mechanistic investigation on the activation of …€¦ · sabine rothbart, erika ember and rudi...

MECHANISTIC INVESTIGATIONS ON THE ACTIVATION OF PEROXIDES

BY MANGANESE COMPOUNDS

MECHANISTISCHE UNTERSUCHUNGEN ZUR AKTIVIERUNG VON PEROXIDEN

AN MANGANVERBINDUNGEN

Der Naturwissenschaftlichen Fakultät der

Friedrich-Alexander-Universität Erlangen-Nürnberg

zur

Erlangung des Doktorgrades Dr. rer. nat.

vorgelegt von

Sabine Rothbart

aus Nürnberg

Als Dissertation genehmigt von der naturwissenschaftlichen Fakultät der

Friedrich-Alexander-Universität Erlangen-Nürnberg

Tag der mündlichen Prüfung: 26.04.2012

Vorsitzender der Promotionskomission: Prof. Dr. Rainer Fink

Erstberichterstatter: Prof. em. Dr. Dr. h. c. mult. Rudi van Eldik

Zweitberichterstatterin: Prof. Dr. Ivana Ivanović-Burmazović

Die vorliegende Arbeit entstand in der Zeit von September 2007 bis Februar 2012 am Department

Chemie und Pharmazie der Friedrich-Alexander-Universität Erlangen Nürnberg.

DANKSAGUNG

Mein besonderer Dank gilt meinem Doktorvater Prof. em. Dr. Dr. h. c. mult. Rudi van Eldik für

das große Interesse an meiner Arbeit und für zahlreiche wissenschaftliche Diskussionen.

Außerdem möchte ich ihm besonders herzlich für die Möglichkeit danken, Teil seiner einmaligen

Arbeitsgruppe und der Zaubervorlesung sein zu dürfen.

Großen Dank schulde ich auch meinen Eltern, Sigrid und Günther Rothbart, für ihre konstante

Unterstützung und ihren unerschütterlichen Glauben an mich.

Des Weiteren möchte ich mich bei folgenden Leuten bedanken, die zum Gelingen dieser Arbeit

beigetragen haben: Dr. Erika Ember für die Zusammenarbeit, Dr. Ralph Puchta für die DFT

Rechnungen, Dr. Achim Zahl für NMR Messungen, Oliver Tröppner für die

massenspektrometrischen Messungen, sowie Prof. U. Zenneck, Dr. Susanne Mossin und Dr. Jörg

Sutter für die Einführung am EPR-Spektrometer. Prof. D. Chatterjee und seinem Team danke ich

für die Durchführung der HPLC Analyse und gewinnbringende Diskussionen.

Natürlich gebührt auch allen aktuellen und ehemaligen Mitgliedern der Arbeitsgruppe van

Eldik ein herzliches und großes Danke für die angenehme Arbeitsatmosphäre, die stete

Hilfsbereitschaft und den unglaublichen Zusammenhalt. Dies gilt insbesondere für: Ariane,

Christoph, Matthias, Peter, Steffi, Lars, Raquel, Simon und Svetlana. Ihr tragt erheblich dazu bei,

dass mir die Zeit meiner Promotion immer in besonders guter Erinnerung bleiben wird. Außerdem

danke ich den fleißigen Zaubervorlesungs-Mitarbeitern der Arbeitskreise Burzlaff und Ivanović-

Burmazović, sowie der Kaffeeraum-Crew für die gute Atmosphäre und Anita Schmitz für die

schöne Zeit und gute Zusammenarbeit im Praktikum „AC-Explodieren“.

Zu guter Letzt möchte ich besonders meinem Christoph sehr herzlich für die stete

Unterstützung und seelische Kraft danken, die er mir während dieser Zeit gegeben hat.

PUBLICATIONS AND CONFERENCE CONTRIBUTIONS

PUBLICATIONS

1. Erika Ember, Sabine Rothbart, Ralph Puchta and Rudi van Eldik, “Metal – ion

catalyzed oxidative degradation of Orange II by H2O2. High catalytic activity

of simple manganese salts”, New J. Chem., 2009, 33, 34-49. The manuscript

is featured on the cover of the January 2009 issue of the New Journal of

Chemistry.

2. Sabine Rothbart, Erika Ember and Rudi van Eldik, “Comparative study of the catalytic activity of

[MnII(bpy)2Cl2] and [Mn2III/IV(µ-O)2(bpy)4](ClO4)3 in the H2O2 induced oxidation of organic dyes

in carbonate buffered aqueous solution”, Dalton Trans., 2010, 39, 3264-3272.

3. Erika Ember, Hanaa Gazzaz, Sabine Rothbart, Ralph Puchta and Rudi van Eldik, “MnII – a

fascinating catalyst: Mechanistic insight into the catalyzed oxidative degradation of organic

dyes by H2O2”, Appl. Catal. B., 2010, 95, 179-191.

4. Sabine Rothbart, Erika Ember and Rudi van Eldik, “Mechanistic studies on the oxidative

degradation of Orange II by peracetic acid catalyzed by simple manganese(II) salts. Tuning the

lifetime of the catalyst”, New J. Chem. 2012, 36, 732-748.

5. Sabine Rothbart and Rudi van Eldik, “Manganese compounds as versatile catalysts for the

oxidative degradation of organic dyes”, Adv. Inorg. Chem. 2012, 65, submitted.

ONGOING PROJECT

Sabine Rothbart and Rudi van Eldik, “High catalytic activity of a Mn-terpy compound in

oxidative dye degradation by peracetic acid”, in preparation.

CONFERENCES AND WORKSHOPS

POSTER. Oxidative degradation of Orange II by H2O2 catalyzed by divalent transition metals,

Inorganic Reaction Mechanism Group Meeting (IRMG-36), March 2007, York, England.

POSTER. Catalytic activity of [MnII(bpy)2Cl2] and [Mn2III/IV(µ-O)2(bpy)4](ClO4)3 in the H2O2 induced

oxidation of organic dyes, Inorganic Reaction Mechanism Group Meeting (IRMG-39), January

2010, Kloster Banz, Germany.

POSTER. MnII – a fascinating oxidation catalyst: Mechanistic insight into the catalyzed oxidative

degradation of organic dyes by H2O2, Inorganic Reaction Mechanism Group Meeting (IRMG-

39), January 2010, Kloster Banz, Germany.

ORAL PRESENTATION. Catalytic activity of [MnII(bpy)2Cl2] and [Mn2III/IV(µ-O)2(bpy)4](ClO4)3 in the

H2O2 induced oxidation of organic dyes, Erlangen-Kraków-Workshop on “Understanding the

mechanisms of chemical processes”, May 2010, Kraków.

LIST OF ABBREVIATIONS

A absorbance

AcOH acetic acid

B / mT magnetic field in millitesla

bpy 2,2’-bipyridine

BHT 2,6-di-tert-butyl-4-methylphenol

tBuOH tert-butanol

c concentration

CHES 2-(cyclohexylamino)ethanesulfonic acid

CV cyclovoltammogram

D value of the zero field splitting parameter

DFT density functional theory

DMSO-d6 hexadeuterodimethyl sulfoxide

E / V potential in volts

EPR electron paramagnetic resonance spectroscopy

ES-MS electrospray mass spectrometry

g g-faktor

h Plank constant (6.62606885 · 10-34 J·s)

h hour

HEPES 4-(2-hydroxyethyl)piperazine-1-ethanesulfonic acid

I / A current in ampère

I nuclear spin

K equilibrium constant

k rate constant

kobs observed rate constant

λ wavelength

LMCT ligand-to-metal charge-transfer

nm nanometer

NMR nuclear magnetic resonance

ν frequency

Mo Morin; 2’,3,4’,5,7-Pentahydroxyflavone

M mol/l

Me3tacn 1,4,7-trimethyl-1,4,7-triazacyclononane

OII Orange II; 4-(2-hydroxy-1-naphthylazo)benzenesulfonic acid sodium salt

PAA peracetic acid

phen 1,10-phenanthroline

PNP p-nitrophenol, 1-hydroxy-4-nitrobenzol

ppm parts per million

S electron spin

s second

t time

T temperature

tacn 1,4,7-triazacyclononane

TAPS [Tris(hydroxymethyl)methyl]aminopropanesulfonic acid

TZ Tartrazine; trisodium (4E)-5-oxo-1-(4-sulfonatophenyl)-4-[(4-sulfonato-

phenyl)hydrazono]-3-pyrazolecarboxylate

terpy 2,2':6',2"-terpyridine

TOF turnover frequency (mol of dye oxidized by mol of catalyst per hour)

TRIS 2-amino-2-hydroxymethylpropanediol

UV/Vis ultraviolet-visible spectrophotometry

TABLE OF CONTENTS

1 Introduction 1

1.1 TRANSITION METAL - MEDIATED OXIDATIVE BLEACH CATALYSIS 3

1.1.1 Iron-based bleaching catalysts 4

1.1.2 Manganese-based bleaching catalysts 5

1.2 OBJECTIVES 10

1.3 REFERENCES AND NOTES 12

2 Metal ion - catalyzed oxidative degradation of Orange II by H2O2. 17

2.1 GERNERAL REMARK 17

2.2 INTRODUCTION 17

2.3 RESULTS AND DISCUSSION 20

2.3.1 General observations 20

2.3.2 Complex-formation between Orange II and MnII 22

2.3.3 CV studies on the complex-formation between Orange II and MnII 26

2.3.4 DFT-calculations 27

2.3.5 Kinetic investigations 30

2.3.5.1 Complex-formation between bicarbonate and MnII 30

2.3.5.2 The effect of the total carbonate concentration 34

2.3.5.3 Reactivity profile as function of pH 39

2.3.5.4 Effect of the [MnII] and [H2O2] on the oxidative reaction course 41

2.3.5.5 Stability of the in situ formed catalyst 44

2.3.6 Mechanistic aspects 45

2.4 CONCLUSIONS 48

2.5 EXPERIMENTAL SECTION 49


3 Comparative study of a MnII-monomer and the corresponding

oxo-bridged Mn2III/IV-dimer 57

3.1 GENERAL REMARK 57

3.2 INTRODUCTION 57


3.3.1 Kinetic measurements of the catalyzed dye degradation with H2O2 59

3.3.2 The reaction of the catalysts with H2O2 in carbonate buffered solution 65

3.3.2.1 EPR-spectroscopic measurements 65

3.3.2.2 UV/Vis spectroscopic measurements 67

3.3.3 In situ formation of the active species in the catalytic oxidation reaction 69

3.3.4 Precursor complex equilibria in solution 71

3.3.5 Mechanistic aspects 73

3.4 CONCLUSION 75



3.7 SUPPORTING INFORMATION 80

4 Metal ion - catalyzed oxidative degradation of Orange II

by peracetic acid 85


4.2 INTRODUCTION 85


4.3.1 Peracetic acid formation and its decomposition at higher pH 87

4.3.2 General observations 89

4.3.3 MnII + PAA – Intermediates formed in the absence of substrate 90

4.3.3.1 UV/Vis spectroscopy 90

4.3.3.2 EPR spectroscopy 97

4.3.4 Comparison of reactivity of different high valent oxo-manganese

species with Orange II 100

4.3.5 Reactivity of different in situ formed intermediates towards Orange II 103

4.3.6 MnII catalyzed degradation of Orange II by PAA 107

4.3.7 Mechanistic interpretation 112

4.3.8 Comparison MnII/PAA vs. MnII/HCO4- system 116

4.4 CONCLUSIONS 118



4.7 SUPPLEMENTARY INFORMATION 126

5 High catalytic activity of a Mn-terpy compound in oxidative dye

degradations with peracetic acid 141


5.2 INTRODUCTION 141


5.3.1 MnII + terpy in solution 143

5.3.2 MnIIterpy + PAA in the absence of dye substrate 144

5.3.2.1 UV/Vis measurements 144

5.3.2.2 EPR measurements 147

5.3.3 MnIIterpy catalyzed dye degradation with PAA 149

5.3.3.1 General observations 149

5.3.3.2 Kinetics of the MnIIterpy catalyzed dye degradation by PAA 151

5.3.3.3 Readily prepared dimers in the catalytic dye degradation by PAA 156

5.3.4 Mechanistic implications 159

5.4 CONCLUSIONS 162



6 Summary 169

7 Zusammenfassung 177

1. Introduction

1

1 INTRODUCTION

The history of bleaching has come a long way – the ancient method of bleaching cotton and

linen practiced in Egypt, Asia and Europe comprised simply the exposure of the fabric to sunlight.

Over the centuries it developed to a complicated array of repetitive, empirically found processes,

like soaking in water and sour milk or boiling in alkaline solution followed by air exposure for

several months. In the 18th century bleaching solutions of potash and lye or of dilute sulphuric acid

were used in Holland and France. Yet, the breakthrough of industrial bleach processes came with

the discovery of the powerful properties of chlorine and hydrogen peroxide – the two main

bleaching systems used nowadays.[1]

In general, the main objective of successful bleaching is the whitening of various substrates in

a homogeneous or heterogeneous chemical reaction. This is either achieved by shifting the

wavelength of light absorbtion outside the range of visible light or by chemical decomposition of

the chromophores’ molecular entity and thereby rendering it water-soluble.[2] For this purpose

different bleaching systems are available, which can be divided into two major classes.[2, 3] On one

hand, the reducing bleaching agents that compris free sulforous acid, bisulfate and hyposulforous

acid and on the other side the class of oxidizing ones, including oxygen, ozone, chlorine,

hypochlorites and peroxides.[3] Chlorine containing compounds are partially used for disinfection

or water treatment and bleaching of delignificated Kraft pulp. This technology is characterized by

high efficiency at relatively low costs. Nevertheless, the major disadvantage of this method is still

the formation of hazardous and non-biodegradable chlorinated side products. The chlorine

bleaching of cellulose pulp, for instance, leaves effluents loaded with chlorophenols,

chloroaliphatics and polychlorinated dioxins and furans.[4]

However, the ever increasing ecological awareness causes a soaring global demand for

environmentally benign procedures. Typical common features of these new “green” procedures

are the reduction of energy demand, the minimization of the total amount of chemicals involved

in the overall process, and in particular the avoidance of toxic side products. Therefore, oxygen-

based bleaching agents are considered to be the chemicals of choice for sustainable bleaching

compositions and are gradually replacing chlorine based technologies.

1. Introduction

2

Although the application of atmospheric oxygen is clearly a long term objective as it would

lead to a further decrease in the chemical loading and a reduction in cost, the scope of molecular

oxygen as an oxidant is at present still narrow.[2, 5] Another environmentally benign and widely

employed oxygen carrier is hydrogen peroxide. It is a cheap, readily available chemical, which

forms only non-toxic water and oxygen as side products during the oxidation process. In order to

facilitate transportation for various applications, hydrogen peroxide is substituted by solid

peroxygens. Laundry cleaning formulations contain sodium percarbonate (PCS) or sodium

perborate (PBS). PCS is only considered as an adduct of carbonate salt and hydrogen peroxide with

the general formula Na2CO3 · 1.5 H2O2, although the in situ formation of the peracid, HCO4- has

been shown by various spectroscopic techniques.[6] PBS on the other hand is a real persalt with a

six-membered heterocyclic di-anion structure.[7a-b] Yet, both peroxygen salts yield a weakly basic

solution of hydrogen peroxide upon dissolving in water.[7c] Other commercially relevant secondary

products of hydrogen peroxide are potassium peroxomonosulfate (Oxone®), mainly used for

chloramine elimination in swimming pools,[8a] and peracetic acid, which has a wide scope of

applications. For instance, it is replacing hypochlorite disinfectants in food industry, brewing,

farming, medicine and waste water treatment, but it is also widely used for domestic laundry

cleaning.[2, 8b]

Oxidative bleaching processes are of major economic importance. For example, the annual

world production of hydrogen peroxide is approximately 2.2 million metric tons, of which 50 % is

used for pulp/paper bleaching and 10 % for textile bleaching.[2] These technologies are

omnipresent in numerous industrially relevant procedures, such as bleaching of stains,[9a] raw

wood pulp,[2, 9b] raw cotton fibers[9a] and waste water treatment.[9b-d] Moreover, they represent an

essential component of various cleaning and refinement processes.Yet, with application in laundry

cleaning[2, 9a] or usage in personal care formulations and detergents,[9a] oxidative bleaching

processes also belong to the everyday life of millions of costumers worldwide.

However, as the kinetics of these oxygen-based bleach systems are sluggish, it is necessary to

activate the oxygen source. Therefore, various technologies to enhance the peroxide efficiency are

known.[2] Besides the application of laccases and peroxidases[10a-b] or activated imines,[10c] the

main research focus lies on peroxide activation by transition metal catalysts. From an

environmental point of view these catalysts are mainly based on iron or manganese complexes,

but also cobalt, vanadium and titanium compounds are known.[9b] Despite the larger number of

1. Introduction

3

commercial applications for transition metal mediated peroxide activation, the underlying reaction

mechanisms still remain unclear. Once the mechanism of action is known, the properties of

potential catalytic systems can be tuned particularly to improve their efficiency by choice of the

ideal reaction conditions. Consequently, a detailed mechanistic understanding represents an

indispensable precondition for the improvement of already implemented catalytic systems as well

as for the development of new efficient and environmentally compatible catalysts.

1.1 TRANSITION METAL - MEDIATED OXIDATIVE BLEACH CATALYSIS

In general, two catalytic pathways are conceivable upon reaction of a peroxide with a

transition metal compound. The transiently formed peroxo- or hydroperoxo-intermediate can

undergo either heterolytic or homolytic O-O-bond scission. A homolytic peroxo bond cleavage

leads to the formation of radicals, which is undesired as these radical species lack selectivity in

their subsequent reactions. Another option is a heterolytic scission of the peroxo bond which

results in the formation of high valent metal-oxo species. Both reaction types are depicted for the

example of H2O2 in Scheme 1.1.

Scheme 1.1 Formation of metal-oxo species in the presence of H2O2 by heterolytic O-O-bond cleavage vs. free radical

generation.

Metal-oxo intermediates often have a greater oxidative power compared to the starting

peroxide and are more selective than free radical species. They represent the one- or two-electron

oxidized form of the starting complex, respectively and following oxygen transfer the starting

1. Introduction

4

catalyst is regenerated. To favor metal based two-electron oxidation over radical type one-

electron processes often requires an electron releasing ligand to stabilize the high-valent metal-

oxo species. Depending on the application, in most cases tailor-made ligand systems are needed to

ensure sufficient selectivity and to expand the catalyst lifetime by avoiding ligand self-oxidation.

Despite of decades of academic and industrial research, there are still ongoing efforts to design

new catalytic transition metal-ligand systems. The main target of these research efforts is to find

model compounds that provide selectivity and efficiency which is comparable to natural enzymes.

Peroxidases, oxygenases or oxidases are the common ideal to activate molecular oxygen or

hydrogen peroxide. During their catalytic cycle all these enzymes form high-valent metal-oxo

species which are considered to be responsible for the subsequent oxidation processes. Many of

these enzymes contain manganese or iron centers, which makes these metals ideal for catalytic

model compounds. Moreover, manganese and iron can play an important role due to their very

rich redox chemistry and their environmental compatibility. Yet, the use of such enzyme mimics is

in many cases connected with a big synthetic effort to obtain adequate ligand systems. In the

following some outstanding and commonly used examples for manganese and iron oxidation and

bleach catalysts are presented.

1.1.1 Iron-based bleaching catalysts

Various classes of iron compounds are known to efficiently catalyze oxygenation and bleach

processes. Besides iron complexes with TPA[11a] (tris(pyridin-2-ylmethyl)amine) or different

pentadentate nitrogen-donor ligands,[11b] one of the most succesful iron systems is the class of

tetraamido macrocyclic ligands developed by Collins et al., the so called TAML systems (Scheme

1.2).[12a]

Scheme 1.2 An iron tetraamidate (TAML) bleaching catalyst.

1. Introduction

5

Since these tetraamidate ligands have a formal charge of -4 they are able to stabilize high-

valent metal species as for example a {MnV=O} complex which was characterized by X-ray and

spectroscopic analyses.[12c-f] In this example the electrophilicity of the oxomanganese center can

be increased by replacing the benzene ring with pyridine, and by subsequent binding of cations

such as lithium to the pyridine.[2, 12f]

A broad range of applications are postulated for this class of complexes, mostly connected to

solution bleaching such as waste water treatment and dye transfer inhibition for laundry

cleaning.[2, 12a,b] Different organic dyes relevant for detergents and waste water treatment can be

bleached in solution with hydrogen peroxide as terminal oxidant.[13a] Moreover, stain removal and

enhanced delignification have also been reported for the FeIII-TAML catalyst.[13b] A variation of

ligand substituents results in a significant alteration of the catalytic properties, such as catalytic

lifetime and hydrolytic stability. Whereas the first generation of TAML catalysts contained ethyl

groups next to the carbonyl moieties, the oxidative stability and thereby the catalytic efficiency

could be increased by replacing them with methyl groups.[13a] Hydrolytic stability was additionally

obtained through substitution of the methyl groups by fluorine, or by introduction of electron

withdrawing groups on the phenyl ring.[12a, 13c, d] No indications for free-radical autoxidation

reactions were found.[2]

Despite recent efforts and ongoing controversy among scientists about many mechanistic

issues and putative reactive species of the FeIII-TAML/oxidant system, the underlying mechanism

of action is still not clarified in detail. Recently a TAML based iron(V)-oxo intermediate which

resembles a proposed reactive intermediate in many catalytic hydrogen peroxide applications has

been isolated and characterized.[13e] However, also monomeric and dimeric (TAML)FeIV=O are

frequently stated as reactive species.[13f, g]

1.1.2 Manganese-based bleaching catalysts

The very first bleach catalyst employed in commercial detergent products was a dinuclear oxo-

bridged manganese compound with the Me3tacn ligand (Scheme 1.3).[2, 14] Wieghardt and co-

workers published this compound in 1988 as a model for manganese-containing enzymes.[2, 14b]

The bleaching activity in combination with hydrogen peroxide is very high for the standard tea-

model stain as well as for wine, fruit, and curry stains.[14a] However, the detergent product

containing this catalyst was withdrawn from the market after it was alleged that the product yields

1. Introduction

6

increased fabric- and dye-damage.[2, 14c] Yet, the catalyst-system is still an integral part of various

machine-dishwashing products where it is responsible for superior removal of tea residues.[2]

Scheme 1.3 Dinuclear Mn(Me3tacn) catalyst.

The mechanistic aspects of the manganese-Me3tacn oxidation reaction were examined for

phenolic compounds and catechol as models for tea stains.[15a-c] The observation of a Mn2III/IV-

dimeric species in EPR spectroscopic measurements was explained by the first step of the reaction,

i.e. a one-electron transfer process from the phenolate ion to the EPR-silent starting complex.[16a-c]

This initial reduction is followed by the appearance of mononuclear MnIV species as confirmed by

further EPR and electrospray mass spectrometry (ES-MS) experiments. Since the dinuclear starting

catalyst is stable against hydrolysis, it was concluded that the initial reduction step is required for

monomer formation. The formation of phenoxyl radicals with Trolox, a water-soluble phenol

derivative, as substrate is observed in both the absence and presence of H2O2.[16b] As the second

step, the oxidation to a MnIV or MnV species is assumed, which can be either mono- or dinuclear.[2]

In general it is suggested that both mononuclear and dinuclear manganese species are

operative.[16a] Contrary, more recent ES-MS experiments indicated the sole formation of

monomeric species in the reaction of [Mn2IV(µ-O)3(Me3tacn)2]2+ with H2O2 and 4-

methoxyphenol.[16c,d]

In addition, Gilbert et al. investigated the catalytic activity towards other substrates like azo-

containing dyes.[16] For this reaction the authors claimed the involvement of monomeric {MnV=O}

species and that such an oxo-manganese species gives rise to two consecutive one-electron

oxidation reactions of the azo dye. Therefore, the catalytic system would resemble the reactivity

of heme-peroxidases (Scheme 1.4). No formation of hydroxyl radicals was observed even with the

use of EPR-sensitive spin traps.[16]

1. Introduction

7

Scheme 1.4 Mechanistic cycle postulated for oxidation of azo dyes by the Mn(Me3tacn) catalyst in the presence of

hydrogen peroxide.[16]

On the other hand, Meunier and co-workers concluded that the oxidation is caused by

hydroxyl radicals. They studied the reactivity of [Mn2IV(µ-O)3(Me3tacn)2]2+/H2O2 with catechol in

the absence and presence of mannitol and found that the formed quinone degrades faster in the

absence than in the presence of mannitol.[2, 17] Although no additional EPR spin-trapping

measurements were conducted, the results were interpreted as the trapping of hydroxyl radicals

by mannitol.[17] Therefore, other explanations may be valid, such as binding of mannitol to the

active species that may prevent the oxidative activity of this system.[2]

Besides efficient catalytic stain and dye bleaching the Mn(Me3tacn) catalyst has in addition

been investigated towards the selective epoxidation with H2O2 as oxidant.[16a] Moreover, the two

mononuclear complexes [MnIV(Me3tacn)(CH3O)3]+ and [MnIV(L)]+ (H3L = 1,4,7-tris(hydroxyethyl)-

1,4,7-triazacyclononane) were studied with regard to their epoxidation activity on the water-

soluble olefins 4-vinylbenzoic acid and styrylbenzoic acid.[2, 16a, 18a] In slightly basic solution, both

monomeric compounds exhibit satisfying total turnover numbers for epoxide formation, but

concomitant H2O2 decomposition occured. By using H218O2 it was evidenced that oxygen transfer

occurs from a coordinated hydrogen peroxide or a derivative species. The epoxide product was

fully 18O-labeled.[18b] Hydroxyl radicals were again excluded, since no hydroxylation of the aromatic

styrene ring was observed.

1. Introduction

8

For the manganese Me3tacn species high epoxidation activities have also been reported in

non-aqueous media, which could achieve total turnover numbers as high as 4000 depending on

the applied conditions.[2, 19a-j] When acetone is used as solvent, the decomposition of hydrogen

peroxide can be suppressed and selectivity towards epoxidation can be increased.[2, 19b, d] High

epoxidation activity was also observed in acetonitrile, provided that additives such as oxalate or

ascorbic acid are present, but no information on the active species is available.[2, 19c, e]

Enantioselective epoxidation of substitued styrenes was achieved by Bolm and co-workers by

using a chiral tacn derivative, 1,4,7-S,S,S-tris(2-hydroxypropyl)-1,4,7-triazacyclononane.[2, 19j] Other

oxidation reactions (Scheme 1.5) include cis-dihydroxylation of cyclooctene,[19h, 20a] oxidation of

benzyl alcohol to benzaldehyde,[20b] oxidation of sulfides to sulfoxides and sulfones,[20c-e] allylic C-H

oxidation of cyclohexene to the alcohol and ketone,[20f] and alkane oxidations by radical

reactions.[2, 20g-l]

Scheme 1.5 Examples of substrate conversions catalyzed by the Mn(Me3tacn) system.

Noteworthy, the dimeric [MnIV(Me3tacn)(CH3O)3]+ complex also epoxidizes cinnamic acid in a

buffer/acetone mixture in the presence of H2O2 as observed for the monomeric manganese

compound with tris(hydroxyethyl)-tacn and also yields fully 18O-labeled epoxide.[18b, 21] Again, ES-

MS signals of mononuclear species were detected, which were assigned to [(Me3tacn)MnIV(OH)3]+

and [(Me3tacn)MnIV(O)(OH)]+ where the latter species would imply a penta-coordinate MnIV

species, which is unusual for a low-spin d3 system.[2]

1. Introduction

9

Despite its excellent epoxidation abilities, it is far more surprising, that the monomeric tacn

derivative [MnIV(L)]+ does not catalyze stain bleaching whereas the dimeric [Mn2IV(µ-

O)3(Me3tacn)2]2+ does.[16a] As a consequence it might be concluded that the chemical processes

which underly the oxygen transfer to an alkene on the one side, and stain bleaching of

polyphenolic chromophores on the other side, are basically different. However, differences in the

applied reaction conditions such as pH (pH 8 vs. 10.5), catalyst stability and feasible interactions

with the stain or garments may account for this observation. Be that as it may, these studies led to

the suggestion that the same {MnV=O} species is involved in the catalysis as described for the

oxidation of phenol and styrene.[2] Electron-rich derivatives of cinnamic acid are converted most

efficiently, which is indicative of an electrophilic {Mn=O} center.[2] Although a coherent

mechanistic understanding is still lacking, it is clear that both typical oxygen-transfer and

hydrogen-abstraction reactions are catalyzed. Based on the reactivity of [Mn2IV(µ-

O)3(Me3tacn)2]2+/H2O2 in aqueous solutions, it is assumed that the observed bleaching activity

primarily originates from epoxidation/oxo transfer, but a proof for this hypothesis has still to be

provided.[2]

Another important class of stain bleaching manganese complexes was patented by Ciba,

Clariant and Henkel.[22] Some of these Schiff-base compounds are specially suitable for low-

temperature bleaching and dye-transfer inhibition which sometimes occurs when colored and

white garments are combined and its prevention is achieved by dye oxidation in solution.[2, 22 a-c]

The manganese metal center in these complexes has the oxidation state III. Some examples of the

salen ligand structures, which together with manganese have been patented as bleach catalysts,

are shown in Scheme 1.6. They are less efficient catalysts than the Mn(Me3tacn) complexes, but

according to the patents, they also cause less dye-fading on the cloths and are easier to

synthesize.[2, 22a-c] These catalysts generally activate hydrogen peroxide to yield an improved tea

stain bleaching, and in some cases they outperform the conventional TAED/percarbonate

bleaching system.[2, 22c-d]

Although non of the patents specifies mechanistic information, the reactivity of manganese

Schiff-base compounds has been studied in non-aqueous solution.[2, 23] For the reaction with

hydrogen peroxide, hypochlorite and iodosobenzene, highly enantioselective oxidations have

been reported in the presence of nitrogen donors such as imidazole.[2] The reactivity is ascribed to

an electrophilic {MnV=O} intermediate, which yields a substrate radical intermediate.[2]

1. Introduction

10

Scheme 1.6 Examples of Schiff-base ligands patented as bleach catalysts in combination with manganese.

1.2 OBJECTIVES

Throughout the last decades a broad variety of manganese based transition metal compounds

has been investigated towards their ability to activate peroxide for numerous homogeneous and

heterogeneous applications as summarized in Chapter 1.1. Yet, there are still several aspects that

require further clarification. In this context, the general goal of this thesis was to contribute to a

more detailed mechanistic understanding of these catalytic processes especially in aqueous

solution. As water is nature’s principle solvent, it is the ideal choice for green chemistry.[12a, 13f, 24]

An unique aspect of aqueous transition-metal chemistry that cannot simply be transferred to

other solvents, is the ability to manipulate the reactivity of various species by a simple change in

pH.[25] Having a degree of control over the lifetime of an intermediate facilitates spectroscopic and

kinetic studies and expands the range of methods and techniques that can be used in mechanistic

investigations.[25]

Despite the complexity of the structure and chemistry in catalytic peroxide activation, the

reactivity of complex catalysts can be sometimes reproduced by much simpler models.[25] In this

context, one objective of this work was to find, investigate and improve catalytic systems for the

homogeneous activation of hydrogen peroxide based on simple manganese salts. Chapter 2

1. Introduction

11

describes the study of the MnII ion catalyzed degradation of the commonly used model substrate

Orange II with H2O2, which revealed the crucial role of the carbonate buffer in the formation of

reactive Mn-bicarbonate complexes and the in situ formation of peroxycarbonate, the actual

active oxidant.

Further, it should be clarified to which extent dimeric, oxo-bridged high-valent species

contribute in the studied bleach reaction with H2O2 as terminal oxidant by applying a well known

ligand system. The enhanced complexity caused by the presence of the ligand and by the presence

of more than one metal center in the molecule, naturally leads to a larger number of possible

intermediates which can now include mixed-metal oxidation states,[26 a] bridging and end-on

superoxide[26b, c]/peroxide[27] groups, or isomeric dioxo- vs. μ-peroxo-forms.[25, 28] In this respect,

Chapter 3 reports kinetic investigations of monomeric and dimerc precursors in the H2O2 induced

degradation of different dye substrates. Again a strong influence of the buffer system on the

activation of H2O2 was detected and discussed in the light of the variation of the system compared

to the simple MnII salts in Chapter 2. In addition, EPR-spectroscopic methods proved to be a

helpful tool to elucidate the nature of the predominant species under catalytic conditions.

A further objective was to extend our work to another commonly used oxidant, peracetic acid

(PAA). The application of peracetic acid marks a special challenge for mechanistic studies, since it

always contains an equilibrium amount of H2O2. In order to address this additional complication, a

careful selection of the reaction conditions is required to lay the foundation for one of the few

mechanistic studies on transition metal catalyzed bleaching with PAA as oxidant. In Chapter 4 the

reaction of MnII salts with PAA was studied by detailed UV/Vis- and concomitant EPR-

spectroscopic experiments in combination with selected reactivity studies to disclose the different

in situ formed intermediates and their relevance to the dye degradation process. Furthermore, our

studies revealed the ambiguous role of the equilibrium concentration of H2O2 on the lifetime of

the catalytic system.

To complete the mechanistic aim of this work, Chapter 5 gives some preliminary insight into

the manganese catalyzed activation of PAA in the presence of a chelate. Contrary to the results

described in Chapter 3 for H2O2, our findings point to the involvement of dimeric, oxo-bridged

species in the dye degradation process.

1. Introduction

12

In summary, the general objective was to provide detailed information on putative reactive

intermediates and the mechanism of action for the activation of two widely used oxidants, namely

hydrogen peroxide and peracetic acid, by manganese systems with and without a defined ligand

system, chosen as simplest representatives for the numerous different peroxide activating

catalysts used in bleaching and oxygenation applications.

1.3 REFERENCES AND NOTES

[1] Suess, H. U. Pulp BleachingToday, Walter de Gruyter 2010.

[2] Hage, R.; Lienke, A. Angew. Chem. Int. Ed. 2006, 45, 206-222.

[3] von Georgievics, G. The Chemical Technology of Textile Fibres - Their Origin, Structure,

Preperation, Washing, Bleaching, Dyeing, Printing and Dressing, Scott. Greenwood, London,

1920.

[4] (a) Daube, A. K.; Karlm, M. R.; Dimmel, D. R.; McDonough, T. J.; Banerjee, S. Environ. Sci.

Technol. 1992, 26, 1324-1326; (b) Nakamata, K.; Ohi, H. J. Wood Sci. 2003, 49, 525-530.

[5] Beller, M. Adv. Synth. Catal. 2004, 346, 107-108.

[6] (a) Flanagan, J.; Jones, D. P.; Griffith, W. P.; Skapski, A. C.; West, A. P. J. Chem. Soc., Chem.

Commun. 1986, 20-21; (b) Lane, B. S.; Vogt, M.; DeRose, V. J.; Burgess, K. J. Am. Chem. Soc.

2002, 124, 11946-11954; (c) Richardson, D. E.; Regino, C. A. S.; Yao, H. R.; Johnson, J. V. Free

Radical Biol. Med. 2003, 35, 1538-1550; (d) Richardson, D. E.; Yao, H.; Frank, K. M.; Bennett,

D. A. J. Am. Chem. Soc. 2000, 122, 1729-1739; (e) Bakhmutova-Albert, E. V.; Yao, H.; Denevan,

D. E.; Richardson, D. E. Inorg. Chem. 2010, 49, 11287-11296.

[7] (a) Carrondo, M. A. A. F. d. C. T.; Griffith, W. P.; Jones, P. D.; Skapski, A. C. J. Chem. Soc.,

Dalton Trans. 1977, 2323-2327; (b) Hansson, A. Acta Chem. Scand. 1961, 15, 934-935; (c)

Galwey, A. K.; Hood, W. J. J. Chem. Soc., Faraday Trans. 1 1982, 78, 2815-2827.

[8] (a) Jakob, H.; Lehmann, T.; Jacobi, S.; Gutewort, S. Ullmann´s Encyclopedia of Industrial

Chemistry, Wiley, 2007; (b) Block, S. Peroxygen Compounds, Philadelphia, 1991.

