]+ is well
established (Table 5.2 entry 4). Surprisingly, under the same conditions, copper (0) shows
reasonable catalytic ability, even though the reaction conditions are apparently heterogeneous
in liquid ammonia (Table 5.2, entry 7).

Normally, 293 copper catalysed C-N bond formations involve ancillary ligands, such as 1,2-
diamines (L1 and L2), 294 1,2-diols (L3-L5), 290e,295amino acids (L6 and L7), 296 2,2‟-biprydine
(L8)297 or 1,10-phenanthroline (L9),286c,298 which can effectively reduce the amount of copper
(I) catalyst required, and achieve benign reaction conditions in some conventionally used
solvents.

Consequently, we measured the kinetics of the copper (I) catalysed amination of iodobenzene
in liquid ammonia at 25 oC, with and without some of these ancillary ligands (L1-L9) (Table
5.3). There is no nucleophilic displacement by any of the ligands (L1 to L9) in the copper (I)
catalysed reaction of iodobenzene in liquid ammonia at room temperature, also the addition of
nucleophiles, such as morpholine and phenoxide ion, give no substitution products other than
aniline. The reaction rates are significantly hampered by added amino acids (L6 and L7) as
ligands (Table 5.3, entries 8 and 9) which is very different from that observed in other
conventional solvents, 296 while 1,2-diamines (L1 and L2), 1,2-diols (L3 and L4), bipyridine
(L8) and phenanthroline (L9) slightly enhanced the reaction rate.

L1 L2 L3 L4 L5

L6 L7 L8 L9

Only L5 leads to a marked rate enhancement (Table 5.3, entry 7). Given that 1,2-diols do not
significantly increase the activity of the catalyst, presumably the reaction rate increase in the
presence of ascorbic acid is due to its ability to keep copper (I) in its reduced state and its
concentration constant. The use of ascorbic acid or its sodium salt as an additive in copper

176
 Results and Discussion

catalysed organic reactions is known, and its function is to retain the catalytic ability of copper
(I) by providing a redox system which effectively reduces copper (II) or (III) back to copper
(I).299 The amino acids alanine (L6) and proline (L7) inhibit the reaction (Table 5.3, entries 8
and 9), but there was no sign of precipitation although the reaction solutions turned blue
indicative of oxidation of Cu (I) to Cu (II). The inhibitory effect of the amino acids may
simply be due to stabilisation of Cu (II) by the ligands and reduction of the electrode potential
(E0).

Table 5.3 The pseudo first order rate constant of copper (I) catalysed amination of iodobenzene with ligands in
liquid ammonia at 25 oCa

entry Cu(I) source ligands 105kobs/s-1

__
1 CuI 3.26b
__
2 Cu(CH3CN)4BF4 3.63

3 CuI L1 4.32

4 CuI L2 3.95

5 CuI L3 4.00

6 CuI L4 4.92

7 CuI L5 19.6b

8 CuI L6 0.037

9 CuI L7 0.025

10 CuI L8 5.23

11 CuI L9 4.91
a
Kinetics conditions: 50mM iodobenzene, 10mM Cu (I) salt, 10mM ligand, 40mg biphenyl as internal standard,
10ml LNH3 at 25 oC. bAverage of 2 runs.

In an attempt to distinguish between the role of ascorbic acid as a ligand and as a general
reductant, the kinetics of the reaction were determined as a function of the concentration of
ascorbic acid. With a constant copper (I) catalyst concentration, the rate of the amination of
iodobenzene significantly increases with added ascorbic acid in liquid ammonia, but when the
concentration of ascorbic acid reaches about twice of catalyst concentration, the rate levels out
and becomes independent of further added ascorbic acid (L5) and (Table 5.4, Figure 5.1).

177
 Results and Discussion

This rate saturation phenomenom implies some sort of association between copper (I) and
ascorbic acid – in particular one or two moles of ascorbate may be complexed with copper (I).
These observations also indicate the optimum ratio of ascorbic acid to copper (I) that is
required for a relatively fast reaction rate. However, the rate enhancement is not great (ca. 5-
fold) and rather than complex formation with Cu (I), the rate of ascorbic acid may simply be to
reduce any copper (II) formed back to active copper (I).

Table 5.4 The effect of ascorbic acid concentration (L5) on the rate of the copper (I) CuI catalysed amination of
iodobenzene in liquid ammonia at 25 oC.

[Iodobenzene]m/M [CuI]/mM [L5]/mM 104kobs/s-1

50 10 0 0.326

50 10 10 1.96

50 10 20 2.15

50 10 40 2.22

50 10 50 2.23

2.0
104kobs/s-1

1.0

0.0
0 0.01 0.02 0.03 0.04 0.05 0.06
[L5]/M

Figure 5.1 The effect of ascorbic acid concentration (L5) on the rate of the copper (I) CuI
(10mM) catalysed amination of iodobenzene in liquid ammonia at 25 oC.

178
 Results and Discussion

Further investigations of the kinetics of the amination of iodobenzene in liquid ammonia

indicates that the rate of the reaction shows a first order dependence on the concentration of
the copper (I) salt, both at lower and higher catalyst concentrations relative to the iodobenzene
concentration (Table 5.5, Figure 5.2). This is consistent with a previous study of the copper
(I) catalysed amidation of aryl iodides.300

Table 5.5 The dependence of the pseudo first order rate constants for the amination of iodobenzene on the copper
(I) catalyst concentration in liquid ammonia at 25 oC

a
[CuI]/mM 104kobs/s-1 b
[CuI]/M 104kobs/s-1

5 1.25 25 5.50

10 2.23 40 8.64

20 3.82 50 10.7

40 9.60 70 13.1
a b
50mM iodobenzene, 50mM ascorbic acid (L5) and various CuI concentration. 10mM iodobenzene, 50mM
ascorbic acid (L5) and various CuI concentration.

16.0

10mM iodobenzene, 25-70mM

CuI, fixed 50mM ascorbic acid.
12.0
104kobs/s-1

8.0

4.0 50mM iodobenzene, 5-40mM

CuI, fixed 50mM ascorbic acid.

0.0
0.00 0.02 0.04 0.06 0.08
[CuI]/M

Figure 5.2 The first order dependence of kobs of the amination of iodobenzene on copper (I)
concentration in liquid ammonia at 25 oC

179
 Results and Discussion

The first order rate constants for the copper (I) catalysed amination of 4-substituted phenyl
iodides in liquid ammonia increase with electron withdrawing substituents (Table 5.6) and
generate a Hammett ρ value of 0.49 (Figure 5.3). The small positive value of ρ indicates the
generation of negative charge in the aryl ring in the transition state.

Table 5.6 The rate of copper (I) catalysed amination of 4-substituted phenyl iodide in liquid ammonia at 25 oCa

substituent σp 104kobs(s-1)
4-MeO -0.27 0.83
4-Me -0.17 1.12
4-H 0 1.25
4-Cl 0.23 1.55
4-NO2 0.78 2.95
a
Reaction conditions: 50mM 4-substituted phenyl iodide, 5 mM CuI and 5mM ascorbic acid (L5) in 10ml liquid
ammonia.

-3.4

-3.6 ρ = 0.49
log10(kobs/s-1)

-3.8

-4.0

-4.2
-0.4 -0.2 0.0 0.2 0.4 0.6 0.8 1.0
σp

Figure 5.3 The Hammett plot for the copper (I) catalysed amination of iodobenzene in liquid
ammonia at 25 oC. Reaction conditions: 0.05M iodobenzene, 5mM CuI and 5mM ascorbic
acid in 10ml liquid ammonia.

180
 Results and Discussion

The relatively small Hammett ρ = 0.49 contrasts with the much larger ρ = 2.3 for the
analogous reactions catalysed by palladium (0) with phosphine ligands in toluene 301
suggesting that the C-I bond is not significantly broken in the transition state for the copper (I)
catalysed amination of 4-substituted phenyl iodides in liquid ammonia.

The activation parameters provide a further characterisation of the copper (I) catalysed
amination of aryl iodides in liquid ammonia. Linear Erying type plot (Table 5.7, Figure 5.4)
gives ΔH‡ = 65.6 kJ mol-1, ΔS‡ = -101.0 J K-1 mole-1 and ΔG‡ = 95.7 kJ mol-1. The entropy of
activation is not as negative as one would expect for a termolecular reaction or one generating
a charged transition state requiring heavy solvation.

-7

-8
ln(kobs/s-1)

-9

-10

-11
3.1 3.2 3.3 3.4 3.5
1000/T(K-1)

Figure 5.4 The Erying plot for the copper (I) catalysed amination of iodobenzene in liquid
ammonia. Reaction conditions: 50mM iodobenzene, 5mM CuI and 5mM ascorbic acid in
10ml liquid ammonia.

Iodide anion is better solvated by liquid ammonia than the other halides, as showed by its
relative small positive Gibbs transfer energy from water to ammonia, compared those larger

181
 Results and Discussion

values for chloride and bromide anions. 302 Therefore it is less likely that the expulsion of
iodide anion from the active complex would be the rate-limiting step as occurs with the
nucleophilic substitution of nitrofluorobenzenes by phenoxide anions in liquid ammonia
(chapter 4).

Table 5.7 The rate for the copper (I) catalysed amination of iodobenzene in liquid ammonia at various
temperaturea

temperature(K) 104kobs(s-1)
318.2 6.11
308.2 2.35
298.2 1.25
288.2 0.39
a
50mM iodobenzene, 5mM CuI and 5mM ascorbic acid (L5) in 10ml liquid ammonia.

To understand the mechanism of these catalysed reactions, it is important to know the active
coordination state of copper (I) under our reaction conditions. The rate of the reaction appears
generally insensitive towards several potential copper (I) ligands, except for the apparent
inhibitory effects of amino acids, and the enhancement brought about by ascorbic acid (Table
5.3). With the large and effectively constant concentration of ammonia (35.5M, 298.2 K),
copper (I) ion is presumably coordinated to ammonia and the added ligands do not compete
with ammonia to form complexes with copper (I) cation.

63
Based on EXAFS and Cu NMR spectroscopy it is claimed that copper (I) ion is three-
coordinated in liquid ammonia, [Cu(NH3)3]+, possibly with a coplanar trigonal geometry. By
contrast, in aqueous ammonia solution, the linear diamminecopper (I) complex, [Cu(NH3)2]+,
is the dominant species relative to [Cu(NH3)3]+,303 and even in highly concentrated aqueous
ammonia solution, only 27% of copper (I) ion adopts three-coordinated coplanar
structure.304,305 Copper (I) ion undergoes disproportionation (Scheme 5.2) in liquid ammonia
but with a small disproportion constant KD = 0.044M-1, the copper (I)-ammonia complex is the
dominant species (>99%) at low concentration of copper (I) liquid ammonia solution (<
0.1M).305

182
 Results and Discussion

KD=0.044 M -1
2 Cu+ Cu Cu2+
LNH3

Scheme 5.2

Copper (I) catalysed amination of iodobenzene in concentrated aqueous ammonia (30% w/w),
at room temperature is much slower than that in liquid ammonia at 25 oC.306 The pseudo first
order rate constant, kobs = 5.0× 10-6s-1 for the copper (I) catalysed amination of iodobenzene in
10ml 30% w/w aqueous ammonia at 25 oC, using 0.5mmol iodobenzene, 1%mole CuI, 1
equivalent ascorbic acid to the copper catalyst. This 40 fold smaller than the analogous
reaction under similar conditions in liquid ammonia (Table 5.3 entry 7). Although the
triamminecopper (I) is apparently the dominant species in liquid ammonia whereas the
diamminecopper (I) is the major complex in aqueous ammonia, it is not known which has the
greater catalytic activity.

In principle Cu+ can act as either an electron donor or acceptor, but based on the importance of
the presence of ascorbic acid, it seems likely that in liquid ammonia Cu + is acting as a
reducing agent and so is itself oxidised to a state that at the end of the reaction requires
reduction for catalysis to continue. Theoretical studies on the cation-π interactions of Cu+ and
benzene indicate that, in the gas phase, Cu + forms a η2 complex with benzene, especially if a
counter-ion is present. 307 Another theoretical study300,301, 308 on the copper (I) catalysed
amidation of aryl halides, suggested a mechanism involving rate-limiting oxidative addition
through a penta-coordinated copper (III) intermediate.

oxidative

dissociation

Scheme 5.3

183
 Results and Discussion

A traditional view of this aromatic nucleophilic substitution reaction would regard the likely
mechanism to involve initial copper (I)-π complex formation to be followed by dissociative or
addition-elimination steps (Scheme 5.3). The latter seems intrinsically unlikely given the
relative weak Lewis acid character of copper (I) and its ability to stabilise the negative charge
on the aromatic ring. However, the dissociation of iodide ion effectively leaves what would
normally be an unstable aromatic carbocation, but in this instance one which could be
stabilised by electron transfers from Cu (I) to form a Cu (II) complex.

Comparison of previous mechanistic studies300,301,309 on the copper (I) catalysed amidation of

aryl halides and current kinetic data, a possible mechanism for the amination of aryl halides is
proposed as in Scheme 5.3. The formation of triamminecopper (I)-aryl ring complex is rate
determining step, small Hammett ρ value and activation parameters support this assumption.

The difference between this mechanism (Scheme 5.3) and the type usually proposed for Pd (0)
catalysed reactions are the timing of bond making and breaking. If Cu (I) is regarded as a
nucleophile then catalysis could occur by a „normal‟ type of addition step to generate a
Meisenheimer type of -complex with negative charge distributed around the aromatic ring
with copper now in the +2 oxidation state, followed by cleavage of the bond to the leaving
group and then a repeat process involving nucleophilic displacement of the Cu2+ by reductive
elimination and regeneration of the Cu (I) catalyst (Scheme 5.4).

oxidative
-
addition

reductive
-
elimination

Scheme 5.4

184
 Results and Discussion

In liquid ammonia, the product of iodide expulsion from the Meisenheimer type intermediate
may well be strongly bound to copper and is represented as a possible trigonal bipyramidal
intermediate. The relatively small Hammett ρ = 0.49 suggests that the C-I bond is not
significantly broken in the transition state and that there is a small generation of negative
charge in the aryl ring which is compatible with the oxidative addition of the copper ion being
rate limiting.

5.1.2 Copper (I) catalysed amination of aryl bromides and chlorides (Scheme 5.5)

1mol% Cu(I)
X = Br or Cl
LNH3,100°C
18 to 35hrs

Scheme 5.5

With 1 mol% CuI as catalyst in liquid ammonia at room temperature, less than 1% of
bromobenzene is converted into aniline after 18 hours, however, under the elevated
temperature, for example, 100 oC, the yield of the reaction increases to 96% after 18 hours
(Table 5.6, entry 1) and the other substituted aryl bromides also can be smoothly converted to
the corresponding anilines with good to excellent yields in liquid ammonia (Table 5.6).

Using the copper (0) powder as a catalyst for the amination of bromobenzene in aqueous
ammonia it was reported that 5mol% of copper (0) was required at 100 oC for 24 hours and
gives a lower yield of aniline (85%), compared with our method. 310 Some bromopyridine
derivatives give excellent yields in liquid ammonia (Table 5.6, entries 12 and 13). 4-t-Butyl
phenyl bromide gives a slightly lower yield (Table 5.6, entry 10), but is still compatible with
that of palladium (II), (Pd[P(o-tol)3]2) catalysed amination with ammonia in 1,4-dioxane (65
o
C, 15 hours, 88% yield).290b Interestingly, the amination of 3-fluoro-4-bromotoluene occurs
only to displace bromide ion (Table 5.6, entry 13).

No diaryl or triaryl amine derivatives, which can occur as a result of further reaction of the
product, are found under elevated temperatures for the copper (I) catalysed amination of aryl
bromides in liquid ammonia. By contrast, these di and triaryl amines by-products are often

185
 Results and Discussion

unavoidable when the strong bases are utilised for the palladium catalysed amination of aryl
bromides or chloride in some conventional solvents, especially under high temperature.288c,290a
At a similar temperature, the yields of the copper (I) catalysed amination of aryl bromides in
liquid ammonia are generally higher than those previously reported by using aqueous
ammonia solution or ammonia surrogates as nitrogen source. Furthermore, the amination of
aryl bromides requires only 1% copper (I) catalyst in liquid ammonia, which is much lower
than those demanded in conventionally used solvents.

Table 5.8 Amination of variety of aryl bromides and chlorides with copper (I) iodide catalyst in liquid ammonia
at elevated temperature in liquid ammoniaa

entry substrate product yield(%)b

1 96

2 95

3 97

4 94

5 86

6 91

7 80

8 96

9 97

10
85

186
 Results and Discussion

entry substrate product yield(%)b

11 99

12 96

13 92

14 5<

15 74c

16 88c

17 63c

a
General reaction conditions unless otherwise noted: 1mmol aryl halides, 1 mole % CuI, 1eq. ascorbic acid to the
copper catalyst, in 1-1.5ml liquid ammonia at 100 oC for 18 hrs. b GC yields, except otherwise noted. c 36 hours,
isolated yields.

Unactivated chlorobenzenes are very inert under similar conditions to that used for the
amination of aryl bromides in liquid ammonia (Table 5.8, entry 5). For example, even at 120
o
C, with 10mol% copper (I) catalyst, less than 5% chlorobenzene is converted into aniline
after 12 hours in liquid ammonia. This is as expected, based on the large dissociation energy
of a sp2 carbon-chlorine bond.311 However, electron-withdrawing groups are activating so that
nitrochlorobenzenes can be smoothly converted into corresponding aniline with only 1mol%
copper (I) catalyst to give moderate yields in liquid ammonia (Table 5.8, entries 15 to 17).

In summary, we have developed a method for the amination of aryl halides in liquid ammonia
using copper (I) catalysis which enables direct synthesis of a number of primary amines with
excellent yields. This method does not require strong base and ligands as additives, which are
often used for the similar metal catalysed amination of aryl halides in the conventionally used
solvents. Most importantly, the amination in liquid ammonia has exclusive selectivity for the
formation of primary amines, even under relative higher temperature. The amount of catalyst
required for the reaction is relatively lower than that generally used, and the convenience of

187
 Results and Discussion

products separation with liquid ammonia as reaction medium indicate its potential industrial
application. The preliminary mechanistic investigation indicates that the rate of the amination
is first order dependence on the concentration of copper (I) catalyst, and the formation of
triamminecopper (I)-aryl ring intermediate is probably the rate limiting step in liquid
ammonia. Due to strong coordination of solvent molecules to the copper (I) ion, the kinetics of
the reaction are generally insensitive to the addition of other conventional ligands in liquid
ammonia.

5.2 Copper (I) catalysed azide-alkyne cycloaddition (CuIAAC)

The 1,3-dipolar cycloaddition between azide and alkyne [3+2] was first discovered in 1893 by
Michael and 70 years later thoroughly developed by Huisgen (and subsequently widely
recognised as Huisgen 1,3-dipolar cycloadditions). 312 The reaction has a high activation
barrier and consequently demanding reaction conditions are usually required, such as elevated
temperature and pressure, and, in addition, the reaction gives a mixture of 1,4- and 1,5-
substituted 1,2,3-triazoles (36 and 37, respectively, Scheme 5.6).

Huisgen cycloaddition

36 37

R1,R2 = Alkyl, aryl CuAAC reaction

36

Scheme 5.6

In 2002, Sharpless313 and Meldal314 independently improved the 1,3-Huisgen cycloaddition by

using copper catalysts and successfully achieved amiable reaction conditions and exquisite
regioselectivity. This variation of the Huisgen cycloaddition, was later described as CuAAC
reactions standing for Cu-catalysed Azide-Alkyne Cycloaddition, or more commonly as “click
chemistry”, 315 which proceeds under ambient temperature and pressure to give exclusively
1,4-substituted 1,2,3-triazoles (36) (Scheme 5.6). This method has become a powerful and
reliable tool for the synthesis of 1,2,3-triazole derivatives which are widely used in

188
 Results and Discussion

pharmaceutical and agrochemical industry. 316 Recently, copper-free click chemistry between
some strained alkynes, such as cyclooctynes, and azides has been reported.317

CuAAC reactions can be carried out in a variety of solvents, such as aqueous alcohol solvents,
pure water, dipolar aprotic solvents or their aqueous solutions, but are often carried out under
a basic atmosphere or, a strong base, such as Et3N or DIPEA (N,N-diisopropylethylamine),
which is required to facilitate the deprotonation of the terminal alkynes.

Liquid ammonia is a basic solvent and we have shown in this thesis that it behaves like a
typical dipolar aprotic solvent in its solvent effects on organic reactions. One of the potential
advantages of using liquid ammonia as reaction medium in the chemical industry is relative
ease in separating the solvent from the products by vapourisation of ammonia.

When phenyl acetylene and benzyl azide are charged together in liquid ammonia at room
temperature without a copper catalyst, there is no reaction even after several days.
Furthermore, in the absence of a copper (I) catalyst, even at elevated temperatures (100 oC),
and after 10 hours agitation, less than 20% of the starting material is converted into a mixture
of 1,4- and 1,5-disubstituted 1,2,3-triazoles (38 and 39, respectively) (Scheme 5.7) in a molar
ratio of 2:1. However, the copper (I) catalysed 1,3-dipolar cycloaddition of phenyl acetylene
and benzyl azide occurs smoothly at ambient temperature in liquid ammonia to give the
regioselective 1,4-disubstituted product (38). This „click reaction‟ in liquid ammonia requires
only 0.5 mol% of copper (I) catalyst and the yields of these reactions are extraordinarily high.
The amount of copper catalyst required for the CuIAAC reactions in liquid ammonia is much
lower than those previously reported in conventionally used solvents318 and both aromatic and
aliphatic alkynes give excellent yields (Table 5.9). In addition, no acetylene dimer (RC≡C)2
is found for the CuIAAC reactions in liquid ammonia, which is often observed when the
reaction is performed in some conventionally used solvents. The 1H NMR spectra of the
products for the equi molar reaction of azides and acetylenes in liquid ammonia show that
pure 1,4-disubstituted 1,2,3-triazoles (38) are obtained quantitatively by the vapourisation of
ammonia after the reactions is complete, and there is no need for the further purification or
additional separation procedures (Appendix C, Figures N30).

189
 Results and Discussion

LNH3, 100°C
10 hrs
38 39
<20%
LNH3, 25°C, 12hrs

0.5mol% CuI, 1eq. ascobic acid

38
~100%

Scheme 5.7

Table 5.9 Copper (I) catalysed „click reactions‟ of a variety of azides and acetylenes in liquid ammonia at room
temperaturea

entry acetylene azide product(37)b yieldc

1 98

2 99

3 97

4 99

5 98

a
General conditions unless otherwise noted are: 1mmol starting azides and 1mmol acetylenes, 0.5 mol% CuI and
1.0 eq. ascorbic acid to the catalyst, 1ml liquid ammonia, at 20 oC for 10hrs. b Products were confirmed by GC-
MS and 1H NMR. c. Isolated yields.

A variety of copper (I) salts can be used as catalyst for the click reaction in liquid ammonia
(Table 5.10). In addition, copper (0) powder can also act as the catalyst, although this appears
to be under heterogeneous conditions and gives a much lower yield (Table 5.10, entry 6), it is
possible that some of the copper(0) dissolves in liquid ammonia to give ammonia solvated
copper(I).

190
 Results and Discussion

Table 5.10 Various copper salts catalysed „click reactions‟ in liquid ammonia at room temperaturea

entry copper source yields of 38(%)b

1 CuI(99.999%) >99

2 CuCl(99%) >99

3 Cu(OAc)(97%) >99

4 Cu(CH3CN)4BF4 (97%) >99

5 Cu(OAc)2(98%)c >99(40d)

6 copper powder 36.9

a
General condition unless otherwise noted are: 1mmol phenyl azide, 1mmol phenyl acetylene, 0.5mol% copper
catalyst, 1eq. ascorbic acid to the copper catalyst, 1ml liquid ammonia for 12 hrs, the product precipitated as
white short crystal in needle form. b GC yields. c 2eq. ascorbic acid to the copper catalyst. d Without ascorbic
acid.

1.40

1.20

1.00
104kobs/s-1

0.80

0.60

0.40

0.20

0.00
0 50 100 150 200
Cu0 mass / mg

Figure 5.5 The observed pseudo first order rate constant as a function of the mass of copper
(0) powder added for the CuAAC reaction in liquid ammonia. (Reaction conditions: 1mmol
phenyl azide, 1mmol phenyl acetylene in 10ml liquid ammonia at 25 oC)

191
 Results and Discussion

The rate of the copper (0) catalysed „click reaction‟ in liquid ammonia is proportional to the
amount of copper powder added, which is compatible with either some copper(0) dissolving or
possibly the reaction occurring on the surface of the copper (0) catalyst in liquid ammonia
(Figure 5.5, Appendix A, Table A44 ).

Generally, copper (II) salts that are used as the catalyst for CuAAC reactions are often done in
aqueous binary solvents, such as in t-butyl alcohol-water system. This process often requires
additional ancillary ligands to avoid the dimerisation of activated acetylene-copper complexes
which occurs under air. In liquid ammonia, with 2 equivalent amount of ascorbic acid, Cu (II)
catalysed CuAAC reaction does not need ancillary ligands and yields the same product as the
CuIAAC reaction (Table 5.10, entry 5). Presumably, copper (II) is reduced to copper (I) by
ascorbic acid in liquid ammonia, as in the absence of ascorbic acid, copper (II) catalysed „click
reactions‟ are slow and the starting materials are partially converted to the 1,4-disubstituted-
1,2,3-triazole in liquid ammonia at room temperature, together with significant amounts of the
acetylene dimer (RC≡C)2.

Some preliminary mechanistic investigations of CuIAAC reaction of benzyl azide and phenyl
acetylene in liquid ammonia show that the observed pseudo first order rate constant for the
reaction has an apparent second order dependence on the copper (I) concentration (Table
5.11).

192
 Results and Discussion

50

40
104kobs/s-1

30

20

10

0
0 4 8 12 16
[CuI]/mM

Figure 5.5a The dependence of the observed rate constant on the concentration of copper (I)
catalyst for the CuIAAC reaction in liquid ammonia at 25 oC (Reaction conditions: 50mM
phenyl azide and 50mM phenyl acetylene, fixed ascorbic acid concentration at 15mM, copper
iodide concentration varies from 2.5-15mM)

The nonlinear relationship between kobs and [CuI] in Figures 5.5a indicates that the reaction in
liquid ammonia has a greater than first order dependence on the catalyst concentration.
Therefore the observed pseudo first order rate constant (k obs) could possibly be expressed by
Equation 5.1.

kobs = k1[CuI] + k2[CuI]2

Thus, kobs/[CuI] = k1 + k2[CuI]

Equation 5.1

The plot of kobs/[CuI] vs. [CuI] shows that there is no significant intercept which indicates
there is no significant first order term in [CuI] in the rate law. The slope of the plot gives the
apparent second order rate constant k2 = 17.4M-1s-1 (Figures 5.5b).

193
 Results and Discussion

3.0

2.5
y = 0.1738x + 0.0036
10kobs/[CuI](M-1s-1) R² = 0.9735
2.0

1.5
k2 = 17.4M-1s-1

1.0

0.5

0.0
0.0 5.0 10.0 15.0 20.0
[CuI]/mM

Figure 5.5b The dependence of the observed rate constant on the concentration of copper (I)
catalyst for the CuIAAC reaction in liquid ammonia at 25 oC (Reaction conditions: 50mM
phenyl azide and 50mM phenyl acetylene, fixed ascorbic acid concentration at 15mM, copper
iodide concentration varies from 2.5-15mM)

Table 5.11 The dependence of the observed rate constant on the concentration of copper (I) catalyst for the
CuIAAC reaction in liquid ammonia at 25 oCa

[CuI]/mM 104kobs./s-1

2.5 1.30
5 4.33
10 15.2
15 41.2
a
Reaction conditions: 50mM phenyl azide and 50mM phenyl acetylene, fixed ascorbic acid concentration at
15mM, copper iodide concentration was varied from 2.5-15mM.

The azide only „clicks‟ with terminal (CH) acetylene groups and there is no CuAAC reaction
between internal alkynes and azide, 319 so, for example, no reaction is observed between

194
 Results and Discussion

diphenylacetylene (PhC≡CPh) or 2-iodoethynylbenzene (PhC≡C_I) with benzyl azide in the

presence of a copper (I) catalyst in liquid ammonia at room temperature. These observations
suggest that the CuAAC reaction in liquid ammonia is a stepwise process involving two metal
centres for the cycloaddition of azides and acetylenes, which is consistent with previous
mechanistic studies of CuAAC in some conventional solvents.320

As described earlier in the copper (I) catalysed amination section, copper (I) salts mainly exist
as tri-coordinated coplanar species, [Cu(NH3)3]+, in liquid ammonia.305 Acetylenes normally
have pKa range from 20 to 25 in water,9 and are neutral in liquid ammonia at room
temperature, as shown by the 1H NMR spectrum of phenylacetylene in liquid ammonia
(Appendix C, Figure N31). However, the theoretical calculations show that copper (I)
coordination to the acetylene increases its acidity by up to 1010-fold in water.320a Compared
with other solvents, the basic nature of liquid ammonia facilitates the deprotonation of
acetylene in the presence of copper (I) ions. Therefore, the formation of a coordinated copper
(I)-acetylide ion complex in liquid ammonia could enhance the rate of reaction. However, it is
worth noting that currently the exact structure of the copper-acetylene complex in liquid
ammonia is still not clear, and we are not sure whether this complex exists in a copper
monomer (E) or dimer form (F) which is crucial for the elucidation of the detailed reaction
mechanism in liquid ammonia.

As stated earlier, the kinetic studies show that the CuIAAC process in liquid ammonia requires
two copper (I) ions, which implies that one of these coordinated with the azide and the other to
the acetylene. The reaction could then proceed via a six- or seven-membered copper-
containing intermediate (H and G, respectively) to give a triazole-copper derivate I, which is
followed by the fast protonation of I by ammonium ion to give the product 36 and the
regenerated copper (I) catalyst (Scheme 5.8).

195
 Results and Discussion

-NH4
NH3
NH3

or

E F
R2N3
NH4
36

_ or _

or

G H

Scheme 5.8
In order to prove the existence of copper-acetylene complexes (E or F) in the CuIAAC
reactions, deuteriated phenylacetylene (PhC≡C_D)321 is used to „click‟ with benzyl azide in
liquid ammonia under the same conditions as introduced in Table 5.9. The 1H NMR spectrum
of the reaction product shows that there is no deuterium in the 1,4-disubstituted-1,2,3-triazole.
In addition, the NMR results also show that there is no deuterium exchange between product
(38) and D2O, together with 0.5% CuI and 1eq. ascorbic acid to the copper catalyst, at room
temperature for overnight (Scheme 5.9).

LNH3, 25°C, 12hrs

0.5mol% CuI, 1eq. ascobic acid

38

0.5mol% CuI, 1eq. ascobic acid

D2O, room temperature, overnight

38

Scheme 5.9

196
 Results and Discussion

These experimental observations described in Scheme 5.9 indicate that the acetylene is
deprotonated and copper atom substitutes the acetylene hydrogen to form E or F as the
possible intermediate for the CuIAAC reactions in liquid ammonia.

In conclusion, the copper (I) catalysed 1,3-Huisgen cycloaddition reaction of azide and
alkynes (CuIAAC) in liquid ammonia requires less catalyst than those in conventionally used
solvents. The excellent yield, exclusive selectivity, and most importantly, the ease of
separation of the product indicate the potential advantages of using liquid ammonia as the
solvent for this reaction. The preliminary mechanistic investigation suggests that CuIAAC
reaction in liquid ammonia is a stepwise process with the initial formation of copper (I)-
acetylide ion complex, followed by its combination with copper (I) coordinated azide.

197
 Appendix A: Tables and Figures

Table A1 The UV absorbance of 3-chlorophenol (10-4M) under various concentration of

KClO4 in LNH3 at room temperature

CKClO4(M) absorbance (317nm) absorbance (257nm)

0 0.0528 0.240

0.005 0.150 0.467

0.01 0.239 0.675

0.05 0.360 0.920

0.1 0.379 1.022

0.2 0.385 1.025

0.3 0.392 1.026

0.4 0.399 1.027

Figure A1 The UV absorbance of 3-chlorophenol (10-4M) under various concentration of

KClO4 (0.005 to 0.4M) in LNH3 at room temperature

198
 Appendix A: Tables and Figures

1.2
257nm
1

0.8
Absorbance

0.6

317nm
0.4

0.2

0
0 0.1 0.2 0.3 0.4 0.5

[KClO4]/M

Figure A2 Various concentration of KClO4 (0.005 to 0.4M) as salt effect on the ionisation
of 3-chlorphenol (10-4M) in LNH3 at room temperature

Table A2 The UV absorbance of 3-chlorophenol (10-4M) under various concentration of

NaCl in liquid ammonia at room temperature

CNaCl(M) absorbance (317nm) absorbance (257nm)

0.01 0.218 0.608

0.05 0.348 0.945

0.1 0.401 1.062

0.15 0.401 1.063

0.2 0.401 1.066

199
 Appendix A: Tables and Figures

Figure A3 The UV absorbance of 3-chlorophenol (10-4M) under various concentration of

NaCl (0.01 to 0.2M) in LNH3 at room temperature

1.2
257nm

0.8
Absorbance

0.6

317nm
0.4

0.2

0
0 0.05 0.1 0.15 0.2 0.25
[NaCl]/M
Figure A4 Various concentration of NaCl (0.01 to 0.2M) as salt effect on the ionisation of
3-chlorphenol (10-4M) in LNH3 at room temperature

200
 Appendix A: Tables and Figures

With 0.1M NaClO4

Figure A5a The UV absorbance of 2.5 × 10-4M 4-nitorphenol with and without salt effects
(0.1M NaClO4) in LNH3 at room temperature

201
 Appendix A: Tables and Figures

Table A3 Linear relationship of pKa of 4-methoxyphenol (3×10-4M) with the square root
of ionic strength (I, KClO4) in LNH3 at room temperature

I=CKClO4(M) I1/2(M1/2) Absorbance at max pKa

0 0.00 0.035 6.59

0.1 0.32 0.063 6.07

0.2 0.45 0.109 5.58

0.3 0.55 0.127 5.44

0.4 0.63 0.168 5.18

0.5 0.71 0.205 4.99

0.8 0.89 0.264 4.74

6.5
y = -2.190x + 6.622
R² = 0.9833
6
pKa

5.5

4.5

4
0.0 0.2 0.4 0.6 0.8 1.0

I1/2/M1/2
Figure A5 Linear relationship of pKa of 4-methoxyphenol (3×10-4M) with the square root
of ionic strength (I, KClO4) in LNH3 at room temperature

202
 Appendix A: Tables and Figures

Table A4 Linear relationship of pKa of phenol (3×10-4M) with the square root of ionic
strength (I, KClO4) in liquid ammonia at room temperature

I=CKClO4(M) I1/2(M1/2) Absorbance at max pKa

0 0.00 0.000 6.02

0.1 0.07 0.140 5.01

0.2 0.10 0.246 4.66

0.3 0.22 0.321 4.16

0.4 0.45 0.349 4.07

0.5 0.54 0.369 4.00

6.5

6
y = -2.996x + 6.023
R² = 0.9822
5.5
pKa

4.5

3.5
0 0.2 0.4 0.6 0.8
I1/2/M1/2

Figure A6 Linear relationship of pKa of phenol (3×10-4M) with the square root of ionic
strength (I, KClO4) in LNH3 at room temperature

203
 Appendix A: Tables and Figures

Table A5 Linear relationship of pKa of 1-naphthol (1×10-4M) with the square root of ionic
strength (I, KClO4) in liquid ammonia at room temperature

I=CKClO4(M) I1/2(M1/2) Absorbance at max pKa

0 0.00 0.152 6.02

0.005 0.32 0.288 5.01

0.01 0.45 0.431 4.66

0.05 0.55 0.681 4.16

0.2 0.63 0.756 4.07

0.3 0.71 0.833 4.00

5.5
y = -3.501x + 4.972
5 R² = 0.8628
pKa

4.5

3.5

2.5
0 0.1 0.2 0.3 0.4 0.5 0.6
I1/2/M1/2

Figure A7 Linear relationship of pKa of 1-naphthol (1×10-4M) with the square root of ionic
strength (I, KClO4) in LNH3 at room temperature

204
 Appendix A: Tables and Figures

Table A6 Linear relationship of pKa of 4-chlorophenol (1×10-4M) with the square root of
ionic strength (I, KClO4) in liquid ammonia at room temperature

I=CKClO4(M) I1/2(M1/2) Absorbance at max pKa

0 0.00 0.056 4.49

0.05 0.07 0.155 3.91

0.1 0.10 0.220 3.48

0.2 0.22 0.281 2.86

0.3 0.32 0.304 2.05

4.5
y = -4.344x + 4.691
4 R² = 0.9483

3.5
pKa

2.5

1.5
0 0.1 0.2 0.3 0.4 0.5 0.6
I1/2/M1/2

Figure A8 Linear relationship of pKa of 4-chlorophenol (1×10-4M) with the square root of
ionic strength (I, KClO4) in LNH3 at room temperature

205
 Appendix A: Tables and Figures

Table A7 Linear relationship of pKa of 3-chlorophenol (1×10-4M) with the square root of
ionic strength (I, KClO4) in liquid ammonia at room temperature