[9] (a) Kirk-Othmer, Encyclopedia of chemical technology, Wiley: New York, 1991; (b) Jones, C. W.

Applications of Hydrogen Peroxide and Derivatives, RSoC: Cambridge, 1999; (c) Aurand, K. Die

Trinkwasserverordnung, Berlin, 1991; (d) Schwarzer, H. Umwelt 1981, 482 - 486.

http://www.google.de/search?hl=de&tbo=p&tbm=bks&q=inauthor:%22Hans+Ulrich+Suess%22

1. Introduction

13

[10] (a) Solomon, E. I.; Sundaram, U. M.; Machonkin, T. E. Chem. Rev. 1996, 96, 2563-2605; (b)

Hansch, C.; Gao, H. Chem. Rev. 1997, 97, 2995-3059; (c) Pocalyko, D. J.; Coope, J. L.; Carchi, A.

J.; Boen, L.; Madison, S. A. J. Chem. Soc., Perkin Trans. 2 1997, 117-121.

[11] (a) Que Jr., L.; Kim, C.; Kim, J.; Zang, Y. WO-A-9748710, 1996. [Chem. Abstr. 1997, 128,

90357]; (b) Feringa, B. L.; Lubben, M.; Hermant, R. M.; Twisker, R. S.; Que Jr., L.

(Unilever),WO-A-9534268, 1994 [Chem. Abstr. 1995, 124, 205657].

[12] (a) Collins, T. J. Acc. Chem. Res. 2002, 35, 782-790; (c) Collins, T. J.; Gordon-Wylie, S.W. J. Am.

Chem. Soc. 1989, 111, 4511-4513; (b) Beach, E. S.; Malecky, R. T.; Gil, R. R.; Horwitz, C. P.;

Collins; T. J. Catal. Sci. Technol. 2011, 1, 437-443; (d) Collins, T. J.; Powell, R. D.; Slebodnick, C.;

Uffelman, E. S. J. Am. Chem. Soc. 1990, 112, 899-901; (e) Workman, J. M.; Powell, R. D.;

Procyk, A. D.; Collins, T. J.; Bocian, D. F. Inorg. Chem. 1992, 31, 1548-1550; (f) Miller, C. M.;

Gordon-Wylie, S. W.; Horwitz, C. P.; Strazisar, S. A.; Peraino, D. K.; Clark, G. R.; Weintraub, S.

T.; Collins, T. J. J. Am. Chem. Soc. 1998, 120, 11540-11541.

[13] (a) Horwitz, C. P.; Fooksman, D. R.; Vuocolo, L. D.; Gordon-Wylie, S. W.; Cox, N. J.; Collins, T. J.

J. Am. Chem. Soc. 1998, 120, 4867-4868; (b) Collins, T. J. ACS Symp. Ser. 2002, 47; (c) Ghosh,

A.; Ryabov, A. D.; Mayer, S. M.; Horner, D. C.; Prasuhn, D. E. Jr., Gupta, S. S.; Vuocolo, L.;

Culver, C.; Hendrich, M. P.; Rickard, C. E. F.; Norman, R. E.; Horwitz, C. P.; Collins, T. J. J. Am.

Chem. Soc. 2003, 125, 12378-12379; (d) Polshin, V.; Popescu, D.-L.; Fischer, A.; Chanda, A.;

Horner, D. C.; Beach, E. S.; Henry, J.; Qian, Y.-L.; Horwitz, C. P.; Lente, G.; Fabian, I.; Münck, E.;

Bominaar, E. L.; Ryabov, A. D.; Collins, T. J. J. Am. Chem. Soc. 2008, 130, 4497-4506; (e) Tiago

de Oliveira, F.; Chanda, A.; Banerjee, D.; Shan, X.; Mondal, S.; Que Jr., L.; Bominaar, E. L.;

Münck, E.; Collins, T. J. Science 2007, 315, 835-838; (f) Popescu, D.-L.; Vrabel, M.; Brausam,

A.; Madsen, P.; Lente, G.; Fabian, I.; Ryabov, A. D.; van Eldik, R.; Collins, T. J. Inorg. Chem.

2010, 49, 11439-11448; (g) Chanda, A.; Shan, X.; Chakrabarti, M.; Ellis, W. C.; Popescu, D. L.;

Tiago de Oliveira, F.; Wang, D.; Que Jr., L.; Collins, T. J.; Münck, E.; Bominaar, E. L. Inorg.

Chem. 2008, 47, 3669-3678.

[14] (a) Favre, F.; Hage, R.; van der Helm-Rademaker, K.; Koek, J. H.; Martens, R. J.; Swarthoff, T.;

van Vliet, M. R. P. (Unilever), EPB-0458397, 1990 [Chem. Abstr. 1991, 116, 154241]; (b)

Wieghardt, K.; Bossek, U.; Nuber, B.; Weiss, J.; Bonvoisin, J.; Corbella, M.; Vitols, S. E.; Girerd,

J.-J. J. Am. Chem. Soc. 1988, 110, 7398-7411; (c) Verrall, M. Nature 1995, 373, 181.

1. Introduction

14

[15] (a) Hage, R.; J. E. Iburg, J. E.; Kerschner, J.; Koek, J. H.; Lempers, E. L. M.; Martens, R. J.;

Racherla, U. S.; Russell, S. W.; Swarthoff, T.; van Vliet, M. R. P.; Warnaar, J. B.; van der Wolf,

L.; Krijnen, L. B. Nature 1994, 369, 637-639; (b) Gilbert, B. C.; Kamp, N. W. J.; Lindsay-Smith, J.

R.; Oakes, J. J. Chem. Soc. Perkin Trans. 2 1997, 2161-2166; (c) Gilbert, B. C.; Kamp, N. W. J.;

Lindsay-Smith, J. R.; Oakes, J. J. Chem. Soc. Perkin Trans. 2 1998, 1841-1844; (d) Gilbert, B. C.;

Kamp, N. W. J.; Lindsay-Smith, J. R.; Oakes, J. Org. Biomol. Chem. 2004, 2, 1176-1180.

[16] Gilbert, B. C.; Lindsay-Smith, J. R.; Newton, M. S.; Oakes, J.; Pons i Prats, R. Org. Biomol. Chem.

2003, 1, 1568-1577.

[17] Sorokin, A.; Fraisse, L.; Rabion, A.; Meunier, B. J. Mol. Catal. A 1997, 117, 103-114.

[18] (a) Quee-Smith, V. C.; DelPizzo, L.; Jureller, S.H.; Kerschner, J.L.; Hage, R. Inorg.Chem. 1996,

35, 6461-6465; (b) Hage, R.; Kerschner, Trends in Inorg. Chem. 1998, 5, 145-159.

[19] (a) de Vos, D. E.; Bein, T. Chem. Commun. 1996, 917-918; (b) de Vos, D. E.; Bein, T.

J.Organomet.Chem. 1996, 520, 195-200; (c) de Vos, D. E.; Sels, B. F.; Reynaers, M.; Subba Rao,

Y. V.; Jacobs, P. A. Tetrahedron Lett. 1998, 39, 3221-3224; (d) Edwards, J-O.; Sauer, M. V. C. J.

Phys.Chem. 1971, 75, 3004-3011; (e) Berkessel, A.; Sklorz, C. A. Tetrahedron Lett. 1999, 40,

7965-7968; (f) de Vos, D. E.; Meinershagen, J.; Bein, T. Angew. Chem. Int. Ed. 1996, 35, 2211-

2213; (g) de Vos, D. E.; Bein, T. J. Am. Chem. Soc. 1997, 119, 9460-9465; (h) de Vos, D.; de

Wildeman, S.; Sels, B. F.; Grobet, P. J.; Jabobs, P. A. Angew. Chem. Int. Ed. 1999, 38, 980-983;

(i) Rao, S. Y. V.; de Vos, D. E.; Bein, T.; Jacobs, P. A. Chem.Commun. 1997, 355-356; (j) Bolm,

C.; Kadereit, D.; Valacchi, M. Synlett 1997, 6, 687-688.

[20] (a) Brinksma, J.; Schmieder, L.; van Vliet, L.; Boaron, R.; Hage, R.; de Vos, D. E.; Alsters, P. L.;

Feringa, B. L. Tetrahedron Lett. 2002, 43, 2619-2622; (b) Handervan, C.; Hage, R.; Feringa, B. L.

Chem. Commun. 1997, 419-420; (c) Barton, D. H. R.; Choi, S.-Y.; Hu, B.; Smith, J. A.

Tetrahedron 1998, 54, 3367-3378; (d) Brinksma, J.; LaCrois, R.; Feringa, B. L.; Donnoli, M. I.;

Rosini, C. Tetrahedron Lett. 2001, 42, 4049-4052; (e) Barker, J. E.; Ren, T. Tetrahedron Lett.

2004, 45, 4681-4683; (f) Vincent, M.; Rabion, A.; Yachandra, V. K.; Fish, R. H. Angew. Chem.

Int. Ed. 1997, 36, 2346-2349 (g) Metal-catalyzed Oxidations of Organic Compounds (Eds.:

Sheldon, R. A.; Kochi, J. K.), Academic Press, NewYork, 1981; (h) Bennur, T. H.; Sabne, S.;

Deshpande, S. S.; Srinivas, D.; Sivasanker, S. J. Mol. Catal. A 2002, 185, 71; (i) Klopstra, M.;

Hage, R.; Kellogg, R. M.;Feringa, B. L. Tetrahedron Lett. 2003, 44, 4581-4584; (j) Shul’pin, G.

B.; Nizova, G. V.; Kozlov, Y. N.; Pechenkina, I. G. New J. Chem. 2002, 26, 1238-1245; (k)

1. Introduction

15

Lindsay-Smith, J. R.; Shul’pin, G. Tetrahedron Lett. 1998, 39, 4909-4912; (l) Woitiski, C. B.;

Kozlov, Y. N.; Mandelli, D.; Nizova, G. V.; Schuchardt, U.; Shul’pin, G. B. J. Mol. Catal. A 2004,

222, 103-119.

[21] Gilbert, B. C.; Lindsay-Smith, J. R.; Mairata i Payeras, A.; Oakes, J.; Pons i Prats, R. J. Mol. Catal.

A 2004, 219, 265-272.

[22] (a) Reinehr, D.; Metzger, G. (Ciba), WO-A-9719162, 1995 [Chem. Abstr. 1997, 127, 67687]; (b)

Bachmann, F.; Dannacher, J.; Hazenkamp, M.; Schlingloff, G.; Richter, G.; Dbaly, H.; Traber, R.

H. (Ciba), WO-A-200105925, 1999 [Chem. Abstr. 2001, 134, 133329]; (c) Hazenkamp, M.;

Bachmann, F.; Makowka, C.; Dubs, M.-J.; Richter, G.; Schlingloff, G.; Dannacher, J.;

Weingartner, P. (Ciba), WO-A-200053712, 1999 [Chem. Abstr. 2000, 133, 239758]; (d) Nestler,

B. (Clariant), EP-B-869171, 1997 [Chem. Abstr. 1998, 129, 291445]; (e) Blum, H.; Nitsch, C.;

Jeschke, P.; Haerer, J.; Pegelow, U. (Henkel), WO-A-9707191, 1997 [Chem. Abstr. 1997, 126,

239902].

[23] (a) Berkessel, A.; Frauenkron, M.; Schwenkreis, T.; Steinmetz, A.; Baum, G.; Fenske, D. J. Mol.

Catal. A 2002, 113, 321-342; (b) Katsuki, T. Coord. Chem. Rev. 1995, 140, 189-214; (c) Lane, B.

S.; Burgess, K. Chem. Rev. 2003, 103, 2457-2473; (d) Linde, C.; Koliai, N.; Norrby, P.-O.;

Akermark, B. Chem. Eur. J. 2002, 8, 2568-2573; (e) Pietikainen, P. J. J. Mol. Catal. A 2001, 165,

73-79.

[24] Anastas, P. T. Crit. Rev. Anal. Chem. 1999, 29, 167-175.

[25] Bakac, A. Inorg. Chem. 2010, 49, 3584-3593.

[26] (a) Kovaleva, E. G.; Neibergall, M. B.; Chakrabarty, S.; Lipscomb, J. D. Acc. Chem. Res. 2007,

40, 475-483; (b) Egan Jr., J. W.; Haggerty, B. S.; Rheingold, A. L.; Sendlinger, S. C.; Theopold, K.

H. J. Am. Chem. Soc. 1990, 112, 2445-2446; (c) Qin, K.; Incarvito, C. D.; Rheingold, A. L.;

Theopold, K. H. Angew. Chem., Int. Ed. 2002, 41, 2333-2335.

[27] (a) Takahashi, Y.; Hashimoto, M.; Hikichi, S.; Akita, M.; Moro-Oka, Y. Angew. Chem., Int. Ed.

1999, 38, 3074-3077; (b) Ho, R. Y. N.; Roelfes, G.; Hermant, R.; Hage, R.; Feringa, B. L.; Que Jr.,

L. Chem. Commun. 1999, 2161-2162; (d) Rahman, A. F. M. M.; Jackson, W. G.; Willis, A. C.

Inorg. Chem. 2004, 43, 7558-7560; (d) Brausam, A.; Maigut, J.; Meier, R.; Szilagyi, P. A.;

Buschmann, H.-J.; Massa, W.; Homonnay, Z.; van Eldik, R. Inorg. Chem. 2009, 48, 7864-7884.

1. Introduction

16

[28] (a) Halfen, J. A.; Mahapatra, S.; Wilkinson, E. C.; Kaderli, S.; Young Jr., V. G.; Que Jr., L.;

Zuberbuhler, A. D.; Tolman, W. B. Science 1996, 271, 1397-1400; (b) Cramer, C. J.; Tolman, W.

B. Acc. Chem. Res. 2007, 40, 601-608.

2. Metal ion - catalyzed oxidative degradation of Orange II by H2O2

17

2 METAL ION - CATALYZED OXIDATIVE DEGRADATION OF

ORANGE II BY H2O2.

2.1 GERNERAL REMARK

The following chapter is based on the original publication: Metal ion – catalyzed oxidative

degradation of Orange II by H2O2. High catalytic activity of simple manganese salts, Erika Ember,

Sabine Rothbart, Ralph Puchta and Rudi van Eldik, New J. Chem., 2009, 33, 34-49.

2.2 INTRODUCTION

Nowadays, one of the major environmental problems concerns the strong increase in

xenobiotic and organic substances that are persistent in the natural ecosystem. Most of these

compounds have an aromatic structure, which makes them highly stable and thus difficult to

degrade.[1a] A significant source of environmental pollution is industrial dye waste due to their

visibility and recalcitrance, since dyes are highly colored and designed to resist chemical,

biochemical and photochemical degradation.[1b] About half of the global production of synthetic

dyes (700,000 t per year) are classified as aromatic azo compounds that have a -N=N- unit as

chromophore in their molecular structure. Over 15 % of textile dyes are lost in waste water

streams during the dyeing operation.[1c] Azo dyes are known to be largely non-biodegradable

under aerobic conditions and to be reduced to more hazardous intermediates under anaerobic

conditions.[1d] The decolourisation of wastewater has acquired increasingly importance in recent

years, however, there is no simple solution to this problem because the conventional

physicochemical methods are costly and lead to the accumulation of sludges.[1e]

One approach to solve these problems would be to develop low-cost, highly efficient, and

environmentally compatible oxidation catalysts on the basis of transition metal complexes.[2a-d]

Recently, photodegradation methods based on TiO2 as a photocatalyst,[2c] beside Fenton

systems,[2d] emerged as one of the most promising technologies and received increasing attention

due to their practical and potential value in environmental protection. However, they are only

successful under specific pH and temperature conditions.


18

Several studies were performed during the last few years in order to find good catalysts for the

oxidative degradation of different organic dyes. From an environmental point of view, first row

transition metals are the most challenging. Highly effective Fe,[3, 4] Co,[5] Cr[6] and Mn[2b, 7] based

oxidation catalysts were developed. In combination with different oxidizing agents, the

decomposition of stable organic substances was possible. A novel highly active and

environmentally benign catalytic system based on Fe-TAML (TAML = tetraamido macrocyclic

ligand) was newly reported by Collins et al.[4] However, besides the tremendous synthetic efforts

to obtain an effective catalytic system, the presence of high concentrations of oxidizing agents is

required in many cases.

Among the possible oxidizing agents, H2O2 is one of the most commonly used owing to its eco-

friendly nature. The use of H2O2 as a green oxidizing agent in these reactions is justified by a low

organic content of the wastewater to be treated and a low reaction temperature, thus requiring

the presence of an adequate catalyst due to the high kinetic activation barrier of such reactions.

Commonly used methods for activation of H2O2 include the formation of reactive peroxyacids from

carboxylic acids and peroxycarboximidic acid from acetonitrile (Payne oxidation),[8a] the generation

of peroxyisourea from carbodiimide in the presence of either a weak acid or a mild base,[8b] or the

use of percarbonate, persulfate or perborate in strongly basic solution.[8c] In order to achieve fast

oxidative transformations, the use of large amounts of co-catalyst additives is often required.[9]

Among these, the use of percarbonate, a versatile oxidizing agent, is preferred for environmental

reasons.[10a, b] Oxidation using environmentally benign oxidants has aroused much interest,[10c-e]

because chemical industry continues to require cleaner oxidation, which is an advance over

environmentally unfavoured oxidations and a step up from more costly organic peroxides.[11]

In this report, we propose a fast and clean catalytic oxidative degradation of Orange II as

model substrate by H2O2 in aqueous carbonate solution under mild reaction conditions, pH 8 - 10

and 25 °C, Scheme 2.1.

Scheme 2.1 Catalytic oxidative degradation of Orange II as model substrate by H2O2 in aqueous carbonate solution

under mild reaction conditions, pH 8 - 10 and 25 °C.


19

Starting from commercially available MnII(NO3)2 in aqueous carbonate solution for catalytic

applications, various aspects of the in situ generation of very reactive high valent manganese

intermediates in the presence of H2O2 were studied.

Baes and Mesmer have shown that manganese salts in aqueous solution are able to form very

reactive aquated intermediates.[12] Moreover, in an alkaline medium, the introduction of a

hydroxy ligand trans to a water ligand is expected to produce more labile OH-Mn-H2O species, and

their formation (Scheme 2.2) is considered to be of major importance for their catalytic activity.

Scheme 2.2 Formation of manganese(II)hexaquacomplex and deprotonation in alkaline medium. L = Cl-, NO3

-, SO4

2-,

CH3COO-.

In the present study, the formation of catalytically inactive MnII(OH)2 species was observed at

higher pH, leading to deactivation of the produced Mn intermediates. The activation of H2O2 in the

presence of manganese salts as a function of pH and carbonate concentration was therefore

monitored using UV/Vis spectrophotometry. In situ formed, high-valent manganese intermediates

are known to be highly unstable in the absence of a spectator ligand. As the study progressed, it

was of importance to investigate the role of the azo dye as a potential coordinating ligand to

stabilize the produced intermediate under different reaction conditions. Electrochemical

measurements and DFT calculations were used to develop a better understanding of the

coordination chemistry of Orange II. The successful implementation of such catalytic systems

becomes a worthwhile objective when issues such as environmental compatibility, high atom

economy, availability, and expenses are considered.[13]


20

2.3 RESULTS AND DISCUSSION

2.3.1 General observations

A series of experiments was performed in order to investigate the in situ generation of the

highly reactive manganese catalyst in the oxidative degradation of Orange II by H2O2 under mild

reaction conditions starting with a simple MnII salt. Oxidation reactions are in generally affected by

the protonation state of the substrate, catalyst and oxidant, and the solvent used. It is further

important to note that the studied organic dye (Orange II) can exist in either one of two

tautomeric forms, or in an equilibrium mixture, depending on the process parameters. This kind of

rapid dynamic equilibrium is relevant as one dye species may be more reactive than the other. Azo

dyes containing a hydroxyl group in the ortho position to the azo group within naphthyl or higher

fused ring systems can exist as azo and hydrazone tautomers,[14a] with the relative amounts

varying with reaction parameters such as solvent and temperature.[14b] Furthermore, in aqueous

solution these species are in a pH dependent equilibrium with a common anion, in which the

negative charge is delocalised throughout the molecule (see Scheme 2.3).[14c]

Scheme 2.3 Orange II, pKa = 11.4, λmax = 484 nm.

These are chemically distinct forms which have characteristically different visible spectra, the

azo form absorbs typically at 400 - 440 nm and the hydrazone form at 475 - 510 nm (see Figure

2.1).[14d] The absorption spectrum of Orange II in an aqueous carbonate solution shows under the

selected reaction conditions (Figure 2.1) one main band at 480 nm, which corresponds to the n →


21

π* transition of the azo form. The other two bands at 300 and 270 nm are attributed to the π →

π* transition of the benzene and naphthalene rings, respectively.[14e]

Figure 2.1 UV/Vis spectrum of 10-4

M Orange II in carbonate buffer solution at pH 8.5.

Figure 2.2 (A) UV/Vis spectrum of a 5 · 10-5

M Orange II carbonate (0.4 M HCO3-) solution at pH 8.5 with a 5 · 10

-5 M

MnII(NO3)2 and 0.01 M H2O2. (B) Comparison of the absorbance changes at 484 nm during the reaction of (A) (―) and

the uncatalyzed reaction of 5 · 10-5

M Orange II with 0.01 M H2O under identical reaction conditions.

Orange II, due to the presence of aromatic groups, is very stable, and in the presence of a

powerful bleaching agent such as H2O2, degradation of dye solutions occurs slowly under specific

reaction conditions. Surprisingly, the oxidation rate was tremendously accelerated by addition of a

simple manganese salt (Figures 2.2 (A) and (B)). Compared to the uncatalyzed reaction of H2O2

with Orange II, a rate enhancement of several orders of magnitude is found, as becomes obvious

(A) (B)


22

from the spectral changes at 484 nm in Figures 2.2 (B). The reactivity of the in situ formed

intermediate was comparable with the catalytic activity of some earlier postulated, well known

manganese bleach catalysts[7, 15a] and manganese porphyrines.[15b] In our work, the formation and

stabilization of the active catalyst was studied in a carbonate buffer solution.

2.3.2 Complex-formation between Orange II and MnII

Ortho-hydroxy aromatic azo dyes, which are bidentate complexing agents, are of considerable

practical and theoretical interest because of their ability to form stable chelate complexes with

several metal ions.[16] It is known that Orange II can act as a chelating agent since the hydroxy and

sulfonate groups allow the formation of a stabilized complex.[16b] Addition of Orange II to a freshly

prepared aqueous carbonate solution of a MnII salt results in significant changes in the UV/Vis

spectrum of Orange II as shown in Figure 2.3.

Figure 2.3 (—) UV/Vis spectrum of a 5 · 10-5

M Orange II carbonate (0.1 M HCO3-) solution at pH 8.5 before mixing with

a 5 · 10-5

M MnII(NO3)2 solution at pH 8.5. (—) UV/Vis spectrum recorded directly after mixing (ca. 5 s delay).

UV/Vis spectra recorded before and after mixing (ca. 5 s delay) of 5 · 10-5 M Orange II with 5 ·

10-5 M MnII(NO3)2 showed a significant increase in absorbance at 480, 310 and 228 nm,

respectively. The differences before and after mixing are not profound at low MnII concentrations.

On increasing the MnII concentration, a continuous increase in ΔAbsλ = 480 nm = A(dye + Mn(II)) - Adye was

observed, indicating the formation of an Orange II∙∙∙∙MnII species according to Equation 2.1. It


23

should be noted that at higher MnII concentration, a precipitate started to form. The value of Keq

was determined through a constant variation of the MnII concentration.

Figure 2.4 (A) Change in absorbance at 480 nm on addition of different concentrations of MnII to 5 · 10

-5 M Orange II in

aqueous carbonate solution (0.2 M HCO3-) at pH 8.5 and 22 °C. (B) Job plot analysis for complex-formation between

Orange II and MnII in aqueous carbonate solution (0.2 M HCO3

-) at pH 8.5. (C) Spectral changes at 480 nm on addition

of different concentrations of Orange II to a freshly prepared 5 · 10-5

M MnII(NO3)2 carbonate solution (0.2 M HCO3

-) at

pH 8.5 and 22 °C. (D) Job plot analysis for the complex formation in aqueous carbonate solution (0.2 M HCO3-).

For a correct determination of the complex-formation constant, independent measurements

were performed at constant manganese concentration where the Orange II concentration was

continuously varied (see Figure 2.4 (C)). Independent measurements were repeated between five

(A) (B)

(C) (D)

(2.1)


24

and eight times. Selected data are shown in Figure 2.4 (A), where the solid line represents a fit of

the data to Equation 2.2.

The values of A0 and A∞ represent the absorbances of Orange II and of the complex Orange

II∙∙∙∙MnII, respectively, and Ax is the absorbance at any MnII concentration.

The value of Keq was calculated from Equation 2.2 to be (2.9 ± 0.9) · 104 M-1, indicating a

relatively weak coordination of the dye to the metal center. Experimentally, through addition of a

4 · 10-5 M MnII(NO3)2 solution to a 5 · 10-5 M Orange II aqueous carbonate solution (0.2 M HCO3-), a

decrease in the pH of the solution from 8.5 to 8.3 was observed, which suggests phenolic proton

release due to MnII coordination to Orange II with the formation of a six-membered ring structure

instead of coordination to the terminal sulfonato group.

At higher concentrations (above ca. 1 · 10-3 M) Orange II forms dimers and higher aggregates in

aqueous solutions,[14c, 17a, b] and has a marked effect on the observed spectra, particularly UV/Vis

and NMR.[17c] A Benesi-Hildebrand treatment of the optical data to determine Keq could not be

applied since the concentration of Orange II and MnII were close to each other.[18]

Scheme 2.4 (A) Proposed structure for a 1:1 Orange II∙∙∙∙MnII complex formed in a carbonate buffer solution at a low

concentration of MnII. (B) Proposed structure for a 1:2 Orange II∙∙∙∙Mn∙∙∙∙Orange II complex formed in a carbonate

buffer solution at a high concentration of Orange II.

(A) (B)

(2.2) ][

][

OrangeIIK1

OrangeIIKAbsAbs

eq

eq


25

Using Job’s method,[19] the stoichiometry of the formed complex could be determined.

According to the data shown in Figures 2.4 (B) and (D), at lower MnII concentration the formation

of a complex with a stoichiometry of 1:1 can be assumed. Upon a further increase in the Orange II

concentration, complexes with a higher stoichiometry are possibly formed (see Scheme 2.4).

Similar structures have been reported earlier by Nadtochenko and Kiwi when a Fe3+ salt was

added to an Orange II solution in acidic medium.[20a] Bauer also reported a TiIV complex, where TiIV

is coordinated by two oxygen atoms from the sulfonato group and the oxygen of the carbonyl

group of the hydrazone tautomer.[20b] In the enzyme manganese peroxidase, the double role of

Orange II as a stabilizer, forming a complex with MnIII, and as a substrate that permits the

regeneration of MnII, was recently postulated by López et al.[16b] Although, the coordination of

organic dyes, viz. Alizarin, Alizarin S[21] and Orange II,[1c] to several transition metal centres has

been known for years, comparatively little is found on their use as potential stabilizing ligands in

oxidative degradation of organic dyes.

The formed Orange II∙∙∙∙MnII complex was isolated and the validity of its composition was

confirmed by elemental analysis. In control experiments the reactivity of the isolated 1:1 Orange

II∙∙∙∙MnII and 2:1 complexes were studied. The isolated complexes exhibit the same catalytic

activity and stability under the experimental conditions employed for the in situ generation of the

complex. Due to the weak coordination mode of the ligand, no differences between the catalytic

activity of the 1:1 and 2:1 complex were found.

In addition, to gain more information on the activation mode of the catalyst, two further

experimental procedures based on different activation and stabilization modes of the activated

catalyst, were followed. In one, the catalytic active species was generated in situ in the carbonate

buffer solution by addition of the desired amount of H2O2, followed by the addition of the

corresponding quantity of Orange II to the reaction mixture. In the other, Orange II was added to

the manganese solution and the formation of an Orange II∙∙∙∙MnII complex was observed. The

decomposition of the dye was initiated through the subsequent addition of H2O2. It is important to

note that the catalytic oxidation of the dye by H2O2 could only be performed in an aqueous

carbonate buffer solution. No other buffer at the same pH, viz. TRIS, TAPS, HEPES or phosphate,

showed the observed catalytic reaction.


26

2.3.3 CV studies on the complex-formation between Orange II and MnII

CV measurements of a 4 · 10-5 M MnII solution in the presence of different Orange II

concentrations were performed in order to determine the interaction between the fully aquated

MnII ions and Orange II present in the reaction mixture. Figure 2.5 (A) shows the results of

MnII∙∙∙∙Orange II complex formation in NaCl electrolyte, performed using a standard three

electrode electrochemical setup. To avoid the oxidation of MnII to MnIV, which precipitates as

MnIVO2, the potential scan was discontinued at +1.0 V, after which the reverse scan from +1.0 to -

0.8 V was started. The CVs of MnaqII in the absence of any coordinating substrate exhibit one quasi-

reversible oxidation peak at E = +0.59 V vs. Ag/AgCl and one quasi-reversible reduction peak at E =

+0.35 V, corresponding to the one electron MnIII/MnII redox couple.

In addition, CV measurements on a freshly prepared 4 · 10-5 M Orange II electrolyte solution at

pH 8.5 and 22 °C were performed. Orange II, as it can be seen in Figure 2.5 (B), undergoes two

electrochemically quasi-reversible one-electron reductions with CV half-wave potentials at Ered1 = -

0.19 V and Ered2 = +0.11 V (vs. Ag/AgCl) with a difference between the cathodic and anodic wave of

0.02 and 0.204 V, respectively. As can be seen from Figure 2.5 (B) the reduction potential of MnIII

decreased from +0.35 V to +0.28 V when Orange II was added to the solution, indicating the

stabilization of MnIII ions. In the presence of a chelating substrate, the generated MnIII complex

becomes more stable and the redox potentials attain lower values.[22]

Figure 2.5 CVs of a 4 · 10-5

M MnII solution (0.1 M NaCl electrolyte) in the presence of different Orange II

concentrations at pH 8.5 (adjusted by addition of NaOH) and 22 °C.

(A) (B)


27

1a 1b

When the concentration of Orange II was increased up to 2 · 10-5 M, the presence of further

reduction peaks along with changes in the oxidation peak intensity were observed, indicating the

formation of other manganese-Orange II species as specified above.

2.3.4 DFT-calculations

To assess the coordination mode of Orange II to the MnII center, DFT (B3LYP/LANL2DZp)

calculations were performed for a series of plausible complexes. Orange II dissociates in aqueous

solution under the selected experimental conditions into an anionic sulfonate group and a cationic

sodium ion. In the presence of an unsolvated SO3- group involving charge transfer from the

electron-rich sulfonate group onto the rest of the molecule, may in general not give satisfactory

DFT results.[14b] Solvent Yellow 14, a model compound for Orange II containing no sulfonate group

was selected for the DFT study of the interaction between the MnII and the chosen azo dye. A

picture of the calculated conformers of the model compound 1 is shown in Figure 2.6.

The optimized geometry of 1a was calculated to be ca. 5.8 kcal mol-1 lower in energy than that

of 1b. Furthermore, the calculated structure of 1a was compared with X-ray structural data of

Solvent Yellow 14.[23] A good agreement between calculated and crystallographically determined

structure was found.

Figure 2.6 Optimized (B3LYP/LANL2DZp) structures of 1a and 1b with a planar geometry and dihedral angles of (a)

180.0° and (b) 178.7° about the azo group, C-N-N-C.

0 kcal mol-1 +5.8 kcal mol-1


28

2a 2b

0 kcal mol-1 +2.9 kcal mol-1

2c

ca. +2 kcal mol-1

According to the UV/Vis and electrochemical data presented above, Orange II can coordinate

to a fully aquated MnII center. Different plausible interaction modes of Solvent Yellow 14∙∙∙∙MnII (2)

and Solvent Yellow 14∙∙∙∙MnII∙∙∙∙Solvent Yellow 14 (3) were studied in detail. Optimized structures

of 2 adopting different coordination modes are presented in Figures 2.7 and 2.8. The studied

organic dye can coordinate to aquated MnII ion by forming two new bonds, one between MnII and

the deprotonated phenolic OH-group of 1a and the second between MnII and one of the azo

nitrogen atoms, leading to the formation of either a planar six-membered (2a) or five-membered

(2b) chelate complex.

Figure 2.7 Optimized structures of complexes 2a and 2b (B3LYP/LANL2DZp).


29

3

Furthermore, a second interaction mode for 2 involving a hydrogen bond between one

coordinated water molecule and the azo nitrogen atom 2c was taken into consideration. The

calculated energies indicate that 2a is energetically favoured over 2b by about 2 kcal mol-1. The N-

N bond length of 1.30 Å for 2a is nearly identical to that found in the free model molecule 1a (1.28

Å), indicating a weak interaction between the nitrogen atoms and the positively charged

manganese center.

In addition to these structures, DFT calculations were performed for a further possible

interaction of a second dye molecule with the MnII center leading to the formation of chelated

MnII innersphere complexes. Similar transition metal complexes of ortho-hydroxy azo dyes were

prepared and characterised by Drew and Landquist.[16c] The introduction of a second dye molecule

is expected to have certain advantages. In addition to the usual stabilization by the chelate-effect,

the introduction of a second molecule of 1a could result in a protecting effect on the coordination

framework. The optimized structure of 3 adopting different coordination modes is presented in

Figure 2.8.

Figure 2.8 Optimized structure of complex 3 (B3LYP/LANL2DZp).

The calculated structure of 3 shows a C2-symmetry and the axial positions are nearly

equivalent. The calculated Mn-N bond lengths in the equatorial plane for the energetically


30

favoured 2a (2.15 Å) and 3 (2.30 Å) are comparable with the X-ray structural data for MnII

complexes with nitrogen containing ligands such as 1,2-bis(imidazol-1-yl)ethane (bim) (2.213 -

2.294 Å),[24a] 2-[N,N-bis(2-pyridylmethyl)-amoniumethyl]-6-[N-(3,5-di-tert-buthyl-2-oxido-benzyl)-

N-(2-pyridylamino)-aminomethyl]-4-methylphenol (H2Ldtb) (2.118 - 2.237 Å)[24b] and 1,4,7-

triazacyclononane (tacn) (2.118 - 2.146).[24c]

As expected, upon coordination of two dye molecules in 3, the N-N bond distance becomes

longer (1.29 Å) than observed in the crystal structure of 1a due to the partial neutralization of the

delocalized negative charge of the nitrogen atom. The elongation of the Mn-O bond trans to the

azo group (Mn-O = 2.38 Å versus 2.06 - 2.27 Å for 2, and Mn-O = 2.27/2.26 Å versus 1.81/2.11 Å for

3) exerts a significant trans influence opposite to the Mn-N bond. The increased lability of the axial

ligand allows the subsequent interaction of the substituted transition metal atom with an oxidant,

leading to the rapid formation of active oxidizing species. Moreover, DFT calculations performed

by Blomberg et al. suggest that in the presence of weak-field ligands for MnII and MnIII, five-

coordination is also accessible whereas MnIV has a much stronger preference for six-

coordination.[25]

2.3.5 Kinetic investigations

2.3.5.1 Complex-formation between bicarbonate and MnII

The reactions between bicarbonate ions (HCO3-) and different manganese species have been

studied for several years, since aquated MnII cations themselves are actually not able to catalyze

H2O2 disproportionation. Depending on the HCO3- concentration in the reaction mixture,

MnII∙∙∙∙HCO3- complexes of different stoichiometry can be formed. Recently, it was suggested that

only the neutral MnII(HCO3-)2 complex can facilitate H2O2 disproportionation.[26] In this study the

complex-formation reaction between MnII and HCO3- was monitored using UV/Vis

spectrophotometric beside CV measurements as a function of carbonate concentration at pH 8.5.

UV/Vis spectra recorded before and after addition of HCO3- to an aqueous MnII solution showed

the formation of a new broad band at 300 nm as illustrated in Figure 2.9 (A). The time course of

the absorption band formation is shown in Figure 2.9 (B).


31

Figure 2.9 (A) UV/Vis spectra of an aqueous 4 · 10-4

M MnII solution before (—) and after (—) addition of 0.4 M HCO3

-

at pH 8.5. (B) Time course of the band formation at 300 nm of an aqueous 2 · 10-4

M MnII solution containing different

amounts of HCO3-.

It can be seen from Figure 2.9 (B) that formation of the manganese carbonate intermediate is

enhanced at higher carbonate concentration. The observed first order rate constants following the

induction period in Figure 2.9 (B), are directly proportional to the [HCO3-] in the range 0.01 - 0.5 M

(see Figure 2.10) with a second order rate constant of (3.6 ± 0.2) · 10-2 M-1 s-1 at 25 °C.

Figure 2.10 Plot of observed first order rate constant (kobs) for the formation of MnII∙∙∙∙HCO3

- versus the bicarbonate

concentration in the presence of 4 · 10-4

M MnII at pH 8.5 and 25 °C.

(A) (B)


32

Moreover, the observed induction period is probably related to the displacement of water

from the first coordination sphere of the fully aquated MnII ion by HCO3- and subsequent

rearrangement of the coordinated ligand, viz. formation of bidentate carbonate complexes. It

should be noted that under these experimental conditions (high carbonate concentration and pH

8.5) insoluble MnIICO3 is formed as a very fine white precipitate at longer reaction times. Its

composition was confirmed by elemental analysis and IR spectroscopy. The reactivity of the

produced intermediate was tested in the oxidative degradation of Orange II by H2O2 at pH 8.5.

During the first 200 s, no change in the reactivity of the in situ formed manganese intermediate

occurs. A significant time dependent loss in catalytic efficiency of the formed MnII∙∙∙∙HCO3-

intermediate was observed after more than 200 s. An irreversible deactivation of the catalyst

occurs within less than 20 min.

On the other hand, no precipitate formation as well as no deactivation of the catalytically

active manganese intermediate could be observed in the presence of a coordinating organic

substrate, i.e. Orange II, over a long period of time (1 - 4 days) in a high carbonate (0.5 M)

containing buffer solution under these conditions. Moreover, this aspect of the stabilization of the

in situ formed active catalyst in the presence of an organic substrate is of considerable practical

interest, because its successful implementation could offer a more efficient alternative for clean

oxidation reactions.

CV measurements of freshly prepared aqueous MnII(NO3)2 solutions were performed in the

presence of different carbonate concentrations in a 0.1 M NaCl electrolyte solution at pH 8.5

(adjusted by careful addition of NaOH) and 22 °C. In the presence of a coordinating substrate, the

displacement of a coordinated water molecule from the manganese coordination sphere takes

place. By coordination of a negatively charged ligand such as HCO3- to a positively charged metal,

the peak potentials are shifted to more negative potentials compared to the fully aquated MnII

(see Figures 2.11 and 2.12).[20b]


33

(2.3)

(2.4)

Figure 2.11 (A) Cyclovoltammograms for 4 · 10-5

M MnII in an aqueous solution of 0.1 M NaCl and different

concentrations of NaHCO3. (B) Typical multiple scan CVs of a 4 · 10-5

M MnII solution in the presence of 0.2 M NaHCO3

and 0.1 M NaCl at pH 8.5 and 25 °C.

On increasing the carbonate concentration in solution a decrease in the peak current intensity

occurs concomitantly with peak broadening because of complexation by carbonate. Typical

multiple scan CVs of a 4 · 10-5 M MnII solution in the presence of 0.2 M NaHCO3 and 0.1 M NaCl at

pH 8.5 and 25 °C is presented in Figure 2.11 (B). In the presence of a chelating substrate, the

generated MnIII complex becomes more stable and the redox potentials attain lower values.

Moreover, at higher carbonate concentrations in the reaction mixture the presence of a second

oxidation peak at E = +0.41 V, attributed to the formation of further complexes such as proposed

in Equation 2.3 and 2.4, was observed.

By plotting the peak potential E as a function of the hydrogen carbonate concentration (see

Figure 2.12), the presence of different complex species at different carbonate concentrations is

revealed.

(A) (B)


34

Figure 2.12 Plot of peak potential E as function of [HCO3-] (E vs. Ag/AgCl electrode). Reaction conditions: [Mn

II] = 4 · 10

-

5 M, [HCO3

-] = 0.1 - 05 M in 0.1 M NaCl electrolyte solution at pH 8.5 and 22 °C.

2.3.5.2 The effect of the total carbonate concentration

The effect of the carbonate concentration on the oxidative degradation of the dye was studied

at a constant pH of 8.5. The total carbonate concentration was varied between 0.05 and 0.5 M. In

the present case, the catalytic reaction leads to a square dependence of kobs on the HCO3-

concentration with a third rate constant (8.3 ± 0.3) · 10-2 M-2 s-1 (see Figure 2.13 (A)), suggesting

that two equivalents of HCO3- are involved in the oxidation mechanism. It is assumed, among

other possibilities, that one equivalent of HCO3- is required for the formation of the more reactive

[MnII(H2O)5(HCO3-)]+ intermediate, and the second equivalent of HCO3

- is necessary for the

formation of the more reactive peroxocarbonate species, known to be a versatile oxidizing agent.

Although the oxidation potential of peroxymonocarbonate and hydrogen peroxide have equal

values (E°(H2O2/H2O) = 1.77 V (vs. NHE), E°(HCO4-/HCO3

-) = 1.8 ± 0.1 V) the higher reactivity of

peroxymonocarbonate compared to that of H2O2 is attributed to carbonate being a better leaving

group than hydroxide.[27d] This is also apparent in the non-catalyzed reaction of H2O2 and Orange II

under variation of the total carbonate concentration (see Figure 2.13 (B)).


35

Figure 2.13 (A) Plot of the observed first order rate constants (kobs) for the MnII catalyzed degradation of Orange II vs.

the [HCO3-]. Reaction conditions: 2 · 10

-5 M Mn

II, 5 · 10

-5 M Orange II, 0.01 M H2O2, pH 8.5, 25 °C. (B) Comparison of the

absorbance changes at 480 nm versus time for the spontaneous, non-catalyzed oxidative degradation of 5 · 10-5

M

Orange II by 0.01 M H2O2 at pH 8.5 and different total carbonate concentrations.

The anionic peracid HCO4- forms in water as well as in mixed organic-water solvents at near

neutral pH values.[27] The structure of peroxymonocarbonate, HOOCO2-, is analogous to other

peroxides such as peroxymonosulfate (HOOSO3-), peroxynitrate (O2NOO-), and peracetic acid

(CH3C(O)OOH), which on the other side can only be prepared by the reaction of their

corresponding acids with H2O2 in strongly acidic aqueous solutions.[28] Borate, like bicarbonate, is

an exception and readily forms reactive peroxides in neutral to mildly alkaline solutions.[27, 29]

The peroxymonocarbonate ion has been isolated in various salts and characterized by Raman

and [27a, 30] NMR-spectroscopy.[27a, d] This inorganic peracid is known to be several orders of

magnitude more reactive toward nucleophilic substrates than H2O2 itself[29] and is formed in a

relatively fast pre-equilibrium (K = 0.32 ± 0.02 M-1)[27a] between hydrogen carbonate ions and H2O2

shown in the overall reaction in Equation 2.5.

Moreover, the reaction of H2O2 and HCO3- to form the more electrophilic HOOCO2

- (HCO4-)

occurs rapidly (t1/2 ≈ 300 s) in 1.76:1 (v/v) ethanol/water at 25 °C.[27d] This step is also regarded to

be a key aspect of several oxidation reactions.[27b-d] A more recent kinetic study performed by

(A) (B)

(2.5)


36

Richardson and coworkers suggests that the mechanism of peroxymonocarbonate formation

proceeds via CO2 as intermediate by reaction of CO2 with H2O2 (perhydration) and its conjugate

base HOO- (base-catalyzed perhydration).[27e]

Figure 2.14 (A) 13

C - NMR spectra recorded for the in situ formation of HCO4- in an aqueous solution of 0.36 M H

13CO3

-

and 1 M H2O2 at pH 8.5 and 25 °C. (B) Time course for the consumption of H13

CO3- and the formation of HCO4 for a

solution of 0.36 M H13

CO3- + 1 M H2O2 and 0.2 M H

13CO3

- + 1 M H2O2. Concentrations were calculated from the relative

peak intensities. Reaction conditions: pH 8.5, 25 °C.

By use of 13C-labeled sodium bicarbonate we were able to follow the in situ formation of

peroxycarbonate in the absence of organic co-solvent. In Figure 2.14 (A) the time dependant

changes of the NMR spectra show a fast appearance of a new signal at 161.7 ppm, attributed to

HOO13CO2-. Figure 2.14 (B) shows the changes in concentration over time for a 0.36 M and 0.2 M

H13CO3- solution, respectively, with 1 M H2O2 at pH 8.5. The concentration values were calculated

from the relative peak intensities of the corresponding NMR-signals. According to Equation 2.6 we

were able to determine the value of the equilibrium constant as K = 0.32 ± 0.02 M-1 and thereby

confirm the literature reported value in purely aquatic solution.[27d]

In the view of these findings we decided to study the influence of carbonate on the manganese

catalyzed oxidation of Orange II by H2O2 and HCO4-, respectively. By performing the oxidation

reactions in the presence of peroxymonocarbonate instead of H2O2 in a 0.5 M carbonate

(A) (B)

(2.6) ]][[

][

3

4

22

-

-

OHHCO

HCOK


37

containing buffer solution at pH 8.5, no difference in the reactivity was observed (Figure 2.15 (A)).

The MnII catalyzed oxidative degradation of Orange II by using HCO4- as an oxidizing agent could be

significantly enhanced through increasing the total carbonate concentration in the reaction

mixture (Figure 2.15 (B)). This can be explained in terms of the equilibrium formulated in Equation

2.5. Based on our experimental observations and aspects reported in the literature[27b] for the MnII

catalyzed oxidation reaction by H2O2 in a carbonate containing solution, the reaction sequence

presented in Scheme 2.5 can be suggested to occur.

Figure 2.15 (A). Spectral changes observed at 480 nm for the 2 · 10-5

M MnII(NO3)2 catalyzed oxidative degradation of

2.5 · 10-5

M Orange II in the presence of (-) 0.01 M H2O2 and (-) 0.01 M HCO4-, respectively, at pH 8.5 and 0.5 M total

carbonate concentration. (B) Comparison of the absorbance changes at 480 nm versus time for the 2 · 10-5

M MnII

catalyzed oxidative degradation of 5 · 10-5

M Orange II by 0.01 M H2O2 at pH 8.5 and different carbonate

concentrations.

Addition of H2O2 to hexaaqua MnII in a carbonate solution leads to significant spectral changes

in the UV/Vis spectra during the reaction (see Figure 2.16 (A)). The initial rapid increase of the

intensity of the broad band at 300 nm, as it is illustrated in Figure 2.16 (A), is attributed to the fast

formation of [MnII(H2O)5(HCO3)]+. An isosbestic point at 330 nm suggests the formation of a new

manganese intermediate by addition of an oxidizing agent, i.e. H2O2.

According to our spectroscopic observations the formed complex with an absorption band at

400 nm could be attributed to a MnIV-η2-peroxycarbonate intermediate.[31, 32a] Based on

spectroscopic observations and data reported in the literature,[32a] the formed intermediate can

be most likely regarded to be a high valent manganese complex. Similar Rh,[32b] Pt[32c] and Fe[32a]

(A) (B)


38

peroxycarbonate complexes have been isolated before and were characterized spectroscopically.

The time course of the absorption band at 400 nm at different pH is illustrated in Figure 2.16 (B).

Scheme 2.5 In situ formation of catalytically active Mn intermediates in the presence of hydrogen peroxide in a

carbonate containing aqueous solution at pH 8.5 and 25 °C.

In the absence of any stabilizing ligand, the formed complex rapidly decomposes with the

formation of catalytically inactive MnIVO2 that precipitates from solution (see Figures 2.16 (A) and

(B)). The decomposition of the active intermediate is accelerated at higher pH (see Figure 2.16

(B)). To ascertain that the formulated reaction steps in Scheme 2.5 are valid under our reaction

conditions, a systematic spectroscopic investigation at different pH values was performed.

Representative data for the reaction course at 400 nm at pH 8.5 and 9.5 are presented in Figure

2.16 (B).

Contrary to our expectations, an increase of one unit in pH resulted in an increase of the

induction period and a decrease in the manganese peroxycarbonate complex formation rate

under the mentioned reaction conditions. This could be partly due to subsequent formation of

MnII(OH)2 precipitates at higher pH and to deprotonation of HCO3- that becomes significant at pH

above 9. This results in a decrease in the HCO3- concentration in the equilibrium presented in

Equation 2.5, reducing the concentration of peroxymonocarbonate present in solution.


39

Figure 2.16 (A) UV/Vis spectra recorded for the reaction of 2 · 10-4

M MnII(NO3)2 with 10

-3 M H2O2 in a 0.5 M HCO3

-

containing solution at pH 8.4 and 25 °C. (B) Comparison of typical absorbance at 400 nm versus time plots at pH 8.5

(—) and 9.5 (—).

2.3.5.3 Reactivity profile as function of pH

The reactivity of the catalytic system is generally influenced by the protonation state of the

substrate, the catalyst and oxidizing agent. In our work the kinetics were studied in 0.4 M NaHCO3

containing buffer solution in the pH range between 8.0 and 9.5 at 25 °C. The pH of the carbonate

buffer solution was adjusted carefully using small amounts of concentrated NaOH solution to

avoid dilution. A typical manganese catalyzed oxidative degradation of Orange II by H2O2 in a

carbonate buffer solution is presented in Figure 2.17 (A). The catalytic degradation is usually

complete within 1 - 10 min depending on the pH of the solution, the catalyst concentration, and

the H2O2 and carbonate concentrations.

The dye decomposition was followed by monitoring the spectral changes at 484 nm. The

depletion of the band at 484 nm is in general correlated with cleavage (heterolytic or homolytic) of

the azo group leading to colorless oxidation products due to the induced discontinuity in the

conjugation of the π-system in the molecule. The inset in Figure 2.17 (A) shows the first spectrum

of Orange II before the addition of the catalyst and H2O2, and the final spectrum recorded after

250 s. A decrease in the intensity of the two other bands at 270 and 300 nm was observed,

showing that further bleaching also occurs under these reaction conditions. The formation of

small, non-toxic and biodegradable organic molecules, i.e. glyoxalic acid, 4-

hydroxybenzenesulfonic acid or acetic acid through a ring-opening reaction is one of the positive

(B) (A)


40

aspects of this process.[4] The isolation and characterization of reaction products is extremely

difficult and requires large synthetic efforts, particularly as different reaction intermediates tend

to react further under these experimental conditions. A comparison of the reaction course at

different pH value is shown in Figure 2.17 (B).

Figure 2.17 (A) First and last UV/Vis spectra of a 2 · 10-5

M MnII(NO3)2 catalyzed oxidative degradation of 5 · 10

-5 M

Orange II by 0.01 M H2O2 in a 0.4 M total carbonate containing solution at pH 8.5 and 25 °C. (B) Comparison of

absorbance at 480 nm vs. time plots for the 2 · 10-5

M MnII(NO3)2 catalyzed oxidative degradation of 5 · 10

-5 M Orange

II by 0.01 M H2O2 in a 0.4 M total carbonate containing solution at different pH values and 25 °C.

The MnII catalyzed decolorization and oxidative decomposition of Orange II was found to be

sensitive to the pH of the solution. According to our experimental data, an increase in pH resulted

in a slight decrease in the reaction rate under the above-mentioned reaction conditions and the

highest reactivity is observed at a pH between 8.2 and 8.6. Increasing the pH to > 9 leads to a

decrease in the oxidation rate for the bicarbonate-activated peroxide, which is presumably the

result of the deprotonation of HOOCO2- to form CO4

2-, a less electrophilic oxidant.[33]

At even higher pH, the decomposition of the peroxide is accelerated and may reduce the

oxidation reaction rate. Contrary to our expectations, the observed rate constants for the

decolorization reaction of Orange II are similar to the destruction rate constants of naphthalene

and benzene rings, long-lived intermediates, under the studied conditions (see Figure 2.18). Thus,

for a complete oxidation of these stable molecules higher concentrations of oxidant and catalyst

are required.

(A) (B)


41

Figure 2.18 Plot of observed rate constant (kobs) calculated for the decolouring reaction followed at 480 and 300 nm,

respectively. Experimental conditions: 2 · 10-5

M MnII(NO3)2, 5 · 10

-5 M Orange II, 0.01 M H2O2, 0.4 M total carbonate

and 25 °C.

A similar screening using MnIICl2, MnII(Ac)2 and MnII(SO4)2 showed identical catalytic activity in

the oxidative degradation of Orange II by H2O2. In all cases, the manganese catalyzed oxidative

degradation of Orange II is favored by moderate alkaline pH values and vanishes completely at

very high or very low (strong acidic) values. According to the experimental observations

mentioned above, the manganese catalyzed oxidative degradation of Orange II by H2O2 in a

carbonate containing solution is considerably inhibited at higher pH values due to the lower

formation of the high valent manganese η2-peroxycarbonate complex (see Figure 2.17 (B)).

2.3.5.4 Effect of the [MnII] and [H2O2] on the oxidative reaction course

To evaluate the effect of the catalyst concentration on the manganese catalyzed oxidative

degradation of Orange II by H2O2 under catalytically relevant experimental conditions, kinetic

studies were performed for solutions in which the carbonate containing water solution with

various amounts of MnII(NO3)2 was added in the presence of 0.01 M H2O2 to a 0.05 M Orange II

solution at 25 °C. The obvious accelerating ability of the HCO3- ions prompted us to study the

catalytic reaction course in more detail at four different carbonate concentrations. The in situ

produced catalyst concentration dependence was studied at 480 nm using in situ UV/Vis

spectroscopic measurements and the kinetic traces could be adequately fitted to a single


42

exponential function. Plots of the observed rate constant as a function of [Mn II] at different

carbonate concentrations are presented in Figure 2.19.

Figure 2.19 MnII(NO3)2 concentration dependence of kobs. Reaction conditions: 5 · 10

-5 M Orange II, 0.01 M H2O2, pH

8.5 and 25 °C.

As it is evidenced the [MnII] dependences of the observed rate constants for the manganese

catalyzed oxidative degradation of Orange II by H2O2 in a low carbonate concentration containing

solution (0.1 - 0.3 M HCO3-) are strongly curved (higher K values, see Table 2.1) and reach a limiting

value at higher catalyst concentration. In contrast, similar data at higher carbonate concentrations

(0.4 - 0.5 M HCO3-) result in a less curved dependence of kobs on the catalyst concentration, i.e.

lower K values (see Table 2.1). The observed rate profile can be explained by the general reaction

mechanism proposed in Scheme 2.5 and simplified in Scheme 2.6. The observed rate law for the

proposed reaction steps is given by Equation 2.7. The calculated k and K values from the non-

linear concentration dependences in Figure 2.19 are summarized in Table 2.1.

Scheme 2.6 Proposed reactions steps for the formation of the catalytically active manganese intermediate in the

presence of H2O2 in a carbonate containing solution.


43

(2.7)

[HCO3-], M k, s-1 K, M-1

0.1 0.0033 34.6 · 103

0.3 0.032 17.6 · 103

0.4 0.051 17.8 · 103

0.5 0.138 15.2 · 103

Table 2.1 The constants k and K for the MnII(NO3)2 catalyzed oxidation of Orange II by H2O2 at pH 8.5 and 22 °C

Figure 2.20 H2O2 concentration dependence of kobs. Reaction conditions: 5 · 10-5

M Orange II, 2 · 10-5

M MnII(NO3)2, pH

8.5 and 25 °C.

The effect of H2O2 on the oxidation reaction course was studied by varying its initial

concentration over a wide range, between 5 – 30 · 10-3 M. At lower H2O2 concentrations (between

1 – 5 · 10-3 M) a fast oxidation reaction occurs in the first few seconds followed by a rapid

consumption of H2O2 resulting finally in a partial and inefficient decolorization of the dye. This

prompted us to study the [H2O2] effect on the catalytic oxidation of the dye at higher

K[Mn(II)]1

kK[Mn(II)]kobs


44

concentrations of H2O2. The kobs values were calculated from a single exponential fit to the

absorbance at 480 nm vs. time plots and showed a linear dependence on the initial H2O2

concentration over the studied concentration range (Figure 2.20).

2.3.5.5 Stability of the in situ formed catalyst

Figure 2.21 (A). Spectral changes observed at 480 nm for the repeated addition of 5 · 10-5

M Orange II to a 2 · 10-5

M

MnII(NO3)2 solution in the presence of 0.01 M H2O2 at pH 8.5 and 0.4 M total carbonate concentration. (B) Spectral

changes observed at 480 nm for a new addition of 5 · 10-5

M Orange II and 0.01 M H2O2 to a 48 h old reaction mixture

containing the catalyst solution under the same experimental conditions as mentioned in (A).

In control experiments the stability of the in situ generated catalyst was studied by repeated

addition of dye and H2O2 to a solution of 2 · 10-5 M MnII(NO3)2 at pH 8.5 (0.4 M HCO3-) and 25 °C

(see Figure 2.21 (A) and (B)). As it can be seen in Figure 2.21 (A), the catalytic cycle could be

repeated several times without any significant loss of activity during the oxidation reaction,

indicating an excellent stability of the in situ formed catalyst. After the fifth cycle the reaction

solution containing the active catalyst was allowed to stay at ambient temperature for 48 h.

Subsequently, the catalytic activity of the in situ formed manganese complex was evaluated again

by performing the oxidation reaction in the presence of freshly added Orange II and H2O2. The

experimental results illustrated in Figure 2.21 (B) provide clear evidence for the high efficiency of

the in situ formed catalyst under the above mentioned experimental reaction conditions.

(A) (B)


45

2.3.6 Mechanistic aspects

Throughout this study, the oxidation reactions were carried out in a thermostated open glass

reactor vessel at ambient temperature in aqueous hydrogen carbonate containing solutions. The

readily available manganese salts, the mild reaction conditions and the operation simplicity and

practicability allow for an easy and green oxidative degradation of the studied organic dye. In

control experiments the catalytic activity of the in situ generated manganese complex was

investigated under an inert atmosphere. By performing the catalytic reaction in a closed glass

reactor under inert reaction conditions, no change in the decomposition reaction rate was

noticed. A comparison of the reaction course carried out under air and inert atmosphere is

illustrated in Figure 2.22. By performing the reaction under inert reaction conditions no significant

differences in the decomposition reaction rate was observed, indicating that HO∙ or HOO∙ radical

formation is not prevalent for this oxidation reaction.

Figure 2.22 Comparison of typical absorbance at 480 nm versus time plots of a 2 · 10-5

M MnII(NO3)2 catalyzed

oxidative degradation of 5 · 10-5

M Orange II by 0.01 M H2O2 in a 0.4 M HCO3- containing solution at pH 8.5 and

ambient temperature performed in the presence of atmospheric oxygen (—) and inert atmosphere (—), respectively.

Taking into account all obtained spectroscopic and kinetic data, the following reaction

schemes can be proposed for the MnII catalyzed oxidative degradation of Orange II by H2O2 in

carbonate solution under catalytically relevant experimental conditions. A key feature of the

proposed reaction mechanism outlined in Scheme 2.7 is that the overall oxidation of Orange II

occurs in a two electron oxidation step leading to the formation of a relatively stable high-valent


46

Mn=O intermediate and transfer of the oxo group to the substrate. Most of the earlier reported

papers on the oxidation reaction catalyzed by several isolated and structurally well defined

manganese complexes have emphasized the formation of a high-valent Mn=O intermediate by the

reaction of manganese with the appropriate oxidant.[11, 34a-e]

According to our observations, HCO3- ions are involved in two catalytically relevant reactions.

HCO3- ions react with aquated MnII present in solution to form a catalytically active Mn-HCO3

-

complex. HCO3- is also involved in a fast equilibrium with H2O2 to form HOOCO2

-, a versatile

heterolytic oxidant. In the following step, through nucleophilic attack of the oxidizing agent on the

MnII center, a MnII-η2-peroxycarbonate complex is formed. The remaining coordination sites in the

first shell will be occupied by water and hydroxyl at a pH between 8 and 10. The principal mode of

the formation of relatively stable high-valent Mn=O intermediates is believed to involve the

heterolytic cleavage of the peroxide bond, as shown in Scheme 2.7. An important role in the

stabilization of the formed Mn=O species is played by the electron donating bicarbonate ions.

Scheme 2.7 Proposed reaction mechanism for the MnII catalyzed oxidative degradation of Orange II by H2O2 in a

carbonate containing aqueous solution at pH between 8 - 9 and 25 °C.


47

This may also account for the unique requirement of HCO3- in the oxidative decomposition of

Orange II catalyzed by simple manganese salts. The further coordination of the substrate followed

by an oxygen transfer step along with the second electron leads to the formation of several

oxidation products and finally to the regeneration of the catalyst. It must be noted that in the

absence of a catalyst, the oxidative degradation of Orange II by addition of an electrophilic

bleaching agent, HOOCO2-, occurs very slowly under certain reaction conditions. The oxidation

mechanism involves a nucleophilic attack of the dye at the electrophilic oxygen of HOOCO2-. In

aqueous solution, proton transfer can lead to the displacement of HCO3- and the slow formation of

oxidized substrate.

Scheme 2.8 Proposed reaction mechanism involving first substrate coordination to MnaqII in a pre-equilibrium step

during the catalyzed oxidative degradation of Orange II by H2O2 in a carbonate containing aqueous solution at pH

between 8 - 9 and 25 °C.

If substrate binding to MnII occurs before the addition of HOOCO2- to the catalyst solution,

following reactions can be assumed to take place during the reaction cycle under the chosen

experimental conditions. In line with the concerns mentioned above, the first step in Scheme 2.8

involves the prior coordination of Orange II to MnII and formation of MnII-Orange II complexes of

different stoichiometry, followed by nucleophilic attack of the oxidant on the MnII center leading

to the formation of Orange II-MnII-peroxycarbonate species. The subsequent scission of the

peroxo bond leads to the formation of high-valent oxo intermediates, as formulated in Scheme


48

2.8. In this case, the formed MnIV=O intermediate is stabilized by Orange II, an electron rich

organic molecule with chelating capacity. The role of Orange II as an axial ligand is also to favor the

homolytic scission of the peroxo bond leading to the MnIV=O intermediate and bicarbonate.

2.4 CONCLUSIONS

A fast and environmentally benign method for the oxidative degradation of a main pollutant

from the dye industry could be achieved using H2O2 in conjunction with catalytic amounts of

relatively non-toxic manganese salts as catalyst precursors in a carbonate containing aqueous

solution under mild reaction conditions. Screening and spectroscopic methods allowed us to study

the catalytic reaction course and to identify some key features of the reaction that reflect upon its

mechanism. Our study revealed that the oxidative degradation of the model substrate Orange II is

catalytic only in carbonate containing aqueous solution. No other buffer containing aqueous

solution could induce the oxidative degradation of Orange II by H2O2 and this led to the implication

of peroxymonocarbonate as a key molecular entity. It was found that the in situ formed high-

valent manganese intermediate possessing one hydrogen carbonate ligand is able to activate

H2O2, but decomposes rapidly with the formation of neutral MnIICO3, which precipitates from

solution as an insoluble white solid. One of the main factors affecting the process efficiency was

the stabilization of the catalytically active Mn complex. Furthermore, by addition of Orange II, the

formation of MnII∙∙∙∙Orange II complexes with different stoichiometry was observed. The

simultaneous σ,π - coordination of the organic dye is well-precedented, and recent DFT studies

support this type of complex formation. The catalytic activity of the formed intermediates was

tested under catalytic reaction conditions.

The kinetic investigations performed at different pH could provide relevant information about

the nature of the oxidizing agent involved in the reaction. It was found that the pH is a critical

issue for the rate of the oxidation process due to its influence on the deprotonation of the

bicarbonate ions, the formation of peroxycarbonate in solution, and the deprotonation of aquated

MnII. The ongoing studies are presently complemented by investigations on different organic

substrates with various functional groups in order to determine the influence of substrate

modification on the catalytic reaction cycle. DFT studies and further kinetic and spectroscopic

investigations should contribute to a better understanding of the catalytic system.


49

2.5 EXPERIMENTAL SECTION

CHEMICALS. Orange II, certified [Acid Orange 7, C.I. 15510, sodium 4-(2-hydroxy-1-

naphthylazo)benzenesulfonate], 99 % was supplied by Sigma-Aldrich and recrystallized from a

Et2O/H2O mixture at 4 °C. Unless otherwise stated, all other dyes were commercially available

(Acros Organics, Germany) and were used without any further purification. Hydrogen peroxide 35

wt. % and sodium percarbonate were of analytical grade and provided by Acros Organics

(Germany). Different manganese salt hydrates (chloride, sulfate, nitrate, acetate, perchlorate)

were of and purchased from Acros Organics (Germany). Sodium bicarbonate, TAPS, TRIS, HEPES

and NaOH were of analytical grade and provided by Acros Organics (Germany). All buffer solutions

were prepared using Millipore Milli-Q purified water. NaH13CO3 (99 atom % 13C) and DMSO-d6

(99.96 % atom % D) were from Sigma-Aldrich.

13C-NMR MEASUREMENTS. All 13C-NMR measurements were performed by using DMSO-d6 internal

standard in a glass capillary. The concentrations of PAA and AcOH were calculated from the

relative peak intensities using the Lorenz fit obtained by the NMRICMA program (developed at the

“Institut de chimie minérale et analytique” of the University of Lausanne in the group of Prof. A. E.

Merbach) of the NMR data. 13C-NMR spectra were recorded at a frequency of 100 MHz on a

Bruker Advance DRX 400WB spectrometer equipped with a superconducting BC-94/89 magnet

system.

CYCLOVOLTAMMETRIC MEASUREMENTS. Cyclovoltammetric (CV) measurements were performed in a

one-compartment three-electrode cell using a gold working electrode (Metrohm) with a

geometrical surface of 0.7 cm2 connected to a silver wire pseudo-reference electrode and a

platinum wire serving as counter electrode (Metrohm). Measurements were recorded with an

Autolab PGSTAT 30 unit at room temperature. The working electrode surface was cleaned using

0.05 μm alumina, sonicated and washed with water every time before use. The working volume of

10 ml was deaerated by passing a stream of high purity N2 through the solution for 15 min prior to

the measurements and then maintaining an inert atmosphere of N2 over the solution during the

measurements. All CVs were recorded for the reaction mixture with a sweep rate of 50 mV s-1 at

25 °C. Potentials were measured in a 0.5 M NaCl/NaOH electrolyte solution and are reported vs.

an Ag/AgCl electrode.


50

DFT-CALCULATIONS. Unrestricted B3LYP/LANL2DZp hybrid density functional calculations,[35a-c,] i.e.,

with pseudo-potentials on the heavy elements and the valence basis set[35] augmented with

polarization functions,[36] were carried out using the Gaussian 03[37] suite of programs. The relative

energies were corrected for zero point vibrational energies (ZPE). The resulting structures were

tested for stability and characterized as minima by computation of vibrational frequencies and the

wave functions were tested for stability. Further energy computations were carried out using the

conductor-like polarizable continuum model (CPCM).[38]

ELEMENTAL ANALYSIS. The measurements were carried out on an elemental analyzer Euro EA 3000

instrument from Hekaltech Gmbh. The analytical method is based on the complete instantaneous

oxidation of the sample by “flash combustion” at 1000 °C, which converts all organic and inorganic

substances into combustion products. The resulting combustion gases are swept into the

chromatographic column by the carrier gas (He) where they are separated and detected by a

thermal conductivity detector.

IR MEASUREMENTS. IR spectra were recorded as KBr pellets using a Mattson Infinity FTIR instrument

(60 AR) at 4 cm-1 resolution in the 400 - 4000 cm-1 range.

SPECTROPHOTOMETRIC TITRATION. UV/Vis spectra were recorded on a Shimdazu UV-2101

spectrophotometer at 25 °C. In the experiments concerning the complexation by different dyes

the tandem cuvette with two separate compartments (0.44 cm path length each), was filled with 1

ml 5 · 10-5 M Orange II stock solution in one, and different concentrations of an aqueous

MnII(NO3)2 solution in the other compartment. The cuvette was placed in the thermostated cell

holder of the spectrophotometer for 10 min. UV/Vis spectra were recorded before and after

mixing the solutions.

SYNTHESIS OF INSOLUBLE MnIICO3. In a 150 ml round flask 3.36 g (0.4 M) NaHCO3 were dissolved in

100 ml doubly distilled water and the pH of the solution was set at 8.5 upon addition of small

amounts of concentrated NaOH solution. To the freshly prepared carbonate solution 1 g (0.04 M)

MnII(NO3)2 was added. The mixture was stirred at room temperature for 15 min during which

MnIICO3 ∙ H2O formed as a white precipitate. The product was filtrated and washed several times

with large amounts of water. IR (KBr pellets): ν (cm-1) 3421 (m), 1416 (vs), 862 (s), 725 (m).

Elemental analysis for Mn1C1H2O4: calculated %C 9.03, %H 1.52. Found: %C 9.38, %H 1.52.


51

SYNTHESIS OF ORANGE II∙∙∙∙MnII COMPLEX. In a 50 ml Schlenk tube 0.014 g (2 · 10-3 M) Orange II were

dissolved in 20 ml doubly distilled water and an aqueous solution of 0.01 g (2 · 10-3 M) MnII(NO3)2

were added dropwise under continuous stirring. The solution mixture was kept for several hours

at room temperature. The formed precipitate was filtered and dried at room temperature. IR (KBr

pellets): ν (cm-1) 3527 (vs), 1619 (s), 1511 (s), 1383 (vs), 1262 (m), 1171 (s), 1120 (s), 1034 (s), 1007

(s), 829 (s), 759 (s), 696 (m), 644 (m), 595 (m). Elemental analysis for Mn1C16H18O10N3S1Na1:

Calculated: %C 36.79, %H 3.47, %N 8.04, %S 6.14, %O 30.63. Found: %C 29.44, %H 3.39, %N 8.05,

%S 4.77, %O 30.31.

SYNTHESIS OF ORANGE II∙∙∙∙MnII∙∙∙∙ORANGE II COMPLEX. An aqueous solution of 0.005 g (1 · 10-3 M)

Mn(NO3)2 was added under continuous stirring to a 0.014 g (2 · 10-3 M) Orange II water solution at

room temperature. The pale yellow precipitate was collected by filtration and dried in air. IR (KBr

pellets): ν (cm-1) 3390 (s), 1619 (s), 1570 (m), 1554 (m), 1520 (vs), 1393 (m), 1260 (m), 1169 (vs),

1119 (vs), 1033 (vs), 1007 (s), 828 (s), 758 (s), 695 (m), 644 (m), 593 (m). Elemental analysis for

Mn1C32H32O15N5S2Na2: Calculated: %C 43.1, %H 3.62, %N 7.23, %S 7.19, %O 26.91. Found: %C

42.99, %H 3.75, %N 7.23, %S 7.00, %O 25.67.

GENERAL PROCEDURE FOR THE H2O2 CATALYZED DYE DEGRADATION REACTIONS. The manganese catalysts were

freshly dissolved in water before use. To a freshly prepared sodium bicarbonate solution, an

adequate amount of NaOH was added to adjust the pH of the solution. Under isothermal

conditions, the desired amount of a concentrated manganese solution was added together with

Orange II, previously dissolved in an aqueous bicarbonate solution, and H2O2. All kinetic data were

obtained by recording time-resolved UV/Vis spectra using a Hellma 661.502 - QX quartz Suprasil

immersion probe attached via optical cables to a 150 W Xe lamp and a multi-wavelength J & M

detector, which records complete absorption spectra at constant time intervals. In a thermostated

open glass reactor vessel equipped with a magnetic stirrer, the freshly prepared catalyst solution

and H2O2 were added to 40 ml of 5 · 10-5 M dye at a pH ranging from 8 to 10 at 25 °C. All kinetic

measurements were carried out under pseudo-first order conditions (i.e. 50 ≤ [oxidant]/[MnII] ≤

1000). The pH of the aqueous solutions was carefully measured using a Mettler Delta 350 pH

meter previously calibrated with standard buffer solutions at two different pH values (4 and 10).