I=CKClO4(M) I1/2(M1/2) Absorbance at max pKa

0 0.00 0.053 4.56

0.005 0.07 0.150 4.12

0.01 0.10 0.239 3.72

0.05 0.22 0.360 2.87

0.1 0.32 0.379 2.56

4.5

y = -6.551x + 4.498
4
R² = 0.9752
pKa

3.5

2.5

2
0 0.1 0.2 0.3
I1/2/M1/2

Figure A9 Linear relationship of pKa of 3-chlorophenol (1×10-4M) with the square root of
ionic strength (I, KClO4) in LNH3 at room temperature

206
 Appendix A: Tables and Figures

Table A8 Linear relationship of pKa of 4-carbomethoxy phenol (5×10-5M) with the square
root of ionic strength (I, KClO4) in liquid ammonia at room temperature

I=CKClO4(M) I1/2(M1/2) Absorbance at max pKa

0 0.00 1.41 4.04

0.01 0.10 1.57 3.84

0.05 0.22 1.71 3.56

0.1 0.32 1.78 3.40

4.1

3.9
y = -2.057x + 4.039
3.8
R² = 0.9978

3.7
pKa

3.6

3.5

3.4

3.3
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35

I1/2/M1/2

Figure A10 Linear relationship of pKa of 4-carbomethoxy phenol (5×10-5M) with the
square root of ionic strength (I, KClO4) in LNH3 at room temperature

207
 Appendix A: Tables and Figures

Table A9 Linear relationship of pKa of 3-nitrophenol (1×10-4M) with the square root of
ionic strength (I, KClO4) in liquid ammonia at room temperature

I=CKClO4(M) I1/2(M1/2) Absorbance at max pKa

0 0.00 0.123 3.63

0.005 0.07 0.136 3.38

0.01 0.10 0.150 2.98

0.05 0.22 0.157 2.60

3.7

3.5
y = -4.697x + 3.611
3.3 R² = 0.9379

3.1
pKa

2.9

2.7

2.5

2.3
0 0.05 0.1 0.15 0.2 0.25

I1/2/M1/2

Figure A11 Linear relationship of pKa of 3-nitrophenol (1×10-4M) with the square root of
ionic strength (I, KClO4) in LNH3 at room temperature

208
 Appendix A: Tables and Figures

Table A10, Figure A12 Titration of 4-carbomethoxy phenol (5×10-5M) with NH4Cl salt in
liquid ammonia at room temperature

[NH4Cl]/M Absorbance
0 1.500

1.00×10-5 1.420

2.50×10-5 1.318

5.00×10-5 1.176

1.25×10-3 0.927

2.50×10-3 0.846

3.75×10-3 0.757

5.00×10-3 0.734

1.00×10-2 0.781

5.00×10-2 0.782

1.6

Kd = 0.65M-1; Ki =1.2×10-4M
1.4
pKa = -lg(Kd•Ki) = 4.1
Absorbance

1.2
Figure A12
1

0.8

0.6
0 0.01 0.02 0.03 0.04 0.05 0.06
NH4Cl concetration/M

209
 Appendix A: Tables and Figures

Table A11, Figure A13 Titration of 4-carbomethoxyphenol (5×10-5M) with NH4Cl salt
(I=0.2M, NaCl) in liquid ammonia at room temperature

[NH4Cl]/M [NaCl]/M Absorbance

0 0.20 1.89

0.01 0.19 1.41

0.03 0.17 1.29

0.04 0.16 1.10

0.05 0.15 1.03

0.10 0.10 0.914

0.15 0.05 0.872

0.20 0 0.874

2.1

1.9 Kd = 0.63M-1; Ki =5.0×10-2M

pKa = -lg(Kd•Ki) = 1.5

1.7
Absorbance

1.5
Figure A13
1.3

1.1

0.9

0.7
0 0.05 0.1 0.15 0.2 0.25
concentration of NH4Cl(I=0.2M, NaCl)/M

210
 Appendix A: Tables and Figures

Table A12, Figure A14 Titration of 4-nitrophenol (2.5×10-5M) with NH4Cl salt in liquid
ammonia at room temperature

[NH4Cl]/M Absorbance
0 0.942

0.01 0.889

0.02 0.875

0.05 0.815

0.10 0.796

0.50 0.792

1.00 0.790

2.00 0.799

0.95

Kd = 5.6M-1; Ki =1.5×10-2M

0.9 pKa = -lg(Kd•Ki) = 1.1

Absorbance

0.85 Figure A14

0.8

0.75
0 0.5 1 1.5 2 2.5
NH4Cl concetration/M

211
 Appendix A: Tables and Figures

Table A13, Figure A15 Titration of 4-nitrophenol (2.5×10-5M) with NH4Cl salt (I=0.2M,
NaCl) in liquid ammonia at room temperature

[NH4Cl]/M [NaCl]/M Absorbance

0 0.20 1.085

0.005 0.195 1.079

0.03 0.17 0.988

0.05 0.15 0.914

0.07 0.13 0.905

0.10 0.10 0.902

0.15 0.05 0.819

0.20 0 0.811

1.15

Kd = 1.9M-1; Ki =0.11M

pKa = -lg(Kd•Ki) = 0.68

1.05
Absorbance

0.95 Figure A15

0.85

0.75
0 0.05 0.1 0.15 0.2 0.25

concentration of NH4Cl(I=0.2M, NaCl)/M

212
 Appendix A: Tables and Figures

Table A14 Solvolysis rate of benzyl chloride in liquid ammonia at different temperature

temperature(K) 104kobs(s-1) 105k2(M-1s-1)

268.2 1.28 0.338

283.2 3.25 0.881
298.2 8.89 2.48
308.2 13.3 3.79

-10

y = -5.103x + 6.428
R² = 0.9968
-11
ln(k2/M-1s-1)

-12

-13
3.2 3.3 3.4 3.5 3.6 3.7 3.8

1000/T(K-1)

Figure A16 Erying plot for the solvolysis of benzyl chloride in LNH3

213
 Appendix A: Tables and Figures

Table A15 Solvolysis rate of 4-chlorobenzyl chloride in liquid ammonia at different

temperature

temperature(K) 104kobs(s-1) 105k2(M-1s-1)

268.2 1.48 0.391

283.2 4.36 1.18
298.2 9.81 2.74
308.2 16.8 4.78

-10

y = -5.136x + 6.734
R² = 0.9986
ln(k2/M-1s-1)

-11

-12

-13
3.2 3.4 3.6 3.8

1000/T(K-1)

Figure A17 Erying plot for the solvolysis of 4-chlorobenzyl chloride in LNH3

214
 Appendix A: Tables and Figures

Table A16 Solvolysis rate of 4-nitrobenzyl chloride in liquid ammonia at different

temperature

temperature(K) 104kobs(s-1) 105k2(M-1s-1)

268.2 2.42 0.636

283.2 6.71 1.82
298.2 15.3 4.27
308.2 22.9 6.52

-9.5

y = -4.841x + 6.129
R² = 0.9969

-10.5
ln(k2/M-1s-1)

-11.5

-12.5
3.2 3.4 3.6 3.8

1000/T(K-1)
Figure A18 Erying plot for the solvolysis of 4-nitrobenzyl chloride in LNH3

215
 Appendix A: Tables and Figures

Table A17 Solvolysis rate of 4-methoxybenzyl chloride in liquid ammonia at different

temperature

temperature(K) 104kobs(s-1) 105k2(M-1s-1)

268.2 2.72 0.717

283.2 7.99 2.16
298.2 19.1 5.33
308.2 32.3 9.23

-9

y = -5.269x + 7.828
R² = 0.9993

-10
ln(k2/M-1s-1)

-11

-12
3.2 3.4 3.6 3.8
1000/T(K-1)

Figure A19 Erying plot for the solvolysis of 4-methoxybenzyl chloride in LNH3

216
 Appendix A: Tables and Figures

Table A18 Solvolysis rate of α-methyl benzyl chloride in liquid ammonia at different
temperature

temperature(K) 105kobs(s-1) 107k2(M-1s-1)

298.2 0.671 1.87

308.2 1.62 4.53
318.2 3.98 11.1

-13

y = -8.443x + 12.816
R² = 0.9995
ln(k2/M-1s-1)

-14

-15

-16
3.1 3.2 3.3 3.4
1000/T(K-1)

Figure A20 Erying plot for the solvolysis of α-methyl benzyl chloride in LNH3

217
 Appendix A: Tables and Figures

Table A19 Salt effects on the solvolysis rates of 4-NFB and 2-NFB in LNH3

substrate temperature salt salt concentration/M 105kobs/s-1

4-NFB 20 NaNO3 0 0.79

0.5 1.32
1 1.55
2 2.32
3 3.17

4-NFB 20 NH4Cl 1 1.42

2 1.88
3 2.28

2-NFB 25 NaNO3 0 21.5

0.5 22.7
1 24.7
2 27.3

Table A20 Calculated ground state free energy of solvation for 4-NFB and 2-NFB in
different solventsa

computational water ethanol DMSO DMF

substrate method

HF/6-31+G* -15.97 -28.02 -24.52 -26.84

4-NFB
DFT/B3LYPb -16.88 -28.71 -24.80 -27.12

HF/6-31+G* -23.98 -35.40 -32.26 -34.47

2-NFB
DFT/B3LYPb -25.05 -36.13 -32.45 -34.66
a
Energy unit: kJ/mol. b basis set: 6-31+G*.

218
 Appendix A: Tables and Figures

Table A21 Solvolysis of 4-NFB in LNH3 at different temperature

T(K) kobs/s-1

293.2 7.86×10-6
333.2 1.24×10-4
353.2 3.90×10-4
373.2 1.51×10-3
393.2 2.06×10-3

-4

y = -6.6769x + 11.062
R² = 0.9909
-7
ln(kobs/s-1)

-10

-13
2.4 2.9 3.4

1000/T(K-1)

Figure A21 Erying plot for the solvolysis of 4-NFB in LNH3

219
 Appendix A: Tables and Figures

Table A22 Solvolysis of 2-NFB in LNH3 at different temperature

T(K) kobs/s-1

263.2 1.16×10-5
298.2 2.13×10-4
308.2 4.02×10-4
318.2 8.32×10-4

-6

y = -6.4715x + 13.224
R² = 0.9997
-8
ln(kobs/s-1)

-10

-12
3.1 3.5 3.9
1000/T(K-1)

Figure A22 Erying plot for the solvolysis of 2-NFB in LNH3

220
 Appendix A: Tables and Figures

Table A23 Solvolysis of 2,4-DFNB in LNH3 at different temperature

T(K) kobs/s-1

288.2 4.14×10-3
296.2 6.29×10-3
308.2 1.30×10-2
318.2 2.15×10-2

-3.5

-4 y = -5.1032x + 12.201
R² = 0.9986

-4.5
ln(kobs/s-1)

-5

-5.5

-6
3.1 3.2 3.3 3.4 3.5

1000/T(K-1)

Figure A23 Erying plot for the solvolysis of 2,4-DFNB in LNH3

221
 Appendix A: Tables and Figures

Table A24 The rate for benzyl chloride with different concentration of phenoxide in liquid
ammonia at 25 oC (I = 0.3 M, KClO4)

[phenoxide]/M [KClO4]/M 103kobs/s-1

0 0.3 1.18
0.05 0.25 2.41
0.1 0.2 3.23
0.2 0.1 4.38
0.3 0 5.30

Table A25 Second order rate constant for 4-substituted benzyl chloride with phenoxide
anion in liquid ammonia at 25 oC

substituent σp 102k2(M-1s-1)

4-MeO -0.27 4.02

4-Me -0.17 1.48
4-H 0 2.01
3-MeO 0.12 2.22

4-Cl 0.23 4.93

4-COOMe 0.45 5.28

4-CN 0.66 13.4

222
 Appendix A: Tables and Figures

Table A26 The rate for 4-methoxybenzyl chloride with different concentration of
phenoxide in liquid ammonia at 25 oC (I = 0.1 M, KClO4)

[PhO-]/M [KClO4]/M 103kobs/s-1

0 0.1 2.07
0.05 0.05 4.18
0.075 0.025 5.52
0.1 0 7.19

Table A27 The rate for the benzyl chloride with different concentration of morpholine in
liquid ammonia at 25 oC

[morpholine]/M 103kobs/s-1

0 8.89
0.1 1.30
0.2 1.60
0.4 2.38
0.75 3.62
1.0 4.18

Table A28 Second order rate constant for 4-substituted benzyl chloride with sodium
triazolate in liquid ammonia at 25 oC

substituent σp 102k2(M-1s-1)
4-MeO -0.27 0.917
4-Me -0.17 0.863
4-H 0 0.942
4-Cl 0.23 1.58
4-COOMe 0.45 1.70
4-NO2 0.78 6.59

223
 Appendix A: Tables and Figures

Table A29 Second order rate constant for 4-substituted benzyl chloride with piperidine in
liquid ammonia at 25 oC

substituent σp 102k2(M-1s-1)
4-Me -0.17 1.79
4-H 0 1.70
4-Cl 0.23 1.34
4-COOMe 0.45 1.27
4-CN 0.66 1.96
4-NO2 0.78 2.06

Table A30 Second order rate constant for 4-substituted benzyl chloride with diethyl
malonate anion in liquid ammonia at 25 oC

substituent σp 103k2(M-1s-1)

4-MeO -0.27 6.92

4-Me -0.17 3.04
4-H 0 2.27
4-Cl 0.23 6.67
4-CN 0.66 10.93
4-NO2 0.78 12.76

224
 Appendix A: Tables and Figures

Table A31 pKa’s of carbon acids in water and DMSO at 25 oC

acid pKa(aq.) pKa(DMSO)

acetone 19.3 26.5

dimedone 5.25 11.2
acetoacetone 9.0 13.3
2-acetylcyclohexanone 10.1 14.1
CH3COCHMeCOCH3 11 15.1
ethyl acetylacetate 10.7 14.2
diethyl malonate 12.9 16.4
CH3COCH2COPh 9.4 14.2
EtCH(CO2Et)2 15.0 18.7
meldrum’s acid 4.8 7.3
ethyl acetate 25.6 29.5

nitroethane 8.6 17.0

1-nitropropane 9.0 17.2
2-nitropropane 7.74 16.8
nitromethane 10.29 17.2
PhCH2NO2 6.88 12.2
(4-MePh)CH2NO2 7.11 12.33
(4-NO2Ph)CH2NO2 5.89 8.62
(3-NO2Ph)CH2NO2 6.3 10.04
(4-CNPh)CH2NO2 6.17 9.31

acetonitrile 28.9 31.3

malononitrile 11.2 11.1
EtO2CCH2CN 11.2 13.1
benzyl cyanide 22.0 21.9
CH3CH2CN 30.9 32.5

225
 Appendix A: Tables and Figures

Table A32 The rate for reaction between 4-NFB and various concentration of sodium
phenoxide in LNH3 at 25 oC (I=0.3M, KClO4)

concentration of phenoxide(M) 103kobs/s-1

0 1.66×10-2
0.05 2.52
0.1 4.81
0.2 6.87
0.3 9.71

Table A33 Solvolysis of 4-NFB in LNH3 at different temperature

T(K) kobs/s-1

293.2 7.86×10-6
333.2 1.24×10-4
353.2 3.90×10-4
373.2 1.51×10-3
393.2 2.06×10-3

226
 Appendix A: Tables and Figures

-4

y = -6.6769x + 11.062
R² = 0.9909
-7
ln(kobs/s-1)

-10

-13
2.4 2.9 3.4
1000/T(K-1)

Figure A24 Erying plot for the solvolysis of 4-NFB in LNH3

Table A34 The second order rate constant for the reaction between 4-NFB and 4-
substituted phenoxide in LNH3 at 25 oC

4-substituted phenoxide pKa(aq.) pKa(LNH3) k2(M-1s-1)

4-MeO 10.27 6.62 0.330

4-t-Bu 10.31 6.67 0.108
4-H 9.99 6.02 0.0528
4-Cl 9.20 4.69 0.0237
4-COOMe 8.47 4.04 5.31×10-4
4-CN 7.95 2.71 4.22×10-5

227
 Appendix A: Tables and Figures

Table A35 Salt effect on the rate of reaction between 4-NFB and phenoxide anion in
LNH3 at 25 oC

[NaNO3]/M 103kobs/s-1

0 5.28
0.2 3.33
0.5 2.90
1 2.38
1.5 1.93

Table A36 The rate for the reaction between 4-NFB and various concentration of
morpholine in LNH3 at 25 oC

morpholine concentration(M) 104kobs(s-1)

0 0.0786
0.2 0.967
0.5 2.275
0.75 3.058

Table A37 A comparison of the second order rate constant for the reaction between
sodium azide and 4-NFB in various solvents

solvent temperature(oC) k2(M-1s-1) krel.

methanol 25 6.3×10-8 1
nitromethane 25 2.0×10-4 3100
liquid ammoniab 25 3.8×10-4 6000
acetonitrile 25 5.0×10-4 8000
DMSO 25 5.0×10-4 8000
DMF 25 2.0×10-3 31000
a
Except the rate in liquid ammonia, all the other data is from ref. 28. b The rate of the reaction was measured
under pseudo first order conditions by using excess of 4-NFB, following the increasing of 4-NAB. The
ratio between 4-NFB and sodium azide was 10:1, less than 1% 4-NAB decomposed under such condition.

228
 Appendix A: Tables and Figures

Table A38 A comparison of the second order rate constants for the reaction between
piperidine and 4-NFB in various solvents

solvent temperature(oC) k2(M-1s-1) krel. reference

benzenea 25 7.8×10-6 1 322

liquid ammonia 25 2.3×10-3 300 this work
tetrahydrofuranb 25 3.9×10-3 500 323
DMSO 25 9.0×10-3 1150 276d
a
The reaction is base catalysed, the second order rate constant increases exponentially with increasing of
piperidine concentration, the rate cited here is when the concentration of piperidine equals 0.92M. b Base
catalysis mechanism was also observed, the rate cited is when the concentration of piperidine equals 1.0 M.

Table A39 The rate for the reaction between 4-NFB and various concentration of sodium
azide in LNH3 at 25 oC (I=3M, NaNO3)

concentration of sodium azide(M) 104kobs/s-1

0 0.248
0.5 0.559
1 0.869
1.5 1.223
2.5 1.810
3 2.158

Table A40 The rate for the solvolysis of 4-NAB under various concentration of KClO4 in
LNH3 at 25 oC

concentration of potassium perchlorate(M) 104kobs/s-1

0 0.0511
0.5 0.848
1 1.78
1.5 2.53

229
 Appendix A: Tables and Figures

Table A41 pKa of phenol in acetonitrile (AN) against its aqueous pKa at 25 oC

phenol pKa(aq.) pKa(AN)

3.5-dinitrophenol 6.66 20.50

4-nitrophenol 7.14 20.90
4-cyanophenol 7.95 22.77
3-nitrophenol 8.35 23.85
3,4-dichlorophenol 8.51 24.06
4-bromophenol 9.36 25.53
3-chlorophenol 9.02 25.04
3-trifluoromethylphenol 9.04 24.90
4-chlorophenol 9.20 25.44
phenol 9.99 27.20
4-methylphenol 10.28 27.45
4-tert-butylphenol 10.31 27.48

230
 Appendix A: Tables and Figures

Table A42 pKa of phenol in DMSO against its aqueous pKa at 25 oC

phenol pKa(aq.) pKa(DMSO)

3,5-dinitrophenol 6.66 10.60
4-nitrophenol 7.14 11.00
4-cyanophenol 7.95 13.01
4-acetylphenol 8.05 13.68
3-nitrophenol 8.36 13.75
3-chlorophenol 9.02 15.83
3-trifluoromethylphenol 9.04 14.30
3-acetylphenol 9.19 15.14
3-fluorophenol 9.28 15.88
4-bromophenol 9.36 15.50
phenol 9.99 16.47
3-methylphenol 10.1 16.86
4-methoxyphenol 10.27 17.58
4-methylphenol 10.28 16.96

231
 Appendix A: Tables and Figures

Table A43 pKa of phenol in methanol against its aqueous pKa at 25 oC

phenol pKa(aq.) pKa(Methanol)

3,4-dinitrophenol 5.42 9.46

3,5-dinitrophenol 6.66 10.20
4-nitrophenol 7.14 11.24
4-cyanophenol 7.95 13.27
3,5-dichlorophenol 8.18 12.94
3-nitrophenol 8.36 12.40
3-bromophenol 9.01 13.30
3-chlorophenol 9.02 13.10
4-chlorophenol 9.20 14.88
4-bromophenol 9.36 14.93
phenol 9.99 14.76
3-methylphenol 10.10 14.48
3,5-dimethylphenol 10.20 14.62
4-tert-butylphenol 10.31 14.52

232
 Appendix A: Tables and Figures

17

16
y = 1.1522x + 3.1998
15 R² = 0.8898

14
pKa (methanol)

13

12

11

10

8
4.5 5.5 6.5 7.5 8.5 9.5 10.5 11.5
pKa (aq.)

Figure A25 pKa of phenol in methanol against its aqueous pKa at 25 oC.

Table A44 The observed rate constant increases linearly with the increasing mass of
copper power added for the CuAAC reaction in liquid ammonia at 25 oCa

Cu0/mg 104kobs./s-1

60 0.48

90 0.70

120 1.06

150 1.23
a
Reaction conditions: 1mmol phenyl azide, 1mmol phenyl acetylene in 10ml liquid ammonia at 25 oC.

233
 Appendix B: Derivations

Derivation 1:

Modeling the influence of added ammonium ion on the ionisation and ion-pairing
behaviour of phenols in liquid ammonia

Two simultaneous equilibria need to be considered: deprotonation of the phenol by

ammonia to give the ion pair, and dissociation of the ion pair to give the free phenolate and
ammonium ions.

ArOH + NH3 

Ki
{ArO-NH4+}ip 
Kd
 ArO- + NH4+

These equilibria can be defined by the equilibrium constants Ki and Kd.

[ArO- NH 4 ]ip [ArO- ][NH 4 ]

Ki  Kd 
[ArOH] [ArO- NH 4 ]ip

We require a smooth function that describes the ionisation process at low (spectroscopic)
concentration, where the only ammonium ion present is that produced by dissociation of
the ion pair, through to conditions where ammonium ion is added to the solution. Solutions
containing stoichiometric initial concentrations of phenol (c) and ammonium ion (Z)
(added as ammonium chloride) are made up. Both equilibria must be satisfied, and from
the mass balance across the phenol/ion pair equilibrium alone.

[ ArOH ] 1 [ IP] Ki
[ArOH]T = [ArOH] + [IP], so  and 
[ ArOH ]T 1  K i [ ArOH ]T 1  K i

Proceeding as usual for a dissociation process:

NH3 + ArOH 

Ki
{ArO-NH4+}ip 
Kd
 ArO- + NH4+

Initial concs. c 0 0 Z

c(1  x) cK i (1  x)
Equilibrium cx cx + Z
1  Ki 1  Ki

(cx )(cx  Z ) x(cx  Z )(1  K i )

Whence K d  
cK i (1  x) K i (1  x)
1  Ki

234
 Appendix B: Derivations

This transforms to the quadratic:

c(1 + Ki)x2 + x(Z + KiZ + KiKd) - KiKd = 0

which can be implemented on Excel with Z as the variable.

For the modeling the influence of added ammonium ion on the ionisation and ion-pairing
behaviour of phenols in liquid ammonia under constant ionic strength (NaCl), the
modeling was based on the approximation that the extinction coefficient of ion pair for
ionised phenol does not change with the nature of counter-ion.

Derivation 2:

The pKa of aminium cations are less than -1 in liquid ammonia at room temperature

The conclusion is based on that the UV absorbance of 1mM 3-chlorophenol in liquid

ammonia at its λmax increases less than 5% when 0.1M triethylamine is added as base.

Consider the equilibrium of ionisation of triethylamine in liquid ammonia:

Et3NH NH3 Et3N NH4

[ Et3N ] [ NH4 ]
Ka(aminium) =
[Et3NH ]

If the added 0.1M triethylamine has little effect on the ionisation of 1mM 3-chlorophenol
as observed, therefore the total ammonium ion concentration can be calculated from the
known initial concentration of 3-chlorophenol and its pKa in liquid ammonia, which is 1.65
× 10-4M.

The NMR results show that amine hydrochloride salts are fully deprotonated in liquid
ammonia at room temperature, therefore, the ratio between free amine and protonated
amine must more than 103 in liquid ammonia. Therefore the pKa (aminium) must less than
-1 in liquid ammonia at room temperature.

235
 Appendix C: NMR Spectra

0.1M in LNH3

Figure N1 1H NMR Spectrum of 0.1M benzylamine in liquid ammonia at 25 oC

0.1M in LNH3

Figure N2 1H NMR Spectrum of 0.1M benzylamine hydrochloride in liquid ammonia at

25 oC

236
 Appendix C: NMR Spectra

0.1M in LNH3

Figure N3 1H NMR Spectrum of 0.1M triethylbenzylammonium chloride in liquid

ammonia at 25 oC

0.1M in LNH3

Figure N4 1H NMR Spectrum of 0.1M piperidine in liquid ammonia at 25 oC

237
 Appendix C: NMR Spectra

 HCl

0.1M in LNH3

Figure N5 1H NMR Spectrum of 0.1M piperidine hydro chloride in liquid ammonia at 25

o
C

1M in LNH3

Figure N6 1H NMR Spectrum of 1 M piperidine in liquid ammonia at 25 oC

238
 Appendix C: NMR Spectra

 HCl

1M in LNH3

Figure N7 1H NMR Spectrum of 1M piperidine hydrochloride in liquid ammonia at 25 oC

Figure N8 1H NMR Spectrum of 0.1M dimedone (5,5-dimethylcyclohexane-1,3-dione) in

liquid ammonia at 25 oC

239
 Appendix C: NMR Spectra

Figure N9 13C NMR Spectrum of 0.1M dimedone (5,5-dimethylcyclohexane-1,3-dione) in

liquid ammonia at 25 oC

Figure N10 DEPT 135 Spectrum of 0.1M dimedone (5,5-dimethylcyclohexane-1,3-dione)

in liquid ammonia at 25 oC

240
 Appendix C: NMR Spectra

Figure N11 1H NMR Spectrum of 0.1M 1-nitropropane in liquid ammonia at 25 oC

Spec.no.16517: Diethyl malonate

4.1308
4.1165
4.1022
4.0880

3.4581

1.1829
1.1686
1.1544

Current Data Parameters

NAME sn16518pj
EXPNO 1
PROCNO 1

F2 - Acquisition Parameters
Date_ 20100914
Time 11.09
INSTRUM av500
PROBHD 5 mm PABBO BB-
PULPROG zg30
TD 65536
SOLVENT DMSO
NS 16
DS 2
3 SWH
FIDRES
10330.578 Hz
0.157632 Hz
AQ 3.1719923 sec
RG 4.5
DW 48.400 usec
DE 6.00 usec
TE 297.7 K
D1 1.00000000 sec
TD0 1

======== CHANNEL f1 ========

NUC1 1H
P1 9.70 usec
PL1 0.10 dB
SFO1 500.1330885 MHz

F2 - Processing parameters
SI 32768
SF 500.1300055 MHz
WDW EM
SSB 0
LB 0.30 Hz
GB 0
PC 1.00

broad peak
4.00

2.15

7.62

10 9 8 7 6 5 4 3 2 1 ppm

Figure N12 1H NMR Spectrum of 0.1M diethyl malonate ester in liquid ammonia at 25 oC

241
 Appendix C: NMR Spectra

Figure N13 1H NMR Spectrum of 0.1M benzylmalonodinitrile ester in liquid ammonia at

25 oC

13
Figure N14 C NMR Spectrum of 0.1M benzylmalonodinitrile ester in liquid ammonia at
o
25 C

242
 Appendix C: NMR Spectra

Figure N15 DEPT 135 Spectrum of 0.1M benzylmalonodinitrile ester in liquid ammonia
at 25 oC

relaxation
time = 1s
5 3

Figure N16 1H NMR Spectrum of 0.1M benzyl cyanide in liquid ammonia at 25 oC,
integration aromatic H : methylene H = 5:3, relaxation time = 1s

243
 Appendix C: NMR Spectra

relaxation 5 2.4
time = 5s 5

Figure N17 1H NMR Spectrum of 0.1M benzyl cyanide in liquid ammonia at 25 oC,
integration aromatic H : methylene H = 5:2.4, relaxation time = 5s

5 2.2
relaxation
time = 10s

Figure N18 1H NMR Spectrum of 0.1M benzyl cyanide in liquid ammonia at 25 oC,
integration aromatic H : methylene H = 5:2.2, relaxation time = 10s

244
 Appendix C: NMR Spectra

13
Figure N19 C NMR Spectrum of 0.1M benzyl cyanide in liquid ammonia at 25 oC

DMSO peak

Figure N20 1H NMR Spectrum of 0.1M malonodinitrile in liquid ammonia at 25 oC

245
 Appendix C: NMR Spectra

DMSO
peak

13
Figure N21 C NMR Spectrum of 0.1M malonodinitrile in liquid ammonia at 25 oC

the partial frozen of solvent

proton exchange at -40oC

Figure N22 1H NMR Spectrum of 0.1M malonodinitrile in liquid ammonia at -40 oC

246
 Appendix C: NMR Spectra

internal standard
acetonitrile peak

Figure N23 1H NMR Spectrum of 0.1M malononitrile with 0.1M NH4Cl and acetonitrile
as internal reference in liquid ammonia at 25 oC

Figure N24 1H NMR Spectrum of 0.1M malononitrile with 1M NaNO3 in liquid ammonia
at 25 oC

247
 Appendix C: NMR Spectra

Figure N25 1H NMR Spectrum of 0.1M malononitrile with 2 eq. NaNH2 in liquid
ammonia at 25 oC

Figure N26 1H NMR Spectrum of 0.1M malononitrile with 2 eq. NaNH2 in liquid
ammonia at 25 oC

248
 Appendix C: NMR Spectra

Figure N27 1H NMR Spectrum of 0.1M benzophenone imine in liquid ammonia at 25 oC

Figure N28 19F NMR Spectrum of 2,4-difluoronitrobenzene in DMSO-d6 at 25 oC

249
 Appendix C: NMR Spectra

14

15

Figure N29 The products analysis for the solvolysis of 2,4-difluoronitrobenzene in liquid
ammonia at 25 oC by 19F NMR spectrum (DMSO-d6)

38

Figure N30 The 1H NMR spectrum for the reaction product of CuIAAC in liquid ammonia
at room temperature [benzyl azide with phenyl acetylene (1:1), after the vapourisation of
ammonia, no purification procedure was performed, the product was directly dissoved in
DMSO-d6 for the NMR study]

250
 Appendix C: NMR Spectra

PhC≡CH

Figure N31 The 1H NMR spectrum of phenylacetylene in liquid ammonia at 25 oC

PhC≡CD, 87% deuterium

incorporation

Figure N32 The 1H NMR spectrum of phenylacetylene-d in DMSO-d6 at 25 oC, 87%

duterium incorporation (ca.)

251
 Appendix D: Safety Protocols

OPERATING PROCEDURE AND RISK ASSESSMENT

Equipment

Several glass pressure vessels of different volume capacity (3 to 35ml); a 30ml glass burette, a
40ml glass tank; a SGE 10ml pressure syringe; several Omnifit TM 2-way valves and 3-way
valves; a pressure UV cell; a pressure NMR glass tube.

General procedure

As described in Experiment section

Properties of liquid ammonia

Vapor pressure of liquid ammonia at 25 oC, 9.89 bar; at 30 oC, 11.7 bar.
Flammability limits in air 15 - 25% v/v (can explode).

Perceived hazards

 Catastrophic failure of vessel with ejection of glass and liquid/vapour into the working
area.

 Splashing with ammonia liquid could lead to burns; inhalation of ammonia vapour at
high concentration will cause unconsciousness.

 Unexpected pressure increase due to gas evolution in unvented vessels.

Risk mitigation

The glassware used is purpose made and the glassblower is aware of the intended use and
required operating pressure. Glassware is pressure tested before use to at least twice the
working pressure. A limit of 20 mm diameter is applied to new glassware.

Glassware under pressure is shielded at all times by Perspex sheet of at least 8 mm thickness.

252
 Appendix D: Safety Protocols

The inventory of liquid ammonia in any glass reaction vessel is restricted to 12 ml., and to 20
ml in feed vessels.

There will be no lone working with pressurised glassware.

All reactions will be assessed for possible exothermicity and generation of permanent gases as
part of the individual experimental risk assessment, and appropriate measures taken e.g.,
control of reactant concentrations and methods of reactant addition.

PPE

Goggles will be worn at all times when manipulating liquid ammonia. A full face shield will
always be available and will be used when judged appropriate.

Pressure Test Standard Procedures

1. Pressure test below 10 bar.

 Connect the vessel to the Omnifit parts and the lid.

 Connect the vessel to an air or nitrogen cylinder with pressure gauges.
 Submerge the vessel in a bucket filled with water
 Gradually increasing the pressure to about 8-10 bar, see if any bubble comes out from
the vessel (MUST WEAR PROTECTION GOGGLES WHEN DOING THIS!).

If no obvious bubbling is observed, close all the Omni valves connected to the vessel, leave
the vessel in water for overnight, and monitor the pressure through a pressure gauge

2. Pressure test above 10 bar.

 Connect the vessel to Omnifit parts and the lid.

 Connect the vessel to an HPLC pump, and make the water as eluent.
 Submerge the vessel in a steel bucket filled with water.
 Set the flow rate of the HPLC pump to 0.5-1ml/min.
 Set the pump pressure up-limit to certain value, generally 20 bar.

253
 Appendix D: Safety Protocols

If the pump automatically shut-down after reaching the up-limit, then the vessel is safe under
that pressure.

Repeat the procedures described above after 2-3 months, keep the records routinely.

Date Signature (Research student)

Signature (Supervisor)

254
 References

References

1
Reichardt, C. Solvents and Solvent Effects in Organic Chemistry, 3rd ed., Wiley-VCH, Verlag GmbH&Co.
KGaA, Weinheim, 2003.
2
Solvents-A Global Strategic Business Report, Global Industry Analysts, Inc., Oct., 2010.
3
For examples: (a) Marcus, Y. Solvent Mixtures: Properties and Selective, Marcel Dekker, Inc., New York,
Basel, 2002. (b) Hynes, J. T. Ann. Rev. Phys. Chem. 1985, 36, 573-597. (c) Kim, Y.; Cramer, C. J.; Truhlar, D.
G. J. Phys. Chem. A, 2009, 113, 9109-9114. (d) Abraham, M. H. Prog. Phys. Org. Chem. 1974, 11, 1-87.
4
(a) Alfonsi, K.; Colberg, J.; Dunn, P. J.; Fevig, T.; Jennings, S.; Johnson, T. A.; Kleine, H. P.; Knight, C.;
Nagy, M. A.; Perry, D. A.; Stefaniak, M. Green Chem. 2008, 10, 31-36. (b) DeSimone, J. M. Science 2002,
297, 799-803. (c) Sheldon, R. A. Green Chem. 2005, 7, 267-278.
5
Lowry, T. H.; Richardson, K. S. Mechanism and Theory in Organic Chemistry, 3rd ed., Happer and Row, New
York, 1987.
6
Isaacs, N. S. Physical Organic Chemistry, 2nd ed., Longman, Essex, 1995.
7
(a) Smallwood, I. M. Handbook of Organic Solvent Properties, John Wiley&Sons Inc., New York, 1996. (b)
Mido, Y.; Taguchi, S.; Sethi, M. S.; Iqubal, S. A. Chemistry in Aqueous and Non-Aqueous Solvents, Discovery
Publishing House, New Delhi, 2001.
8
Lagowski, J. J. Synth. React. Inorg., Met.-Org., Nano-Met. Chem. 2007, 37, 115-153.
9
Anslyn, E. V.; Dougherty, D. A. Modern Physical Organic Chemistry, University Science Books, Sausalito,
California, 2005.
10
For examples: (a) Kolthoff, I. M.; Chantoni, M. K.; Popov, A. I.; Caruso, J. A.; Steigman, J. Acid-Base
Equilibria in Non-Aqueous Solvents, Kolthoff, I. M.; Elving, P. J., Eds., Treatise on Analytical Chemistry, 2nd
ed., John Wiley, New York, 1979. Part I, Vol. 2. (b) Bell, R. P. The Proton in Chemistry, Methuen, London,
1959. (c) King, E. J. Acid-Base Equilibria, Pergamon Press, Oxford, 1965.
11
Shedlovsky, T.; Kay, R. L. J. Phys. Chem. 1956, 60, 151-155.
12
Kolthoff, I. M.; Chantooni Jr., M. K.; Bhowmik, S. J. Am. Chem. Soc. 1968, 90, 23-28.
13
Jover, J.; Bosque, R.; Sales, J. QSAR Comb. Sci. 2007, 26, 385-397.
14
For review, see: Kol’tsov, A. I.; Kheifets, G. M. Russ. Chem. Rev. 1971, 40, 773-788.
15
For examples, see: (a) Eistert, B.; Geiss, F. Tetrahedron 1959, 7, 1-9. (b) Eistert, B.; Reiss, W. Chem. Ber.
1954, 87- 92. (3) Eistert, B.; Schank, K. Tetrahedron lett. 1964, 5, 429-432.
16
Russell, P. B.; Mentha, J. J. Am. Chem. Soc. 1955, 77, 4245-4247.
17
(a) Beak, P. Acc. Chem. Res. 1977, 10, 186-192. (b) Beak, P.; Covington, J. B.; Zeigler, J. M. J. Org. Chem.
1978, 43, 177-178. (c) Maquestiau, A.; Haverbeke, Y. V.; DeMeyer, C.; Katritzky, A. R.; Cook, M. J.; Page,
A. D. Can. J. Chem. 1975, 53, 490-496. (d) Cook, M. J.; El-Abbady, S.; Katritzky, A. R.; Guimon, C.; Pfister-
Guillouzo, G. J. Chem. Soc., Perkin Trans. 2 1977, 1652-1656. (e) Nowak, M. J.; Lapinski, L.; Fulava, J.; Les,
A.; Adamowicz, L. J. Phys. Chem. 1992, 96, 1562-1569. (f) Frank, J.; Katritzky, A. R. J. Chem. Soc., Perkin
Trans. 2 1976, 1428-1431.