The kinetics of the oxidation reaction was monitored at the λmax of the corresponding dye. First

order rate constants, where possible, were calculated using Specfit/32 and Origin (version 7.5)

software. To estimate the effect of the catalyst and H2O2 concentrations on the catalytic reaction


52

at different carbonate concentrations, stopped-flow kinetic measurements were carried out

additionally using an SX.18MV stopped-flow instrument from Applied Photophysics.


[1] (a) Mielgo, I.; López, C.; Mareira, M. T.; Feijoo, G.; Lema, J. M. Biotechnol. Prog. 2003, 19, 325-

331; (b) Hastie, J.; Bejan, D.; Teutli-Leon, M. N.; Bunce, J. Ind. Eng. Chem. Res. 2006, 45, 4898-

4909; (c) Park, H.; Choi, W. J. Photochem. Photobiol. 2003, 159, 241-247; (d) Baughman, G. L..;

Weber, E. J. Environ. Sci. Technol. 1994, 28, 267-276; (e) Robinson, T.; McMullan, G.;

Marchant, R.; Nigam, P. Bioresource Technol. 2001, 77, 247-255.

[2] (a) Collins, T. J. Acc. Chem. Res. 1994, 27, 279-285; (b) Hage, R.; Iburg, J. E.; Kerschner, J.;

Koek, J. H.; Lempers, E. L. M.; Martens, R. J.; Racherla, U. S.; Russel, S. W.; Swarthoff, T.; van

Vliet, M. R. P.; Warnaar, J. B.; van der Wolf, L.; Krijnen, B. Nature 1994, 369, 637-639; (c)

Bhattacharyya, A.; Kawi, S.; Ray, M. B. Catal. Today 2004, 98, 431-439; (d) Ramirez, J. H.;

Maldonado-Hodar, F. J.; Perez-Cadenas, A. F.; Moreno-Castilla, C.; Costa, C. A.; Madeira, L. M.

Appl. Catal. B 2007, 75, 312-323.

[3] Ramirez, J. H.; Costa, C. A.; Madeira, L. M.; Mata, G.; Vicente, M. A.; Rojas-Cervantes, M. L.;

López-Peinado, A. J.; Martin-Aranda, R. M. Appl. Catal. B 2007, 71, 44-56.

[4] Chahbane, N.; Popescu, D.-L.; Mitchell, D. A.; Chanda, A.; Lenoir, D.; Ryabov, A. D.; Schramm,

K.-W.; Collins, T. J. Green Chem. 2007, 9, 49-57.

[5] (a) Zhiyong, Y.; Bensimon, M.; Laub, D.; Kiwi-Minsker, L.; Jardim, W.; Mielczarski, E.;

Mielczarski, J.; Kiwi, J. J. Mol. Catal. A 2007, 272, 11-19; (b) Chen, X.; Qiao, X.; Wang, D.; Lin, J.;

Chen, J. Chemosphere 2007, 67, 802-808.

[6] Kim, W.; Hur, S. G.; Swang, S. J.; Park, H.; Choi, W.; Choy, J. H. Adv. Funct. Mater 2007, 17,

307-314.

[7] Wieprecht, T.; Xia, J.; Heinz, U.; Dannacher, J.; Schlingloff, G. J. Mol. Catal. A 2003, 203, 113-

128.

[8] (a) Payne, G. B.; Deming, P. H.; Williams, P. H. J. Org. Chem. 1961, 26, 659-663; (b) Majetich,

G.; Hicks, R. Synlett 1996, 649-651; (c) McKillop, A.; Sanderson, W. R. Tetrahedron 1995, 51,

6145-6166.


53

[9] Barker, J. E.; Ren, T. Tetrahedron Lett. 2004, 45, 4681-4683.

[10] (a) Oakes, J. Eur. Pat. Appl. 145091, 1985; (b) Feng, X.-M.; Wang, Z.; Bian, N.-S.; Wang, Z.-L.

Inorg. Chim. Acta 2007, 360, 4103-4110; (c) Kim, C.; Chen, K.; Kim, J.; Que Jr., L. J. Am. Chem.

Soc. 1997, 119, 5964-5965; (d) de Vos, D. E.; Bein, T. J. Chem. Soc., Chem. Commun. 1996,

917-918; (e) Battioni; P.; Renaud, J. P.; Bartoli, J. F.; Rein-Artiles, M.; Fort, M.; Mansuy, D. J.

Am. Chem. Soc. 1988, 110, 8462-8470.

[11] Nakayama, N.; Tsuchiya, S.; Ogawa, S. J. Mol. Catal. A 2007, 277, 61-71.

[12] Baes, C. F.; Mesmer, J. R.; Mesmer, R. E. Am. J. Sci. 1981, 281, 935-962.

[13] Liu, S.-Y.; Nocera, D. N. Tetrahedron Lett. 2006, 47, 1923-1926.

[14] (a) Zollinger, H. Color Chemistry, 3rd ed., Wiley-VCH: Weinheim, Germany, 2003; (b) Abbott, L.

C.; Batchelor, S. N.; Oakes, J.; Gilbert, B. C.; Whitwood, A. C.; Smith, J. R. L.; Moor, J. N. J. Phys.

Chem. 2005, 109, 2894-2905. (c) Hodges, G. R.; Lindsay-Smith, J. R.; Oakes, J. J. Chem. Soc.,

Perkin Trans. 2 1998, 3, 617-628; (d) Oakes, J.; Gratton, P. J. Chem. Soc., Perkin Trans. 2 1998,

9, 1857-1864; (e) Feng, W.; Nansheng, D.; Helin, H. Chemosphere 2000, 41, 1233-1238.

[15] (a) Lindsay-Smith, J. R.; Gilbert, B. C.; Mairata i Payeras, A.; Murray, J.; Lowdon, T. R.; Oakes,

J.; Pons i Prats, R.; Walton, P. H. J. Mol. Catal. A 2006, 251, 114-122; (b) Tokuda, J.; Oura, R.;

Iwasaki, T.; Takeuchi, Y.; Kashiwada, A.; Nango, M. Coloration Technol. 2000, 116, 42-47.

[16] (a) Yoshida, T.; Sawada, S. Bull. Chem. Soc. Jpn. 1975, 48, 345-346; (b) López, C.; García-

Monteagudo, J. C.; Moreira, M. T.; Feijoo, G.; Lema, J. M. Enzyme Microbiol. Technol. 2007,

42, 70-75; (c) Drew, H. D. K.; Landquist, J. K. J. Chem. Soc. 1938, 292-304.

[17] (a) Reeves, R. L.; Maggio, M. S.; Harkaway, S. A. J. Phys. Chem. 1979, 83, 2359-2368; (b)

Asakura, T.; Ishida, M. J. Colloid. Interface Sci. 1989, 130, 184-189; (c) Murakami, K. Dyes

Pigm. 2002, 53, 31-43.

[18] Benesi, H.; Hildebrand, H. J. Am. Chem. Soc. 1949, 71, 2703-2807.

[19] Job, P., Compt. Rend. 1925, 180, 928-930.

[20] (a) Nadtochenko, V.; Kiwi, J. J. Chem. Soc., Faraday Trans. 1997, 93, 2373-2778; (b) Bauer, C.;

Jacques, P.; Kalt, A. Chem. Phys. Lett. 1999, 307, 397-406.

[21] Sychev, A. Y.; Pfannmueller, U.; Isak, V. G. Zh. Fiz. Khim. 1983, 8, 1974-1978.


54

[22] (a) Lesht, D.; Bauman, J. E. Inorg. Chem. 1978, 17, 3332-3334; (b) Dasgupta, J.; Tyryshkin, A.

M.; Kozlov, Y. N.; Klimov, V. V.; Dismukes, G. C. J. Phys. Chem. B 2006, 110, 5099-5111.

[23] Olivieri, A. C.; Wilson, R. B.; Paul, I. C.; Curtin, D. Y. J. Am. Chem. Soc. 1989, 111, 5525-5532.

[24] (a) Zhu, X.; Li, B.; He, X.; Zhang, Y. J. Chem. Crystallogr. 2005, 35, 443-446; (b) Bortoluzzi, A. J.;

Neves, A.; Couto, R. A. A.; Perelta, R. A. Acta Crystallogr., Sect. C: Cryst. Struct. Commun.

2006, C62, m27; (c) Bennur, T.H.; Srinivas, D.; Sivasanker, S.; Puranik, V. G. J. Mol. Catal. A

2004, 219, 209-216.

[25] Blomberg, M. R.; Siegbahn, P. E. M.; Styring, S.; Babcock, G. T.; Aakermark, B.; Korall, P. J. Am.

Chem. Soc. 1997, 119, 8285-8292.

[26] Tikhonov, K. G.; Zastrizhanaya, O. M.; Kozlov, Y. N.; Klimov, V. V. Biochem. 2006, 71, 1270-

1277.

[27] (a) Flanagan, J.; Jones, D. P.; Griffith, W. P.; Skapski, A. C.; West, A. P. J. Chem. Soc., Chem.

Commun. 1986, 20-21; (b) Lane, B. S.; Vogt, M.; DeRose, V. J.; Burgess, K. J. Am. Chem. Soc.

2002, 124, 11946-11954; (c) Richardson, D. E.; Regino, C. A. S.; Yao, H. R.; Johnson, J. V. Free

Radical Biol. Med. 2003, 35, 1538-1550; (d) Richardson, D. E.; Yao, H.; Frank, K. M.; Bennett,

D. A. J. Am. Chem. Soc. 2000, 122, 1729-1739; (e) Bakhmutova-Albert, E. V.; Yao, H.; Denevan,

D. E.; Richardson, D. E. Inorg. Chem. 2010, 49, 11287-11296.

[28] (a) Greenspan, F. P. J. Am. Chem. Soc. 1946, 68, 907; (b) Ball, E.; Edwards, J. O. J. Am. Chem.

Soc. 1956, 78, 1125-1129; (c) Keith, W. G.; Powell, R. E. J. Chem. Soc. A 1969, 90-98.

[29] Swern, D. Organic Peroxides, Ed. Wiley, New York, 1970, 313.

[30] Jones, D. P.; Griffith, W. P. J. Chem. Soc., Dalton Trans. 1980, 2526-2532.

[31] MnIV in the presence of further Mn(II) leads to the formation of Mn(III), a highly unstable

oxidant, according to the reaction: MnIV + MnII 2 MnIII

[32] (a) Hashimoto, K.; Nagatomo, S.; Fujnami, S.; Furutachi, H.; Ogo, S.; Suzuki, M.; Uehara, A.;

Maeda, Y.; Watanabe, Y.; Kitagawa, T. Angew. Chem., Int. Ed. 2002, 114, 1202-1205; (b)

Aresta, M.; Tommasi, I.; Quaranta, E.; Fragale, C.; Mascetti, J.; Tranquille, M.; Galan, F.;

Fouassier, M. Inorg. Chem. 1996, 35, 4252-4260; (c) Hayward, P. J.; Blake, D. M.; Wilkinson,

G.; Nyman, C. J. J. Am. Chem. Soc. 1970, 92, 5873-5878.

[33] Balagam, B.; Richardson, D. E. Inorg. Chem. 2008, 47, 1173-1178.


55

[34] (a) Khenkin, A. M.; Kumar, D.; Shaik, S.; Neumann, R. J. Am. Chem. Soc. 2006, 128, 15451-

15460; (b) Siegbahn, P. E. M.; Carbtree, R. H. J. Am. Chem. Soc. 1999, 121, 117-127; (c) Kondo,

M.; Mitsui, T.; Ito, S.; Kondo, Y.; Ishigure, S.; Dewa, T.; Yamashita, K.; Nakamura, J.; Oura, R.;

Nango, M. J. Colloid Interface Sci. 2007, 310, 686-689; (d) Shul´pin, G. B.; Matthes, M. G.;

Romakh, V. B.; Barbosa, M. I.; Aoyagi, J. L. T.; Mandelli, D. Tetrahedron 2008, 64, 2143-2152;

(e) Anastasi, A. E.; Walton, P. H.; Lindsay-Smith, J. R.; Sameera, W. M. C.; McGrady, J. E. Inorg.

Chim. Acta 2008, 361, 1079-1086.

[35] (a) Becke, A. D. J. Phys. Chem. 1993, 97, 5648-5652; (b) Lee, C.; Yang, W.; Parr, R. G. Phys. Rev.

B, 1988, 37, 785-789; (c) Stephens, P. J.; Devlin, F. J.; Chabalowski, C. F.; Frisch, M. J. J. Phys.

Chem. 1994, 98, 11623-11627; (d) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 270-283; (e)

Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 284-298; (f) Hay, P. J.; Wadt, W. R. J. Chem.

Phys. 1985, 82, 299-310.

[36] Huzinaga, S. (Ed.), Gaussian Basis Sets for Molecular Calculations, Elsevier, Amsterdam 1984.

[37] Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.;

Montgomery Jr., J. A.; Vreven, T.; Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.;

Tomasi, J.; Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.;

Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima,

T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.;

Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.;

Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg,

J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.;

Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.;

Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.;

Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M.

W.; Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian, Inc., Wallingford

CT, 2004.

[38] (a) Barone, V.; Cossi, M. J. Phys. Chem. A 1998, 102, 1995-2001; (b) Cossi, M.; Rega, N.;

Scalmani, G.; Barone, V. J. Comp. Chem. 2003, 24, 669-681.

2. Metal ion-catalyzed oxidative degradation of Orange II by H2O2

56

3. Comparative study of a MnII-monomer and the corresponding oxo-bridged Mn2III/IV-dimer

57

3 COMPARATIVE STUDY OF A MNII-MONOMER AND THE

CORRESPONDING OXO-BRIDGED MN2III/IV-DIMER

3.1 GENERAL REMARK

The following chapter is based on the original publication: Comparative study of the catalytic

activity of [MnII(bpy)2Cl2] and [Mn2III/IV(µ-O)2(bpy)4](ClO4)3 in the H2O2 induced oxidation of organic

dyes in carbonate buffered aqueous solution, Sabine Rothbart, Erika Ember and Rudi van Eldik,

Dalton Trans. 2010, 39, 3264-3272.

3.2 INTRODUCTION

The development of transition metal complexes as effective catalysts for a wide range of

oxidative transformation reactions and the understanding of the latter processes still represent a

fundamental challenge to inorganic chemists. Inspired by biological systems and under

consideration of environmental aspects, the complexes of choice are bearing an iron or

manganese active site. Manganese complexes with their redox rich chemistry are well-known to

act as outstanding oxygenation catalysts. During recent years many studies have provided

important insight into the relevant manganese catalyzed mechanisms of processes such as

photosynthetic water oxidation,[1] bleaching[2] or epoxidation.[2a, 3] Nevertheless, the nature of the

reactive intermediates in these processes remains to be clarified. Several studies give credence to

the proposal that oxo-bridged MnIV or MnV complexes[1a, 3, 4] are involved as key intermediates. A

fundamental class of complexes among these are µ-oxo-bridged dinuclear complexes of high-

valent MnIII and MnIV, since their involvement in epoxidation,[2a] bleaching[2a] and oxidation of

hydrocarbons[5] and alcohols[5, 6] is known from the literature. As a result, a manifold of applicable

high-valent MnIII and MnIV model complexes have been synthesized, characterized and

investigated towards their catalytic activity in oxygenation reactions.[7] However, there still

remains a vivid debate in the literature regarding the nuclearity of the oxidizing species, since

other studies also endorse monomeric species in higher oxidation states as reactive intermediates

in manganese catalyzed oxidation and epoxidation reactions.[8, 9]


58

Previously we reported kinetic and mechanistic studies on the manganese ion catalyzed

oxidation reaction of various highly stable organic dyes by addition of H2O2 to a MnII containing

buffer solution.[10] The results indicated the formation of high valent MnIV=O as a key catalytic

intermediate by addition of H2O2 to a MnII containing buffer solution.[10] In the view of this finding

we extended our work to elucidate the role of bis-µ-oxo-dimanganese(III,IV) species as

intermediates. It is generally acknowledged that these complexes can be formed from MnII

complexes with labile coordination sites in the presence of a stabilizing ligand and an oxidizing

agent like hydrogen peroxide in basic aqueous media.[11] To have a closer look at the nature of

putative intermediates and the rate-limiting step of the reaction sequence, a comparative study

on the reactivity of a [Mn2III/IV(µ-O)2]3+-dimer and its readily accessible mononuclear analogous

MnII complex in the hydrogen peroxide assisted catalyzed oxidation of Orange II was performed.

As the study progressed, it was important to check our results also for other dyes as model

substrates. We selected p-nitrophenol and Morin, a dye of polyphenolic structure (Scheme 3.1),

which is a common element of chromophores present in fruit, vegetable and tea and therefore an

adequate model substrate for bleach processes.[12]

Scheme 3.1 Structure of the model substrates Orange II, Morin and p-nitrophenol.

A commonly used oxidant with proximity to many biological systems and with beneficial

environmental properties, like no additional redundant waste from organic peroxides or it

economically price, is hydrogen peroxide. Although H2O2 is thermodynamically a potent oxidant its

activity is often kinetically hindered under mild experimental conditions as low temperature or


59

atmospheric pressure. Hence, several methods for the activation of H2O2 involve the use of

conventional catalysts along with co-catalytic additives. In our previous work we showed the

importance of the in situ formed peroxycarbonate anion,[10] HOOCO2-, which is known to be

several orders of magnitude more reactive than hydrogen peroxide itself.[13] Well-established is

also the application of reactive peroxyacids, formed from carboxylic acids, or the use of persulfate

or perborate in basic solution.[14]


3.3.1 Kinetic measurements of the catalyzed dye degradation with H2O2

Typical spectral changes observed during the reaction of Orange II as substrate with hydrogen

peroxide in the presence of a catalyst are exemplarily presented by the inset of Figure 3.1. The

decrease of the absorbance with time reflects oxidative degradation of the substrate and can

hence be considered as a measure for the catalytic activity of the complexes. It is noteworthy that

not only decolourization at 484 nm, which corresponds to the dominating hydrazone tautomer n -

π* transition at this pH,[15] was observed but also destruction of the stable naphthalene subunit of

the substrate, shown by the absorbance decrease at 300 nm (see inset of Figure 3.1) with identical

observed rate constants. The rate enhancement in the presence of the catalyst becomes evident

when it is compared with the non-catalyzed oxidation of Orange II by H2O2 under the same

reaction conditions. When the catalysts were used in equimolar concentrations in terms of the

manganese content, surprisingly no difference in the catalytic activity of the µ-oxo-bridged

dinuclear and the mononuclear MnII complex was observed (Figure 3.1). This also holds true under

conditions of catalytic conversion, i.e. tenfold substrate excess compared to the catalyst

concentration (see Figure 3.5). According to our experiments, the observed rate constants for the

oxidation of the naphthalene subunit and the destruction of the azo linkage are very similar. To

verify this correlation the reaction was also performed with Morin and p-nitrophenol as substrate.

For both dyes the reaction course was monitored at a wavelength of 400 nm, which corresponds

to the absorbance of the resorcinol subunit in Morin.[16] In agreement with the results obtained for

Orange II, the oxidation course with Morin and p-nitrophenol as substrate show identical reactivity

for the in situ formed catalyst as well as for the readily prepared catalysts [MnII(bpy)2Cl2] and

[Mn2III/IV(µ-O)2(bpy)4](ClO4)3∙2H2O. In the case of flavonoidic compounds such as Morin, it is known

that upon oxidation an initial increase followed by the oxidative decay at 330 nm can be observed,


60

whereas at 400 nm the decay of the phenolic part starts immediately. This is due to the extension

of the π-system of the dye, which results in an initial intensification of the corresponding

absorbance, as it is often observed for flavonoidic dyes. (for typical spectral changes and kinetic

traces see S3.1 and S3.2, Supporting Information). For all three substrates, the [Mn2III/IV(µ-O)2]3+-

dimer and the mononuclear analogous MnII complex gave the same results.

Figure 3.1 Observed spectral changes and kinetic traces recorded at 484 nm for the (·····) uncatalyzed degradation of

Orange II in comparison to the catalyzed reaction. Reaction conditions: 0.1 M HCO3-, 5 · 10

-5 M Orange II, 0.015 M

H2O2, pH 9.0, room temp.. (—) 4 · 10-5

M [MnII(bpy)2Cl2], (—) 2 · 10

-5 M [Mn2

III/IV(µ-O)2(bpy)4](ClO4)3, (—) 4 · 10

-5 M

[MnII(bpy)2Cl2] in the presence of

tBuOH, (—) 2 · 10

-5 M [Mn2

III/IV(µ-O)2(bpy)4](ClO4)3 in the presence of

tBuOH.

In order to exclude the involvement of free radical processes the experiment was repeated in

the presence of a strong radical scavenger like tBuOH. It reacts with hydroxyl radicals by

generation of inert intermediates, which cause termination of the radical chain reaction.

Although hydroxyl radicals might be present during the H2O2 induced reaction, they do not

participate in the oxidation process catalyzed by [MnII(bpy)2Cl2] and [Mn2III/IV(µ-O)2(bpy)4](ClO4)3,

since in the presence of tBuOH no negative effect on the reaction course was observed (Figure

3.1).

The influence of pH on the observed rate constant for the decolourization of Orange II (Orange

II: pKa = 11.4[17]) catalyzed by [MnII(bpy)2Cl2] and [Mn2III/IV(µ-O)2(bpy)4](ClO4)3 (Figure 3.2), shows

that both catalysts reach a maximum reactivity at a pH of about 8.7, indicating that both catalysts

must involve the same reactive intermediate in the catalytic cycle independent of the pH. At

higher pH the observed rate constants show a drastic decrease. This observation is consistent with


61

earlier studies [10, 18] and may be due to deprotonation of HOOCO2- to form the less electrophilic

oxidant CO42-.[18] The additionally performed measurements of the oxidation reaction of p-

nitrophenol at different pH values for either of the catalysts (Figure S3.3, Supporting Information)

confirm the assumption that the observed pH profile is caused by the difference in reactivity of the

oxidizing species and not by substrate effects. Due to the enhanced decomposition of hydrogen

peroxide at lower pH, originating from the parallel catalase-like reaction and leading to incomplete

decolourization of the solution, further investigations were performed at a slightly higher pH of

9.0.

Figure 3.2 Observed rate constants measured at 484 nm as a function of pH. Reaction conditions: 0.1 M HCO3-, 0.01 M

H2O2, 5 · 10-5

M Orange II, room temp., (■) 4 · 10-5

M [MnII(bpy)2Cl2], (▲) 2 · 10

-5 M [Mn2

III/IV(µ-O)2(bpy)4](ClO4)3.

Earlier investigations on oxidation reactions with H2O2 in bicarbonate buffered solution

revealed the importance of the in situ formed peroxycarbonate anion,[10, 18, 19, 20, 21] HOOCO2-,

which is known to be a stronger oxidant than hydrogen peroxide itself.[13] Despite the low

formation constant of HOOCO2- (viz. K = 0.32 ± 0.02 M-1 [21]), the latter is formed in a relatively fast

pre-equilibrium step.[21] The effect of the carbonate concentration on the oxidation course was

studied in a total carbonate concentration range between 0.01 and 0.4 M at a constant pH of 9.0.

This unambiguously has a strong effect on the observed rate constants for the catalyzed oxidation

of Orange II by either of the complexes (see Figure 3.2). The second order dependence of kobs on

the bicarbonate concentration leads to almost identical rate constants for the µ-oxo bridged and

the mononuclear MnII complexes determined by linear fit of the correlation presented in Figure


62

3.3, viz. for MnII(bpy)2Cl2 (15.1 ± 0.4) · 10-2 M-2 s-1 and for [Mn2III/IV(µ-O)2(bpy)4]3+ (16.9 ± 0.5) · 10-2

M-2 s-1. Control experiments at different ionic strengths with additional NaNO3 or NaCl at constant

bicarbonate concentrations showed no influence of ionic strength or the counter ion on the

observed reaction courses (see Figure S3.4, Supporting Information).

Figure 3.3 Observed second order dependence on the total carbonate concentration for the oxidation of Orange II by

both complexes measured at 484 nm. Reaction conditions: 5 · 10-5

M Orange II, 0.01 M H2O2, pH 9.0, room temp., (—

■—) 4 · 10-5

M MnII(bpy)2Cl2, (—▲—) 2 · 10

-5 M [Mn2

III/IV(µ-O)2(bpy)4](ClO4)3 · 2H2O.

This observation is consistent with earlier findings[10] and can be interpreted in terms of two

equivalents of bicarbonate that are involved in catalytically relevant reaction steps of the

oxidation mechanism of both catalysts. One is the mentioned relatively fast pre-equilibrium

between hydrogen peroxide and bicarbonate,[18, 20, 21] that results in the formation of the

peroxycarbonate anion. Whereas control experiments at pH 9 in the absence of any carbonate

revealed no catalytic effect, which confirms the crucial role of in situ formed peroxycarbonate for

the observed catalyzed oxidation reaction. Therefore the oxidation reaction was additionally

performed under identical reaction conditions without any carbonate present in CHES buffered

solution at pH 9.0. In Figure 3.4 the catalyzed oxidation reaction of Orange II for either of the

catalysts without carbonate present (c) is compared to the same experiments in 0.1 M (b) and 0.3

M (a) carbonate buffered solution under identical reaction conditions. Although the reaction of

the manganese catalysts with HOO- (note pKa of H2O2 > 9.0) could be in general possible, the

tremendous rate enhancing effect of carbonate on the catalyzed oxidation reaction of Orange II

becomes evident for either of the catalysts. This points to the crucial pre-equilibrium of


63

peroxycarbonate formation resulting in the second order dependence of the observed reaction

rate on the total carbonate concentration.

Figure 3.4 Kinetic traces recorded for the oxidation of 5 · 10-5

M Orange II by 4 · 10-5

M [MnII(bpy)2Cl2] and 2 · 10

-5 M

[Mn2III/IV

(µ-O)2(bpy)4](ClO4)3 in the presence of 0.01 M H2O2 and (a) 0.1 M HCO3- at pH 9.0 (b) 0.3 M HCO3

- at pH 9.0,

compared to (c) the same reaction in the absence of any carbonate in 0.1 M CHES buffer. Reactions conditions: pH 9.0,

room temp..

The participation of the second equivalent of bicarbonate could be due to coordination of an

electron donating bicarbonate ion to an aquated MnII-bipyridine precursor complex to form a

more reactive MnII-bipyridine-bicarbonate species or to stabilize the oxidatively formed high-

valent MnIV=O intermediate. It was shown before that monocarboxylate ions, i.e. acetate,

bicarbonate and formate, enhance the catalytic activity in the H2O2 induced epoxidation reactions,

but do not considerably disturb the first coordination sphere of well defined MnII complexes.[22]

Whereas dicarboxylate additives such as oxalate have been reported to act as strong ligands to the

MnII catalyst, this leads to lower reactivity in the epoxidation reaction.[22] Consequently, a

displacement of the chelating 2,2’-bipyridine ligands by bicarbonate/carbonate is unlikely, but a

monodentate binding mode is in principle possible if easily exchangeable coordination sites exist,

as in the case of the earlier reported catalytic activity of simple metal salts.[10]

Performing kinetic measurements under variation of the initial hydrogen peroxide

concentration (between 2.5 · 10-3 and 5.0 · 10-2 M) enabled us to evaluate the effect of H2O2 on

the reaction course of the catalytic oxidation of the different substrates. At high catalyst

concentration and a small hydrogen peroxide excess, the strong hydrogen peroxide consumption


64

(catalase-like reaction) resulted in an inefficient decolourisation due to the loss of pseudo-first-

order reaction conditions. In order to suppress this effect, we investigated the [H2O2] influence on

the dye degradation at catalyst concentrations of 5 · 10-6 M for [MnII(bpy)2Cl2] and 2.5 · 10-6 M for

[Mn2III/IV(µ-O)2(bpy)4](ClO4)3. Both catalysts again show identical rates within the experimental

error, when used at equimolar manganese content, confirming the suggestion that in both cases

the same catalytically active intermediate is generated. A good linear correlation between the

corresponding kobs value and the oxidant concentration was observed (see Figure 3.5).

Figure 3.5 Plot of the observed rate constants determined at 484 nm as a function of [H2O2] for both studied

complexes. Reaction conditions: 0.1 M HCO3-, 5 · 10

-5 M Orange II, pH 9.0, room temp., (solid lines) 5 · 10

-6 M

[MnII(bpy)2Cl2], (dotted lines) 2.5 · 10

-6 M [Mn2

III/IV(µ-O)2(bpy)4](ClO4)3 · 2H2O.

Substrate (λmax / nm) k / M-1s-1 (MnII(bpy)2Cl2) k / M-1s-1 ([Mn2III/IV(µ-O)2(bpy)4](ClO4)3)

OII (484) (7.1 ± 0.2) · 10-2 (7.2 ± 0.2) · 10-2

Mo (400) (4.5 ± 0.1) · 10-1 (4.5± 0.1) · 10-1

PNP (400) (1.10 ± 0.06) · 10-2 (1.12 ± 0.05) · 10-2

Table 3.1 Second order rate constants for both studied complexes in the hydrogen peroxide assisted oxidative

degradation of different dyes. Reaction conditions: 0.1 M HCO3

-, 5 · 10

-5 M dye, 5 · 10

-6 M [Mn

II(bpy)2Cl2], 2.5 · 10

-6 M

[Mn2III/IV

(µ-O)2(bpy)4](ClO4)3, pH 9.0, room temperature.


65

From the slope of the plot with a zero intercept the second order rate constants for the µ-oxo-

bridged dinuclear and the mononuclear MnII complex for all dyes were determined to be identical

within the experimental error (Table 3.1). The values of k for the oxidation of the phenolic group in

Morin are more than one order of magnitude higher than for the oxidation of p-nitrophenol. This

trend can be accounted for in terms of the electronic properties of the benzene core that is

oxidized. Whereas Morin bears two activating, i.e. electron donating, OH-substituents, p-

nitrophenol contains only one OH- and in addition one strongly deactivating NO2-substituent. As a

consequence, p-nitrophenol is less nucleophilic and a less favourable substrate for an oxidizing

species such as an electrophilic, high-valent manganese complex, which is reflected by the large

difference in the second-order rate constants for both dyes (Table 3.1).

To evaluate the effect of the catalyst concentration on the oxidative degradation of Orange II

by H2O2, kinetic studies over a wide concentration range were performed under catalytically

relevant conditions. Since the concentration of the co-catalyst carbonate has such a severe effect

on the oxidative reaction course, the experiment was also conducted at a higher [HCO3-] and

showed similar tendencies (see Figures S3.5, Supporting Information). The kobs values for

[MnII(bpy)2Cl2] and [Mn2III/IV(µ-O)2(bpy)4](ClO4)3 plotted as a function of the manganese

concentration (note that the dimer has double the Mn content) showed identical saturation

kinetics and reached a limiting value at high catalyst concentrations.

The observed trend of the resulting kobs dependencies of [MnII(bpy)2Cl2] and [Mn2III/IV(µ-

O)2(bpy)4](ClO4)3 at different carbonate content can be attributed to a pre-equilibrium step, i.e. a

rapid equilibration reaction prior to the rate-determining reaction step, involving the

peroxycarbonate anion and one equivalent of manganese, which finally results in a mononuclear

high-valent Mn-oxo species.

3.3.2 The reaction of the catalysts with H2O2 in carbonate buffered solution

3.3.2.1 EPR-spectroscopic measurements

In order to obtain more information on the high-valent manganese species formed in the

presence of H2O2 in bicarbonate containing solution, EPR spectroscopic measurements were

performed. Samples were transferred to a 3 mm EPR quartz tube immediately after mixing the


66

reactants and frozen in liquid nitrogen to quench the reaction. The X-band EPR spectra at 6 K of

bicarbonate (0.1 M) and hydrogen peroxide (0.04 M) containing solutions of 1 · 10-4 M

[MnII(bpy)2Cl2] and 5 · 10-5 M [Mn2III/IV(µ-O)2(bpy)4](ClO4)3, exhibit a similar set of features over a

wide range from 50 to 500 mT (see Figure 3.6).

Figure 3.6 X-band EPR spectrum of 0.1 M bicarbonate containing solutions at pH 9.0 of (—) 1 · 10-4

M [MnII(bpy)2Cl2]

and (—) 5 · 10-5

[Mn2III/IV

(µ-O)2(bpy)4](ClO4)3 with 0.04 M H2O2 immediately after mixing. EPR conditions: 8.98 GHz, 6 K,

1 mW microwave power, modulation amplitude 20 mT.

Both spectra show well resolved and slightly anisotropic sextets of the characteristic 55Mn

hyperfine lines (I = 5/2) at g ≈ 2, which is consistent with a high-spin S = 5/2 MnII. In addition,

broad and unresolved signals of lower amplitude at g ≈ 4-5 could be observed for both catalysts,

which is indicative of a mononuclear MnIV complex. Several examples with similar g-values for

mononuclear MnIV species have been reported before[19, 23] and some also show a lower amplitude

feature at g ≈ 2.[24] If more oxidant is used an increase in the characteristic MnIV signal could be

observed (see Figure S3.6, Supporting Information), which is in agreement with our experimental

results for the catalytic oxidation reaction.

In the present case strong EPR spectral features exhibited by the MnIV=O intermediate are

observed in the g ≈ 2 region with only minor components at lower field. In general EPR spectra of

d3 MnIV ions in an axial field (E/D = 0) are often difficult to interpret on account of the dependence

on the magnitude of the zero-field splitting parameters.[25] If the axial zero-field splitting

parameter D is high, the spectrum is dominated by the lower field signal and shows only minor g ≈

2 contribution, whereas small D values result in spectra with inverted signal intensity for these two


67

signals. For instance, this is the case for sulfur-containing thiohydroxamate[25b] and

dithiocarbamate[25c] MnIV complexes. In addition we performed EPR experiments in the presence

of tBuOH, i.e. a radical scavenger, to check whether the presence of MnIII or the participation of

radical processes can account for the observed intermediates and therefore the oxidation

reaction. In any case no difference in the observed results was found regardless whether a radical

scavenger was present in the reaction mixture or not (for EPR spectra see Figure S3.6, Supporting

Information).

Throughout our studies no multi-line signals as known for mixed-valent oxo-bridged

manganese complexes could be observed, indicating that these species are not prevalent in the

catalytic solution. This is in agreement with recent results for acetylacetone-based Schiff bases of

manganese which show that in catalytic alkene epoxidation with H2O2 and carbonate present in

acetone/MeOH mixtures, the catalytic centre is a mononuclear Mn complex.[22] The presence of

residual MnIII/IV dimers cannot be absolutely excluded since the excessively present MnII causes a

higher peak intensity at equal concentration levels than the dimer. Nevertheless, the obtained EPR

results in conjunction with the observed reactivity pattern provide a strong indication that a

monomeric MnIV complex is present as potential oxidizing species in the reaction solution and

most notably for both catalysts. The presence of monomeric MnII indicates that in both cases the

key step of the oxidation process is a two-electron oxidation of a monomeric MnII precursor to a

MnIV=O intermediate.

3.3.2.2 UV/Vis spectroscopic measurements

The reaction of [MnII(bpy)2Cl2] or [Mn2III/IV(µ-O)2(bpy)4](ClO4)3 with hydrogen peroxide was also

studied with stopped-flow rapid scan UV/Vis spectroscopy. Since the concentration of possible

high-valent manganese species are rather low in solution, higher concentrations of the manganese

catalysts had to be used to obtain more significant spectral changes. Therefore, a small amount of

freshly prepared catalyst stock solution was added to a 0.1 M bicarbonate containing buffer

solution at pH 9.0 and room temperature, and spectra were recorded at time intervals of 0.2 s

over the first 5 seconds of the reaction after mixing. The spectral changes that accompany the

reaction of [MnII(bpy)2Cl2] with H2O2 are presented in Figure 3.7 and show the formation and

partial decay of an absorption band at 445 nm. It is supposed that this transition arises from a

LMCT process. Similar spectral assignments that the lower energy bands in monomeric MnIV


68

complexes arise from LMCT transitions have been reported before for a complex with a tacn

derived ligand[26a] or for [MnIV(bpy)(N3)4].[26b]

Figure 3.7 Spectra recorded for the reaction of 0.2 · 10-3

M [MnII(bpy)2Cl2] with 5 · 10

-3 M H2O2 in 0.1 M HCO3

- solution

at pH 9.0, room temperature. Inset: Corresponding kinetic trace at 445 nm.

In this case it is probably an oxo to MnIV LMCT band, since the band at 450 nm was not

observed in the absence of H2O2. Based on spectroscopic observations, the rapid formation and

decay of a band at 445 nm could be attributed to the intermediate formation of a mononuclear

MnIV-oxo intermediate. The subsequent partial decay of the observed species is attributed to the

parallel oxidation of hydrogen peroxide which is accompanied by gas evolution as no other

substrate was present (Figure 3.7 inset).

The reaction of [Mn2(µ-O)2(bpy)4](ClO4)3 with H2O2/HCO3- was studied in a similar way by use

of rapid scan UV/Vis spectroscopy, although it was not possible to use stopped-flow techniques

with aqueous stock solutions due to the insufficient stability of the µ-oxo-bridged core in aqueous

solution over longer time scale.[27] The spectral changes within the first few seconds after mixing

of the dimer with a H2O2/bicarbonate containing solution were much more intense, but it cannot

be excluded that a similar band at 445 nm underlies the spectral changes of the remaining di-µ-

oxo bridged dimer. (Figure S3.7, Supporting Information). The absence of the characteristic 16-line

signal in the performed EPR experiments indicates that the dimer is no longer the predominant

species under catalytically relevant conditions. Presumably it is rapidly converted to a

mononuclear manganese species due to the presence of bicarbonate/carbonate, which causes the


69

dinuclear species to dissociate,[28] and due to the excess hydrogen peroxide being able to reduce

the bis-µ-oxo-bridged dinuclear complex.[9]

With respect to the UV/Vis measurements and the data obtained by means of EPR

spectroscopy, we conclude that a key step of the oxidation reaction for both catalysts is the two

electron oxidation of a monomeric MnII-precursor complex to form an high-valent MnIV=O species,

which could act as potential oxidizing species.

3.3.3 In situ formation of the active species in the catalytic oxidation reaction

In addition, the in-situ formation of the catalytically active species was checked by performing

the oxidation reaction of the different substrates when MnIICl2 and ligand (2,2'-bipyridine) were

added simultaneously to the solution in a molar ratio of 2:1 (bipyridine:MnII). The 2:1

stoichiometry would in principle allow the rapid formation of the µ-oxo-bridged dimers on

stabilizing manganese in its higher oxidation states. Compared to the kinetic traces obtained with

the synthesized catalysts [MnII(bpy)2Cl2] and [Mn2III/IV(µ-O)2(bpy)4](ClO4)3, no difference in the

catalytic reaction course was found for any of the three substrates when such a mixture instead of

the synthesized catalysts was used. This observation, exemplarily shown for Orange II in Figure 3.8

(see Figures S3.1 and S3.2, Supporting Information for the corresponding experiments with Morin

and p-nitrophenol), implies the formation of the same oxidising intermediate regardless whether

the catalyst is formed in situ or used in an isolated form.

The catalytic activity of [MnII(bpy)2Cl2] and [Mn2III/IVO2(bpy)4](ClO4)3 is higher than the activity

of the simple MnIICl2 salt under identical reaction conditions, since complexation of the MnII ion by

a chelating ligand is in general advantageous for the catalytic activity. From the fact that the same

catalytically active form is accessible by in situ preparation of the complex, and the known

catalytic effect of simple MnII salt on the oxidative degradation of various model substrates,[10] the

question arises what the actual MnII precursor form of the active intermediate is that accounts for

the observed oxidation process.


70

Figure 3.8 Kinetic traces recorded at 484 nm for the catalyzed degradation of Orange II. Reaction conditions: 0.1 M

HCO3-, 5 · 10

-5 M Orange II, 0.015 M H2O2, pH 9.0, room temp., (—) 4 · 10

-5 M [Mn

II(bpy)2Cl2], (—) 2 · 10

-5 M [Mn2

III/IV(µ-

O)2(bpy)4](ClO4)3, (—) 4 · 10-5

M MnIICl2·4H2O and 8 · 10

-5 M 2,2'-bipyridine added to the reaction mixture.

This becomes more evident when the oxidation reaction of Orange II is performed with a MnII

salt in the presence of different concentrations of the bipyridine ligand. Increasing concentrations

of the ligand lead to faster reaction rates as shown in Figure 3.9, which can be ascribed to the

stabilization of the MnII-precursor complex in slightly basic solution, and thereby the stabilization

of a high-valent Mn-oxo intermediate. At higher bipyridine concentration the rapid consumption

of hydrogen peroxide disrupts the oxidative degradation process. This is supported by the

observation that the catalytic reaction can be started again by addition of a fresh amount of

hydrogen peroxide (Figure 3.9), i.e. the disruption is not due to catalyst deactivation, since the

initial oxidation rate remains unchanged at ligand to metal ratios higher than 2:1. The higher

coordination number reached at higher bipyridine concentrations favours decomposition of H2O2

above substrate oxidation. Similar observations were made for the other model substrates Morin

and p-nitrophenol (see Figure S3.8, Supporting Information).

In the case of aquated MnII ions in the presence of bipyridine different complexes, viz. 1:1, 1:2

and 1:3 (M:L), are expected to be formed in solution. Working with a 1:2 stoichiometry seemed

very appropriate for our purpose, since the existence of free coordination sites on the precursor

complex is considered to be of importance for the catalytic activity and allows as a matter of

principle the coordination of hydrogen peroxide and the subsequent formation of µ-oxo-bridged

dimers.


71

Figure 3.9 Observed kinetic traces recorded at 484 nm for the degradation of Orange II. Reaction conditions: 0.1 M

HCO3-, 0.5-4 % CH3CN, 5 · 10

-5 M Orange II, 0.015 M H2O2, pH 9.0, room temp., 4 · 10

-5 M Mn

IICl2 and different

concentrations of 2,2'-bipyridine from 0 to 1.6 · 10-4

M added to the reaction mixture.

It was also shown before that in the absence of a chelating ligand in bicarbonate containing

solution, catalytically active Mn-bicarbonate complexes of different stoichiometry can be

formed.[10] As a consequence, several differently substituted MnII precursor complexes with

bipyridine and bicarbonate ligands may in principle be able to catalyse the oxygenation reaction.

However, the formal metal-to-ligand ratio in solution and the stoichiometry in the solid state may

differ from the complex species that are finally responsible for the observed catalytic behaviour.

This has to be considered particularly in the case of MnII complexes, since they are known to be

kinetically labile.[29]

3.3.4 Precursor complex equilibria in solution

Consequently, we investigated the formation of potential MnII precursor complexes with

bipyridine in carbonate buffered solution by the method of continuous variation of Job.[30]

Therefore the change in absorbance between before and after mixing of different molar fractions

of MnII and ligand was studied in 0.1 M carbonate buffer solution at pH 9.0 in the presence of a

small amount of acetonitrile (5 %) to improve the solubility of the bipyridine ligand. Although the

absorbance changes are rather small (Figure 3.9, inset), a non-linearity in the Job-plot correlation

(Figure 3.10) at a molar fraction of 0.5 can be observed.


72

Figure 3.10 Observed spectral changes at 280 nm before and after mixing of different molar fractions of MnIICl2 and

bipyridine, (■) overall concentration 1 · 10-4

M, (●) overall concentration 5 · 10-5

M. Reaction conditions: 5% CH3CN,

0.1 M HCO3-, pH 9.0, room temperature. Inset: Observed spectral changes before and after mixing of Mn

IICl2 and

bipyridine at a molar ratio of 0.65.

This indicates a hidden local extremum as expected for a 1:1 complex. However the broad

maximum at a molar fraction of about 0.7 clearly points to the formation of a 2:1 complex in

solution. Based on this finding the corresponding kinetic data for the MnII catalyzed oxidation

reaction in the presence of different bipyridine concentrations can be interpreted in the following

way: An increase in the ligand concentration shifts the equilibrium from the 1:1 to the 1:2

bipyridine substituted precursor complex which shows a higher catalytic reactivity. Additional

ligand beyond a stoichiometry of 1:2 especially favors the parallel hydrogen peroxide

decomposition and causes disruption of the catalytic cycle. This indicates that the 1:2 form is the

catalytically active precursor species responsible for the oxidation process and emphasizes the

importance of labile coordination sites for a high catalytic activity. In contrast, the 1:3 form favors

parallel decomposition (catalase) of H2O2, which presumably occurs via an outer-sphere electron

transfer process. As expected, UV/Vis experiments verified that the chloride ions are substituted

by water as soon as the [MnII(bpy)2Cl2] complex is dissolved, so that the catalyst precursor can be

described as aqua or hydroxo form of MnII bearing two bipyridine ligands.


73

3.3.5 Mechanistic aspects

On the basis of the collected data, the following mechanistic conclusions can be drawn to

provide a basis for the understanding of the catalytic activation of H2O2 by manganese complexes

in carbonate buffered aqueous solution. Of significant importance is the observation that in this

catalytic system, i.e. aqueous bicarbonate buffered solution in a pH range of 8-10, it does not

make a difference if a [Mn2III/IV(µ-O)2]3+-dimer or its mononuclear analogous MnII complex is used.

Both catalysts show identical catalytic oxidative reactivity for different kinds of dyes, indicating

that the same oxidizing intermediate is formed under all conditions. Rapid-scan UV/Vis and EPR

spectroscopy provided evidence for this active intermediate being a mononuclear high-valent

MnIV-oxo species. Kinetic and EPR data could be verified in the presence of a radical scavenger,

which rules out an involvement of radical species in the catalytic oxidation process. This in turn

indicates that MnIII species are not relevant for the observed reactivity, since their presence would

be expected in the case of a radical mechanism. The parallel presence of MnII in the EPR spectra

led to the conclusion that the key feature of the reaction of either of the catalysts is a two electron

oxidation process of a MnII precursor complex to form a high-valent MnIV-oxo intermediate.

Accordingly, the main reaction sequence of the studied reaction can be simplified as presented in

Scheme 3.2.

Above all, it is essential to consider the role of peroxycarbonate as a more powerful oxidant

than H2O2 itself.[13] As shown before, this oxidant is formed in situ in an pre-equilibrium process

between hydrogen peroxide and bicarbonate.[18-21] Furthermore, in the case of the Mn catalyzed

substrate oxidation by H2O2, no oxidative reactivity was observed in the absence of a carbonate

buffer.[10] It is proposed that the peroxycarbonate anion results in the formation of a manganese-

η2-peroxycarbonate intermediate as the consequence of a fast nucleophilic attack of the

peroxycarbonate anion on the MnII center. Similar Rh[31] and Fe[32] peroxycarbonate complexes

have been isolated before.

The observed saturation kinetics for the catalytic reaction course implies that this reaction

starts with a pre-equilibrium step. Subsequently, the manganese-η2-peroxycarbonate is believed

to undergo a proton induced heterolytic O-O-bond cleavage to produce a high-valent MnIV=O

intermediate. Since such a MnIV=O intermediate has a high oxidation ability, it will react with

substrate in an oxygen transfer process to lead to the oxidation products and completion of the

catalytic cycle by regeneration of the MnII precursor complex.


74

Scheme 3.2 Proposed reaction cycle for the catalyzed oxidative degradation of a substrate by H2O2 (S = substrate, L =

water or bicarbonate) by a possible MnII precursor complex formed from [Mn

II(bpy)2Cl2] or [Mn2

III/IV(µ-

O)2(bpy)4](ClO4)3.

Of significant importance is the drastic enhancement of the oxidation activity by the presence

of bicarbonate/carbonate as co-catalyst. Its contribution is firstly related to the in situ formation of

the highly reactive peroxycarbonate and secondly to its capability to serve as electron donating

ligand to form more reactive precursor complexes like it was shown for the manganese ion

catalyzed activation of H2O2 in the absence of any other ligand.[9] Literature findings[22] and the Job

plot analysis support the assumption that the bipyridine ligands in the catalytically active

precursor form are not displaced by bicarbonate. The reactions performed with different

MnII:bipyridine ratios are in favour of an MnII(bpy)2 complex form as precursor for the catalytically

active intermediate in the oxidation process, whereas the two remaining coordination sites are

most likely easily exchangeable bicarbonate and/or aqua and hydroxo ligands.


75

3.4 CONCLUSION

In summary, we were able to demonstrate by a well studied example that an elaborated oxo-

bridged catalyst in higher oxidation states is not always required for efficient catalysis. Instead, the

same results can be achieved by the use of [MnII(bpy)2Cl2] as catalyst to oxidize various organic

substrates in aqueous solution under mild conditions. Moreover, this catalytic activity is also

accessible by simple in situ preparation of the 1:2 complex from a MnII salt and the ligand within

the reaction mixture. In addition, the present study provides insight into mechanistic aspects and

the nature of the oxidizing species occurring during the reaction of [MnII(bpy)2Cl2] and [Mn2III/IV(µ-

O)2(bpy)4](ClO4)3 with hydrogen peroxide in aqueous carbonate containing solution. As a result we

were able to show that the key feature of both catalysts is a two electron oxidation from a MnII

precursor to a high-valent MnIV=O intermediate, which represents the potential oxidizing species.



naphthylazo)benzenesulfonate], 99 % was supplied by Sigma-Aldrich and recrystallized from a

Et2O/H2O mixture at 4 °C. Unless otherwise stated, all other dyes were commercially available

(Acros Organics, Germany) and were used without any further purification. Hydrogen peroxide 35

wt. % was of analytical grade and provided by Acros Organics (Germany). Analytical grade

Hydrogen peroxide 30 wt. % was supplied by Sigma-Aldrich. Different manganese salt hydrates

(chloride, sulfate, nitrate, acetate, perchlorate) were of analytical grade and purchased from Acros

Organics (Germany). Sodium bicarbonate and CHES, were of analytical grade and provided by

Acros Organics (Germany). All buffer solutions were prepared using Millipore Milli-Q purified

water. Analytical grade 2,2'-bipyridine, CH3CN, tBuOH and sodium hydroxide were purchased from

Acros Organics (Germany).

MANGANESE CATALYSTS. [MnII(bpy)2Cl2] and [Mn2III/IV(µ-O)2(bpy)4](ClO4)3∙2H2O were synthesized as

reported before[27] and confirmed by elemental analysis. Anal. calculated for [MnII(bpy)2Cl2]: C:

54.82, H: 3.68, N: 12.79; Found: C: 54.91, H: 3.70, N, 12.83. Anal. calculated for [Mn2III/IV(µ-

O)2(bpy)4](ClO4)3∙2H2O: C: 43.64, H: 3.30, N: 10.18; Found: C: 43.62, H: 3.29, N, 10.14. The

oxidation state of [Mn2III/IV(µ-O)2(bpy)4](ClO4)3∙2H2O was verified by EPR spectroscopy.


76

GENERAL PROCEDURE FOR THE H2O2 CATALYZED DYE DEGRADATION REACTIONS. The manganese catalysts were

freshly dissolved in water (in the case of Mn2III/IV(µ-O)2(bpy)4](ClO4)3∙2H2O the catalyst was freshly

dissolved in acetonitrile) before use. To a freshly prepared sodium bicarbonate solution, an

adequate amount of NaOH was added to adjust the pH of the solution. Under isothermal

conditions, the desired amount of a concentrated manganese solution was added together with

Orange II, previously dissolved in an aqueous bicarbonate solution, and H2O2.

All kinetic data were obtained by recording time-resolved UV-Vis spectra using a Hellma

661.502 - QX quartz Suprasil immersion probe attached via optical cables to a 150 W Xe lamp and

a multi-wavelength J & M detector, which records complete absorption spectra at constant time

intervals. In a thermostated open glass reactor vessel equipped with a magnetic stirrer, the freshly

prepared catalyst solution and H2O2 were added to 40 ml of 5 · 10-5 M dye at a pH ranging from 8

to 10 at 25 °C. The pH of the aqueous solutions was carefully measured using a Mettler Delta 350

pH meter previously calibrated with standard buffer solutions at two different pH values (4 and

10). The kinetics of the oxidation reaction was monitored at the λmax of the corresponding dye.

First order rate constants, where possible, were calculated using Specfit/32 and Origin (version

7.5) software. To estimate the effect of the catalyst and H2O2 concentrations on the catalytic

reaction at different carbonate concentrations, stopped-flow kinetic measurements were carried

out additionally using an SX.18MV stopped-flow instrument from Applied Photophysics.

EPR MEASUREMENTS. EPR spectroscopy was performed using a JEOL continuous wave spectrometer

JES-FA200 equipped with a X-band Gunn diode oscillator, a cylindrical mode cavity and a helium

cryostat. Samples were transferred to a 3 mm EPR quartz tube immediately after mixing the

reactants and frozen in liquid nitrogen to quench the reaction before data collection. Typical

spectrometer conditions were 8.98 GHz microwave frequency, 1 and 2 mW microwave power,

respectively, and 20 mT modulation amplitude.

SPECTROPHOTOMETRIC TITRATION. UV/Vis spectra were recorded on a Shimdazu UV-2101

spectrophotometer at 25 °C. A tandem cuvette with two separate compartments was filled with

stock solutions of MnIICl2 in water/acetonitrile (20:1) solution in one, and bipyridine in 0.2 M

carbonate buffer/acetonitrile (20:1) in the other compartment. The absorbance at 280 nm was

studied for different molar fractions of manganese and bipyridine before and 120 s after mixing of

the reactants.


77


[1] (a) Jacobsen, E. N.; Ojima, I. (Ed.), Catalytic Asymmetric Synthesis, VCH Publishers, New York,

1993, 159; (b) Katsuki, T. Coord. Chem. Rev. 1995, 140, 189-214; (c) Limburg, J.; Vrettos, J. S.;

Liable-Sands, L. M.; Rheingold, A. L.; Crabtree, R. H.; Brudvig, G. W. Science 1999, 283, 1524-

1527; (d) Yachandra, V. K.; Sauer, K.; Klein, M. P. Chem. Rev. 1996, 96, 2927-2950; (e)

Ruettinger, W.; Dismukes, G. C. Chem. Rev. 1997, 97, 1-24.

[2] (a) Hage, R.; Iburg, J. E.; Kerschner, J.; Koek, J. H.; Lempers, E. L. M.; Martens, R. J.; Racherla,

U. S.; Russel, S. W.; Swarthoff, T.; van Vliet, M. R. P.; Warnaar, J. B.; van der Wolf, L..; Krijnen,

B. Nature 1994, 369, 637-639; (b) de Boer, J. W.; Browne, W. R.; Brinksma, J.; Alsters, P. L.;

Hage, R.; Feringa, B. L. Inorg. Chem. 2007, 46, 6353-6372.

[3] (a) Groves, J. T.; Stern, M. K.; J. Am. Chem. Soc. 1987, 109, 3812-3814; (b) Groves, J. T.; Stern,

M. K.; J. Am. Chem. Soc. 1988, 110, 8628-8638.

[4] (a) Limburg, J.; Vrettos, J. S.; Chen, H.; de Paula, J.C.; Crabtree, R. H.; Brudvig, G. W. J. Am.

Chem. Soc. 2001, 123, 423-430; (b) Barton, D. H. R.; Choi, S.-Y.; Hu, B.; Smith, J. A.

Tetrahedron 1998, 54, 3367-3378; (c) Tagore, R.; Chen, H.; Crabtree, R. H.; Brudvig, G. W. J.

Am. Chem. Soc. 2006, 128, 9457-9465.

[5] (a) Tetard, D.; Rabion, A.; Verlhac, J.-B.; Guilhem, J. J. Chem. Soc., Chem. Commun. 1995, 5,

531-532; (b) Romakh, V. B.; Therrien, B.; Suess-Fink, G.; Shul’pin, G. B. Inorg. Chem. 2007, 46,

1315-1331; (c) Vincent, J. M.; Menage, S.; Lambeaux, C.; Fontecave, M. Tetrahedron Lett.

1994, 34, 6287-6290; (d) Fish, R. H.; Fong, R. H.; Oberhausen, K. J.; Konings, M. S.; Vega, M. C.;

Christou, G.; Vincent, J. B.; Buchanan, R. M. New. J. Chem. 1992, 16, 727-733.

[6] Handervan, C.; Hage, R.; Feringa, B. L. Chem. Commun. 1997, 5, 419-420.

[7] (a) Towle, D. K.; Botsford, C. A.; Hodgson, D. J. Inorg. Chim. Acta 1988, 141, 167-168; (b) Oki,

A. R.; Glerup, J.; Hodgson, D. J. Inorg. Chem. 1990, 29, 2435-2441; (c) Schindler, S.; Walter, O.;

Pederson, J. Z.; Toflund, H. Inorg. Chim. Acta 2000, 303, 215-219.

[8] (a) Yin, G.; Buchalova, M.; Danby, A. M.; Perkins, C. M.; Kitko, D.; Carter, J. D.; Scheper, W. M.;

Busch, D. H. Inorg. Chem. 2006, 45, 3467-3474; (b) Adam, W.; Roschmann, K. J.; Saha-Möller,

C. R.; Seebach, D. J. Am. Chem. Soc. 2002, 124, 5068-5073. (c) Finney, N. S.; Pospisil, P. J.;


78

Chang, S.; Palucki, M.; Konsler, R. G.; Hansen, K. B.; Jacobsen, E. N. Angew Chem., Int. Ed.

1997, 36, 1720-1723.

[9] Murphy, A.; Stack, T. D. P. J. Mol. Catal. A 2006, 251, 78-88.

[10] (a) Ember, E.; Rothbart, S.; Puchta, R.; van Eldik, R. New J. Chem. 2009, 33, 34-49; (b) Ember,

E.; Gazzaz, H.; Rothbart, S.; Puchta, R.; van Eldik, R. Appl. Catal. B. 2010, 95, 179-191.

[11] Suzuki, M.; Tokura, S.; Suhura, M.; Uehara, A. Chem. Lett. 1988, 3, 477-480.

[12] (a) Blum, H.; Mayer, B.; Pegelow, U. Patent Appl. DE 19620267, 1996; (b) Blum, H.; Kramer, R.

Patent Appl. WO 0032731, 1999.

[13] Swern, D. Organic Peroxides, Wiley, New York, 1970, 313.

[14] Lindsay-Smith, J. R.; Gilbert, B. C.; Mairata i Payeras, A.; Murray, J.; Lowdon, T. R.; Oakes, J.;

Pons i Prats, R.; Walton, P. H. J. Mol. Catal. A 2006, 251, 114-122.

[15] Oakes, J.; Gratton, P. J. Chem. Soc., Perkin Trans. 2 1998, 9, 1857-1864.

[16] (a) Boudet, A. C.; Cornard, J. P.; Merlin, J. C. Spectrochim. Actat A 2000, 56, 829-839; (b)

Castro, G. T.; Blanco, S. E. Spectrochim. Acta A 2004, 60, 2235-2241.

[17] Sabnis, R. W. Handbook of Acid-base Indicators, CRC Press (Taylor and Francis Group),

London, 2008.

[18] (a) Balagam, B.; Richardson, D. E. Inorg. Chem. 2008, 47, 1173-1178; (b) Siegel, A.; Siegel, H.

Metal ions in biological systems, Marcel Dekker, New York, 1999, Vol. 37, 330.

[19] Flanagan, J.; Jones, D. P.; Griffith, W. P.; Skapski, A. C.; West, A. P. J. Chem. Soc., Chem.

Commun. 1986, 20-21.

[20] Lane, B. S.; Vogt, M.; DeRose, V. J.; Burgess, K. J. Am. Chem. Soc. 2002, 124, 11946-11954.

[21] Richardson, D. E.; Yao, H.; Frank, K. M.; Bennett, D. A. J. Am. Chem. Soc. 2000, 122, 1729-

1739.

[22] Stamatis, A.; Doutsia, P.; Vartzouma, C.; Christoforidis, K. C.; Deligiannakis, Y.; Louloudi, M. J.

Mol. Catal. A 2009, 297, 44-53.

[23] (a) Romain, S.; Baffert, C.; Duboc, C.; Leprêtre, J.-C.; Deronzier, A.; Collomb, M.-N. Inorg.

Chem. 2009, 48, 3125-3131; (b) Parsell, T. H.; Behan, R. K.; Green, M. T.; Hendrich, M. P.;


79

Borovik, A. S. J. Am. Chem. Soc. 2006, 128, 8728-8729; (c) Campbell, K. A.; Lashley, M. R.;

Wyatt, J. K.; Nantz, M. H.; Britt, R. D. J. Am. Chem. Soc. 2001, 123, 5710-5719.

[24] Saadeh, S. M.; Lah, M. S.; Pecoraro, V. L. Inorg. Chem. 1991, 30, 8-15.

[25] (a) Pedersen, E.; Toftlund, H. Inorg. Chem. 1974, 13, 1603-1612; (b) Pal, S.; Ghosh, P.;

Chakravorty, A. Inorg. Chem. 1985, 24, 3704-3706; (c) Brown, K. L.; Golding, R. M.; Healy, P.

C.; Jessop, K. J.; Tennant, W. C. Aust. J. Chem. 1974, 27, 2075-2081.

[26] (a) Belal, A. A.; Chaudhuri, P.; Fallis, I.; Farrugia, L. J.; Hartung, R.; Macdonald, N. M.; Nuber,

B.; Peacock, R. D.; Weiss, J.; Wieghardt, K. Inorg. Chem. 1991, 30, 4397-4402; (b) Dave, B. C.;

Czernuszewicz, R. S. J. Coord. Chem. 1994, 33, 257-269.

[27] (a) McCann, S.; McCann, M.; Rev. Casey, M.T.; Jackman, M.; Devereux, M.; McKee, V. Inorg.

Chim. Acta 1998, 279, 24-29; (b) Cooper, S. R.; Calvin, M. J. Am. Chem. Soc. 1977, 99, 6623-

6630.

[28] Limburg, J.; Vrettos, J. S.; Crabtree, R. H.; Brudvig, G. W.; de Paula, J. C.; Hassan, A.; Barra, A.-

L.; Duboc-Toia, C.; Collomb, M.-N. Inorg. Chem. 2001, 40, 1698-1703.

[29] Hogg, R.; Wilkins, R. C. J. Chem. Soc. 1962, 341-350.

[30] Job, P. Ann. Chim. 1928, 9, 113.

[31] Aresta, M.; Tommasi, I.; Quaranta, E.; Fragale, C.; Mascetti, J.; Tranquille, M.; Galan, F.;

Fouassier, M. Inorg. Chem. 1996, 35, 4254-4260.

[32] Hashimoto, K.; Nagatomo, S.; Fujnami, S.; Furutachi, H.; Ogo, S.; Suzuki, M.; Uehara, A.;

Maeda, Y.; Watanabe, Y.; Kitagawa, T. Angew. Chem., Int. Ed. 2002, 41, 1202-1205.


80

3.7 SUPPORTING INFORMATION

Figure S3.1 Observed spectral changes (inset A) and kinetic traces recorded at 400 nm (A) and 330 nm (B) for the

degradation of Morin. Reaction conditions: 0.1 M HCO3-, 5 · 10

-5 M Morin, 0.015 M H2O2, pH 9.0, room temp., (—) 4 ·

10-5

M [MnII(bpy)2Cl2], (—) 2 · 10

-5 M [Mn2

III/IV(µ-O)2(bpy)4](ClO4)3, (—) 4 · 10

-5 M Mn

IICl2 + 8 · 10

-5 M bpy.

Figure S3.2 Kinetic traces recorded at 400 nm for the degradation of p-nitrophenol. Reaction conditions: 0.1 M HCO3-,

5 · 10-5

M p-nitrophenol, pH 9.0, room temp., (A): 0.015 M H2O2, (—) 4 · 10-5

M [MnII(bpy)2Cl2], (—) 2 · 10

-5 M

[Mn2III/IV

(µ-O)2(bpy)4](ClO4)3, (—) 4 · 10-5

M MnIICl2 + 8 · 10

-5 M bpy. (B): 0.01M H2O2, (—) 5 · 10

-6 M [Mn

II(bpy)2Cl2], (—)

2.5 · 10-6

M [Mn2III/IV

(µ-O)2(bpy)4](ClO4)3, (—) 5 · 10-6

M MnIICl2 + 1 · 10

-5 M bpy and corresponding spectral changes

(inset B).

(A) (B)

(A) (B)


81

Figure S3.3 (A) Kinetic traces recorded at 400 nm for the degradation of p-nitrophenol at different pH values. (B) pH

dependence of the initial rate of p-nitrophenol bleaching (■) 4 · 10-5

M [MnII(bpy)2Cl2], (●) 2 · 10

-5 M [Mn2

III/IV(µ-

O)2(bpy)4](ClO4)3. Reaction conditions: 0.1 M HCO3-, 5 · 10

-5 M p-nitrophenol, 0.01 M H2O2, room temp..

Figure S3.4 Kinetic traces recorded for the oxidation of 5 · 10-5

M Orange II by (A) 4 · 10-5

M [MnII(bpy)2Cl2] and 2 · 10

-5

M [Mn2III/IV

(µ-O)2(bpy)4](ClO4)3 in the presence of 0.1 M HCO3- and 0.015 M H2O2, compared to the same reaction in

the presence of 0.2 M NaNO3. (B) 4 · 10-5

M [MnII(bpy)2Cl2] in the presence of 0.1 M NaCl. Reactions followed at 484

nm, pH 9.0, room temp., over 20 min.

(A) (B)


82

Figure S3.5 Observed rate constants measured at 484 nm as a function of [MnII(bpy)2(H2O)2]

2+ (A) and [Mn2

III/IV(µ-

O)2(bpy)4]3+

(B) concentration at different total carbonate concentrations. Reaction conditions: 0.01 M H2O2, 5 · 10-5

M

Orange II, pH 9.0, room temp., (■) 0.1 M HCO3-, (▲) 0.5 M HCO3

- .

Figure S3.6 X-band EPR spectra of 0.1 M bicarbonate containing solutions of H2O:tBuOH (1:1). Reaction conditions: pH

9.0, room temp., (A) 1 · 10-4

M [MnIIbpy2Cl2] and (B) 5 · 10

-5 [Mn2

III/IV(µ-O)2(bpy)4](ClO4)3 with (—) 0.01 M and (—) 0.04

M H2O2 immediately after mixing. EPR conditions: 8.98 GHz, 6 K, 1 mW microwave power, modulation amplitude 20

mT.

(A) (B)

(A) (B)


83

Figure S3.7 Spectral changes recorded for the reaction of 0.05 · 10-3

M [Mn2(µ-O)2(bpy)4](ClO4)3 with 5 · 10-3

M H2O2 in

0.1 M HCO3- solution. Spectra recorded in time intervals of 0.2 s for the first two seconds, pH 9.0, room temp..

Figure S3.8 Kinetic traces recorded at 400 nm for the in situ oxidation of Morin (A) and p-nitrophenol with 4 · 10-5

M

MnIICl2 and increasing amount of 2,2'-bipyridine added to the reaction mixture. Reaction conditions: 0.1 M

bicarbonate, between 0.5-4 % CH3CN, 5 · 10-5

M dye, 0.015 M H2O2, pH 9.0, room temp..

(A) (B)


84

4. Metal ion - catalyzed oxidative degradation of Orange II by peracetic acid

85

4 METAL ION - CATALYZED OXIDATIVE DEGRADATION OF

ORANGE II BY PERACETIC ACID

4.1 GENERAL REMARK

The following chapter is based on the original publication: Mechanistic studies on the oxidative

degradation of Orange II by peracetic acid catalyzed by simple manganese(II) salts. Tuning the

lifetime of the catalyst, Sabine Rothbart, Erika Ember and Rudi van Eldik, New J. Chem. 2012, 36,

732-748.

.

4.2 INTRODUCTION

Persistent and non biodegradable organic waste still marks one of the main environmental

problems of our time. Significant ecological impact is caused by industrial dye waste since over 15

% of textile dyes are lost in waste water streams during the dyeing operation.[1] A recent review

emphasized the high cost of disposing the high volumes of dye effluent and that 128 tons of dyes

are released daily to the global environment.[2] About half of the global production of synthetic

colorants (700.000 t per year) is classified as aromatic azo compounds.[3] Textile dyes in general

are designed to resist chemical, biochemical and photochemical degradation. In aerobic processes,

azo compounds are known to be largely non-biodegradable, whereas under anaerobic conditions

they can be reduced to even more hazardous intermediates.[4] The increasing ecological awareness

stimulated an active field of scientific research dedicated to new and ecologically worthwhile

oxidation processes for the catalytic decomposition of environmental pollutants[5]and dyes[6] in

water. Among the various transition metals for catalytic oxidation, manganese is of particular

interest as one of the most efficient and environmentally benign elements. There are various

manganese complexes with different salen,[7] porphyrin,[8] tacn (1,4,7-triazacyclononane)[6e, 9] or

aromatic N-donor ligands[10] known to efficiently catalyze the oxidation of a wide range of

substrates. However, several limitations including elaborated synthetic methods, long reaction

times and substrate scope, have still to be resolved. Recent attention has therefore been focused

http://dict.leo.org/ende?lp=ende&p=Ci4HO3kMAA&search=ecologically&trestr=0x8004

http://dict.leo.org/ende?lp=ende&p=Ci4HO3kMAA&search=worthwhile&trestr=0x8004


86

(4.1)

(4.2)

on metal complexes that use cheap and clean oxidants to bring about efficient oxidation under

mild reaction conditions.[11] Among these, H2O2 or peracetic acid are commonly used green

oxidants due to their eco-friendly nature. The only by-products formed are water and oxygen or

acetic acid, respectively.

In order to develop efficient and simple pre-catalysts for the oxidative degradation of

structurally different organic dyes by H2O2, our earlier work focused on the reactivity of simple

MnII salts under mild reaction conditions (Equation 4.1).[12]

It is known that simple MnII salts are able to form very reactive aquated intermediates in

aqueous solution.[13] Moreover, in slightly alkaline medium, the introduction of a hydroxy ligand

trans to a water ligand is expected to produce more labile HO-Mn-H2O species which are

considered to be of major importance for the catalytic activity.

Detailed mechanistic investigations revealed the in situ formation of percarbonate (HOOCO2-)

as a key molecular entity under the selected experimental conditions.[12, 14] Percarbonate can

coordinate to the MnII center and lead to the formation of a quasi-stable MnII-η2-percarbonate

complex, which subsequently undergoes heterolytic cleavage of the peroxide O-O bond to form

MnIV=O intermediates. The nature of the produced reactive intermediate was confirmed by low

temperature EPR measurements under catalytic reaction conditions.[12a, b] Although simple MnII

ions could efficiently catalyze the oxidative degradation of a large number of organic substrates

under mild reaction conditions, the use of rather high concentrations of pre-catalyst and

bicarbonate in solution was required.[12a, b]

In an attempt to overcome these limitations, we now report our findings for the Mn II catalyzed

degradation of Orange II with peracetic acid (PAA) under mild reaction conditions. PAA is formed

in an equilibrium reaction of hydrogen peroxide and acetic acid catalyzed by sulfuric acid

(Equation 4.2).[15]


87

Similar to hydrogen peroxide, PAA is often used as a safe and environmentally friendly oxidant,

since the released side-products, viz. water and acetic acid, are non-toxic unlike organic peroxides

or other oxidants used. In contrast to hydrogen peroxide, mechanistic investigations with

peracetic acid are very rare. Yet, a strong advantage of peracetic acid compared to hydrogen

peroxide is its lower tendency to undergo catalase-like reactions, which could result in rapid

oxidant decomposition under catalytic conditions. These attractive oxidizing properties make the

transition metal catalyzed activation of peracetic acid a worthwhile objective in chemical

oxygenation processes.


4.3.1 Peracetic acid formation and its decomposition at higher pH

It is generally accepted that at a pH equal to the pKa of PAA (8.2)[16], the hydrolysis of PAA to

acetic acid and H2O2, and the spontaneous decomposition strongly complicate the H2O2-PAA

equilibrium, which causes poor reproducibility of the data.[16a, 17] For a better understanding of the

reactivity and in particular the stability of PAA as oxidant, we investigated its in situ formation and

hydrolysis by means of 13C-NMR spectroscopy (see Supporting Information). When H3C-13COOH

and H2O2 were mixed in water, the slow formation of a new 13C-NMR signal, attributed to H3C-

13C(O)OOH, was observed at 175.3 ppm (see Figure 4.1).