255
 References
18
Cram, D. J.; Rickborn, B.; Kingsbury, C. A.; Haberfield, P. J. Am. Chem. Soc. 1961, 83, 3678-3687.
19
Menschutkin, N. Z. Phys. Chem. 1890, 6, 41-57.
20
For the establishment of transition state theory, see: (a) Eyring, H.; Polanyi, M. Z. Phys. Chem. Abt. B, 1931,
12, 279-311. (b) Eyring, H. J. Chem. Phys. 1935, 3, 107-116. (c) Evans, M. G.; Polanyi, M. Trans. Faraday
Soc. 1935, 31, 875-894. For the reviews on transition state theory, see: (a) Truhlar, D. G.; Hase, W. L.; Hynes,
J. T. J. Phys. Chem. 1983, 87, 2664-2682. (b) Laidler, K. J.; King, M. C. J. Phys. Chem. 1983, 87, 2657-2664.
(c) Levine, R. D. Molecular Reaction Dynamics, Cambridge University Press, Cambridge, 2005. (d) Henriksen,
N. E.; Hansen, F. Y. Theories of Molecular Reaction Dynamics, Oxford University Press, 2008.
21
For the reviews on non-equilibrium transition state theory, see: (a) Hynes, J. T. Ann. Rev. Phys. Chem. 1983,
36, 573-597. (b) Simon, J. D. Acc. Chem. Res. 1988, 21, 128-143. (c) Carpenter, K. B. Angew. Chem. In. Ed.
1998, 37, 3340-3350. (d) Bertran, J.; Burgos, F. S. J. Chem. Edu. 1984, 61, 416-417.
22
Hammett, L. P. Physical Organic Chemistry: Reaction Rates, Equilibria and Mechanisms, McGraw Hill, New
York, 1940.
23
Hughes, E. D.; Ingold, C. K. J. Chem. Soc. 1935, 244-255.
24
Pocker, Y. Nucleophilic Substitutions at a Saturated Carbon Atom in Non-Hydroxylic Solvents, Porter G.;
Stevens, B. Eds., Progress in Reaction Kinetics, Pergamon Press, Oxford, 1961, Vol. 1.
25
Westaway, K. C. Can. J .Chem. 1978, 56, 2691-2699.
26
(a) Kim, H. J.; Hynes, J. T. J. Am. Chem. Soc. 1992, 114, 10508-10528. (b) Kim, H. J.; Hynes, J. T. J. Am.
Chem. Soc. 1992, 114, 10528-10537. (c) Mathis, J. R.; Kim, H. J.; Hynes, J. T. J. Am. Chem. Soc. 1993, 115,
8248-8246.
27
Smith, S. G.; Fainberg, A. H.; Winstein, S. J. Am. Chem. Soc. 1961, 83, 618-625.
28
Cox, B. G. Morden Liquid Phase Kinetics, Compton, R. G. Ed., Oxford Chemistry Primers, Oxford University
Press, New York, 1994.
29
Acevedo, O.; Jorgensen, W. L. Org. Lett. 2004, 6, 2881-2884.
30
For the rate of azide anion with n-butyl bromide, see: Delpuech, J. J. Tetrahedron Lett. 1965, 6, 2111-2115; for
the rate of azide anion with 4-nitrofluorobenzene, see: Alexander, R.; Ko, E. C. F.; Parker, A. J.; Broxton, T. J.
J. Am. Chem. Soc. 1968, 90, 5049-5069; for the rate of piperidine with 4-nitrofluorobenzene, see: Suhr, H.
Chem. Ber. 1964, 97, 3277-3283.
31
For examples, see: (a) Breydo, L. P.; Shevchenko, A. A.; Kost, O. A. Russ. Chem. Bull. 1997, 46, 1339-1343.
(b) Ehlerova, J.; Trevani, L.; Sedlbauer, J.; Tremaine, P. J. Solution Chem. 2008, 37, 857-874. (c) Rosenberg,
L. S.; Jackson, J. L. Drug Dev. Ind. Pharm. 1989, 15, 373-386. (d) Belokon, Y. N.; Faleev, N. G.; Belikov, V.
M. Russ. Chem. Bull. 1969, 18, 949-955.
32
Kirkwood, J. G. J. Chem. Phys. 1934, 2, 351-361.
33
Tommila, E. Acta Chem. Scand. 1959, 13, 622-623. -
34
(a) Grunwald, E.; Winstein, S. J. Am. Chem. Soc. 1948, 70, 846-854. (b) Fainberg, A. H.; Grunwald, E. J. Am.
Chem. Soc. 1956, 78, 2770-2777. (c) Fainberg, A. H.; Grunwald, E. J. Am. Chem. Soc. 1957, 79, 1597-1602.

256
 References

(d) Fainberg, A. H.; Grunwald, E. J. Am. Chem. Soc. 1957, 79, 1602-1608. (e) Fainberg, A. H.; Grunwald, E. J.
Am. Chem. Soc. 1957, 79, 1608-1612. (f) Winstein, S.; Fainberg, A. H. J. Am. Chem. Soc. 1957, 79, 5937-5950.
35
Raber, D. J.; Bingham, R. C.; Harris, J. M.; Fry, J. L.; Schleyer, P. v. R. J. Am. Chem. Soc. 1970, 92, 5977-
5981.
36
For examples, see: (a) Bentley, T. W.; Carter, G. E. J. Am. Chem. Soc. 1982, 92, 5741-5747. (b) Kevill, D. N.;
Kamil, W. A.; Anderson, S. W. Tetrahedron Lett. 1982, 23, 4635-4638.
37
(a) Grunwald, E.; Winstein, S.; Jones, H. W. J. Am. Chem. Soc. 1951, 73, 2700-2707. (b) Winstein, S.;
Fainberg, A. H.; Grunwald, E. J. Am. Chem. Soc. 1957, 79, 4146-4155.
38
(a) Drougard, D; Decroocq, D. Bull. Soc. Chem. Fr. 1969, 2972. (b) Auriel, M.; Hoffmann, E. D. J. Am. Chem.
Soc. 1975, 97, 7433-7437.
39
Reichardt, C. Empirical Parameters of Solvent Polarity and Chemical Reactivity, Ratajczak, H.; Orville-
Thomas, W. J. Eds., Molecular Interactions, Wiley, Chichester, 1982, Vol. 3, Chapter 5.
40
Kosower, E. M. J. Am. Chem. Soc. 1958, 80, 3253-3260.
41
Dimroth, K.; Reichardt, C.; Siepmann, T.; Bohlmann, F. Liebigs Ann. Chem. 1963, 661, 1-37.
42
For the establishment of α scale, see: Kamlet, M. J.; Taft, R. W. J. Am. Chem. Soc. 1976, 98, 2886-2894; for
the establishment of β scale, see: Kamlet, M. J.; Taft, R. W. J. Am. Chem. Soc. 1976, 98, 377-383.
43
Kamlet, M. J.; Abboud, J. L.; Taft, R. W. J. Am. Chem. Soc. 1977, 99, 6027-6038.
44
For reviews, see: (a) Nicholls, D. Inorganic Chemistry in Liquid Ammonia, Clark R. J. H. Ed., Topic in
Inorganic and General Chemistry, Monograph 17, Elsevier Scientific Publishing Company, Amsterdam, 1979.
(b) Yost, D. M.; Russell, Jr., H. Systematic Inorganic Chemistry, Prentice-Hall, New York, 1946, chapter 4. (c)
Hildebrand, J. H. J. Chem. Edu. 1948, 25, 74-77.
45
Wells, A. F. Structural Inorganic Chemistry, 5th ed, Clarendon Press, Oxford, 1987.
46
Swalen, J. D.; Ibers, J. A. J. Chem. Phys. 1962, 36, 1914-1918.
47
Olovsson, I.; Templeton, D. H. Acta. Cryst. 1959, 21, 832-836.
48
Kruh, R. F.; Petz, J. I. J. Chem. Phys. 1964, 41, 890-891.
49
Dougal, J. C.; Gans, P.; Gill, J. B. J. Chem. Soc., Faraday Trans. 1 1988, 84, 657-664.
50
(a) Franklin, E. C. The Nitrogen System of Compounds, Reinhold Publishing Co., 1935. (b) Jolly, W. L. The
Inorganic Chemistry of Nitrogen, Benjamin, New York, 1964.
51
Sciaini, G. ; Marceca, E.; Fernandez-Prini, R. J. Chem. Phys. 2008, 112, 11990-11995.
52
(a) Impey, R. W.; Sprik, M.; Klein, M. L. J. Am. Chem. Soc. 1987, 109, 5900-5904. (b) Goldman, N.; Fellers,
R. S.; Brown, M. G.; Braly, L. B.; Keoshin, C. J.; Leforestier, C.; Saykally, R. J. J. Chem. Phys. 2002, 116,
10148-10163.
53
Dvorko, G. F.; Ponomareva, E. A.; Kulik, N. I. Russ. Chem. Rev. 1984, 53, 547-560.
54
Gutmann, V. Coord. Chem. Rev. 1976, 18, 225-255.
55
Herlem, M.; Popov, A. I. J. Am. Chem. Soc. 1972, 94, 1431-1434.
56
For examples, see: (a) Nelson, Jr., D. D.; Fraser, G. T.; Klemperer, W. Science 1987, 238, 1670-1674. (b)
Luehurs, D. C.; Brown, R. E.; Godbole, K. A. J. Solution Chem. 1989, 18, 463-469. (c) Swift, T. J.; Marks, S.

257
 References

B.; Sayre, W. G. J. Chem. Phys. 1966, 44, 2797-2801. (d) Tongraar, A.; Kerdcharoen, T.; Hannongbua, S. J.
Phys. Chem. A, 2006, 110, 4924-4929.
57
Marcus, Y. Pure Appl. Chem. 1983, 55, 977-1021.
58
Howard, Jr., D. H.; Friedrich, F.; Browne, A. W. J. Am. Chem. Soc. 1934, 56, 2332-2340.
59
Lagowski, J. J.; Moczygemba, G. A. Acid and Basic Solutions, Lagowski, J. J. Ed., The Chemistry of Non-
Aqueous Solutions, Academic Press, 1967.
60
Kraus, C. A. J. Am. Chem. Soc. 1907, 29, 1557-1571.
61
Johnson, W. C. Chem. Rev. 1940, 62, 1-2.
62
Kraus, C. A.; Bray, W. C. J. Am. Chem. Soc. 1913, 35, 1315-1434.
63
Audrieth, L. F.; Kleinberg, J. Non-Aqueous Solvents, John Wiley&Sons Inc., New York, 1953, Chapter 3.
64
Watt, G. W.; Leslie, W. B.; Moore, T. E. Chem. Rev. 1942, 31, 525-536. (b) Watt, G. W.; Leslie, W. B.;
Moore, T. E. Chem. Rev. 1943, 32, 219-229.
65
Smith, H. Organic Reactions in Liquid Ammonia, Jander, G.; Spandau, H.; Addison, C. C. Eds., Chemistry in
Nonaqueous Ionizing Solvents, Vol. 1, Part 2, John Wiley&Sons Inc., New York-London, 1963.
66
Dauben, W. G.; Wolf, R. E. J. Org. Chem. 1970, 35, 374-379.
67
Dauben, W. G.; Deviny, E. J. J. Org. Chem. 1966, 31, 3794-3798.
68
Bauld, N. L. J. Am. Chem. Soc. 1962, 84, 4345-4347.
69
Campbell, K. N.; Eby, L. T. J. Am. Chem. Soc. 1941, 63, 2683-2685.
70
Paquette, L. A. ; Nelson, N. A. J. Org. Chem. 1962, 27, 2272-2274.
71
Huffman, J. W.; Charles, J. T. J. Am. Chem. Soc. 1968, 90, 6486-6492.
72
Birch, A. J. Quart. Rev. 1950, 4, 69-93.
73
Birch, A. J.; Smith, H. Quart. Rev. 1958, 12, 17-33.
74
G. W. Watt, Chem. Rev. 1950, 46, 317-379.
75
Rieger, P. H.; Bernal, I.; Reinmuth, W. H.; Fraenkel, G. K. J. Am. Chem. Soc. 1963, 85, 683-693.
76
Buick, A. R.; Kemp, T. J.; Neal, G. T.; Stone, T. J. Chem. Commun., (London), 1968, 1331-1332.
77
Brown, J. K.; Burnham, D. R.; Rogers N. A. J. J. Chem. Soc. B 1969, 1149-1154.
78
Kuda, L.; Boekelheide, V. Org. Syn. 1962, 43, 22.
79
Schubert, W. M.; Rabinovitch, B. S.; Larson, N. R.; Sims, V. A. J. Am. Chem. Soc.1952, 74, 4950-4951.
80
Watt, G. W.; Knowles, C. M.; Morgan, L. O. J. Am. Chem. Soc. 1947, 69, 1657-1659.
81
Schultz, A. G.; Macielag, M. J. Org. Chem. 1986, 51, 4983-4987.
82
Kaiser, E. M.; Edmonds, C. G.; Grubb, S. D.; Smith, J. W.; Tramp, D. J. Org. Chem. 1971, 36, 330-335.
83
Drake, N. L.; Smith, W. M. Org. Syn. 1955, Coll. Vol. 3, 219.
84
Hauser, C. R.; W. Swamer, F. W.; Adams, J. T. Org. React. 1954, 8, 59.
85
Flahaut, J.; Miginiac, P. Helv. Chim. Acta 1978, 61, 2275-2285.
86
Zoss, A. O.; Hennion, G. F. J. Am. Chem. Soc. 1941, 63, 1151-1153.
87
For details of benzyne formation and mechanism development, see: (a) Bergstrom, F. W.; Wright, R. E.;
Chandler, C.; Gilkey, W. A. J. Org. Chem. 1936, 1, 170-178. (b) Roberts, J. D.; Simmons, Jr., H. E.; Carlsmith,

258
 References

L. A.; Vaughan, C. W. J. Am. Chem. Soc. 1953, 75, 3290-3291. (c) Bunnett, J. F.; Brotherton, T. K. J. Org.
Chem. 1958, 23, 904-906.
88
Graaff, G. B. R. D.; Hertog, H. J. D.; Melger, W. C. Tetrahedron Lett. 1965, 6, 963-968.
89
Dunnavant, W. R.; Hauser, C. R. J. Org. Chem. 1960, 25, 503-507.
90
Hauser, C. R.; Shivers, J. C.; Skell, P. S. J. Am. Chem. Soc. 1945, 67, 409-412.
91
Shivers, J. C.; Dillon, M. L.; Hauser, C. R. J. Am. Chem. Soc. 1947, 69, 119-123.
92
Atherton, J. H. personal communications.
93
C. K. Ingold, Structure and Mechanism in Organic Chemistry, G. Bell and Sons Ltd., London, 1953, 339.
94
Markova, G. S.; Shatenshtein, A. I. Compt. Rend. Acad. Sci. U. R. S. S. 1942, 35, 68-70.
95
Braun, J. V.; Pinkernelle, W. Chem. Ber. 1937, 70, 1230-1240.
96
Lemons, J. F.; Anderson, R. C.; Watt, G. W. J. Am. Chem. Soc. 1941, 63, 1953-1956.
97
McCasland, G. E.; Clark, R. K.; Carter, H. E. J. Am. Chem. Soc. 1949, 71, 637-642.
98
McManus, S. P.; Larson, C. A.; Hearn, R. A. Synth. Commun. 1973, 3, 177-180.
99
Strain, H. R. J. Am. Chem. Soc.1932, 54, 1221-1228.
100
Dragunov, S. S. J. Gen. Chem. (U. S. S. R.) 1939, 9, 693-712.
101
Strain, H. R. J. Am. Chem. Soc.1930, 52, 820-823.
102
Pinck, L. A.; Hilbert, G. E. J. Am. Chem. Soc. 1934, 56, 490-491.
103
US 5679855, BASF Aktiengesellschaft, issue date: Oct. 21, 1997.
104
Hine, J. Physical Organic Chemistry, McGraw-Hill Book Co., New York, 1956.
105
Audrieth, L. F.; Kleinberg, J. J. Org. Chem. 1938, 3, 312-316.
106
Kizner, T. A.; Shteingardts, V. D. J. Org. Chem. U. S. S. R. 1984, 20, 1089-1096.
107
Shtark, A. A.; Kizner, T. A.; Shteingarts, V. D. J. Org. Chem. U. S. S. R. 1982, 18, 2321-2317.
108
Politanskaya, L. V.; Malysheva, L. A.; Beregovaya, I. R.; Bagryanskaya, I. Y.; Gatilov, Y. V.; Malykhin , E.
V.; Shteingardts, V. D. J. Fluorine Chem. 2006, 126, 1502-1509.
109
Ostrowicz, A.; Baloniak, S.; Makosza, M; Rykowski, A. Tetrahedron Lett. 1992, 33, 4787-4790.
110
Charushin, V. N.; Chupakhin, O. N. Pure Appl. Chem. 2004, 76, 1621-1631.
111
Rossi, R. A.; Rossi, R. H. D. Aromatic Substitution by the SRN1 Mechanism, ACS Monograph 178, The
American Chemical Society, Washington, DC, 1983.
112
(a) Manzo, P. G.; Palacios, S. M.; Alonso, R. A.; Rossi, R. A. Org. Prep. Proced. Int. 1995, 27, 660-663. (b)
Austin, E.; Alonso, R. A.; Rossi, R. A. J. Org. Chem. 1991, 56, 4486-4489.
113
For examples, see: (a) Appl, M. Ammonia: Principles and Industrial Practice, Wiley-VCH, Weinheim, 1999.
(b) Appl, M. Ammonia, Ullmann's Encyclopedia of Industrial Chemistry, Wiley-VCH, Weinheim, 2006.
114
For examples, see: (a) Jorgensen, W. L.; Ibrahim, M. J. Am. Chem. Soc. 1980, 102, 3309-3315. (b) Kraus, C.
A. Chem. Rev. 1940, 26, 95-104. (c) Schmidt, F. C.; Sottysiak, J.; Kluge, H. D. J. Am. Chem. Soc. 1936, 58,
2509-2510. (d) Fernelius, W. C.; Bowman, G. B. Chem. Rev. 1940, 26, 3-48. (e) Jolly, W. A. Chem. Rev.
1952, 50, 351-361.
115
For examples, see: (a) Birch, A. J. Pure Appl. Chem. 1996, 68, 553-556. (b) Schon, I. Chem. Rev. 1984, 84,
287-297. (c) Fernelius, W. C.; Watt, G. W. Chem. Rev. 1937, 20, 195-258.

259
 References
116
For examples, see: (a) Harris, T. M.; Harris, C. M. J. Org. Chem. 1966, 31, 1032-1035. (b) Flahaut, J.;
Miginiac, P. Helv. Chim. Acta 1978, 61, 2275-2285. (c) Vaughn, T. H.; Hennion, G. F.; Vogt, R. R.;
Nieuwland, J. A. J. Org. Chem. 1937, 2, 1-22. (d) Baldinger, L. H.; Nieuwland, J. A. J. Am. Chem. Soc. 1933,
55, 2851-2853.
117
For details of the source of ammonia, see: http://www.boconline.co.uk/products/products_by_type/
refrigerant_gases /natural_refrigerants/ammonia_r717.asp
118
Kunert, M.; Dinjus, E.; Nauck, M.; Sieler, J. Chem. Ber./Recueil 1997, 130, 1461-1465.
119
Kazhemekaite, M.; Yuodvirshis, A.; Vektarene, A. Chem. Heterocycl. Compd. 1998, 34, 252-253.
120
Akhtar, M. H.; Oehlschlager, A. C. Tetrahedron 1970, 26, 3245-3263.
121
Maddani, M. R.; Moorthy, S. K.; Prabhu, K. R. Tetrahedron 2010, 66, 329-333.
122
Mallory, F. B.; Smith, P. A. S.; Boyer, J. H. Org. Synth. 1963, Coll. Vol. 4, 74.
123
Ueno, M.; Hori, C.; Suzawa, K.; Ebisawa, M.; Kondo, Y. Euro. J. Org. Chem. 2005, 70, 1965-1968.
124
Shea, K. J.; Kim, J. S. J. Am. Chem. Soc. 1992, 114, 4846-4855.
125
Giles, D. E.; Parker, A. J. Aust. J. Chem. 1973, 26, 273-299.
126
Woehrle, G. H.; Brown, L. O.; Hutchison, J. E. J. Am. Chem. Soc. 2005, 127, 2172-2183.
127
Oberster, A. E.; Farhat, K.; Kibler, R. W.; Cook, W. S.; Setoodeh, S. Y.; Smith Jr., G. E. P. Can. J. Chem.
1967, 45, 195-201.
128
Rout, L.; Saha, P.; Jammi, S.; Punniyamurthy, T. Adv. Synth. Catal. 2008, 395-398.
129
Kirby, A. J.; Jencks, W. P. J. Am. Chem. Soc. 1965, 87, 3209-3216.
130
Schaap, W. B.; McKinney, P. S. Anal. Chem. 1964, 36, 29-35.
131
Street, L. J.; Baker, R.; Davey, W. B.; Guiblin, A. R.; Jelley, R. A.; Reeve, A. J.; Routledge, H.; Sternfeld, F.;
Watt, A. P. J. Med. Chem. 1995, 38, 1799-1810.
132
Romagnoli, R.; Baraldi, P. G.; Cruz-Lopez, O.; Cara, C. L.; Carrion, M. D. ; Brancale, A.; Hamel, E.; Chen,
L.; Bortolozzi, R.; Basso, G.; Viola, G. J. Med. Chem. 2010, 53, 4248-4258.
133
Kitazaki, T.; Ichikawa, T.; Tasaka, A.; Hosono, H.; Matsushita, Y.; Hayashi, R.; Okonogi, K.; Itoh, K. Chem.
Pharm. Bull. 2000, 48, 1935-1946.
134
Ali, M. A.; Saha, P.; Punniyamurthy, T. Synthesis 2010, 908-910.
135
Dey, A.; Desiraju, G. R. Chem. Commun. 2005, 2486-2488.
136
Bunce, R. A.; Easton, K. M. Org. Prep. Proced. Int. 2004, 36, 76-81.
137
Ma, D.; Cai, Q. Org. Lett. 2003, 5, 3799-3802.
138
Dale. J.; Vikersveen, L. Acta Chem. Scand. 1988, 42B, 354-361.
139
Abramovitch, R. A.; Challand, S. R.; Scriven, E. F. V. J. Org. Chem. 1972, 37, 2705-2710.
140
Droumaguet, B. L.; Mantovani, G.; Haddleton, D. M.; Velonia, K. J. Mater. Chem. 2007, 17, 1916-1922.
141
Suwa, T.; Sugiyama, E.; Shibata, I.; Baba, A. Synthesis 2000, 6, 789-800.
142
Tillack, A.; Rudloff, I.; Beller, M. Euro. J. Org. Chem. 2001, 523-528.
143
Zhang, M.; Moore, J. D.; Flynn, D. L.; Hanson, P. R. Org. Lett. 2004, 6, 2657-2660.
144
Verras, A.; Kuntz, I. D.; Montellano, P. R. O. J. Med. Chem. 2004, 47, 3572-3579.
145
Claramunt, R. M.; Elguero, J.; Garceran, R. Heterocycles 1985, 23, 2895-2906.

260
 References
146
(a) Matsunaga, H.; Ishizuka, T.; Kunieda, T. Tetrahedron Lett. 2005, 46, 3645-3648. (b) Coulbeck, E.; Eames,
J. Tetrahedron Asymm. 2007, 18, 2313-2325.
147
When the enantiomerically rich chlorides treated with general flash column workup procedure (Merck KGaA
Silica Gel 60, EtOAc/n-Hexane=4:1 as eluent), exothermic was observed once the concentrated crude
chloride was contacted with the silica. The chiral GC results showed that the chloride had been decomposed
back into initial alcohol after the column, with loss of ee.
148
Qiu, R.; Zhang, G.; Zhu, Y.; Xu, X.; Shao, L.; Li, Y.; An, D.; Yin, S. Chem. Eur. J. 2009, 15, 6488-6494.
149
Sakai, N.; Moriya, T.; Konakahara, T. J. Org. Chem. 2007, 72, 5920-5922.
150
Guo. K.; Thompson, M. J.; Chen, B. J. Org. Chem. 2009, 74, 6999-7006.
151
Curtin, D. Y.; Wilhelm, M. J. Org. Chem. 1958, 23, 9-12.
152
Ward, R. S.; Cooks, R. G.; Williams, D. H. J. Am. Chem. Soc. 1969, 91, 2727-2732.
153
Paventi, M.; Hay, A. S. J. Org. Chem. 1991, 56, 5875-5882.
154
Vowinkel, E.; Bartel, J. Chem. Ber. 1974, 1221-1227.
155
Zhang, M.; Flynn, D. L.; Hanson, P. R. J. Org. Chem. 2007, 72, 3194-3198.
156
Bhalerao, D. S.; Mahajan, U. S.; Chaudhari, K. H.; Akamanchi, K. G. J. Org. Chem. 2007, 72, 662-665.
157
Wen, X. J. Chem. Res. 2005, 10, 663-664.
158
Dai, H.; Liu, W.; Xu, H.; Yang, L.; Lu, M.; Zheng, Y. Chem. Pharm. Bull. 2009, 57, 84-86.
159
Baggaley, K. H.; Fears, R.; Hindley, R. M.; Morgan, B.; Murrell, E.; Thorne, D. E. J. Med. Chem. 1977, 20,
1388-1393.
160
Kuwano, R.; Kusano, H. Org. Lett. 2008, 10, 1979-1982.
161
Fife, T. H.; De, N. C. J. Am. Chem. Soc. 1974, 96, 6158-6165.
162
Kunishima, M.; Nakata, D.; Tanaka, S.; Hioki, K.; Tani, S. Tetrahedron 2000, 56, 9927-9935.
163
Hein, J. E.; Tripp, J. C.; Krasnova, L.; Sharpless, K. B.; Fokin, V. V. Angew. Chem. Int. Ed. 2009, 48, 8018-
8021.
164
Gans, P.; Gill, J. B.; Maclnnes, Y. M.; Reyner, C. Spectrochim. Acta 1986, 42A, 1349-1354.
165
For the details of pressure NMR tube, see: http://www.wilmad-labglass.com/group/2025
166
Information about Scientist 3.0 TM: www.sciencesoftware.com/pages.php?pageid=137
Information about Berkeley Madonna: www.berkeleymadonna.com
167
Gansow, D. A.; Burke, A. R.; La Mar, G. N. J. Chem. Soc., Chem. Commun. 1972, 456-457.
168
For examples, see: (a) Billiau, F; Belloni, J. Nature 1976, 263, 47-48. (b) Lwenje, S.; Tuttle, Jr., T. R.;
Golden. S. J. Phys. Chem. 1984, 88, 3799-3804.
169
For examples, see: (a) Norris, A. R. Can. J. Chem. 1967, 45, 2703-2709. (b) Sauer, A.; Wasgestian, F.;
Barabasch, B. J. Chem. Soc., Perkin Trans. 2 1990, 1317-1320. (c) Millot, F.; Terrier, F. Bull. Soc. Chim. Fr.
1969, 2692. (d) Sekiguchi, S.; Fujisawa, S.; Ando, Y. Bull. Chem. Soc. Jpn. 1976, 49, 1451-1452. (e)
Hasegawa, Y.; Abe, T. Chem. Lett. 1972, 1, 985-988.
170
For examples, see: (a) Chudek, J. A.; Foster, R. J. Chem. Soc., Perkin Trans. 2 1979, 628-633. (b) Chudek, J.
A.; Foster, R. J. Chem. Soc., Perkin Trans. 2 1982, 511-512. (c) Chudek, J. A.; Ellingham, R. A.; Foster, R.

261
 References

J. Chem. Soc., Perkin Trans. 2 1985, 1477-1978. (d) Chudek, J. A.; Foster, R.; Marr, A. W. J. Chem. Soc.,
Perkin Trans 2, 1987, 1341-1344.
171
For examples, see: (a) Bernasconi, C. F.; Muller, M. C.; Schmid, P. J. Org. Chem. 1979, 44, 3189-3196. (b)
Miller, R. E.; Wynne-Jones, W. F. K. J. Chem. Soc. 1959, 2375-2384. (c) Foster, R.; Mackie, R. K.
Tetrahedron 1961, 16, 119-129. (d) Foster, R.; Mackie, R. K. Tetrahedron 1962, 18, 161-168. (e) Bernasconi,
C. F. J. Org. Chem. 1970, 35, 1214-1216.
172
Van der Plas, H. C.; Woiniak, M. Croat. Chim. Acta 1986, 59, 33-49.
173
De Bie, D. A.; Geurtsen, B.; Van der Plas, H. C. J. Org. Chem. 1985, 50, 484-487.
174
(a) Wozniak, M.; Baranski, A.; Szpakiewicz, B. Libigs. Ann. Chem. 1991, 851-856. (b) Szpakiewicz, B.;
Grzegozek, M. Russ. J. Org. Chem. 2004, 40, 869-872.
175
For details, see: (a) Lagowski, J. J. Pure Appl. Chem. 1971, 25, 429-456. (b) Cuthrell, R. E.; Fohn, E. C.;
Lagowski, J. J. Inorg. Chem. 1966, 5, 111-114.
176
For examples: (a) Kraus, C. A.; Bray, W. C. J. Am. Chem. Soc. 1913, 35, 1315–1434. (b) Badoz-lambling, J.;
Herlem, M.; Thiebault, A. Anal. Lett. 1969, 2, 35-39. (c) Birchall, T.; Jolly, W. L. J. Am. Chem. Soc. 1966,
83, 5439–5443.
177
The aqueous pKa of phenols are from: (a) Jover, J.; Bosque, R.; Sales, J. QSAR Comb. Sci. 2007, 26, 385-397.
(b) Serjeant, E. P.; Dempsey, B. Ionization Constants of Organic Acids in Aqueous Solution, Pergamon Press,
New York, 1979.
178
Klipatrick, M. Ann. NY Acad. Sci. 1949, 51, 673-679.
179
For the pKa of phenols in DMSO and acetonitrile, see: Izutsu, K. Acid-Base Dissociation Constants in Dipolar
Aprotic Solvents, Blackwell Scientific Publications, Oxford, 1990.
180
For pKa of phenols in methanol, see: Roy K.; Popelier, P. L. A. J. Phys. Org. Chem. 2009, 22, 186-196.
181
Strictly speaking, the extinction coefficient of free ion and ion pair are different, but in a negligible difference.
We found that the maximum absorbance wavelength of ion pair is slightly lower than free ion in liquid
ammonia, which is about 5-10nm, depending on the nature of phenol.
182
(a) Birchall, T; Jolly, W. L. J. Am. Chem. Soc. 1966, 88, 5439-5443. (b) Birchall, T; Jolly, W. L. J. Am.
Chem. Soc. 1965, 87, 3007-3008. (c) Birchall, T; Jolly, W. L. Inorg. Chem. 1966, 5, 2177-2179. (d)
Zoltewicz, J. A.; Helmick, L. S. J. Org. Chem. 1973, 38, 658-662.
183
Cremlyn, R. J.; Osborne, A. G.; Warmsley, J. F. Spectrochim. Acta A 1996, 52, 1423-1432.
184
Breitmaier. E; Voelter, W. Carbon-13 NMR spectroscopy, 3rd ed., Wiley-VCH Publishers, New York, 1987.
185
Zoltewicz, J. A.; O'Halloran, J. K. J. Org. Chem. 1974, 39, 89-92.
186
Crowley, P. J.; Leach, M. R.; Meth-Cohn, O.; Wakefield, B. J. Tetrahedron Lett. 1986, 27, 2909-2912.
187
Weizman, H. J. Chem. Edu. 2008, 85, 294-296.
188
Bryant, R. G. Ann. Rev. Phys. Chem. 1978, 29, 167-188.
189
For details about carbanion 13C NMR shifts, see: (a) Abbotto, A.; Bradamante, S.; Pagani, G. A. J. Org. Chem.
1993, 58, 449-455. (b) Vogt, H. H.; Gompper, R. Chem. Ber. 1981, 114, 2884-2897. (c) Barchiesi, E.;
Bradamante, S.; Ferraccioli, R.; Pagani, G. A. J. Chem. Soc., Perkin Trans. 2 1990, 375-383.
190
Gansow, D. A.; Burke, A. R.; La Mar, G. N. J. Chem. Soc., Chem. Comm. 1972, 456-457.

262
 References
191
Williams, E. A.; Cargioli, J. D.; Ewo, A. J. Chem. Soc., Chem. Comm. 1975, 366-367.
192
Bibas, H.; Moloney, D. W. J.; Neumann, R.; Shtaiwi, M.; Bernhardt, P. V.; Wentrup, C. J. Org. Chem. 2002,
67, 2619-2631.
193 1
H NMR of benzophenone imine in liquid ammonia at 25oC: δH (LNH3): 7.42-7.75 (m, 10H), 10.42 (s, 1H)
(Appendix C: Figure N27).
194
Richard, J. P.; Williams, G.; Gao, J. J. Am. Chem. Soc. 1999, 121, 715-726.
195
Bordwell, F. G. Acc. Chem. Res. 1988, 21, 456-463.
196
Bug, T.; Mayr, H. J. Am. Chem. Soc. 2003, 125, 12980-12986.
197
Bentley, T. W.; Jones, R. O.; Kang, D. H.; Koo, I. S. J. Phys. Org. Chem. 2009, 22, 799-806.
198
(a) Hyne, J. B.; Wills, R.; Wonkka, R. E. J. Am. Chem. Soc. 1962, 84, 2914-2919. (b) Tommila, E. Acta
Chem. Scand. 1966, 20, 923- 936. (c) Kwun, O. C.; Kyong, J. B.; Choi, K. J. J. Kor. Chem. Soc. 1986, 30,
301-306. (d) Kwun, O. C.; Kim, J. R.; Ryu, J. C. J. Kor. Chem. Soc. 1981, 25, 152-159.
199
Bentley, T. W.; Koo, I. S.; Norman, S. J. J. Org. Chem. 1991, 56, 1604-1609.
200
Liu, K. T.; Duan, Y. F.; Hou, S. J. J. Chem. Soc., Perkin Trans. 2 1998, 2181-2185.
201
For examples, see: (a) Chung, D. S.; Kim, C. K.; Lee, B. S.; Lee, I. Tetrahedron 1993, 49, 8359-8372. (b)
Kevill, D. N.; Ismail, N. H. J.; D’Souza, M. J. J. Org. Chem. 1994, 59, 6303-6312. (c) Amyes, T. L.; Richard,
J. P. J. Am. Chem. Soc. 1990, 112, 9507-9512.
202
For examples, see: (a) Fujio, M.; Susuki, T.; Goto, M.; Tsuji, Y.; Yatsugi, K.; Kim, S. H.; Ahmed, G. A. W.;
Tsuno, Y. Bull. Chem. Soc. Jpn. 1995, 68, 673-682. (b) Dietze, P.; Jencks, W. P. J. Am. Chem. Soc. 1989, 111,
5880-5886.
203
Markova, G. S.; Shatenshtein, A. I. Dokl. Akad. Nauk SSSR 1942, 35, 74.
204
For examples, see: (a) Liu, K. T.; Chang, L. W.; Yu, D. G.; Chen, P. S.; Fan, J. T. J. Phys. Org. Chem. 1997,
10, 879-884. (b) Shiner, Jr. V. J.; Buddenbaum, W. E.; Murr, B. L.; Lamaty, G. J. Am. Chem. Soc. 1968, 90,
418-426. (c) Traylor, T. G.; Ware, J. C. J. Am. Chem. Soc. 1967, 89, 2304-2315.
205
Scott, J. M.; Robertson, R. E. Can. J. Chem. 1972, 50, 167-175.
206
Page, M. I. Chem. Soc. Rev. 1973, 2, 295-323.
207
Smyth, C. P. Dielectric Behaviour and Structure, McGraw-Hill Book Co. Inc., New York, 1955.
208
The second rate constant for 4-nitrofluorobenzene react with sodium azide in liquid ammonia is similar to
those observed in some typical dipolar aprotic solvents, for details see: Appendix A: Table A37.
209
March, J. Advanced Organic Chemistry, 4th ed., Wiley-Interescience, New York, 1992, page 261, and the
references cited within.
210
Bentley, T. W.; Koo, I. S.; Choi, H.; Llewellyn, G. J. Phys. Org. Chem. 2008, 21, 251-256.
211
Ingold, C. K. Structure and Mechanism in Organic Chemistry, Cornell University Press, Ithaca, New York,
1953, 346-350.
212
For examples, see: (a) Hojo, M.; Ueda, T.; Ueno, E.; Hamasaki, T.; Fujimura, D. Bull. Chem. Soc. Jpn. 2006,
79, 751-760. (b) Robertson, R. E. Can. J. Chem. 1953, 31, 589-601.
213
Hojo, M.; Ueda, T.; Inoue, S.; Kawahara, Y. J. Chem. Soc., Perkin Trans. 2 2000, 1735-1740.
214
Hansch, C.; Leo, A; Taft, R. W. Chem. Rev. 1991, 91, 165-195.