In order to accelerate the formation of 13C-labeled PAA, 0.1 M H2SO4 was added as catalyst.

13C-NMR experiments revealed that under the selected experimental conditions PAA reached its

maximum concentration after approximately 50 h and remained constant over longer periods of

time (see Figures S4.1 (A) and (B), Supporting Information). The observed rate constants at pH = 1

for the formation of PAA is (1.78 ± 0.05) · 10-5 s-1 at 25 °C. To study the hydrolysis of PAA to form

H2O2 and 13C-AcOH, a fully equilibrated solution of 13C-labeled PAA (formed from 2.5 M 13C-AcOH,

2.5 M H2O2 and 0.1 M H2SO4) was diluted in 0.1 M H2SO4 to 25 % of the initial concentration and

the set-in of the new equilibrium position was followed by 13C-NMR spectroscopy (see Figure S4.1

(C) and (D), Supporting Information).


88

(4.3)

Figure 4.1 13

C - NMR spectra recorded for the in situ formation of PAA in an aqueous solution of 2.5 M 13

C-AcOH and

2.5 M H2O2 in the presence of 0.1 M H2SO4 at 25 °C. Inset: The development of the H3C-13

C(O)OOH NMR signal at

175.3 ppm as a function of time.

From the determined equilibrium concentrations (x1 = [13C-PAA]eq, x2 = [13C-AcOH]eq) the

average equilibrium constant of K = 1.3 ± 0.1 at 25 °C was calculated for reaction 4.2 according to

Equation 4.3:

Even though there is only limited literature data available for the equilibrium constant and the

reported values differ as a function of temperature, viz. 3.7 (20 °C), 4.0 (30 °C), 4.3 (35 °C),[18a] 1.21

(0 °C), 1.20 (5 °C), 1.29 (35 °C),[18b] 2.10 (20 °C),[18c] our value fits in quite well with the data

published most recently.[18b, c]

An equilibrated solution of 13C-labeled PAA (produced from 2.5 M 13C-AcOH, 2.5 M H2O2 and

0.1 M H2SO4) was diluted to 50 % of the initial concentration with NaOH to a pH of approximately

10 and the decomposition reaction was monitored by 13C-NMR at 25 °C (Figure S4.2 (A) and (B),

Supporting Information). The observed rate constant is (2.3 ± 0.1) · 10-4 s-1 with a half-life of 3000

s. Consequently, it can be concluded that PAA can be considered to be stable in aqueous solution

under the experimental conditions selected for the studied degradation of Orange II, for which the

21220

1201

xx-)O(Hc

xO)(HcxK

][

][


89

half-life is at most 100 s. These findings are in agreement with reports in the literature on the

formation and stability of PAA based on redox titrations.[18]

4.3.2 General observations

In order to study the catalytic degradation of Orange II by PAA, a series of measurements was

performed at pH 9.5 and 25 °C. Figure 4.2 (A) shows the UV/Vis spectral changes that accompany

the catalytic degradation reaction.

Figure 4.2 (A) UV/Vis spectral changes observed during the 1 · 10-5

M MnII catalyzed oxidative degradation of 5 · 10

-5 M

Orange II by 0.01 M oxidant at pH 9.5 (0.05 M NaHCO3 buffer) and 25 °C (for the sake of clarity only every tenth

spectrum is shown). (B) Comparison of the absorbance vs. time traces at 484 nm for the oxidative degradation of 5 ·

10-5

M Orange II catalyzed by 1 · 10-5

M MnII and 0.01 M PAA or 0.01 M H2O2 at pH 9.5 (0.05 M NaHCO3 buffer) and 25

°C.

It is obvious that under these reaction conditions the bleaching reaction is not only limited to

the cleavage of the azo-linkage (484 nm), destruction of the more stable aromatic subunits (310

nm) also occurs.[19] Although PAA is a strong oxidant, its spontaneous non-catalyzed reaction with

the dye substrate is about 500 times slower compared to the reactivity in the presence of the Mn II

catalyst, as is apparent from the corresponding observed rate constants, viz. kobs cat. = 6.05 · 10-2 s-1

vs. kobs non-cat. = 1.28 · 10-4 s-1 at 25 °C (see Figure 4.2 (B)).

Consequently, the spontaneous reaction between PAA and Orange II does not significantly

contribute to the determined degradation rates for the MnII catalyzed reactions. Addition of an

excess of EDTA to the non-catalytic reaction mixture only slightly reduced the spontaneous non-

(B) (A)


90

catalyzed reaction with the dye substrate due to scavenge of possible trace metals (as for instance

adventitious Mn) in the stock solutions (for comparison see Figure S4.3, Supporting Information).

When the results are compared to the MnII catalyzed oxidative degradation of Orange II by H2O2 as

oxidant under identical experimental conditions (see Figure 4.2), it is obvious that the rate

constant in the presence of PAA as oxidant is several orders of magnitude higher (kobs = 6.05 · 10-2

s-1 for PAA, and kobs = 7.92 · 10-4 s-1 for H2O2[12a] at 25 °C). This surprising result clearly suggests

fundamental differences in the activation of the two peroxides by MnII ions. Since the

experimental data did not allow the determination of pseudo-first-order rate constants under all

experimental conditions, the initial rate method was used in most cases to further quantify the

obtained results. If not stated otherwise, the MnII salt used in this study was MnIICl2 · 2H2O, since

no influence of the anion on the reaction course was observed, as is obvious from a comparison of

the Orange II degradation rate for different MnII salt catalysts (see Figure S4.4, Supporting

Information).

The possible contribution of free radicals to the remarkable efficiency of the MnII catalyzed

degradation of Orange II by PAA could be excluded by the use of tBuOH and BHT as radical

scavenger. Although hydroxyl radicals might be present during the PAA induced reaction, they do

not participate in the studied oxidation process, since in the presence of tBuOH or BHT no negative

effect on the reaction course was observed (see Figure S4.5, Supporting Information). In further

control experiments it could be confirmed that neither acetate (since commercial PAA is a

equilibrated mixture of H2O2 and acetic acid), nor the counter ion of the MnII salts affects the

performance of the catalyzed dye degradation (see Figure S4.4, Supporting Information). The

influence of ionic strength was found to be negligible over a wide concentration range (see Figure

S4.6, Supporting Information).

4.3.3 MnII + PAA – Intermediates formed in the absence of substrate

4.3.3.1 UV/Vis spectroscopy

In order to follow the speciation of Mn during the catalytic cycle, we analyzed the catalytic

solution in detail in the absence of added substrate by means of UV/Vis and EPR spectroscopy.

From the observed spectral changes and the corresponding EPR spectra recorded at different time

intervals some preliminary conclusions can be drawn. As shown in Figure 4.3, the reaction of


91

aquated MnII proceeds in two different phases. In the first phase of the reaction (Figure 4.3 (A),

inset) a new species with weak absorbance bands at approximately 405 and 470 nm is formed.

This intermediate persists for several seconds. Similar spectra with a band at 470 nm are often

attributed to MnIII species.[20] However, absorbance bands in the region of approximately 450 nm

are also considered to be a result of an oxo to MnIV charge transfer transition of a possible MnIV=O

intermediate.[12b, c, 21] In the second phase of the reaction, the first intermediate is rapidly

converted to permanganate (band at 525 nm) and colloidal MnIVO2 (band at 350 nm) (Figure 4.3

(B)).

Figure 4.3 UV/Vis spectra recorded for the reaction of 1 · 10-4

M MnII with 2.5 · 10

-2 M PAA in a 0.05 M NaHCO3

containing buffer solution at pH 9.5 and 25 °C. (A) Reactions studied over 300 s. Inset: first 25 s. (B) Biphasic behavior

of the kinetic traces at 350 nm and 525 nm.

Due to the broad absorbance of colloidal MnIVO2 over the whole spectral range (with

characteristic bands at 312 and 350 nm[22]), the exact quantification of the generated products

became difficult. However, the generation of the pink permanganate ion is evident from the five

characteristic absorbance bands in the range from 500-570 nm. In general, the formation of

permanganate is known to also involve short-lived MnVI and MnV intermediates, which tend to

undergo disproportionation to permanganate and MnIVO2 in the absence of a large excess of OH-

.[23, 24] This stands in direct contrast to the behavior of the MnII/H2O2/HCO3- system under

comparable conditions, in which the oxidation state of manganese did not exceed that of Mn IV=O

and no permanganate was formed.[12]

(A) (B)


92

The observation of sigmoid-shaped kinetic traces as shown in Figure 4.3 (B), which are marked

by the beginning of the second reaction phase, i.e. formation of colloidal MnO2 (which partially

precipitates at higher [Mn] during the course of the reaction in phase II) and permanganate,

remind of typical autocatalytic pathways that are common features of chemical reactions involving

higher oxidation states of manganese.[25, 26, 27] It was found that during the initial phase the parallel

existence of MnVIIO4- and MnII, as well as MnIII and MnIV[27] during the MnII catalyzed

decomposition reaction of PAA, resulted in the formation of permanganate and Mn IVO2 depending

on the initial reactant concentration. In neutral to basic media, the autocatalytic contribution

occurring in oxidation reactions by permanganate are generally attributed to the existence of Mn IV

species as autocatalytic reaction product.[25, 28, 29]

Catalyst and oxidant influence on the biphasic reaction behavior. In the absence of substrate the

studied reaction is very sensitive to variation of the reactant concentration. Since the use of

stopped-flow techniques is hardly possible due to the formation of MnIVO2 precipitates and the

way the catalytic mixture is prepared via a pH jump, reliable results can only be obtained with in

situ UV/Vis spectroscopy. Nevertheless, upon variation of the metal to catalyst ratio some trends

in the biphasic reaction behavior could be observed. A comparison of the spectral changes for

different [MnII] at constant [PAA] (Figure 4.4), clearly shows that the length of the first reaction

phase at pH 9.5 strongly depends on the [MnII] present in solution.

Figure 4.4 Kinetic traces recorded at 350 nm for the reaction of 0.01 M PAA with different [MnII]. Inset: 1 · 10

-4 M Mn

II

with different [PAA]. Reaction conditions: 0.05 M NaHCO3 buffer, pH 9.5 and 25 °C.


93

If the pre-catalyst concentration is kept constant (Figure 4.4, inset), an increase in the [PAA]

leads only to minor changes in the biphasic behavior, but the amount of the species formed during

phase I increased, as indicated by the higher absorbance. Higher [MnII] on the other hand resulted

in a much shorter reaction time during phase I (see Figure 4.4). A higher overall [Mn] probably

causes a faster consumption or decomposition of the equilibrium content of H2O2, omnipresent in

peracetic acid solutions, which as a consequence shortens the induction period. The reducing

character of H2O2 towards high valence metal species also represents another likely explanation

for the biphasic reaction behavior. As long as H2O2 is present in the reaction mixture it maintains a

steady-state concentration of the lower valence Mn species, i.e. MnIII and MnIV, by reduction of

the rapidly formed higher valence manganese intermediates. Once all H2O2 has depleted, the lack

of the back reaction causes immediate accumulation of colloidal MnIVO2 and permanganate as the

only stable high valence reaction products. This suggestion prompted us to further investigate the

role of hydrogen peroxide in the catalytic system.

H2O2 as reductant. Commercially available PAA is not a pure peroxide, but an equilibrated mixture

of acetic acid (45 %), hydrogen peroxide (6 %) and water with sulfuric acid as catalyst. Thus, PAA

solutions always contain a significant amount of hydrogen peroxide. The effect of an extra amount

of hydrogen peroxide is also apparent in the UV/Vis spectral changes observed in the absence of a

substrate. If MnII and PAA react in the presence of additional 2 · 10-3 M of H2O2, the first phase of

the reaction is extended, while the formation of colloidal MnIVO2 and permanganate is delayed for

50 to 100 seconds (see Figure 4.5).

This behavior can be interpreted in terms of a continuous fast reduction of the rapidly in situ

formed high valence manganese species (MnVIIO4- and colloidal MnIVO2) as long as H2O2 is present

in the catalytic reaction mixture. Once the available hydrogen peroxide has been used, the

concentration of the high valence Mn species increases. However, since MnVIO42- and MnVO4

3- are

very unstable below pH 14, the known disproportionation chemistry will apply and lead to the

accumulation of colloidal MnIVO2 and MnVIIO4- as final products (Scheme 4.1).

http://dict.leo.org/ende?lp=ende&p=Ci4HO3kMAA&search=disproportionation&trestr=0x2001


94

Figure 4.5 Absorbance changes at 350 nm for the reaction of 2 · 10-5

M MnII with 0.01 M PAA (0.05 M NaHCO3 buffer,

pH 9.5, 25 °C) without and with different amounts of added H2O2.

Scheme 4.1 Reactions suggested to account for the disproportionation of high valence Mn-oxo anions.

Transiently formed colloidal manganese dioxide is also known to be reduced by hydrogen

peroxide.[30] However, depending on the experimental conditions, either partial or total reduction

of MnIVO2 occurs. Whereas under acidic conditions total reduction of MnIVO2 to MnII is possible, it

is reported that under neutral to slightly basic conditions only partial reduction of the colloid

occurs.[31] It is generally accepted that these processes proceed via non-stoichiometric hydrated

MnIVO2 with interstitial metal ions of lower valence like MnIII, leading to an overall increased

amount of MnIII present in the catalytic mixture.[31, 32, 33] According to studies by Perez-Benito and

co-workers, partial reduction of MnIVO2 only takes place on the surface of the colloid.[31] The result

is a monomolecular MnII oxide layer on the surface of a mixed MnIVO2-MnIIO colloid. Both the MnIV


95

and MnII oxides in a mixed MnIVO2-MnIIO colloid can react to form MnIII oxide according to Scheme

4.2, which may also account for the formation of the absorbance shoulder at 470 nm during the

initial stage of the reaction (see Figure 4.3 (A), inset).[31]

Scheme 4.2 Possible reduction pathway of MnIV

O2 with H2O2.

In order study the role of hydrogen peroxide in the catalytic cycle, we investigated the

reduction of the reaction products MnVIIO4- and colloidal MnIVO2 with H2O2 as a function of pH

under reaction conditions comparable to those for the catalytic degradation reaction. Upon

addition of H2O2 to a colloidal MnIVO2 containing solution at pH 9.5, a fast decrease in the broad

and characteristic absorbance between 300 and 400 nm was observed (see Figure S4.7 (A),

Supporting Information). As expected, the reduction of colloidal MnIVO2 was not complete and the

extent to which the reduction proceeds depends strongly on the pH (see Figure S4.7 (B),

Supprorting Information). This is due to the behavior described above and the formation of a

precipitate at higher pH, which results in incomplete reduction. In contrast, the reaction of H2O2

with MnVIIO4- was found to be accelerated with increasing pH (see Figure S4.8, Supporting

Information), which presumably originates from the increase in the formation of HOO- on

approaching the pKa of hydrogen peroxide (pKa = 11.67)[34] and the thereby increased reducing

ability. Although oxidation reactions by permanganate, often autocatalytic, have been studied for

quite a long time,[35, 36] they are not fully understood. Nevertheless, the results clearly indicate the

occurrence of a reaction of the in situ formed intermediates with hydrogen peroxide, which most

likely represents a parallel degradation reaction in the presence of a dye.

Role of pH. Another factor that drastically influences the biphasic reaction behavior of MnII with

PAA, is the pH of the solution. The maximum rate of permanganate and MnIVO2 formation was

found at a pH around 8.5, where no induction period is observed. Surprisingly, a pH of 8.5 is not

where the best reactivity is observed, which implies that another oxidizing intermediate than

permanganate is responsible for the excellent reactivity at pH ≈ 9.5. Increasing the pH from 8.8 to

10.1 causes a drastic delay in the formation of the high valence oxidation products (see Figure


96

S4.9, Supporting Information). This could, however, also be due to the enhanced reducing

properties of H2O2 at higher pH since it cumulatively dissociates to form HOO- (pKa = 11.67)[34], i.e.

another potential reducing agent for higher valence manganese species such as MnVIIO4-.

Influence of the total carbonate concentration. Earlier studies on the MnII ion catalyzed

degradation of Orange II by H2O2 revealed the in situ formation of percarbonate (HOOCO2-) from

bicarbonate and hydrogen peroxide as a key molecular entity under the selected reaction

conditions.[12] Furthermore, the role of bicarbonate as possible ligand that facilitates the

heterolytic O-O-bond scission and thereby MnIV=O formation was discussed.[12] Since PAA is

strongly acidic, and HCO3-/CO3

2- was used as buffer system, the formation of possible

intermediates was also investigated as a function of the total carbonate concentration.

Control experiments already excluded any influence of ionic strength (see Figure S4.6) so that

under identical experimental conditions the difference in the formation of intermediates is directly

attributed to the presence of bicarbonate and carbonate. Figure 4.6 shows the UV/Vis spectral

changes that accompany the reaction of MnII with PAA at two different [total carbonate], viz. 0.05

M (Figure 4.6 (A) and 0.4 M (Figure 4.6 (B)).

Figure 4.6 UV/Vis spectral changes that accompany the reaction of 2 · 10-5

M MnII with 0.01 M PAA at pH 9.5 and 25 °C

in (A) 0.05 M NaHCO3 buffer and (B) 0.4 M NaHCO3 buffer. Inset: UV/Vis spectra recorded at t = 300 s for the reaction

as a function of [total carbonate].

On following the spectral changes at 350 nm (colloidal MnIVO2) and 525 nm (MnVIIO4-) (see

Figure S4.10, Supporting Information) under variation of the [total carbonate], only minor changes

(A) (B)


97

in the biphasic behavior were observed. Yet, it is evident that the distribution of the in situ formed

high-valance manganese intermediates changes drastically as a function of [HCO3-/CO3

2-]. The

inset in Figure 4.6 (B) presents UV/Vis spectra recorded at a reaction time of 300 s for a [total

carbonate] range from 0.025 to 0.4 M. Whereas colloidal MnIVO2 is favored at lower buffer

concentration, the formation of MnVIIO4- is enhanced at higher [total carbonate]. A reasonable

explanation might be that HCO3-/CO3

2- can coordinate to the metal center and thereby facilitate

either a further oxidation by PAA or open up alternative reaction pathways leading to the

enhanced formation of permanganate. The outstanding role of bicarbonate ligands in the

beneficial manipulation of the MnIII/MnII redox couple by the formation of highly reactive Mn-

bicarbonate complexes has been discussed before.[12a, 37] The present results provide strong

evidence that a change in [HCO3-/CO3

2-] should result in drastic changes in the degradation

reaction of the dye (see further discussion).

4.3.3.2 EPR spectroscopy

The above described information on how to tune the biphasic behavior of the reaction of MnII

with PAA, enabled us to select appropriate reaction conditions that allowed sufficient time to

investigate the nature of the formed intermediates in more detail. In an attempt to further

characterize the possible reactive intermediates, samples were taken directly from the

concomitant UV/Vis measurements at different reaction times (marked by the arrows in Figure 4.7

(A)), immediately frozen to quench the reaction and analyzed by perpendicular EPR spectroscopy

at 10 K.

The initial MnII catalyst containing buffer solution without any oxidant present, showed the

typical six-line pattern of MnII (I = 5/2, S = 5/2) centered at g ≈ 2 (see Figure 4.7 (B), first

spectrum). Formation of hydroxo species in basic medium and complex-formation with

bicarbonate further reduces the symmetry of the ligand field around the MnII ion and thereby

causes a weaker intensity of the six-line EPR signal compared to the more symmetric fully aquated

MnII ion as described before.[12b] Figure 4.7 (B) shows the successive development of the X-band

EPR spectrum upon addition of MnII to a PAA/H2O2 containing buffer solution at pH 9.8. The

decrease in the overall spectral features compared to the spectrum in the absence of oxidant

indicates that a small amount of the Mn catalyst is in an EPR-silent form, which according to the


98

UV/Vis spectral changes is most likely a MnIII species. However, neither Mn2III/III nor Mn2

IV/IV

dimeric species are expected to give an EPR spectrum.[38]

Figure 4.7 (A) UV/Vis spectra recorded during the reaction of 1 · 10-4


-2 M PAA and 5 · 10

-3 M H2O2

in 0.05 M NaHCO3. The first phase of the reaction was extended by the addition of 5 · 10-3

M H2O2. Inset: kinetic traces

at 350 and 525 nm. Arrows mark the time at which different samples were taken for the EPR spectra in (B). Reaction

conditions: 1 · 10-4


-2 M PAA and 5 · 10

-3 M H2O2 at pH 9.8 and 25 °C. (B) X-band EPR spectra

recorded at 10 K for 1 · 10-4

M MnII in carbonate buffer and at different time intervals after the addition of 2.5 · 10

-2 M

PAA and 5 · 10-3

M H2O2 in 0.05 M NaHCO3. EPR conditions: 8.95 GHz, 10 K, 1 mW microwave power, modulation

amplitude 400 mT.

In the first spectrum recorded 20 s after mixing, still some MnII is found. The lack of the typical

well resolved MnII sextet at g ≈ 2 is attributed to the enhanced formation of manganese-hydroxo

species, which results in a broader and weaker transition of the MnII signal at g ≈ 2. Moreover, two

further signals at g ≈ 2 and g ≈ 4 can be observed. On one hand the sharp g ≈ 2 transition might be

due to a very low [MnVIO42-], since the spectrum slightly resembles the one obtained for a readily

prepared, dark green MnVIO42- solution (d1, S = 1/2) (see Figure S4.11, Supporting Information).

However, the lack of the characteristic absorbance band at 630 nm for MnVIO42- in the

corresponding UV/Vis spectra of the intermediate during the initial phase implies that only a

minor [MnVIO42-] is present due to the equilibrium concentration of H2O2 omnipresent in PAA. On

the other hand, the characteristic signals at g ≈ 2 and g ≈ 4 can also originate from a high-spin MnIV

(S = 3/2) species in an octahedral environment having D < hν, which also has a higher amplitude

feature at g ≈ 2.[39, 40] In general, EPR spectra of d3 MnIV ions in an axial field (E/D = 0) are often

difficult to interpret, since they strongly depend on the magnitude of the zero-field splitting

(A) (B)


99

parameters.[41] A large axial zero-field splitting parameter D accounts for a spectrum dominated by

the g ≈ 4 signal, as for complexes with hard oxygen-rich catecholate[42a] and sorbitalate[42b] ligands.

If D is small, the g ≈ 2 signal dominates with relatively weak low field signals. This is seen for

example for sulfur-containing thiohydroxamate[41b] and dithiocarbamate[43] manganese(IV)

complexes. The assignment of a MnIV intermediate also correlates with the observed UV/Vis

spectra observed during the initial reaction stage, in which the absorbance bands in the region of

450 nm can be interpreted as a result of an oxo to MnIV LMCT band.[12b, c, 21] Consequently, it is

more likely to be the result of an in situ formed MnIV species showing g ≈ 2 and g ≈ 4 signals during

the initial phase of the reaction of MnII with PAA.

From the subsequent EPR spectra (Figure 4.7 (B)) taken with a delay of 75 and 120 s for the

same reaction mixture, it is apparent that the concentration of the in situ formed MnIV

intermediate increases during the time of the first reaction phase. With the beginning of the

second reaction phase (130 s) this MnIV intermediate is not detected anymore, whereas the

corresponding UV/Vis spectra show the formation of MnVIIO4- and precipitation of colloidal

MnIVO2. Although the UV/Vis and EPR results indicate the existence of monomeric MnIII and

monomeric MnIV species during the initial phase of the reaction, no evidence for the formation of

bis µ-oxo bridged MnIII/MnIV dimers was found. For such a strongly coupled dimer (MnIII(µ-

O)2MnIV) a very characteristic 16-line EPR signal at g ≈ 2 would be expected.[44] The EPR spectra

recorded in the presence of tBuOH as radical scavenger show a similar behavior, indicating that

free radical processes do neither interfere in the catalytic dye degradation nor in the reaction of

the MnII catalyst with PAA in the absence of substrate (see Figure S4.12, Supporting Information).

Very similar results were obtained when the reaction was carried out at a higher [total carbonate]

(see Figure S4.13, Supporting Information).

For a better understanding of the influence of the equilibrium content of H2O2 on the in situ

formed intermediates, EPR spectroscopic experiments were also performed in the presence of an

excess of H2O2. As is evident from the UV/Vis spectral changes recorded for the reaction between

MnII and PAA in the presence of a large excess of H2O2 (see Figure 4.8 (A)), formation of MnVIIO4-

and colloidal MnIVO2 does not occur on a catalytically relevant time scale. Instead, the

disappearance of the characteristic MnII sextet at g ≈ 2 and the concomitant formation of the MnIV

species are delayed for several seconds compared to the experiments in the absence of an excess

H2O2. Moreover, the intensity of the above described MnIV signals at g ≈ 2 and g ≈ 4 is decreased,


100

implying that less MnIV is formed (see Figure 4.8 (B)). Consequently, there is a reductive influence

of the excess of H2O2 on the in situ formed MnIV intermediate, which shifts the complex

oxidation/reduction equilibria in solution to the side of the MnII pre-catalyst and thereby causes

delayed formation of higher valence-species, such as the MnIV intermediate.

Figure 4.8 (A) UV/Vis spectra recorded for the reaction of 1 · 10-4

M MnII with 0.01 M PAA and 0.05 M H2O2 in a 0.05 M

NaHCO3 containing buffer solution at pH 9.8 and 25 °C. Inset: kinetic traces at 350 nm and 525 nm, arrows mark the

time at which samples were taken for EPR spectroscopic measurements. (B) X-band EPR spectra recorded at 10 K for 1

· 10-4

M MnII in carbonate buffer and at different time intervals after the addition of 0.01 M PAA and 5 · 10

-2 M H2O2 in

0.05 M NaHCO3. EPR condidions: 8.95 GHz, 10 K, 1 mW microwave power, modulation amplitude 400 mT.

4.3.4 Comparison of reactivity of different high valent oxo-manganese species

with Orange II

In the absence of PAA oxidant. Since the reaction of peracetic acid with aquated manganese ions

in slightly basic solution leads to the rapid in situ generation of several high-valence manganese-

oxo intermediates, we tested the different species separately for their degradation ability towards

the dye substrate in the absence of an oxidant. Following the spectral changes during the

stoichiometric (1:1) reaction of Orange II with the freshly prepared O3MnV(OH)2-, MnVIO42- and

MnVIIO4- species (Figure 4.9) under the same experimental conditions, it is obvious that the activity

of the different oxo complexes decreases drastically (Figure 4.9, inset) with decreasing oxidation

state of the metal center.

(A) (B)


101

Figure 4.9 UV/Vis spectra of 2 · 10-5

M high valence manganese-oxo anions. Inset: comparison of the spectral changes

at 484 nm during the stoichiometric reaction of 2 · 10-5

M Orange II with 2 · 10-5

M O3MnV(OH)

2-(a), Mn

VIO4

2- (b) and

MnVII

O4- (c).

[45] Reaction conditions: 0.05 M NaHCO3 buffer, pH 9.5 and 25 °C.

This is not surprising since Pode and Waters already reported that the order of reactivity of

high valence Mn-oxo anions towards organic substrates decreases markedly along the series

MnVIIO4- > MnVIO4

2- > MnVO43-.[46] Although permanganate is a strong oxidant, its full oxidative

potential in the studied reaction with Orange II develops at higher pH values as can be seen from

the pH dependence in (Figure S4.14, Supporting Information). If the degradation reaction is

performed with manganate(VI) and hypomanganate(V) under identical experimental conditions, it

becomes clear that permanganate is the stronger oxidant. Without any primary oxidant present,

one equivalent of Orange II is oxidized by two equivalents of MnVIIO4- (see the remaining

absorbance of Orange II at 484 nm, Figure 4.9, inset).

Moreover, during the first 50 s of the oxidation by manganate(VI), a build-up of the

absorbance at 484 nm can be observed. This is due to the background disproportionation reaction

during which permanganate is formed. At the same time the Orange II degradation starts as soon

as sufficient permanganate is formed in the disproportionation process. The time scale of these

absorbance changes fits well with the disproportionation of manganate(VI) in the absence of

substrate under identical experimental conditions (see Figure S4.15 (A), Supporting Information).

Moreover, a comparison of the pH dependence of the disproportionation of manganate(VI) in the

stoichiometric reaction with Orange II clearly shows that no reaction between MnVIO42- and


102

Orange II occurs under conditions where manganate(VI) disproportionation is avoided, i.e. at

higher pH (see Figure S4.15 (B), Supporting Information).

Upon addition of hypomanganate(V), no absorbance decrease was observed, probably

because of the high instability of O3MnV(OH)2- in non-basic media (pKa = 13.7[23]). Control

experiments at higher pH confirmed this order of reactivity. If the pH is high enough to ensure the

stability of hypomanganate(V) and manganate(VI) for several seconds, no reaction with the dye

substrate occurred (see Figure S4.16, Supporting Information). Consequently, it is assumed that

despite their possible transient presence, hypomanganate(V) and manganate(VI) play a negligible

role as active species in the studied MnII catalyzed dye degradation of Orange II with PAA.

As expected, colloidal MnIVO2 showed no reaction with the substrate (Figure S4.17, Supporting

Information). Yet, oxo-manganese(IV) compounds are known to be versatile homogeneous

oxygenation species,[12, 47] but in the absence of a stabilizing ligand they rapidly agglomerate to

give insoluble MnIVO2 and thus lose their oxygenation ability. Hence, the in situ generation of

transient, soluble MnIV-oxo species was investigated by stoichiometric reduction of MnVIIO4- with

sulfite and hydrogen peroxide, and immediate addition of dye substrate as soon as the MnVIIO4-

band vanished. Figure S4.18 (A) (Supporting Information) presents the UV/Vis spectra of MnVIIO4-

after reduction with sulfite, and after immediate addition of Orange II to the obtained Mn IV

species. Although MnVIIO4- is no longer present after reduction, a decrease in the characteristic

absorbance of Orange II is observed in the reaction with substrate (see Figure S4.18 (B),

Supporting Information). The rather moderate reactivity of the in situ formed transient MnIV

species towards Orange II is attributed to the rapid formation of a MnIVO2 precipitate. Yet, it

exhibits oxygenating capabilities and can represent a catalytically active species.

In the presence of PAA oxidant. The reactivity of freshly prepared Mn-oxo species towards

Orange II was tested in the presence of PAA. Surprisingly, the dye degradation catalyzed by the

same amount of O3MnV(OH)2-, MnVIO42-, MnVIIO4

- and MnII under identical experimental

conditions, showed no significant difference within the experimental error limits (see Figure 4.10).

This strongly suggests that the rate-limiting step of the catalytic dye degradation process does not

involve the underlying disproportionation chemistry of high valence Mn-oxo anions. Moreover,

the latter species must be rapidly reduced by the equilibrium amount of H2O2 present in PAA.


103

Figure 4.10 Comparison of the absorbance changes observed at 484 nm during the 1 · 10-5

M Mn catalyzed

degradation of 5 · 10-5

M Orange II by 5 · 10-3

M PAA with different Mn complexes as catalyst. Reaction conditions:

0.05 M NaHCO3 buffer, pH 9.5 and 25 °C.

The catalytic activity of colloidal MnIVO2 as starting material in the presence of PAA was found

to be significantly lower than that of MnVIIO4-, MnVIO4

2- and MnII (Figure 4.10). Hence, the surface

activation of the oxidant by coordination to colloidal MnIVO2 particles is unlikely to be responsible

for the outstanding dye degradation reactivity of the MnII/PAA system.

4.3.5 Reactivity of different in situ formed intermediates towards Orange II

The careful selection of reaction conditions enabled us to perform comparative measurements

on the catalytic performance of different in situ produced reactive intermediates. In the

experimental set-up the addition of the substrate Orange II to the reaction mixture of MnII and

PAA was carried out at different points of time during the reaction course under catalytic

conditions. A comparison of the observed initial rates for the degradation of Orange II yielded

some helpful information about the redox states through which the catalytic system cycled during

the course of the reaction. Figure 4.11 (A) presents the UV/Vis spectral changes observed for the

reaction of MnII with PAA in the absence of substrate, and the corresponding kinetic traces

recorded at 350 nm (colloidal MnIVO2) and 525 nm (MnVIIO4-) at pH 9.6 (inset in Figure 4.11 (A))

with the earlier mentioned biphasic behavior. If Orange II is added after different time intervals,

the initial oxidation rates change significantly. It is obvious that the catalytic mixture containing


104

different Mn-oxo intermediates formed during the reaction of MnII with PAA, shows different

reactivity (Figures 4.11 (B) and 4.12).

Figure 4.11 (A) Spectral changes recorded for the reaction of 1 · 10-5

M MnII with 0.01 M PAA. Inset: kinetic traces at

350 and 525 nm. (B) Identical reaction conditions as in (A) but with different delay times for substrate addition (5 · 10-5

M Orange II) followed at the maximum absorbance of Orange II at 484 nm. Reaction conditions: 0.05 M NaHCO3

buffer, pH 9.6 and 25 °C.

A plot of the initial rate of dye degradation vs. the delay time for the addition of Orange II is

shown in Figure 4.12 (■) along with the corresponding absorbance changes at 350 and 525 nm in

the inset of Figure 4.12. At first an increase in reactivity is observed up to the point shortly before

the end of the first phase. As a matter of fact, about 185 s after the start of the reaction (under the

selected experimental conditions), the initial degradation activity is three times higher than at the

beginning. As soon as colloidal MnIVO2 and MnVIIO4- accumulate in the reaction mixture, the

catalytic performance almost vanishes, although a large excess of oxidant is still present in

solution. Thus, it can be assumed that the most reactive intermediates are formed during the first

phase of the reaction, more precisely near to its end, which in turn coincides with the amount of

hydrogen peroxide present in the reaction mixture, as discussed above. If the experiment is

repeated in the presence of a small aliquot of additional H2O2 right from the beginning of the

reaction, the initial phase is extended, but the above described behavior remains to be identical

(Figure 4.12 ▲).

(A) (B)


105

Figure 4.12 Comparison of the initial degradation rates determined at 484 nm for the 1 · 10-5

M MnII catalyzed

reaction with 0.01 M PAA with different delay times for the addition of 5 · 10-5

M Orange II and different amounts of

H2O2. Inset shows the corresponding absorbance changes at 350 and 525 nm in the absence of substrate. Reaction

conditions: 0.05 M NaHCO3 buffer, pH 9.6 and 25 °C.

If H2O2 is added shortly before the beginning of the second phase (Figure 4.12 ●), the further

oxidation of the catalyst is again avoided and delayed such that the reactivity increases at the end

of the first phase (t ≈ 880 s) and decreases with the occurrence of permanganate and Mn IVO2. This

is a strong indication for the reductive influence of H2O2 during the first phase of the reaction.

In order to elucidate what causes the drop in reactivity as well as the insufficient substrate

bleaching in the second phase when MnVIIO4- and colloidal MnIVO2 are formed, it was attempted to

selectively reduce the latter species by addition of H2O2. In the presence of substrate, the sluggish

activity of the catalytic system at the stage where MnVIIO4- and colloidal MnIVO2 are formed could

be partially reactivated by the addition of a small aliquot of H2O2 (see Figure S4.19, Supporting

Information). In addition, selective reduction of most of the intermediates present in the second

phase (MnIVO2 or MnVIIO4-) is also achieved in the absence of substrate (Figure 4.13 (A)).

A comparison of the catalytic reactivity at different points of time of the reaction of MnII with

PAA revealed a drastic decrease as soon as MnVIIO4- and colloidal MnIVO2 are formed during the

second phase (Figures 4.13 (A) and (B), 1 and 2). However, if under identical reaction conditions

the Orange II degradation is performed after the partial selective reduction of these intermediates

with H2O2, the catalytic activity of the system is almost fully restored (Figure 4.13 (B), 3).