263
 References
215
Dewick, P. M. Essentials of Organic Chemistry: For Students of Pharmacy, Medicinal Chemistry and
Biological Chemistry, John Wiley&Sons Ltd., West Sussex, 2006.
216
Kornblum, N.; Ackermann, P.; Manthey, J. W.; Musser, M. T.; Pinnick, H. W.; Singaram, S.; Wade, P. A. J.
Org. Chem. 1988, 53, 1475-1481.
217
Hughes, E. D.; Ingold, C. K.; Scott, A. D. J. Chem. Soc. 1937, 1201-1208.
218
For examples, see: (a) Flores-Parra, A.; Suarez-Moreno, P.; Sanchez-Ruiz, S. A.; Tlahuextl, M.; Jaen-Gaspar,
J.; Tlahuext, H.; Salas-Coronado, R.; Cruz, A.; Noth, H.; Contreras, R. Tetrahedron Asymm. 1998, 9, 1661-
1671.[20% ee] (b) Durrat, F.; Sanchez, M. V. ; Couty, F.; Evano,G. ; Marrot, J. Eur. J. Org. Chem. 2008, 19,
3286-3297. (c) Ashby, E. C.; DePriest, R. N.; Goel, A. B.; Wenderoth, B.; Pham, T. N. J. Org. Chem. 1984,
49, 3545-3556.[57% ee] (d) Mangold J. B.; Abdel-Monem, M. M. J. Med. Chem. 1983, 26, 66-71.[79.8% ee,
but a different synthetic route]
219
(a) Ji, P.; Atherton, J. H.; Page, M. I. J. Chem. Soc., Faraday Discuss. 2010, 145, 15-25. (b) Ji, P.; Atherton, J.
H.; Page, M. I. J. Org. Chem. 2011, 76, 1425-1435.
220
The rate enhancement, with increasing of water content in aqueous binary organic solvents is far more
pronounced than in aqueous liquid ammonia solution, which indicates a change of reaction mechanism from
bimolecular to unimolecular. For details, see: Anantaraman, R.; Saramma, K. Can. J. Chem. 1967, 4, 1770-
1777.
221
For examples, see: (a) Marcus, Y.; Kamlet, M. J.; Tafts, R. W. J. Phys. Chem. 1988, 92, 3613-3622. (b)
Parker, A. J. Quart. Rev. Chem. Soc. 1962, 16, 163-187.
222
(a) Markova, G. S.; Shatenshtein, A. I. Compt. Rend. Acad. Sci., U. R. S. S. 1942, 35, 68-70. (b) Glukhovstev,
M. N.; Bach, R. D.; Laiter, S. J. Org. Chem. 1997, 62, 4036-4046.
223
Vlasov, V. M. Russ. Chem. Rev. 2003, 72, 681-703.
224
For examples, see: (a) Bevan, C. W. L.; Bye, G. C. J. Chem. Soc. 1951, 3091-3094. (b) Bolto, B. A.; Miller, J.
Aust. J. Chem. 1956, 9, 74-82. (c) Bamkole, T. O.; Hirst, J.; Udoessien, I. J. Chem. Soc., Perkin Trans. 2 1972,
110-114. (d) Bamkole, T. O.; Hirst, J.; Udoessien, I. J. Chem. Soc., Perkin Trans. 2 1973, 2114-2119. (e)
Pietra, F.; Del Cima, F. J. Org. Chem. 1968, 33, 1411-1416. (f) Pietra, F.; Del Cima, F. Tetrahedron Lett.
1966, 37, 4453-4457.
225
Bunnett, J. F.; Morath, R. J. J. Am. Chem. Soc. 1955, 77, 5051-5055.
226
Bolto, B. A.; Miller, J.; Williams, V. A. J. Chem. Soc. 1955, 2926-2929.
227
Bondi, A. J. Phys. Chem. 1964, 68, 441-451.
228
Wang, X.; Salaski, E. J.; Berger, D. M.; Powell, D. Org. Lett. 2009, 11, 5662–5664.
229
For examples, see: (a) Bolton, R.; Kazeroonian, S. M.; Sandall, J. P. B. J. Fluorine Chem. 8, 1976, 471-480.
(b) Nishida, S. J. Org. Chem. 1967, 32, 2695-2697. (c) Bolton, R.; Sandall, J. P. B. J. Chem. Soc., Perkin
Trans. 2 1978, 141-144.
230
A redundant process of ICI in mid-1980s, personal communications with John H Atherton. The condition for
4-nitrochlorobenzene to be able to react with 40 wt% aq. ammonia was in a pipe reactor at 235oC, 214 atm, 40
mins to the completion.
231
Blumenstein, J. J.; Ukachukwu, V. C.; Mohan, R. S.; Whaled, D. L. J. Org. Chem. 1993, 58, 924-932.

264
 References
232
At 25oC, hydrolysis rate of benzenesulfonyl chloride in water at 25 oC is 3.07 × 10-3 s-1, for details, see:
Bentley, T. W.; Jones, R. O.; Kang, D. H.; Koo, I. S. J. Phys. Org. Chem. 2009, 22, 799-806.
233
Strain, H. H. J. Am. Chem. Soc. 1930, 52, 820-822. (180oC, 4 hours, large excess of NH4Cl, only 20-30%
yields for acetophenone and benzophenone)
234
Personal communications with John H Atherton. The reaction temperature was 100 oC with TiO2 as catalyst in
a flow reactor, and the conversion rate can reach as high as 99% for benzophenone within hours.
235
For examples, see: (a) Westaway, K. C. Can. J. Chem. 1978, 56, 2691-2699. (b) Westaway, K. C.; Lai, Z.
Can. J Chem. 1987, 67, 345-349. (c) Fang, Y.; MacMillar, S.; Eriksson, J.; Kolodziejska-Huben, M.; Dybala-
Defratyka, A.; Paneth, P.; Matsson, O.; Westaway, K. C. J. Org. Chem. 2006, 71, 4742-4747.
236
For the HSAB theory, see: Pearson, R. G. J. Am. Chem. Soc. 1963, 85, 3533-3539.
237
Parker, A. J. Rates of Bimolecular Substitution Reactions in Protic and Dipolar Aprotic Solvents, Gold, V.
Ed., Advances in Physical Organic Chemistry, Academic Press, London, 1967, Vol. 5.
238
Extrapolated data, see: Vlasov, V. M. Russ. Chem. Rev. 2006, 75, 765-796.
239
Tsai, Y. H. Nucleophilic displacement reactions of benzyl chlorides, Ph.D thesis, University of Ottawa, 1968.
240
Fang, Y.; Lai, Z.; Westaway, K. C. Can. J. Chem. 1998, 76, 758-764.
241
For examples, see: (a) Parker, A. J. Chem. Rev. 1969, 69, 1-32. (b) Miller, J.; Parker, A. J. J. Am. Chem. Soc.
1961, 83, 117-123.
242
For details, see: (a) R. Breslow, K. Groves and M. U. Mayer, Org. Lett. 1999, 1, 117-120. (b) Madhavan, K.;
Srinivasan, V. S.; Venkatasubramanian, N. Proc. Indian Aead. Sci. 1979, 88, 329-335.
243
Ballinger, P.; Long, F. A. J. Am. Chem. Soc. 1960, 82, 795-798.
244
βnuc=0.28 in methanol (r2=0.9310) and 0.17 in 2-propanol/water [(%v/v) =1:1] (r2=0.9859) system at 25oC. For
details, see: ref. 242b.
245
For examples, see: (a) Fuchs, R.; Carlton, D. M. J. Org. Chem. 1962, 27, 1520-1523. (b) Young, P. R.; Jencks,
W. P. J. Am. Chem. Soc. 1979, 101, 3288-3294. (c) Tsuno, Y.; Fujio, M. The Yukawa-Tsuno Relationship in
Carbocationic Systems, Bethell, D. Ed., Advances in Physical Organic Chemistry, Elsevier Scientific Inc.,
1999, Vol. 32, 267-385.
246
The Hammett σ constant for 3-methoxy group is 0.12, and this value is used as σp here. For details, see:
Dolman, D.; Stewart, R. Can. J. Chem. 1967, 45, 903-910.
247
(a) Kornblum N.; Lurie, A. P. J. Am. Chem. Soc. 1959, 81, 2705-2715. (b) Kornblum, N.; Seltzer R.;
Haberfield, P. J. Am. Chem. Soc. 1963, 85, 1148-1154.
248
Tehan, B. G.; Lloyd, E. J.; Wong, M. G.; Pitt, W. R.; Gancia, E.; Manallack, D. T. Quant. Struct. Act. Relat.
2002, 21, 473-485.
249
For examples, see: (a) Wamhoff, H. Comprehensive Heterocyclic Chemistry, Katritzky, A. R.; Rees, C. W.
Eds., Pergamon Press, Oxford, 1984, Vol. 5. (b) Poyla, J. P. Comprehensive Heterocyclic Chemistry,
Katritzky, A. R.; Rees, C. W. Eds., Pergamon Press, Oxford, 1984, Vol. 5. (c) Kartritzky, A. R.; Chang, H.
X.; Yang, B. Synthesis 1995, 5, 503-505.
250
For examples of 1,2,4-triazolate as an ambident nucleophile, see: (a) Potts, K. T. Chem. Rev. 1961, 61, 87-127.
(b) Bentley, T. W.; Jones, R. V. H.; Wareham, P. J. Tetrahedron Lett. 1989, 30, 4013-4016. (c) Sano, S.;

265
 References

Tanba, M.; Nagao, Y. Heterocycles 1984, 38, 481-486. For examples of benzotriazolate as an ambident
nucleophile, see: Le, Z. G.; Chen, Z. C.; Hu, Y.; Zheng, Q. G. J. Chem. Res. 2004, 344-346.
251
(a) Begtrup, M.; Larson, P. Acta Chem. Scand. 1990, 44, 1050-1057. (b) Smith, K.; Small, A.; Hutchings, M.
Chem. Lett. 1990, 19, 347-350.
252
Buncel, E.; Um, I. K. Tetrahedron 2004, 60, 7801-7825, and the references cited within.
253
Bentley, T. W.; Howle, L. M.; Wareham, P. J.; Jones, R. V. H. Tetrahedron 1992, 48, 7869-7878.
Also a comparison of this reaction in different solvents was investigated, a selectivity of 10:1 between 7 and 8
in DMF at 85oC, 4:1 in methanol under reflux condition were found.
254
For examples, see: (a) Ko, E. C. F.; Parker, A. J. J. Am. Chem. Soc. 1968, 90, 6447-6543. (b) Ritter, R. D.;
Krueger, J. H. J. Am. Chem. Soc. 1970, 92, 2316-2321.
255
(a) Bordwell, F. G.; Hughes, D. L. J. Am. Chem. Soc.1986, 108, 5991-5997. (b) Bordwell, F. G.; Hughes, D.
L. J. Org. Chem. 1980, 45, 3314-3320. (c) Bordwell, F. G.; Hughes, D. L. J. Am. Chem. Soc.1984, 106, 3234-
3240. (d) Vlasov, V. M.; Os’kina, I. A. Russ. J. Org. Chem. 2002, 38, 1767-1780.
256
Kraus, C. A.; White, G. F. J. Am. Chem. Soc. 1922, 45, 768-778.
257
Bordwell, F. G.; Cripe, T. A.; Hughes, D. L. Nucleophilicity, Harris, J. M.; McManus, S. P. Eds., Advances in
Chemistry, American Chemical Society, Washington, DC, 1987.
258
(a) Kresge, A. J. Can. J. Chem. 1974, 52, 1897-1903. (b) Kresge, A. J.; Drake, D. A.; Chiang, Y. Can. J.
Chem. 1974, 52, 1889-1896.
259
(a) Walters, E. A.; Long, F. A. J. Am. Chem. Soc. 1969, 91, 3733-3739. (b) Hibbert, F; Long, F. A.; Walters,
E. A. J. Am. Chem. Soc. 1971, 93, 2829-2835. (c) Hibbert, F; Long, F. A. J. Am. Chem. Soc. 1971, 93, 2836-
2840.
260
Jobe, D. J.; Westaway, K. C. Can. J. Chem. 1993, 71, 1353-1361.
261
Extrapolated data under heterogeneous condition in ionic liquid solvent, for details, see: Wheeler, C.; West,
K. N.; Liotta, C. L.; Eckert, C. A. Chem. Commun. 2001, 887-888.
262
Bug, T.; Lemek, T.; Mayr, H. J. Org. Chem. 2004, 69, 7565-7576.
263
(a) Galli, C.; Pau, T. Tetrahedron 1998, 54, 2893-2904. (b) Wiberg, K. B.; Rablen, P. R. J. Org. Chem. 1998,
63, 3722-3730.
264
Kingsbury, C. A. J. Org. Chem. 1964, 29, 3262-3270.
265
(a) Bevan, C. W. L.; Foley, A. J.; Hirst, J.; Uwanu, W. O. J. Chem. Soc. B 1970, 794-797. (b) Annulli, A.;
Mencarelli, P.; Stegel, F. J. Org. Chem. 1988, 49, 4065-4067.
266
(a) Bartoli, G.; Todesco, P. E. Tetrahedron Lett. 1968, 9, 4867-4870. (b) Bartoli, G.; Ciminale, F.; Todesco,
P. E. J. Org. Chem. 1975, 40, 872-874.
267
For examples, see: (a) Bowden, K.; Cook, R. S. Tetrahedron Lett. 1970, 11, 249-250. (b) Shtark, A. A.;
Kizner, T. A.; Shteingarts, V. D. Russ. J. Org. Chem. 1981, 18, 2321-2327.
268
Ho, K. C.; Miller, J.; Wong, K. W. J. Chem. Soc. B 1966, 310-314.
269
Page. M. I.; Jencks, W. P. Proc. Nat. Acad. Sci. 1971, 68, 1678-1683.
270
Nudelman, N. S. SNAr Reactions of Amines in Aprotic Solvents, Patai, S. Ed., The Chemistry of Amino,
Nitroso, Nitro and Related Groups, Supplement F2, John Wiley & Sons, New York, 1996, chapter 26.

266
 References
271
Lagowski, J. J.; Moczeygemba, G. A. Acidic and Basic Solvents, Lagowski, J. J. Ed., The Chemistry of Non-
Aqueous Solution, Academic Press, 1967, Vol. 2, chapter 7.
272
At 25oC, the second order rate constant for the reaction between 4-nitrochlorobenzene and sodium phenoxide
in liquid ammonia at 25oC is 2.51×10-7 M-1s-1.
273
The second order rate constant for 2-NFB with phenoxide in liquid ammonia at 25oC is 9.95×10-2 M-1s-1.
274
For examples, see: (a) Durantini, E.; Zingaretti, L.; Anunziata, J. D.; Silber, J. J. J. Phys. Org. Chem. 1992, 5,
577-566. (b) Nudelman, N. S.; Silvana Alvaro, C. E.; Savini, M.; Niotra, V.; Yankelevich, J. Coll. Czech.
Chem. Commun. 1998, 64, 1583-1593. (c) Bamkole, T. O.; Hirst, J.; Onyido, I. J. Chem. Soc., Perkin Trans. 2
1982, 889-893. (d) Hirst, J.; Onyido, I. J. Chem. Soc., Perkin Trans. 2 1984, 711-715. (e) Crampton, M. R.;
Emokpae, T. A.; Isanbor, C. Euro. J. Org. Chem. 2007, 1378-1383. (f) Isanbor, C.; Emokpae, T. A. Int. J.
Chem. Kinet. 2008, 40, 125-135. (g) Akinyele, E. T.; Onyido, I. J. Chem. Soc., Perkin Trans. 2 1988, 1859-
1861.
275
(a) Mancini, P. M. E.; Martinez, R. D.; Vottero, L. R.; Nudelman, N. S. J. Chem. Soc., Perkin Trans. 2 1984,
1133-1138. (b) Mancini, P. M. E.; Martinez, R. D.; Vottero, L. R.; Nudelman, N. S. J. Chem. Soc., Perkin
Trans. 2 1987, 951-954.
276
For examples, see: (a) Parker, A. J. Chem. Rev. 1969, 69, 1-32. (b) Parker, A. J. Quart. Rev. Chem. Soc. 1962,
16, 163-187. (c) Miller, J.; Parker, A. J. J. Am. Chem. Soc. 1961, 83, 117-123. (d) Suhr, H. Liebigs Ann.
Chem. 1967, 701, 101-106.
277
Coetzee, J. F; Padmanabhan, G. R. J. Am. Chem. Soc. 1965, 87, 5005-5010.
278
For the details of Brønsted βnuc value for the reactions with a SET mechanism, see: Terrier, F.; Mokhtari, M.;
Goumont, R.; Halle, J.; Buncel, E. Org. Biomol. Chem. 2003, 1, 1757-1763.
279
The decomposition of 4-NAB using 15NH4Cl in liquid ammonia (50% 15N enriched NH3) shows that the
nitrogen gas released from the reaction is not 15N enriched, nut the content of 15N from the gas sample is the
same as that found in nitrogen gas with a natural abundance of 15N. The gas analysis was performed by HR
GC-MS, with manual injection of the gas samples.
280
The oxidative decomposition of aromatic hydrazine is known, for details, see: (a) Stroh, H.; Ebert, B. Chem.
Ber. 1964, 97, 2335-2341. (b) Hoffman, R. V.; Kumar, A. J. Org. Chem. 1984, 49, 4014-4017. (c) Itoh, T.;
Matsuya, Y.; Nagata, K.; Ohsawa, A. Tetrahedron Lett. 1997, 38, 4117-4120.
281
For examples, see: (a) Gritsan, N. P.; Pritchina, E. A. Russ. Chem. Rev. 1992, 61, 500-516. (b) Budyka, M. F.
Russ. Chem. Rev. 2008, 77, 709-723. (c) Herbranson, D. E.; Hawley, M. D. J. Org. Chem. 1990, 55, 4297-
4303.
282
For examples, see: (a) Liang, T. Y.; Schuster, G. B. J. Am. Chem. Soc. 1986, 108, 546-548. (b) Liang, T. Y.;
Schuster, G. B. J. Am. Chem. Soc. 1987, 109, 7803-7810. (c) Albini, A.; Bettinetti, G.; Minoli, G. J. Chem.
Soc., Perkin Trans. 2 1999, 2803-2807. (d) Liang, T. Y.; Schuster, G. B. Tetrahedron Lett. 1986, 26, 3325-
3328. (e) Odum, R. A.; Wolf, G. J. Chem. Soc., Chem. Comm. 1973, 360-361. (f) Odum, R. A.; Aaronson, A.
M. J. Am. Chem. Soc. 1969, 91, 5680-5681.
283
Bordwell, F. G.; Hughes, D. L. J. Am. Chem. Soc. 1986, 108, 5991-5997.
284
Bartoli, G.; Todesco, P. E. Acc. Chem. Res. 1977, 10, 125-132.

267
 References
285
(a) Vogt, P. F.; Gerulis, J. J. Amines, Aromatic, Ullmann's Encyclopedia of Industrial Chemistry, Wiley-VCH,
Weinheim, 2005. (b) Hunger, K. Industry Dyes: Chemistry, Properties and Applications, Wiley-VCH,
Weinheim, 2003. (c) Rappoport, Z. Chemistry of Anilines, Part 1, in Patai Series: The Chemistry of
Functional Groups, Ed., Rappoport, Z., John Wiley & Sons Ltd, Chichester, West Sussex, 2007. (d)
Lawrence, S. A. Amines: Synthesis, Properties and Application, Cambridge University Press, Cambridge,
2004.
286
For examples, see: (a) Jaime-Figueroa, S.; Liu, Y.; Muchowski, J. M.; Putman, D. G. Tetrahedron Lett. 1998,
39, 1313-1316. (b) Mann, G.; Hartwig, J. F.; Driver, M. S.; Fernandez-Rivas, C. J. Am. Chem. Soc. 1998,
120, 827-828. (c) Kiyomori, A.; Marcoux, J.; Buchwald, S. L. Tetrahedron Lett. 1999, 40, 2657-2660. (d)
Crasa, G. A.; Viciu, M. S.; Huang, J.; Nolan, S. P. J. Org. Chem. 2001, 66, 7729-7737. (e) Barluenga, J.;
Aznar, F.; Valdes, C. Angew. Chem. Int. Ed. 2004, 43, 343-345. (f) Xu, L.; Zhu, D.; Wu, F.; Wang, R.; Wan,
B. Tetrahedron, 2005, 61, 6553-6560. (g) Altman, R. A.; Fors, B. P.; Buchwald, S. L. Nature Protocols 2007,
2, 2881-2887. (h) Ogata, T.; Hartwig, J. F. J. Am. Chem. Soc. 2008, 130, 13848-13849. (i) Zhao, H.; Qiao, R.
J. Org. Chem. 2010, 75, 3311-3316.
287
(a) Widenhoefer, R. A.; Buchwald, S. L. Organometallics 1996, 15, 2755-2763. (b) Widenhoefer, R. A.;
Buchwald, S. L. Organometallics 1996, 15, 3534-3542. (c) Paul, F.; Patt, J.; Hartwig, J. F. Organometallics
1995, 14, 3030-3039.
288
(a) Lang, F.; Zewge, D.; Houpis, I. N.; Volante, R. P. Tetrahedron Lett. 2001, 42, 3251-3254. (b) Willis, M.
C. Angew. Chem. Int. Ed. 2007, 46, 3402-3404. (c) Shen, Q.; Hartwig, J. F. J. Am. Chem. Soc. 2006, 128,
10028-10029.
289
For examples: (a) Wolfe, J. P.; Ahman, J.; Sadighi, J. P.; Singer, R. A.; Buchwald, S. L. Tetrahedron Lett.
1997, 38, 6367-6370. (b) Wolfe, J. P.; Tomori, H.; Sadighi, J. P.; Yin, J. J.; Buchwald, S. L. J. Org. Chem.
2000, 65, 1158-1174. (c) Huang, X. H.; Buchwald, S. L. Org. Lett. 2001, 3, 3417-3419. (d) Huang, X. H.;
Anderson, K. W.; Zim, D.; Jiang, L.; Klapars, A.; Buchwald, S. L. J. Am. Chem. Soc. 2003, 125, 6653-6655.
(e) Lee, S.; Jorgensen, M.; Hartwig, J. F. Org. Lett. 2001, 3, 2729-2732. (f) Lee, D. Y.; Hartwig, J. F. Org.
Lett. 2005, 7, 1169-1172.
290
For examples: (a) Surry, D. S.; Buchwald, S. L. J. Am. Chem. Soc. 2007, 129, 10354-10355. (b) Vo, G. D.;
Hartwig, J. F. J. Am. Chem. Soc. 2009, 131, 11049-1106. (c) Wang, D.; Cai, Q.; Ding, K. Adv. Synth. Catal.
2009, 351, 1722-1726. (d) Schulz, T.; Torborg, C.; Enthaler, S.; Schaffner, B.; Dumrath, A.; Spannenberg,
A.; Neumann, H.; Borner, A.; Beller, M. Chem. Eur. J. 2009, 15, 4528-4533. (e) Guo, Z.; Guo, J.; Song, Y.;
Wang, L.; Zou, G. Appl. Organometal. Chem. 2009, 23, 150-153. (f) Xia, N.; Taillefer, M. Angew. Chem.
2009, 121, 343-345. (g) Wu, Z.; Jiang, Z.; Wu, D.; Xiang, H.; Zhou, X. Eur. J. Org. Chem. 2010, 1854-1857.
(h) Jiang, L.; Lu, X.; Zhang, H.; Jiang, Y.; Ma, D. J. Org. Chem. 2009, 74, 4542-4546.
291
Similar reaction in other solvent system requires 5-20mole% copper catalyst together with a variety of
ancillary ligands, for selected examples, see: (a) Lu, Z.; Twieg, R. J. Tetrahedron Lett. 2005, 46, 2997-3001..
(b) Xu, H.; Wolf, C. Chem. Commun. 2009, 1715-1717. (c) Kim, J.; Chang, S. Chem. Commun. 2008, 3052-
3054. (d) Cortes-Salva, M.; Nguyen, B.; Cuevas, J.; Pennypacker, K. R.; Antilla, J. C. Org. Lett. 2010, 12,
1316-1319. (e) Surry, D. S.; Buchwald, S. L. Chem. Sci. 2010, 1, 13-31. (f) CuO nanoparticle (1.26 mole%)

268
 References

catalysed C-N cross coupling of amines with iodobenzene was also reported, for details, see: Rout, L.; Jammi,
S.; Punniyamurthy, T. Org. Lett. 2007, 9, 3397-3399. (g) The conversion of bromopyridine into
aminopyridine under 0.5 mole% Cu2O as catalyst in ethyl glycol solution of ammonia was also reported, but
the higher reaction temperature is required (80oC), for details, see: Lai, F.; Zewge, D.; Houpis, I. N.; Volante,
R. P. Tetrahedron Lett. 2001, 42, 3251-3254.
292
(a) Davies, M. B. Polyhedron 1992, 11, 285-321. (b) Creutz, C. Inorg. Chem. 1981, 20, 4449-4452.
293
There are also some ligand-free copper catalysed amination reactions, for examples, see: (a) Wu, X.; Darcel,
C. Eur. J. Org. Chem. 2009, 4753-4756. (b) Kantam, M. L.; Rao, B. P. C.; Choudary, B. M.; Reddy, R. S.
Synlett 2006, 14, 2195-2198. (c) Chang, J. W. W.; Xu, X.; Chan, P. W. H. Tetrahedron Lett. 2007, 48, 245-
248. (d) Sperotto, E.; Vries, J. G.; Klink, G. P. M.; Koten, G. Tetrahedron Lett. 2007, 48, 7366-7370. (f) Zhu,
L.; Guo, P.; Li, G.; Lan, J.; Xie, R.; You, J. J. Org. Chem. 2007, 72, 8535-8538.
294
For examples, see: (a) Klapars, A.; Antilla, J. C.; Huang, X. H.; Buchwald, S. L. J. Am. Chem. Soc. 2001, 123,
7727-7729. (b) Philips, D. P.; Zhu, X.; Lau, T.; He, X.; Yang, K.; Liu, H. Tetrahedron Lett. 2006, 47, 7137-
7296. (c) Liu, Y.; Bai, Y.; Zhang, J.; Li, Y.; Jiao, J.; Qi, X. Eur. J. Org. Chem. 2007, 6084-6088. (d) Alcalde,
E.; Dinarès, I.; Rodríguez, S.; Miguel, C. G. Eur. J. Org. Chem. 2005, 1637-1643. (e) Altenhoff, G.; Glorius,
F. Adv. Synth. Catal. 2004, 346, 1661-1664. (f) Hosseinzadeh, R.; Sarrafi, Y.; Mohadjerani, M.;
Mohammadpourmir, F. Tetrahedron Lett. 2008, 49, 840-843. (g) Zheng, N.; Buchwald, S. L. Org. Lett. 2007,
9, 4749-4751. (h) Viirre, R. D.; Evindar, G.; Batev, R. A. J. Org. Chem. 2008, 73, 3452-3459. (i) Barros, O.
S. R.; Nogueira, C. W.; Stangherlin, E. C.;. Menezes, P. H; Zeni, G. J. Org. Chem. 2006, 73, 1552-1557.
295
For examples, see: (a) Kwong, F. Y.; Klapars, A.; Buchwald, S. L. Org. Lett. 2002, 4, 581-584. (b) Job, G. E.;
Buchwald, S. L. Org. Lett. 2002, 4, 3703-3706.
296
For examples, see: (a) Ma, D.; Zhang, Y.; Zhao, J.; Wu, S.; Tao, F. J. Am. Chem. Soc. 1998, 120, 12459-
12467. (b) Ma, D.; Cai, Q. Acc. Chem. Res. 2008, 41, 1450-1460. (c) Yeh, V. S. C.; Wiedeman, P. E.
Tetrahedron Lett. 2006, 47, 6011-6016. (d) Guo, L.; Li, B.; Huang, W.; Pei, G.; Ma, D. Synlett 2008, 1833-
1837. (e) Deng, W.; Wang, Y.; Zou, Y.; Liu, L.; Guo, Q. Tetrahedron Lett. 2004, 45, 2311-2315.
297
Jiang, B.; Tian, H.; Huang, Z.; Xu, M. Org. Lett. 2008, 10, 2737-2740.
298
For examples, see: (a) Goodbrand, B.; Hu, N. J. Org. Chem.1999, 64, 670-674. (b) Altman, R. A.; Koval, E.
D.; Buchwald, S. L. J. Org. Chem. 2007, 72, 6190-6199. (c) Hosseinzadeh, R.; Tajbakhsh, M.; Mohadjerani,
M.; Mehdinejad, H. Synlett, 2004, 1517-1520. (d) Altman, R. A.; Buchwald, S. L. Org. Lett. 2007, 9, 643-
646. (e) Altman, R. A.; Buchwald, S. L. Org. Lett. 2006, 8, 2779-2782.
299
For examples, see: (a) Andersen, J.; Madsen, U.; Bjorkling, F.; Liang, X. Synlett 2005, 14, 2209-2213. (b)
Goriya, Y.; Ramana, C. V. Tetrahedron 2010, 66, 7642-7650.
300
(a) Strieter, E. R.; Blackmond, D. G.; Buchwald, S. L. J. Am. Chem. Soc. 2005, 127, 4120-4121. (b) Strieter,
E. R.; Bhayana, B.; Buchwald, S. L. J. Am. Chem. Soc. 2009, 131, 78-88.
301
Amatore, C.; Pfluger, F. Organometallics 1990, 9, 2276-2282.
302
Marcus, Y. Pure Appl. Chem. 1983, 55, 977-1021.

269
 References
303
For details of stability constant of copper (I) ion in aqueous ammonia, see: (a) Bjerrum, J. Acta Chem. Scand.,
Ser. A 1986, 40, 233-235. (b) Braish, T. F.; Duncan, R. E.; Harber, J. J.; Steffen, R. L.; Stevenson, K. L.
Inorg. Chem.1984, 23, 4072-4075. (c) Horváth, O.; Stevenson, K. L. Inorg. Chem. 1989, 28, 2548-2551.
304
Lamble, G.; Moen, A.; Nicholson, D. G. J. Chem. Soc., Faraday Trans. 1994, 90, 2211-2213.
305
Nilsson, K. B.; Persson, I. J. Chem. Soc., Dalton Trans. 2004, 1312-1319.
306
kobs = 5.0× 10-6s-1 for the copper (I) catalysed amination of iodobenzene in 30% w/w aqueous ammonia at
25oC. reaction conditions: 0.5mmol iodobenzene, 1%mole CuI, 1eq ascorbic acid to the copper catalyst, in
10ml 30% w/w aqueous ammonia solution.
307
(a) Zhang, S.; Liu, L.; Fu, Y.; Guo, Q. X. J. Mol. Struct. (Theochem) 2005, 757, 37-46. (b) Turner, R. W.;
Amma, E. L. J. Am. Chem. Soc. 1966, 88, 1877-1882. (c) Rodesiler, P. F.; Amma, E. L. J. Chem. Soc., Chem.
Comm. 1974, 599-600.
308
(a) Tye, J. W.; Weng, Z.; Johns, A. M.; Incarvito, C. D.; Hartwig, J. F. J. Am. Chem. Soc. 2008, 130, 9971-
9983. (b) Zhang, S.; Liu, L.; Fu, Y.; Guo, Q. Organometallics 2007, 26, 4546-4554.
309
(a) Tye, J. W.; Weng, Z.; Johns, A. M.; Incarvito, C. D.; Hartwig, J. F. J. Am. Chem. Soc. 2008, 130, 9971-
9983. (b) Zhang, S.; Liu, L.; Fu, Y.; Guo, Q. Organometallics 2007, 26, 4546-4554.
310
Jiao, J.; Zhang, X.; Chang, N.; Wang, J.; Wei, J.; Shi, X.; Chen, Z. J. Org. Chem. 2011, 76, 1180-1183.
311
For examples: (a) Zhang, R.; Li, X.; Zhang, X. Chin. J. Chem. Phys. 2009, 22, 235-240. (b) Blanksby, S. J.;
Ellison, G. B. Acc. Chem. Res. 2003, 36, 255-263.
312
(a) Michael, A. J. Prakt. Chem. 1893, 48, 94-95. (b) Huisgen, R. Angew. Chem. Int. Ed. 1963, 2, 565-598. (c)
Huisgen, R. 1,3-Dipolar Cycloaddition Chemistry, Ed. Padwa, A., Wiley, New York, 1984, 1-176.
313
Rostovtsev, V. V.; Green, L. G.; Fokin,V. V.; Sharpless, K. B. Angew. Chem. Int. Ed. 2002, 41, 2596-2599.
314
Tornoe, C.; Christensen, C.; Meldal, M. J. Org. Chem. 2002, 67, 3057-3064.
315
For the recent reviews of CuAAC, see: (a) Mamidyala, S. K.; Finn, M. G. Chem. Soc. Rev. 2010, 39, 1252-
1261. (b) Hua, Y.; Flood, A. H. Chem. Soc. Rev. 2010, 39, 1262-1271. (c) Le Droumaguet, C.; Wang, C.;
Wang, Q. Chem. Soc. Rev. 2010, 39, 1233-1239. (d) Hanni, K. D.; Leigh, D. A. Chem. Soc. Rev. 2010, 39,
1240-1251. (e) Hein, J. E.; Fokin, V. V. Chem. Soc. Rev. 2010, 39, 1302-1315. (f) Holub, J. M.;
Kirshenbaum, K. Chem. Soc. Rev. 2010, 39, 1325-1337. (g) Meldal, M.; Tornøe, C. W. Chem. Rev. 2008,
108, 2952-3015. (h) Bock, V. D.; Hiemstra, H.; van Maarseveen, J. H. Eur. J. Org. Chem. 2006, 51-68.
For the definition of ‘Click Chemistry’, see: (a) Kolb, H. C.; Finn, M. G.; Sharpless, K. B. Angew. Chem. Int.
Ed. 2001, 40, 2004-2021. (b) Broggi, J.; Joubert, N.; Aucagne, V. Nucleos. Nucleot. Nucl. Acids 2007, 26,
779-783.
316
For the application of 1,2,3-triazoles, see: Kolb, H. C.; Sharpless, K. B. Drug Discov. Today 2003, 8, 1128-
1137. (b) Chen, M. D.; Lu, S. J.; Yuag, G. P.; Yang, S. Y.; Du, X. L. Heterocycl. Commun. 2000, 6, 421-423.
(c) Manfredini, S.; Vicentini, C. B.; Manfrini, M.; Bianchi, N.; Rutigliano, C; Mischiati, C; Gambari, R.
Bioorg. Med. Chem. 2000, 8, 2343-2346.
317
For details, see: (a) Lutz, J. F. Angew. Chem. Int. Ed. 2008, 47, 2182-2184. (b) Becer, C. R.; Hoogenboom, R.;
Schubert, U. S. Angew. Chem. Int. Ed. 2009, 48, 2-11, and the references cited within. (c) Jewett, J. C.;

270
 References

Sletten, E. M.; Bertozzi, C. R. J. Am. Chem. Soc. 2010, 132, 3688-3690. (d) Baskin, J. M.; Bertozzi, C. R.
Aldrichim. Acta 2010, 43, 15-23.
318
Normally the amount of copper catalyst used in the CuAAC was from range from 5-10%mole. For details,
see: ref.315
319
There are experimental evidence indicating that internal alkynes show no activity in CuAAC reactions, for
details, see: refs. 313 and 314.
320
For examples, see: (a) Himo, F.; Lovell, T.; Hilgraf, R.; Rostovtsev, V. V.; Noodleman, L.; Sharpless, K. B.;
Fokin, V. V. J. Am. Chem. Soc. 2005, 127, 210-216. (b) Rodinov, V. O.; Fokin, V. V.; Finn, M. G. Angew.
Chem. Int. Ed. 2005, 44, 2210–2215.
321
The preparation of deuterium rich phenylacetylene (PhC≡C-D): 1.02g phenylacetylene was agitated in 5ml
sodium deuteroxide (30%w/w) at 40oC for overnight; the reaction mixture was neutralized with concentrated
HCl, and then extracted with DCM, dried with NaSO4. The solvent was removed in vacuum and the residue
liquid was distilled under pressure (2-3 mmHg, 45-50oC) to give 0.9g (89% yield) colourless liquid. The 1H
NMR spectrum showed that the deuterium incorporation was 87% (ca.) (Appendix C, Figure N32).
322
Pietra, F. Tetrahedron Lett. 1963, 4, 1535-1540.
323
Benito-Lopez, F.; Verboom, W.; Kakuta, M.; Gardeniers, J. (Han) G. E.; Egberink, R. J. M.; Oosterbroek, E.
R.; Bergb, A. van den; Reinhoudta, D. N. Chem. Commun. 2005, 2857-2859.

271