106

Figure 4.13 (A) Absorbance changes at 350 and 525 nm for the reaction of 1 · 10-5

M MnII with 0.01 M PAA, followed

by reduction with 2 · 10-3

M H2O2 after 250 s. Arrows mark the addition of 5 · 10-5

M Orange II for the corresponding

reactivity test in (B). (B) Absorbance changes at 484 nm for the different in situ formed intermediates in phase 1 at

150 s, phase 2 at 250 s and after reduction with 2 · 10-3

M H2O2 at 330 s. Reaction conditions: 0.05 M NaHCO3 buffer,

pH 9.60 and 25 °C.

In order to study the influence of H2O2 on the in situ formed intermediates in the second phase

of the reaction, we investigated the UV/Vis spectral changes that accompany the reduction. To the

reaction mixture containing 2 · 10-5 M MnII and 0.01 M PAA, 2 · 10-3 M H2O2 (pH 9.6, 25 °C) was

added in the second phase of the reaction after 50 s where the formation of colloidal MnIVO2 and

MnVIIO4- was complete. The UV/Vis spectra in Figure 4.14 show the rapid and complete reduction

of permanganate as well as the reduction of approximately 50 % of the colloidal MnIVO2 by H2O2

under the selected experimental conditions. The incomplete reduction of MnIVO2 by H2O2 is due to

the already discussed pH effects and causes a loss in catalytically active Mn and thus a slight

decrease in reactivity.

Nevertheless, the result is surprising since one would expect the highest reactivity at the point

of complete formation of MnVIIO4-, the strongest oxidant among the high valence Mn-oxo anions.

Furthermore, it is obvious from the corresponding reactivity test that the catalytic activity of the

system is restored after the reduction by H2O2. Consequently, the reactivity drop at this stage is

not caused by the depletion of the PAA oxidant, but by the over-oxidation of MnII to MnVIIO4- due

to the depletion of H2O2 as reducing agent present in stock solutions of PAA. The oxidation of the

catalyst to the MnVIIO4- state negatively affects the catalytic performance. This also implies that

(A) (B)


107

the in situ formed MnVIIO4- is not the most reactive species in the oxidative degradation of Orange

II, and that for efficient catalysis the oxidation state of manganese should not exceed that of MnIV.

Figure 4.14 UV/Vis spectrum recorded after 2 · 10-3

M H2O2 was added to the in situ formed intermediates during the

reaction of 2 · 10-5

M MnII with 0.01 M PAA before and after the addition. Reaction conditions: 0.05 M NaHCO3 buffer,

pH 9.5 and 25 °C. Inset shows the corresponding absorbance changes at 350 and 525 nm.

4.3.6 MnII catalyzed degradation of Orange II by PAA

pH dependence. The oxidative degradation of Orange II by PAA was studied in the pH range 7.0 to

12.6 (± 0.1) to investigate the catalytic activity of the in situ formed high-valence Mn intermediates

as a function of pH. Figure 4.15 reports the dependence of the initial rate of the degradation

reaction on the pH of the solution.

The results show a bell shaped profile with a maximum reactivity at a pH between 9 and 10. As

could be expected, very low reaction rates are observed at pH 7.5 where peracetate is present in

its protonated form and can only weakly interact with MnII. The first inflection at pH ≈ 8.5

coincides rather well with the pKa of PAA, viz. 8.2,[16] which suggests the necessity of PAA to form

CH3C(O)OO- to reach a high catalytic activity. In addition, it may be related to the faster reduction

of permanganate, rapidly formed in the presence of PAA, by H2O2 to form the actual catalytic

species, viz. MnIV=O, which shows an apparent pK value around 8.8 (see Figure S4.8 (B)). Upon

increasing the pH above pH 9.5, the decrease in catalytic activity can be ascribed to the formation


108

of catalytically inactive Mn intermediates such as Mn(OH)2, MnCO3 and insoluble MnIVO2.

Alternatively, it may be due to the disproportionation of manganate(VI) that shows a similar

decrease in reactivity on increasing the pH with an apparent pK value around 10.5.

Figure 4.15 pH dependence of the initial rate for the 1 · 10-5

M MnII catalyzed oxidative degradation of 5 · 10

-5 M

Orange II by 5 · 10-3

M PAA in 0.05 M NaHCO3 at 25 °C.

Variation of the [pre-catalyst] and [PAA]. To obtain more information on the underlying reaction

mechanism, the initial rate for the degradation of Orange II was studied under optimal conditions

(pH 9.5 and 25 °C) as a function of the [PAA] for two pre-catalyst concentrations. For this purpose

the initial [PAA] was varied in the range of 0.5 – 50 · 10-3 M. The observed rate profile (see Figure

4.16 (A)) shows typical saturation behavior and reaches a limiting rate at high oxidant

concentration. The saturation effect at higher [PAA] implies the formation of a pre-equilibrium

between the oxidant and the MnII precursor, prior to the rate-limiting step of the oxidative

degradation of Orange II. This is in line with a mechanism involving coordination of the

deprotonated peroxide to the MnII center and transformation of the thereby formed peroxo

complex into a catalytically relevant species.

At a constant [PAA], a good linear dependence of the initial rate on the initial pre-catalyst

concentration was observed (see Figure 4.16 (B)). The slope of the plot (with a zero intercept) was

found to be (1.47 ± 0.05) · 10-1 s-1 at 25 °C, which corresponds to a turnover frequency of 530 h-1

under the selected reaction conditions.


109

Figure 4.16 (A) Dependence of the initial rate of the Mn(II) catalyzed degradation reaction of Orange II on the [PAA] at

2 · 10-5

and 2 · 10-6

M MnII. (B) Dependence of the initial rate on the [Mn

II]. Reaction conditions: 0.05 M NaHCO3, pH

9.5, 5 · 10-5

M Orange II, 25 °C, (A) 2 · 10-5

M and 2 · 10-6

M MnII, (B) 5 · 10

-3 M PAA.

Influence of bicarbonate buffer. In our earlier work on the MnII catalyzed decomposition of

organic dyes by H2O2, the role of the bicarbonate buffer turned out to be of major importance.[12]

The second order dependence of the observed reaction rate on the [total bicarbonate] suggested

the involvement of HCO3- in two different ways. The first was the role of bicarbonate in the in situ

generation of peroxycarbonate, whereas the second was ascribed to the formation of highly

reactive manganese(II)-bicarbonate complexes under the selected experimental conditions (pH 8.5

- 9).[12a, b] However, when PAA is used as primary oxidant, this situation changes completely. Figure

S4.20 (Supporting Information) shows the dependence of the initial reaction rate on the total

carbonate concentration for PAA as oxidant. The oxidative degradation is obviously decelerated at

higher [total bicarbonate], which on one hand could be caused by enhanced formation of

insoluble MnCO3 at this pH. But on the other hand, if the results are compared to the reaction of

MnII with PAA in the absence of Orange II, an unequivocal tendency is observed. As described

above, higher [total bicarbonate] resulted in the formation of more MnVIIO4-. This suggests that the

more MnVIIO4- is formed, the less reactive the catalytic system is, which in turn emphasizes that

permanganate is not the actual catalytically oxidizing species.

Influence of [H2O2]. Figure 4.17 illustrates the change in the initial bleaching rate upon

addition of H2O2 to the catalytic reaction mixture.

(B) (A)


110

Figure 4.17 Dependence of the initial rate of the MnII catalyzed degradation reaction of Orange II by PAA on the added

[H2O2]. Inset: Corresponding kinetic traces recorded at 484 nm for the reaction of 1 · 10-5

M MnII with 0.01 M PAA and

5 · 10-5

M Orange II at different concentrations of added H2O2. Reaction conditions: 1 · 10-5

M MnII, 0.01 M PAA, 5 · 10

-5

M Orange II, 0.05 M NaHCO3, pH 9.5 and 25 °C.

The results show that higher [H2O2] inhibit the bleach reaction. However, it was shown that

small quantities of H2O2 reactivated the catalytic system by acting as reducing agent for MnVIIO4-

and colloidal MnIVO2 formed during the second, less reactive phase of the reaction. Thus, on the

one hand the catalytic degradation reaction is hindered by excess of additional H2O2, but on the

other hand it benefits from the presence of a low [H2O2]. This seemingly inconsistent behavior can

be accounted for by the EPR spectroscopic experiments for the reaction of Mn II with PAA in the

presence of an excess H2O2 (see Figures 4.8 (A) and (B)). It was evidenced that an excess of H2O2

delayed the in situ formation of the MnIV intermediate, while the MnII precursor persists longer in

solution most likely in a catalase-like reaction of the MnIV species to MnII. However, a small

amount of H2O2 avoids the visible formation of higher-valence species such as permanganate.

Hence, the ambivalent role of H2O2 can be interpreted in terms of stabilizing a low steady-state

concentration of a reactive MnIV=O species, whereas an excess of H2O2 pushes the complex

oxidation by PAA/reduction by H2O2 equilibrium back to the side of the MnII starting compound.

Catalytic cycles. In control experiments the stability of the in situ generated catalyst was

studied by repeated addition of substrate to a solution of MnII with an excess of PAA (see Figure

4.18). The catalytic cycle could be repeated about five times by sole addition of new portions of

dye substrate without any significant loss of activity. Yet, at the stage of the sudden formation of


111

colloidal MnIVO2 and MnVIIO4-, the catalytic degradation performance decreased drastically,

although a large excess of PAA oxidant was still present in solution. Addition of small aliquots of

H2O2 as reducing agent restored the catalytic activity. In this manner, several cycles could be

completed up to the depletion of PAA.

Figure 4.18 Absorbance changes recorded at 484 nm during the repeated addition of 5 · 10-5

M Orange II to a 1 · 10-4

M MnII and 0.025 M PAA containing solution. Asterisks mark the reactivation by addition of 5 · 10

-3 M H2O2. Reaction

conditions: 0.05 M NaHCO3, pH 9.5 and 25 °C.

As a first step to the possible use for larger substrate quantities, we performed a preliminary

experiment with a higher substrate to catalyst ratio, as required for putative applications. Within

only 100 s complete oxidative degradation of the 50 fold excess of Orange II (5 · 10-4 M) was

achieved with only 10 µM MnII catalyst, which implies an even higher TOF of about 1800 h-1. The

experimental results provide clear evidence for a highly efficient catalytic turnover during the

reactive first phase of the MnII/PAA system and demonstrate its possible application for larger

substrate quantities (see Figure S4.21, Supporting Information).

In order to gain further mechanistic insight, the reaction intermediates and final products formed

during the reaction process were examined by HPLC and mass spectrometric methods.HPLC

product analysis was performed immediately after decoloration of the reaction solution

containing MnII, PAA and Orange II (typical experimental conditions as in Figure 4.2 (A), for which

the reaction is over in 100 s). The selected HPLC protocol was very similar to that used for the

analysis of the degradation of Orange II by H2O2 in the presence of [RuIII(edta)H2O]- as catalyst.[48]


112

A typical chromatogram is shown in Figure S4.22 (Supporting Information). The broad peak at 3.00

- 3.82 min can be assigned to an overlap of short-chain organic acids including phthalic acid, oxalic

acid and glycolic acid that have retention times of 3.13, 3.45 and 3.79 min, respectively, under the

selected experimental conditions. Although we were not able to resolve the broad signal at 3.82

min any further by variation of the chromatographic parameters, these results are in agreement

with that found in earlier studies.[48, 6d]

We further performed mass spectrometric analysis to compare the degradation products of

Orange II with that of other well studied catalytic systems. The corresponding spectral changes

during this reaction at 484 nm are shown in Figure S4.23 (Supporting Information). Oxalic acid and

glycolic acid could not be detected by this method since they are known to decompose at higher

temperatures. However, the presence of phthalic acid and 4-hydroxybenzosulfonate could be

demonstrated (see Figure S4.24, Supporting Information), which is again in good agreement with

literature reported degradation products.[6d]

4.3.7 Mechanistic interpretation

Although from a mechanistic point of view, the actual catalytically active intermediates are

quite difficult to pin down, some valid conclusions can be drawn on the basis of the available

experimental results. The activation of PAA certainly involves a labile manganese aqua-hydroxo

species and the deprotonated form of the peroxide, H3CCOOO-, as can be concluded from the pH

sensitive rate profile for the overall dye degradation reaction.

The reaction of these components results in the transient formation of an acetylperoxo-MnII

intermediate. Such peroxo intermediates are often proposed to precede the formation of high-

valence manganese-oxo species invoked as active oxidants for synthetic manganese catalysts.

Recent developments suggest that more attention should be paid to the role of these metal-

peroxo species in general, since they may be more important than considered before.[49] However,

in the case of the MnII catalyzed oxidative degradation of Orange II, the involvement of catalytic

active MnII-peroxo species is excluded, since the EPR experiments show only negligible presence of

MnII species during the reactive phase. In general, two possible reaction pathways are conceivable

for a MnII-peroxo intermediate. A heterolytic cleavage of the peroxo bond will lead to a MnIV-oxo

species, whereas a homolytic bond cleavage will yield a MnIII intermediate and an organic radical.

Since the formation of both MnIII and MnIV species could be evidenced in the first phase of the


113

reaction by UV/Vis and EPR spectroscopy, it has to be considered that both reactions may occur.

Nevertheless, the interference of free radical processes in the degradation of the dye was

excluded by the use of tBuOH and BHT as radical scavengers, which had no negative effect on the

degradation reaction. The presence of MnIII at this stage could also be due to the manifold dis- and

synproportionation reactions of manganese ions in the absence of a stabilizing ligand. Control

experiments showed no degradation activity of MnIII itself towards the studied dye substrate (see

Figure S4.25, Supporting Information). It is known that the reduction of colloidal MnIVO2 in non-

acidic media results only in partial reduction on the colloidal surface, which is a monomolecular

MnII-oxide layer on the surface of a mixed MnIVO2-MnIIO colloid.[28] Finally, this mixed colloid may

form to some extent MnIII-oxide, which causes the formation of an absorbance shoulder at 470

nm. Taking all this into account, it is suggested that the first oxidation step mainly proceeds via O-

O bond heterolysis to give a high-valence MnIV=O species, which was evidenced by EPR

spectroscopy. Furthermore, a slight increase in the UV/Vis absorbance, as well as in the MnIV EPR

signal, indicate a small increase in the concentration of this MnIV intermediate during the first

reaction phase.

The presence of an excess of PAA would allow further oxidation of the intermediate MnIV

species and thereby enable the formation of higher-valence species like MnV intermediates,

MnVIO42- and MnVIIO4

-. The manganese-oxo anions of MnV and MnVI are highly unstable at pH

values below 14,[23] so that they rapidly disproportionate to yield permanganate and colloidal

MnIVO2 as final products as shown by UV/Vis spectroscopy.

Despite the presence of an excess of PAA, the degradation rate slows down drastically as soon

as the visible formation of permanganate occurs, which points to the requirement of lower-

valence manganese species rather than permanganate for efficient catalysis. Consequently, the

catalytically active intermediate must be formed in the initial reactive phase. During this reaction

stage only MnIV and a small amount of MnIII and MnII were detected. Moreover, the increasing

reactivity towards the dye substrate along with the concomitant increase in the [MnIV

intermediate], strongly suggest that the latter is the actual catalytically active intermediate formed

during the reaction of MnII with PAA. This is in agreement with earlier studies on the peroxide

activation ability of MnII ions, where the observed reactivity was also attributed to a transiently

formed MnIV=O species.[12a, b]


114

Although MnIV=O species are commonly postulated as reactive intermediates in numerous

oxygenation reactions, they tend to yield MnIVO2 precipitates in the absence of a stabilizing ligand.

Since the first initial phase with seemingly constant intermediate distribution is very unlikely to be

the result of a stable MnIV=O compound in the absence of a stabilizing ligand, there has to be an

ongoing back reaction that avoids the formation of MnIVO2 as well as the accumulation of MnVIIO4-.

It was evidenced that this back reaction is caused by the equilibrium concentration of H2O2

omnipresent in PAA solutions. H2O2 acts as reducing species for the higher-valence manganese

intermediates such as colloidal MnIVO2 and MnVIIO4- under the experimental conditions of this

study. Thereby, a small [H2O2] keeps the complex oxidation/reduction/disproportionation

equilibria of manganese on the side of the lower-valence MnIII and MnIV species during the initial

phase of the reaction of MnII with PAA. This is supported by the fact that in the UV/Vis spectra

recorded during the initial reaction stage no indication for the formation of O3MnV(OH)2-, MnVIO42-

or MnVIIO4- was found. If the system is selectively reduced by H2O2 at the stage of the second

phase, its catalytic performance is almost completely restored, which strongly supports the

observation that the Mn species produced in the initial phase are crucial for the catalytic activity.

While small [H2O2] are required to maintain the catalytic activity of the system, larger

concentrations (5 - 40 · 10-3 M) of additional H2O2 negatively affect the catalytic performance. This

observation was further clarified by detailed EPR measurements on the reaction of MnII with PAA

in the presence of an excess of additional H2O2. This resulted in a delayed formation of MnIV

species during which the MnII precursor persisted much longer in solution. Furthermore, gas

evolution was observed in the absence of Orange II and with excess H2O2. It is suggested that this

is due to a catalase-like reaction between the in situ formed MnIV species and hydrogen peroxide.

This suggestion also provides an explanation for the increasing reactivity towards the dye

substrate during the initial phase of the reaction. With progressive H2O2 consumption, the back

reaction to MnII becomes less important and the concentration of the reactive MnIV=O slightly

increases. However, as soon as H2O2 has been depleted, MnVIIO4- and colloidal MnIVO2 accumulate

and the catalytic reactivity is lost. Hence, the role of H2O2 in the MnII/PAA system is an ambivalent

one. On the one hand, at low [H2O2] it benefits the catalytic dye degradation by preventing two

contra productive processes. The catalytic life-time is extended by avoiding the rapid over-

oxidation of the Mn catalyst to MnVIIO4- and by preventing catalytic deactivation when MnIVO2 is

formed. On the other hand, higher [H2O2] inhibit the catalytic dye degradation. It pushes the


115

complex oxidation by PAA/reduction by H2O2 equilibrium back to the side of the MnII starting

compound, which thereby represents a parallel reaction to the desired dye degradation. A

simplified mechanistic scheme to account for the described observations is presented in Scheme

4.3.

Scheme 4.3 Simplified reaction pathways leading to the formation of reactive MnIV

=O species and permanganate in

the MnII catalyzed oxidative degradation of Orange II by PAA (H3CC(O)OO

-) including the role of the equilibrium

content of H2O2 as reductive species (S = Orange II).

Although reactivity tests with synthetic samples of hypomanganate(V), manganate(VI) and

permanganate(VII) under comparable experimental conditions revealed that permanganate is the

most reactive species for stoichiometric oxidation, they are not relevant to the catalytic process

that occurs during the reaction of the MnII catalyst with PAA. Consequently, it has to be concluded

that the remarkably high reactivity of the MnII/PAA system towards the dye substrate Orange II at

pH 9.5 is due to a well balanced sequence of oxidation (by PAA) and reduction (by H2O2) reactions

to maintain an ideal steady state concentration of an highly reactive MnIV-oxo intermediate for

efficient catalysis of dye degradation. This sequence of events is outlined schematically in Scheme

4.3.

It is known that ortho-substituted azo dyes may act as ligands for metal centers and their

chelating abilities strongly depend on the nature and number of substituents especially adjacent


116

to the azo linkage.[12a, 50] Earlier studies reported that a specific complexation mode of ortho-

dihydroxy substituted azo dyes to MnII was required for efficient oxidation catalysis.[50b, d]

However, mono-ortho or unsubstituted dyes showed no reaction which was attributed to their

lower metal binding constants.[50b, d] Consequently, two further azo dyes were selected to clarify to

what extent dye coordination is an essential prerequisite for dye oxidation. Methyl Orange

contains no ortho substituent and thus represents a monodentate ligand, and Calmagite is an o,o’-

dihydroxy azo dye with high binding affinity. Under identical reaction conditions, a very definite

tendency was observed for these dyes (see Figure S26, Supporting Information). Whereas for

Methyl Orange almost no oxidation was observed, the introduction of one hydroxo group adjacent

to the azo bridge in Orange II already resulted in an enhanced performance. Furthermore, a

markedly faster decrease in the characteristic azo absorbance occurred in the case of Calmagite.

These findings are in line with earlier results and strongly suggest that coordination of the dye to

the metal center occurs during the reaction course and is indispensable for efficient catalysis.[12,

50b, d]

Whether the dye substrate coordinates to the MnII precursor or a high-valence form such as

the reactive MnIV=O intermediate, remains to be resolved. If these effects require dye

coordination to manganese in its pre-catalyst MnII form, no reaction or a slower reaction should

occur when the substrate is added to the MnII/PAA containing reaction mixture with a delay of a

several seconds, where our experiments show that MnII is no longer present in the reaction

mixture (see EPR experiments). In fact, the reaction rate for the degradation of Orange II increases

as more MnII is converted to MnIV=O (see Figure 4.12), suggesting that coordination occurs to the

MnIV intermediate. In general, the tendency of the electron rich dye molecule to bind to a metal

centre is expected to increase with increasing electrophilicity of the metal center. Thus it can be

argued that the coordinating influence of the different ortho-substituted dyes can be ascribed to

complex-formation with a high-valence MnIV intermediate rather than with the MnII precursor

form of the catalyst.

4.3.8 Comparison MnII/PAA vs. MnII/HCO4- system

In earlier investigations on the MnII catalyzed oxidative degradation of organic dyes by H2O2, it

was found that a crucial aspect of the catalytic system was the in situ formation of

peroxycarbonate as actual oxidant.[12] Peroxycarbonate formation is known to proceed in a rapid


117

equilibration process, however, with an unfavorable formation constant, viz. K = 0.32 ± 0.02 M-1

for solvent mixtures[14a] and K = 0.33 ± 0.02 M-1 for pure aqueous solution.14b Therefore, high

[H2O2] and [total carbonate] are required to obtain an adequate equilibrium concentration of

HCO4- in the catalytic dye degradation with MnII/H2O2/HCO3

- at pH 8.5. Furthermore, it was shown

that the crucial reaction step involves a two electron oxidation of the MnII precursor to a MnIV-oxo

intermediate, but the oxidation state of the metal did not exceed MnIV, and MnVIIO4- formation

never occurred.[12] EPR studies on the MnII/HCO4- system disclosed the presence of a very low

concentration of MnIV, while the main spectroscopic features pointed to the dominance of MnII

species in the catalytic reaction mixture (see Figure 4.19).

Figure 4.19 Comparison of the EPR spectra for MnII/H2O2/HCO3

-12b vs. Mn

II/PAA. Inset: Amplification of the g = 4 signal

for MnII/H2O2/HCO3

-. Conditions for Mn

II/H2O2/HCO3

-: 1 · 10

-4 M Mn

II, 0.02 M H2O2 in 0.4 M HCO3

-, directly after mixing

at pH 8.5 and 25 °C. EPR spectra: 7 K, 9.4 GHz, 2 mW microwave power. Conditions for MnII/PAA: 1 · 10

-4 M Mn

II, 2.5 ·

10-2

M PAA and 5 · 10-3

M H2O2 at pH 9.8 and 25 °C. EPR spectra: 8.96 GHz, 10 K, 1 mW microwave power.

On the other hand, the use of the readily accessible peroxide PAA enables a significantly faster

catalytic degradation at lower [catalyst], [oxidant] and [buffer], during which the formation of

higher-valence Mn-oxo species such as permanganate is observed. Moreover, the MnII/PAA

system suffers less from the undesired catalase-like parallel reaction due to the lower [H2O2]

present. Yet, it was shown that H2O2 is indispensable as reducing agent for efficient catalysis in the

PAA/MnII system, which in turn emphasizes the decisive role of the MnIV species in the catalytic

degradation of Orange II by PAA.


118

A comparison of the EPR results obtained for both systems, i.e. MnII/H2O2/HCO3- [12b] vs.

MnII/PAA, points to a much higher concentration of the transiently formed MnIV species when PAA

is used. For MnII/H2O2/HCO3- the excess H2O2 used, which is required for the more efficient

formation of HCO4-, enhances the reduction of the in situ formed MnIV=O to MnII, such that only a

very low [MnIV] is constantly present in the catalytic solution. The MnII/PAA system benefits from

the readily accessible peroxoacetic acid, so that more MnIV=O is formed and the lower equilibrium

concentration of the reducing H2O2 keeps it at a higher steady-state concentration as compared to

the MnII/H2O2/HCO3- system. Thus, it is suggested that both reaction systems, i.e. MnII/H2O2/HCO3

-

and MnII/PAA, involve the same catalytically active MnIV=O intermediate, and that through the

described effects of H2O2 the different steady state concentrations of MnIV=O account for the

observed difference in catalytic activity.

4.4 CONCLUSIONS

In conclusion, we were able to gain more insight into the complex chemical and mechanistic

behavior, and the reactivity of the different in situ formed species that occur during the reaction of

simple MnII salts with PAA in weakly basic solution. The kinetics of the catalytic oxidative dye

degradation was investigated in detail for Orange II as substrate. Our kinetic studies lead to the

suggestion that a highly reactive mixture is formed upon addition of PAA to a weakly basic solution

of MnII. The reaction shows a biphasic behaviour which is very sensitive to the selected conditions

and concentrations of the complex catalytic system. By careful selection of the reaction

conditions, we were able to closely investigate the different reaction steps by means of UV/Vis

and EPR spectroscopy, and thereby disclose the different intermediates and their relevance to the

catalytic dye degradation process. Selective reactivity studies of the different in situ formed

intermediates, as well as readily prepared high-valence manganese species, provided further

information on the possible reactive species. Moreover, the omnipresent equilibrium content of

H2O2 in PAA solutions was shown to play a crucial role as reductive species in the catalytic cycle,

presumably by avoiding precipitation of inactive MnIVO2 and over-oxidation of the catalyst,

thereby extending the lifetime of the catalytic system.

On the basis of the experimental results in the presence and absence of substrate, a simplified

reaction scheme to account for the observed behaviour is presented. The key feature of the Mn II


119

catalyzed degradation of Orange II by PAA is suggested to involve a complex oxidation by

PAA/reduction by H2O2 reaction sequence in which a steady state equilibrium of the reactive

MnIV=O intermediate sets in. A comparison of the results of the presented detailed kinetic

investigations led to the conclusion that in both systems the same catalytically active intermediate

accounts for the oxidative degradation of the substrate Orange II. The much higher degradation

rate found in the presence of PAA can be ascribed to the more efficient formation of Mn IV=O than

in the presence of HCO4-, as supported by the EPR measurements.



naphthylazo)benzenesulfonate], 99 % was supplied by Sigma-Aldrich and recrystallized from an

EtOH/H2O mixture at 4 °C.[51 ] Peracetic acid 39 wt. %, H2O2 30 %, as well as CH3-13COOH were of

analytical grade and provided by Sigma-Aldrich. Analytical grade tBuOH, BHT (2,6-di-tert-butyl-4-

methylphenol), Calmagite (1-(1-hydroxy-4-methyl-2-phenylazo)-2-naphthol-4-sulfonic acid), and

Methyl Orange (4-[4-(dimethylamino)phenylazo]benzenesulfonic acid, sodium salt) were used. All

other chemicals were of the highest purity commercially available and used without any further

purification. Carbonate buffer solutions were prepared using Millipore Milli-Q purified water.

Stock solutions of manganate(VI)[52] and colloidal MnIVO2[25] were prepared according to the

literature and monitored by UV/Vis (MnVIO42-: ε610 nm = 1500 M-1cm-1 [23]) and EPR spectroscopy,

respectively. Hypomanganate(V) solutions were prepared by careful reduction of manganate(VI)

with Na2SO3 as reported earlier.[23] The concentration of the hypomanganate(V) solutions was

estimated by the molar extinction coefficient as published before.[45]

GENERAL PROCEDURE AND PH JUMP TECHNIQUE. The different manganese salts (MnIICl2 · 2H2O, MnIISO4 ·

H2O, MnII(NO3)2 · 4H2O, MnII(ClO4)2 · 4H2O and MnII(O2CCH3)2 · 4H2O) were freshly dissolved in

water before use. To a freshly prepared 0.05 M sodium bicarbonate solution, an adequate amount

of NaOH was added to adjust the pH in a way that the subsequent addition of a specific [PAA] gave

the desired pH. The reaction was started under isothermal conditions by addition of small aliquots

of a concentrated manganese stock solution together with Orange II to the PAA containing buffer

solution. The catalytic reaction was followed by in situ UV/Vis spectroscopy.


120

INSTRUMENTATION AND EQUIPMENT. All kinetic data were obtained by recording time-resolved UV/Vis

spectra using a Hellma 661.502 – QX quartz Suprasil immersion probe attached via optical cables

to a 150 W Xe lamp and a multi-wavelength J & M detector, which records complete absorption

spectra at constant time intervals. Kinetic measurements were carried out under pseudo-first-

order conditions. The pH of the PAA containing aqueous carbonate solution was carefully

measured and adjusted using a Mettler Delta 350 pH meter previously calibrated with standard

buffer solutions at two different pH values (4 and 10). The kinetics of the Orange II degradation

reaction was monitored at 484 nm. First-order rate constants, where possible, were calculated

using Specfit/32 and Origin (version 7.5) software.

13C-NMR measurements were performed by using DMSO-d6 as internal standard in a glass

capillary. The [PAA] and [AcOH] were calculated from the relative peak intensities using the Lorenz

fit obtained by the NMRICMA program (developed at the Institut de chimie minérale et analytique,

University of Lausanne, in the group of Prof. A. E. Merbach) from the NMR data. 13C-NMR spectra

were recorded at a frequency of 100 MHz on a Bruker Advance DRX 400WB spectrometer

equipped with a superconducting BC-94/89 magnet system.

Perpendicular mode EPR spectra were recorded on an X-band Joel Jes Fa 200 spectrometer

equipped with a cylindrical mode cavity and a liquid helium cryostat. Samples were taken from the

investigated solutions and immediately frozen to quench the reaction. The EPR measurements

were performed in quartz tubes at 10 K (9.45 GHz, 1 mW microwave power). Data analyses were

done with the Jes-Fa Series software package.

HPLC analysis of the oxidation products of Orange II was performed using a Waters (M 515 &

PDA) HPLC equipped with a photodiode array detector and a Symmetry C18 (5 m, 100 A) column.

Resultant solutions collected after 4 min and 60 min of the reaction were subjected to HPLC

analysis using a mobile phase of HPLC grade water-acetonitrile mixture (70:30 v/v) at a flow rate of

0.5 mL/min. HPLC parameters were quantified and optimized with authentic samples of naphthalic

acid, glycolic acid and oxalic acid prior to the analysis.

Mass spectrometric analysis of the oxidation products of ORII were performed on an UHR-TOF

Bruker Daltonik (Bremen, Germany) maXis, an ESI-TOF mass spectrometer capable of a resolution

of at least 40 000 fwhm used by the group of Prof. Ivana Ivanović-Burmazović at the University of

Erlangen-Nürnberg. The mass spectrometric detection was carried out in the negative-ion mode


121

with a source voltage of 5500 V, a flow rate of 500 μL/h, and the drying gas (N2) was kept at 180

°C. The instrument was calibrated prior to every experiment via direct infusion of the Agilent ESI-

TOF low concentration tuning mixture. The sample was taken after completion of the reaction

performed under the standard conditions (1 · 10-5 M MnII, 0.01 M PAA, 0.05 M NaHCO3 buffer, pH

9.5). The HCO3-/CO3

2- buffer was removed by acidification with HCl and the pH was readjusted

with NaOH prior to the analysis.


[1] (a) Park, H.; Choi, W. J. Photochem. Photobiol. A 2003, 159, 241-247; (b) Brown, D.,

Laboureur, P. Chemosphere 1987, 12, 397-404; (b) Goszczynski, S.; Paszczynki, A.; Pasti-

Grigsby, M. B.; Crawford, R. L.; Crawford, D. L. J. Bacteriol. 1994, 176, 1339-1347; (c)

Chivukula, M.; Spadaro, J. T.; Renganathan, V. Biochemistry 1995, 34, 7765-7772; (d) Spadaro,

J. T.; Renganathan, V.; Arch. Biochem. Biophys. 1994, 312, 301-307; (e) Spadaro, J. T.; Isabelle,

L.; Renganathan, V. Environ Sci & Tech. 1994, 28, 1389-1393.

[2] (a) Zollinger, H. Color Chemistry, 3rd ed., Wiley-VCH: Weinheim, Germany, 2003; (b) Beach, E.

S.; Malecky, R. T.; Gil, R. R.; Horwitz, C. P.; Collins; T. J. Catal. Sci. Technol. 2011, 1, 437-443.

[3] Baughman, G. L.; Weber, E. J.; Environ. Sci. Technol. 1994, 28, 267-276.

[4] Hastie, J.; Bejan, D.; Teutli-Leon, M.; Bunce, N. J. Ind. Eng. Chem. Res. 2006, 45, 4898-4909.

[5] (a) Ellis, W. C.; Tran, C. T.; Roy, R.; Rusten, M.; Fischer, A.; Ryabov, A. D.; Blumberg, B.; Collins,

T. J. J. Am. Chem. Soc. 2010, 132, 9774-9781. (b) Kluson, P.; Drobek, M.; Krejcikova, S.; Krysa,

J.; Kalaji, A.; Cajthaml, T.; Rakusan, J. Appl. Catal. B 2008, 80, 321-326. (c) Wieprecht, T.; Xia,

J.; Heinz, U.; Dannacher, J.; Schlingloff, G. J. Mol. Catal. A: Chem. 2003, 203, 113-128. (d)

Ramirez, J. H.; Maldonado-Hódar, F. J.; Pérez-Cadenas, A. F.; Moreno-Castilla, C.; Costa, C. A.;

Madeira, L. M. Appl. Catal. B 2007, 75, 312-323. (e) Hage, R.; Iburg, J. E.; Kerschner, J.; Koek, J.

H.; M. Lempers, E. L.; Martens, R. J.; Racherla, U. S.; Russell, S. W.; Swarthoff, T.; van Vliet, M.

R. P.; Warnaar, J. B.; van der Wolf, L.; Krijnen, B. Nature 1994, 369, 637-639.

[6] (a) Oakes, J.; Gratton, P. J. Chem. Soc., Perkin Trans. 2 1998, 1857-1864. (b) Nadtochenko, V.;

Kiwi, J. J. Chem. Soc., Faraday Trans. 1997, 93, 2373-2378. (c) Oakes, J.; Gratton, P; Weil, I. J.

Chem. Soc., Dalton Trans. 1997, 3805-3809. (d) Chahbane, N.; Popescu, D.-L.; Mitchell, D. A.;

Chanda, A.; Lenoir, D.; Ryabov, A. D.; Schramm, K.-W.; Collins, T. J. Green Chem. 2007, 9, 49-


122

57. (e) Gilbert, B. C.; Lindsay-Smith, J. R.; Newton, M. S.; Oakes, J.; Ponts i Parts, R. Org.

Biomol. Chem., 2003, 1, 1568-1577.

[7] (a) Song, Y. J.; Lee, S. H.; Park, H. M.; Kim, S. H.; Goo, H. G.; Eom, G. H.; Lee, J. H.; Lah, M. S.;

Kim, Y.; Kim, S.-J.; Lee, J. E.; Lee, H.-I.; Kim, C. Chem. Eur. J. 2011, 17, 7336-7344. (b) Song, F.

Wang, C.; Falkowski, J. M.; Ma, L.; Lin, W. J. Am. Chem. Soc. 2010, 132, 15390-15398. (c)

Palucki, M.; Finney, N. S.; Pospisil, P. J.; Guler, M. L.; Ishida, T.; Jacobsen, E. N. J. Am. Chem.

Soc., 1998, 120, 948-954.

[8] (a) Arunkumar, C.; Lee, Y.-M.; Lee, J. Y.; Fukuzumi, S.; Nam, W. Chem. Eur. J. 2009, 15, 11482-

11489. (b) Crestoni, M. E.; Fornarini, S.; Lanucara, F. Chem. Eur. J. 2009, 15, 7863-7866. (c)

Song, W. J.; Seo, M. S.; de Beer George, S.; Ohta, T.; Song, R.; Kang, M.-J.; Tosha, T.; Kitagawa,

T.; Solomon, E. I.; Nam, W. J. Am. Chem. Soc. 2007, 129, 1268-1277. (d) Rose, E.; Andrioletti,

B.; Zrig, S.; Quelquejeu-Etheve, M. Chem. Soc. Rev., 2005, 34, 573-583.

[9] (a) Wieghardt, K.; Schmidt, W.; Nuber, B.; Weiss, J. Chem. Ber. 1979, 112, 2220-2230. (b)

Romakh, V. B.; Therrien, B.; Süss-Fink, G.; Shul’pin, G. B. Inorg. Chem. 2007, 46, 1315-1331; (c)

Mandelli, D.; Kozlov, Y. N.; Golfeto, C. C.; Shul'pin, G. B. Catal. Lett. 2007, 118, 22-29; (d)

Sibbons, K. F.; Shastri, K.; Watkinson, M. Dalton Trans. 2006, 645-661; (e) Gilbert, B. C.;

Lindsay Smith, J. R.; Mairata y Payeras, A.; Oakes, J. Org. Biomol. Chem. 2004, 2, 1176-1180;

(f) Bennur, T. H.; Srinivas, D.; Sivasanker, S.; Puranik, V. G. J. Mol. Cat. A 2004, 219, 209-216;.

(g) Hage, R.; Kerschner, J. Trends in Inorg. Chem. 1998, 5, 145-159.

[10] De Boer, J. W.; Browne, W. R.; Brinksma, J.; Alsters, P. L.; Hage, R.; Feringa, B. L. Inorg. Chem.

2007, 46, 6353-6372.

[11] (a) Shilov, A. E.; Shul’pin, G. B. Chem. Rev. 1997, 97, 2879-2932; (b) Sheldon, R. A. J. Chem.

Technol. Biotechnol. 1997, 68, 381-388.


E.; Gazzaz, H. A.; Rothbart, S.; Puchta, R.; van Eldik, R. Appl. Catal. B. 2009, 95, 179-191; (c)

Rothbart, S.; Ember, E.; van Eldik, R. Dalton Trans. 2010, 39, 3264-3272.

[13] Baes, C. F.; Mesmer, R. E. Am. J. Sci. 1981, 281, 935-962.

[14] (a) Richardson, D. E.; Yao, H.; Frank, K. M.; Bennett, D. A. J. Am. Chem. Soc. 2000, 122, 1729-

1739. (b) Bakhmutova-Albert, E. V.; Yao, H.; Denevan, D. E.; Richardson, D. E. Inorg. Chem.

2010, 49, 11287-11296.


123

[15] Silbert, L. S.; Organic Peroxides (Ed:D . Swern), Wiley-Interscience, New York 1970.

[16] (a) Evans, D. F.; Upton, M. W. J. Chem. Soc., Dalton Trans. 1985, 1151-1158 (b) Koubek, E.;

Haggett, M. L.; Battaglia, C. J.; Ibne-Rasa, K. M.; Pyun, H. Y.; Edwards, J. O. J. Am. Chem. Soc.

1963, 85, 2263-2268 (c) Yuan, Z. ; Ni, Y.; Heiningen, A. R. P. Can. J. Chem. Eng. 1997, 75, 37-

41.

[17] (a) Ball, R. E.; Edwards, J. O. J. Am.Chem. Soc. 1956, 78, 1125-1129; (b) Awad, M. I.; Harnoode,

C.; Tokuda, K.; Ohsaka, T. Anal. Chem. 2001, 73, 1839-1843.

[18] (a) Shapilov, O. D. Doctoral (Pharm.) Dissertation, Moscow, 1978; (b) Doguzova, I. A. Cand.

Sci. (Chem.) Dissertation, Moscow, 2000. (c) Dul’neva, L. V.; Moskvin, A. V. Russian Journal of

General Chemistry, 2005, 75, No. 7, 1125-1130.

[19] Feng, W.; Nansheng, D.; Helin, H. Chemosphere 2000, 41, 1233-1238.

[20] (a) Jee; J.-E.; Pestovsky, O.; Bakac; A. Datlton Trans. 2010, 39, 11636-11642. (b) Wells, C. F.,

Davies, G. J. Chem. Soc. A. 1967, 1858-1861.

[21] (a) Wei, Z.; Cady, C. W.; Brudvig, G. W.; Hou, H. J. M. J. Photochem. Photobiol. B 2011, 104,

118-125; (b) Chen, H.; Tagore, R.; Das, S.; Incarvito, C.; Faller, J. W.; Crabtree, R. H.; Brudvig,

G. W. Inorg. Chem. 2005, 44, 7661-7670; (c) Gamelin, D. R.; Kirk, M. L.; Stemmler, T. L.; Pal, S.;

Armstrong, W. H.; Penner-Hahn, J. E.; Solomon, E. I. J. Am. Chem. Soc. 1994, 116, 2392-2399;

(d) Belal, A. A.; Chaudhuri, P.; Fallis, I.; Farrugia, L. J.; Hartung, R.; MacDonald,N. M.; Nuber,

B.; Peacock, R. D; Weiss, J.; Wieghardt, K. Inorg. Chem. 1991, 30, 4397-4402.

[22] Lin, Z.-X.; Huang, T. T. Ind. Eng. Chem. Res. 1987, 26, 2148-2151.

[23] Rush, J. D.; Bielski, B. H. J. Inorg. Chem. 1995, 34, 5832-5838.

[24] Leavason, W.; McAuliffe, C. A. Coord. Chem. Rev. 1972, 2, 353-384.

[25] (a) Perez-Benito, J. F. J. Phys. Chem. C 2009, 113, 15982-15991. (b) Abel, E. Monatheft für

Chemie 1949, 80, 455-462.

[26] (a) Bourgougnon, A. J. Am. Chem. Soc. 1889, 11, 94-98; (b) Doona, C. J.; Schneider, F. W. J.

Am. Chem. Soc. 1993, 115, 9683-9686; (c) Li, Y.; Zhao, Y.; Zhu, Z.; Anal. Sci., 2011, 27, 37-41;

(d) Ladbury, J. W.; Cullis, C. F. Chem. Rev. 1958, 58, 403-438.

[27] Popov, E.; Eloranta, J.; Hietapelto, V.; Vuorenpalo, V.-M.; Aksela, R.; Jäkärä, J. Holzforschung

2005, 59, 507-513.


124

[28] Perez-Benito, J. F.; Arias, C.; Amat, E. A. J. Collo. Interf. Scienc. 1996, 177, 288-297.

[29] Perez-Benito, J. F. J. Collo. Interf. Scienc. 2002, 248, 130-135.

[30] Petlicki, J.; Palusova, D.; van de Ven, T. G. M. Ind. Eng. Chem. Res. 2005, 44, 2002-2010.

[31] Perez-Benito, J. F.; Arias, J. Collo. Interf. Scienc. 1992, 152, 70-84.

[32] Baral, S.; Lume-Pereira, C.; Janata, E.; Henglein, A. J Phys. Chem. 1985, 89, 5779-5783.

[33] Kanungo, S.B.; Parida, K. M.; Sant, B. R. Electrochim. Acta 1981, 26, 1157-1167.

[34] Evan, M. G.; Uri, N Trans. Faraday Soc. 1949, 45, 224-236.

[35] Simoyi, R. H.; Keeper, P. D.; Epstein, I. R.; Kustin, K. Inorg. Chem. 1986, 25, 583-542.

[36] (a) Casado, J.; Lizaso-Lamsfus, J. An. R. Soc. Es. Fis. Quim. 1971, 63, 739-748. (b) Senent, S.;

Casado, J.; Lizaso, J. An. R. Soc. Es. Fis. Quim. 1971, 67, 1133-1144.

[37] (a) Kozlov, Y. N.; Zharmukhamedov, S. K.; Tikhonov, K. G.; Dasgupta, J.; Kazakova, A. A.;

Dismukes; G. C.; Klimov, V. V. Phys. Chem. Chem. Phys. 2004, 6, 4905-4911; (b) Dasgupta, J.;

Tyryshkin, A. M.; Kozlov, Y. N.; Klimov, V. V.; Dismukes, G. C. J. Phys. Chem. B 2006, 110, 5099-

5111.

[38] Sarneski, J. E.; Didiuk, M.; Thorp, H. H.; Crabtree, R. H.; Brudvig, G. W.; Faller, J. W.; Schulte,

G. K. Inorg. Chem. 1991, 30, 2833-2835.

[39] (a) Saadeh, S. M.; Lah, M. S.; Pecoraro, V. L. Inorg. Chem. 1991, 30, 8-15. (b) John, R. P.;

Sreekanth, A.; Prathapachandra Kurup, M. R.; Fun, H.-K. Polyhedron 2005, 24, 601-610.

[40] Lane, B. S.; Vogt, M.; DeRose, V. J.; Burgess, K. J. Am. Chem. Soc. 2002, 124, 11946-11954.

[41] (a) Duboc, C.; Collomb, M.-N. Chem. Commun. 2009, 2715-2717; (b) Pedersen, E.; Toftlund, H.

Inorg. Chem. 1974, 13, 1603-1612; (c) Pal, S.; Ghosh, P.; Chakravorty, A. Inorg. Chem. 1985,

24, 3704-3706; (d) Hempel, J. C.; Morgan, L. O.; Lewis, W. B. Inorg. Chem. 1970, 9, 2064-2072.

[42] (a) Magers, K. D.; Smith, G. C.; Sawyer, D. T. Inorg. Chem. 1980, 19, 492-496. (b) Richens, D.

T.; Sawyer, D. T. J. Am. Chem. Soc. 1979, 101, 3681-3683.

[43] Brown, K. L.; Golding, R. M.; Healy, P. C.; Jessop, K. J.; Tennant, W. C. Aust. J. Chem. 1974, 27,

2075-2081.

[44] Cooper, S. R.; Dismukes, G. C.; Klein, M. P.; Calvin, M. J. Am. Chem. Soc. 1978, 100, 7248-7252.

http://www.sciencedirect.com/science/journal/00134686


125

[45] Note that according to ref. [23] MnV-oxo anion is protonated below its pKa of pH 13.7 and in

the form of O3MnV(OH)2-.

[46] Pode, J. S. F.; Waters, W. A. J. Chem. Soc. 1956, 717-725.

[47] (a) Garcia-Bosch, I.; Company, A.; Cady, C. W.; Styring, S.; Browne, W. R.; Ribas, X.; Costas, M.

Angew. Chem. Int. Ed. 2011, 50, 5648-5653; (b) Sawant, S. C.; Wu, X.; Cho, J.; Cho, K.-B.; Kim,

S. H.; Seo, M. S.; Lee, Y.-M.; Kubo, M.; Ogura, T.; Shaik, S.; Nam, W. Angew. Chem. Int. Ed.

2010, 49, 8190-8194; (c) Adam, W.; Roschmann, K. J.; Saha-Möller, C. R.; Seebach, D. J. Am.

Chem. Soc. 2002, 124, 5068-5073; (d) Finney, N. S.; Pospisil, P. J.; Chang, S.; Palucki, M.;

Konsler, R. G.; Hansen, K. B.; Jacobsen, E. N. Angew. Chem. Int. Ed. 1997, 36, 1720-1723; (e)

Murphy, A.; Stack, T. D. P. J. Mol. Cat. A 2006, 251, 78-88.

[48] Chatterjee, D.; Ember, E.; Pal, U.; Ghosh, S.; van Eldik, R. Dalton. Trans. 2011, 40, 10473-

10480.

[49] (a) Mukherjee, A.; Cranswick, M. A.; Chakrabarti, M.; Paine, T. K.; Fujisawa, K.; Münck, E.; Que

Jr., L. Inorg. Chem. 2010, 49, 3618-3628; (b) Geiger, R. A.; Chattopadhyay, S.; Day, V. W.;

Jackson, T. A. J. Am. Chem. Soc. 2010, 132, 2821-2831; (c) Yin, G.; Buchalova, M.; Danby, A.

M.; Perkins, C. M.; Kitko, D.; Carter, J. D.; Scheper, W. M.; Busch, D. H. Inorg. Chem. 2006, 45,

3467-3474.

[50] (a) Yoshida, T.; Sawada, S. Bull. Chem. Soc. Jpn. 1975, 48, 345-346; (b) Oakes, J.; Gratton, P. J.

Chem. Soc., Dalton Trans. 1998, 9, 3805-3809. (c) Madeira, M. T.; Feijoo, G.; Lema, J. M.

Enzyme Microbiol. Technol. 2007, 42, 70-75; (d) Sheriff, T. S.; Cope, S.; Ekwegh, M. Dalton

Trans. 2007, 44, 5119-5122.

[51] Theodoridis, A.; Maigut, J.; Puchta, R.; Kudrik, E. V.; van Eldik, R. Inorg. Chem. 2008, 47, 2294-

3013.

[52] Carrington, A; Symons, M. C. R. J. Chem. Soc. 1956, 3373-3380.


126

4.7 SUPPLEMENTARY INFORMATION

Figure S4.1 Time course for the consumption of AcOH (A) and the formation of PAA (B). Experimental conditions: 2.5

M AcOH, 2.5 M H2O2, 0.1 M H2SO4 and 25 °C. (C) Time course for the reformation of AcOH and the consumption of

PAA (D). Experimental conditions: 0.041 M PAA (produced from 2.5 M AcOH, 2.5 M H2O2, 0.1 M H2SO4 at 25 °C), 0.1 M

H2SO4 and 25 °C. The concentration profiles were calculated from the Lorenz fit of the 13

C-NMR signals at 179.2 ppm

(H3C-13

COOH) and 175.3 ppm (H3C-13

C(O)OOH).

(A) (B)

(C) (D)


127

Figure S4.2 Time courses for the formation of AcOH (A) and decomposition of PAA (B) at pH ≈ 10 and 25 °C. The

concentration profiles were calculated from the Lorenz fit of the 13

C-NMR signals at 179.2 ppm (H3C-13

COOH) and

175.3 ppm (H3C-13

C(O)OOH).

Figure S4.3 Comparison of the absorbance changes at 484 nm in the uncatalyzed reaction of 5 · 10-5

M Orange II with

5 · 10-3

M PAA (—), with 5 · 10-3

M PAA and 5 · 10-3

M EDTA (—), and in the presence of 1 · 10-5

M MnII catalyst (—).

Reaction conditions: 0.05 M NaHCO3 buffer, pH 9.5 and 25 °C.

(A) (B)


128

Figure S4.4 (A) Influence of acetate concentration on the reactivity of 1 · 10-5

M MnII with 0.01 M PAA and 5 · 10

-5 M

Orange in 0.05 M NaHCO3 buffer at pH 9.5 and 25 °C. Absorbance change followed at 484 nm. (B) Comparison of

different MnII salts used as starting material for the catalytic decomposition of 5 · 10

-5 M Orange II with 1 · 10

-5 M Mn

II

and 0.01 M PAA in 0.05 M NaHCO3 buffer at pH 9.5 and 25 °C. Absorbance change followed at 484 nm.

Figure S4.5 Comparison of the reactivity of 1 · 10-5

M MnII with 0.01 M PAA and 5 · 10

-5 M Orange II in the absence (—)

and presence of 0.02 M tBuOH (—) and 0.02 M BHT (—) as radical scavenger followed at 484 nm. Reaction conditions:

0.05 M NaHCO3 buffer, pH 9.5 and 25 °C.

(A) (B)


129

Figure S4.6 Effect of ionic strength (adjusted with different concentrations of NaNO3 on the catalytic decomposition of

5 · 10-5


M MnII and 0.01 M PAA in 0.05 M NaHCO3 buffer at pH 9.5 and 25 °C. Absorbance

change followed at 484 nm.

Figure S4.7 (A) UV/Vis spectral changes that accompany the reaction of 2 · 10-5

M colloidal MnIV

O2 with 2 · 10-3

M H2O2

(0.05 M NaHCO3 buffer) at pH 9.5 and 25 °C. Inset: Absorbance changes at 350 nm. (B) Observed rate constants for the

reduction of 2 · 10-5

M colloidal MnIV

O2 by 1 · 10-3

M H2O2 followed at 350 nm. Reaction conditions: 0.05 M

bicarbonate or phosphate buffer at 25 °C.

(A) (B)


130

Figure S4.8 (A) Absorbance change at 525 nm for the reaction of 2 · 10-5

M MnVII

O4- with 5 · 10

-4 M H2O2 as a function

of pH. (B) Corresponding observed rate constants as a function of pH (induction period was neglected). Reaction

conditions: stopped-flow experiments, 0.05 M bicarbonate or phosphate buffer at 25 °C.

Figure S4.9 Absorbance changes at 350 nm (A) and 525 nm (B) for the reaction of 2 · 10-5

M MnII with 0.01 M PAA in

0.05 M NaHCO3 as a function of pH at 25 °C.

(A) (B)

(A) (B)


131

Figure S4.10. Absorbance changes at 350 nm (A) and 525 nm (B) for the reaction of 2 · 10-5

M MnII with 0.01 M PAA at

different total carbonate concentrations. Reaction conditions: pH 9.5 and 25 °C.

Figure S4.11. X-band EPR spectra recorded at 10 K for 1 · 10-4

M MnVI

O42-

in 10 M KOH (—) and for the first sample

taken after mixing 1 · 10-4


-2 M PAA and 5 · 10

-3 M H2O2, 0.05 M carbonate buffer at pH 9.8 and 25

°C (—). EPR conditions: 8.95 GHz, 10 K, 1 mW microwave power, modulation amplitude 400 mT for (—) and 40 mT for

(—).

(B) (A)


132

Figure S4.12 (A) X-band EPR spectra recorded at 10 K for 1 · 10-4

M MnII in carbonate buffer (—) and at different time

intervals after the addition of 0.025 M PAA and 5 · 10-3

M H2O2 in 0.05 M NaHCO3 with 20 % of tBuOH. (B) Kinetic

traces at 350 and 525 nm. Arrows mark the time at which different EPR samples were taken. Reaction conditions: 1 ·

10-4

M MnII, 0.025 M PAA, 5 · 10

-3 M H2O2, pH 9.8 and 25 °C. EPR conditions: 8.95 GHz, 10 K, 1 mW microwave power,

modulation amplitude 400 mT.

Figure S4.13 (A) X-band EPR spectra recorded at 10 K for 1 · 10-4

M MnII at different time intervals after the addition of

0.025 M PAA and 5 · 10-3

M H2O2 in 0.2 M carbonate buffer (B) Kinetic traces at 350 and 525 nm. Arrows mark the time

at which different EPR samples were taken. Reaction conditions: 1 · 10-4

M MnII, 0.025 M PAA, 5 · 10

-3 M H2O2, pH 9.8

and 25 °C. EPR conditions: 8.95 GHz, 10 K, 1 mW microwave power, modulation amplitude 400 mT.

(A) (B)

(B) (A)


133

Figure S4.14 (A) Absorbance changes recorded at 484 nm for the reaction of 5 · 10-5

M MnVII

O4- with 5 · 10

-5 M Orange

II as a function of pH. (B) Initial rate as a function of pH. Reaction conditions: 0.05 M NaHCO3 buffer and 25 °C.

Figure S4.15 (A) UV/Vis spectral changes recorded for the disproportionation of 2 · 10-5

M MnVI

O42-

. Inset: kinetic

traces at 525 (—) and 610 nm (—). Reaction conditions: 0.05 M NaHCO3 buffer, pH ≈ 9.5 and 25 °C. (B) Initial rate of

the reaction of 2 · 10-5

M MnVI

O42-

with 5 · 10-5

M Orange II as a function of pH. Inset: observed rate constants for the

disproportionation of 2 · 10-5

M MnVI

O42-

as a function of pH. Reaction conditions: 2 · 10-5

M MnVI

O42-

, 0.05 M NaHCO3

buffer, 25 °C.

(A) (B)

(B) (A)


134

Figure S4.16 Comparison of the absorbance changes at 465 nm during the reaction of 5 · 10-5


M MnVII

O4- (), Mn

VIO4

2- () and O3Mn

V(OH)

2- () at pH 12.0. Reaction conditions: 0.05 M NaHCO3 and 25 °C.

Figure S4.17 Absorbance change at 484 nm for the reaction of 1 · 10-5

M colloidal MnIV

O2 with 5 · 10-5

M Orange II.

Reaction conditions: 0.05 M NaHCO3 buffer, pH 9.5 and 25 °C.


135

Figure S4.18 (A) UV/Vis spectra of 1 · 10-4

M MnO4- (—) after the reduction by 4 · 10

-4 M SO3

2- (—) and immediately

after addition of 5 · 10-5

M Orange II (—). (B) Absorbance changes at 484 nm that accompany the reaction of the in situ

generated MnIV

via reduction of MnVII

O4-. Reaction conditions: 1 · 10

-4 M Mn

VIIO4

-, 0.05 M NaHCO3 buffer, pH 9.5 and

25 °C.

Figure S4.19 UV/Vis absorbance changes recorded at 484 nm showing the reactivation of the catalytic system by

addition of 1 · 10-3

M H2O2. Reaction conditions: 1 · 10-5

M MnII, 0.01 M PAA, 5 · 10

-5 M Orange II, 0.05 M NaHCO3

buffer at pH 9.6 and 25 °C.

(B) (A)


136

Figure S4.20 Dependence of the initial rate of the MnII catalyzed degradation reaction on the total carbonate

concentration. Reaction conditions: 1 · 10-5

M MnII, 0.01 M PAA, 5 · 10

-5 M Orange II, 25 °C and pH 9.5.

Figure S4.21 (A) UV/Vis spectral changes recorded during the reaction of 5 · 10-4

M Orange II with 2.5 · 10-2

M PAA

catalyzed by 1 · 10-5

M MnII. (B) Corresponding absorbance vs. time plot at 484 nm. Reaction conditions: 0.05 M

NaHCO3 buffer at pH 9.5 and 25 °C (During the first 40 s of the measurement the absorbance remains above the

detection limit of the detector due to the high Orange II concentration).

(B) (A)


137

Figure S4.22 HPLC product analysis for the degradation of Orange II by PAA in the presence of MnII. Experimental


M Orange II, 1 · 10-5

M MnII and 0.01 M PAA at pH 9.5 and 25 °C.

Figure S4.23 Absorbance vs. time traces at 484 nm for the oxidative degradation Orange II catalyzed by 1 · 10-5

M MnII

and 0.01 M PAA at pH 9.5 (0.05 M NaHCO3 buffer) and 25 °C. Substrate addition was carried out in five consecutive

steps each consisting of 5 · 10-5

M Orange II.


138

Figure S4.24 Mass spectra of mono-protonated phthalate (m/z = 165,0184) (A) and 4-hydroxybenzosulfonate (m/z =

172,9904) (B) as products of the degradation reaction in Figure S4.23.

165.0184

166.0217

-MS, 2.5-7.3min #(208-605)

165.0193

166.0227

C8O4H5, M ,165.020

10

20

30

40

50

Intens.

[%]

0

20

40

60

80

100

[%]

164.5 165.0 165.5 166.0 166.5 167.0 m/z

172.9904

173.9935174.9871

-MS, 2.5-7.3min #(208-605)

172.9914

173.9947174.9873

C6SO4H5, M ,172.990

10

20

30

40

Intens.

[%]

0

20

40

60

80

100

[%]

172.5 173.0 173.5 174.0 174.5 175.0 175.5 m/z

(A)

(B)

Experiment

Simulation

Experiment

Simulation


139

Figure S4.25 Absorbance changes at 484 nm in the reaction of 1 · 10-4

M MnIII

(OAc)3 with 5 · 10-5

M Orange II. Reaction

conditions: 0.05 M NaHCO3, pH 9.5 and 25 °C.

Figure S4.26 Comparison of the absorbance changes recorded at λmax of the corresponding dye during the reaction of

5 · 10-5

M dye with 1 · 10-5

M MnII (A) and 1 · 10

-6 M Mn

II (B), respectively, and 5 · 10

-3 M PAA containing solution for

Methyl Orange 465 nm (), Orange II 484 nm () and Calmagite 612 nm (). Reaction conditions: 0.05 M NaHCO3,

pH 9.5 and 25 °C.

(A) (B)


140

5. High catalytic activity of a Mn-terpy compound in oxidative dye degradations with peracetic acid

141

5 HIGH CATALYTIC ACTIVITY OF A MN-TERPY COMPOUND IN

OXIDATIVE DYE DEGRADATIONS WITH PERACETIC ACID

5.1 GENERAL REMARK

The following chapter is based on ongoing work: High catalytic activity of manganese

terpyridine in the oxidative catalytic dye degradation with PAA, Sabine Rothbart and Rudi van

Eldik, in preparation.

5.2 INTRODUCTION

Significant ecological impact is caused by industrial dye waste since over 15 % of textile dyes

are lost in waste water streams during the coloring operation.[1] A recent review emphasized the

high cost involved in disposing the high volumes of dye effluent and that 128 tons of dyes are

released daily to the global environment.[2] About half of the global production of synthetic

colorants (700.000 t per year) are classified as aromatic azo compounds which are often designed

to be resistant to chemical, biochemical and photochemical degradation.[3] The increasing

ecological awareness stimulated an active field of scientific research dedicated to new and

ecologically worthwhile oxidation processes for the catalytic decomposition of environmental

pollutants[4] and dyes[5] in water. Among the various transition metals for catalytic oxidation,

manganese is of particular interest as one of the most efficient and environmental benign

elements. There are various manganese complexes with different salen,[6] porphyrin,[7] tacn[8] or

aromatic N-donor ligands[9] known to efficiently catalyze the oxidation of a wide range of

substrates. However, several limitations including elaborated synthetic methods, long reaction

times and substrate scope, have still to be resolved. Recent attention has therefore been focused

on metal complexes that use cheap and clean oxidants like H2O2 or peracetic acid (PAA) to bring

about efficient oxidation under mild reaction conditions.[10]

Our earlier work focused on the reactivity of simple MnII salts and Mn complexes in the H2O2

induced catalytic degradation of various organic dye substrates.[11] Detailed kinetic and

mechanistic investigations revealed the in situ generation of percarbonate (HCO4-)][12] as a key

http://dict.leo.org/ende?lp=ende&p=Ci4HO3kMAA&search=ecologically&trestr=0x8004

http://dict.leo.org/ende?lp=ende&p=Ci4HO3kMAA&search=worthwhile&trestr=0x8004


142

molecular oxidant from H2O2 and HCO3-, which leads to the formation of a reactive MnIV=O

species.[11a, b] Despite the high efficiency combined with the striking simplicity of the system, the

use of rather high concentrations of oxidant and bicarbonate was required.[11 a, b]

As reported recently these limitations could be overcome by the use of PAA.[13] By a detailed

UV/Vis and EPR spectroscopic investigation of the different reaction steps and intermediates, it

was evidenced that the key feature of the MnII ion catalyzed dye degradation by PAA involves a

complex oxidation by PAA vs. reduction by H2O2 (omnipresent in commercial PAA) reaction

sequence in which a steady state equilibrium of the reactive MnIV=O intermediate sets in. The

much higher degradation rate found in the presence of PAA was ascribed to the higher steady

state availability of the reactive MnIV=O compared to that in the use of HCO4- or H2O2/HCO3

-.

Scheme 5.1 Structure of the model substrates Orange II (OII), Tartrazine (TZ)and p-nitrophenol (PNP).

We now report an extension of the MnII/PAA system by application of the simple organic

chelate terpyridine (terpy). The much higher catalytic activity in the presence of the terpyridine

chelate in the Mn catalyzed oxidative degradation by PAA, was studied for various different

organic dyes (Scheme 5.1) to gain deeper insight into the mechanism of action of this intriguing

reaction. Besides Orange II (OII), a commonly used model system, we also chose highly stable dye

substrated as Tartrazine (TZ), which is one of the top three dyes by total sales and widely

employed as food colorant, in textile and paper applications[2] ,and p-nitrophenol (PNP) as

benchmark for the catalytic performance.


143

(5.1)

(5.2)


5.3.1 MnII + terpy in solution

In general the tridentate ligand terpyridine is able to form mono (1:1) and bis (1:2) complexes

with various transition metals (Equations 5.1 and 5.2). The equilibrium constant K1 for the

formation of the mono complex has been reported before.[14] We additionally performed a

spectrophotometric titrations under reaction conditions (phosphate buffer, pH 7.0) which reflect

those of the catalytic degradation reaction (results not shown) and the thereby obtained

equilibrium constant K1 = 1.1 · 104 M-1 fits rather well within the literature known range for K1 ((2.5

- 3.0) · 104 M-1)[14]. The fact that K1 appears to be smaller in our case is attributed to the use of the

H2PO4-/HPO4

- buffer, since phosphate anions could also form relatively stable metal-phosphate

complexes causing deviations in the known equilibrium of Equation 5.1. Since a second

coordination (K2) can not be excluded, a typical Job-plot analysis was performed to gain more

information on the distribution of the different species in solution.

The inset in Figure 5.1 exemplarily shows spectral changes before and after mixing of a MnII

solution with terpyridine in a phosphate containing buffer solution. A plot of the maximum

absorbance changes vs. the molar fraction reveals that the maximum is about at a molar fraction

of 0.58, which indicates that in solution a mixture of mono- and bis-terpyridine-MnII complexes is

formed (Figure 5.1). The deviation of the linear behavior at about a molar fraction of 0.5 is

attributed to an underlying local maximum of the mono compound. The lack of labile coordination

sites in a possible bis-MnIIterpy compound would be disadvantageous for the activation of the

oxidant, thus it is assumed that only the in situ generated mono terpyridine complex is relevant to

the catalytic reaction.


144

Figure 5.1 Absorbance change at two wavelengths for different molar fractions of MnII and terpy at pH 7.0. Inset:

UV/Vis spectral changes before (MnII in water with terpy in buffer/CH3CN) and after mixing at a

([terpy]/([terpy]+[MnII]) of 0.3. Reaction conditions: [Mn

II]+[terpy] = 1 · 10

-4 M, 0.05 M phosphate buffer with 10 %

CH3CN, pH 7.0, 25 °C.

5.3.2 MnIIterpy + PAA in the absence of dye substrate

5.3.2.1 UV/Vis measurements

Upon following the UV/Vis spectral changes of the reaction of MnIIterpy with PAA in the

absence of substrate (Figure 5.2 (A)) a rapid reaction is observed which seems to consist of several

parallel reaction steps. The fact that no isosbestic points are observed already points to the

complexity of this reaction. However, a closer look at the UV/Vis spectra directly after mixing

implies the presence of oxo-bridged high valence manganese species which have been reported

before in the reaction of Mn precursors with oxone (HSO5-).[15] UV/Vis characteristics of both

species, i.e. µ-oxo-bridged Mn2III/IV and Mn2

IV/IV can be found, as is evident from a comparison with

the spectral features of the readily prepared dimers (see Figures 5.3 (A)).[16, 17]

Similar to literature findings, the decomposition of these intermediates to MnVIIO4- (five

characteristic bands at 500-575 nm) was observed [15] after different reaction times depending on

the [PAA] and [Mn] (results not shown). Since the formation of free permanganate involves


145

decomplexation of the terpy ligand and no significant amounts of free ligand is found after the

reaction (see 296 nm), it can be assumed that ligand degradation also occurs during the reaction

of MnIIterpy with PAA.

Figure 5.2 (A) UV/Vis spectral changes recorded during the reaction of MnIIterpy (Mn

II:terpy = 1:2) with PAA. Inset:

aborbance vs. time plots recorded at different wavelengths. Reaction conditions: 2 · 10-5

M MnII, 4 · 10

-5 M terpy, 2 ·

10-3

M PAA, 0.05 M phosphate buffer, pH 7.0, 25 °C. (B) Plot of kobs vs. the delay time of the addition of 5 · 10-5

M OII

to a mixture of MnIIterpy and PAA at 5 °C. Inset: aborbance vs. time plot recorded at 336 nm for the same reaction as

in (A) at 5 °C. Reaction conditions: 2 · 10-5

M MnII, 4 · 10

-5 M terpy, 2 · 10

-3 M PAA, 5 · 10

-5 M OII, 0.05 M phosphate

buffer, pH 7.0, 5 °C.

Figure 5.3 (A) UV/Vis spectra of 5 · 10-5

M readily prepared [Mn2III/IV

(µ-O)2terpy2(H2O)2](NO3)3 (—) and [Mn2IV/IV

(µ-

O)2terpy2(SO4)2] (—) in 0.05 M phosphate buffer at pH 7.0 and 25 °C. (B) UV/Vis spectra of the intermediate species

formed during the reaction of MnIIterpy (Mn

II:terpy = 1:2) with PAA, reaction conditions: 2 · 10

-5 Mn

II, 4 · 10

-5 M terpy,

2 · 10-3

M PAA, phosphate buffer, pH 7.0, 25 °C.

(A) (B)

(A) (B)


146

In order to clarify what the catalytically relevant phases of the reaction are, we performed an

experiment where a dye substrate (OII) is added after different delay times in the reaction of

MnIIterpy with PAA. A comparison of the thereby obtained rates could give some helpful

information on the different reaction phases and their relevance to the catalytic dye degradation

reaction. To slow down the reaction and gain sufficient time for several of these reactivity tests,

we chose a reaction temperature of 5 °C. The results presented in Figure 5.2 (B) show that with

the decomposition of the intermediates of the first reaction phase a concomitant loss in the

catalytic activity towards the dye substrate is observed. This is attributed to the above described

MnVIIO4- formation and ligand degradation, and leads to the suggestion that the first reaction

phase is more important to the studied reaction of the catalytic dye degradation with PAA.

Contrary to our earlier studies on the activation of PAA by MnII ions, we were not able to

resolve the complex behavior in particular during the first phase of the reaction by variation of the

different reaction parameters, i.e. [catalyst], [PAA], [H2O2], pH or temperature (results not shown).

This, unfortunately, made it impossible to obtain stable reaction phases for more detailed

investigations and thereby disclose the mechanism of action of the dimer formation upon reaction

of the MnII precursor with PAA. However, it is well established that the reaction of a subequivalent

of oxidant with a 1:1 mixture of terpy in aqueous solution results in the formation of [Mn2III/IV(µ-

O)2terpy2(H2O)2]3+, in a way that it can be isolated and crystalographically characterized (see

Scheme 5.2).[16]

Scheme 5.2 Structure of the proposed µ-oxo dimers with the Mn oxidation states either III/IV (overall charge 3+) or

IV/IV (overall charge 4+).

Furthermore, the reaction of [Mn2III/IV(µ-O)2terpy2(H2O)2]3+ in the presence of an excess of the

two-electron oxygen-atom donor HSO5- has extensively been studied before and serves as a model

system for the oxygen-evolving complex which catalyzes the conversion of water to dioxygen.[15, 16,


147

18, 22] In order to elucidate if there are analogies in the reaction with PAA and HSO5-, the

experiments were repeated with KHSO5/H2O2 under comparable conditions to mimic the presence

of the [H2O2]eq in PAA. The observed UV/Vis spectral changes of the reaction of the MnII precursor

with KHSO5/H2O2 (Figures 5.4 (A) and (B)) clearly resemble the ones observed with PAA as oxidant.

Figure 5.4 (A) UV/Vis spectral changes during the reaction of MnIIterpy (Mn

II:terpy = 1:2) with a HSO5

-/H2O2 mixture

and (B) corresponding absorbance vs. time plots at 336 nm and 410 nm. Reaction conditions: 2 · 10-5

M MnII, 4 · 10

-5 M

terpy, 2 · 10-3

M HSO5-, 2.5 · 10

-4 M H2O2, 0.05 M phosphate buffer, pH 7.0, 25 °C.

The small decay following the first increase in absorbance can be attributed to the

disproportionation of the [H2O2]eq in PAA by the high-valence di-µ-oxo core.[11c, 19, 20] In general,

oxo-bridged Mn2III/IV and Mn2

IV/IV dimers can be considered as oxidized forms of typical catalase

model compounds and are known to cause H2O2 dismutation.[19] Consequently, it is concluded,

that the reaction of the MnII precursor with PAA proceeds in analogy to the oxidation by KHSO5

besides the fact that the reaction is complicated by the [H2O2]eq in PAA. Since it was hardly

possible from the UV/vis data alone to determine whether the dimeric form is predominantly

Mn2III/IV or Mn2

IV/IV under catalytic conditions, further EPR spectroscopic experiments were

performed.

5.3.2.2 EPR measurements

The initial MnII catalyst containing buffer solution without any oxidant present showed an

expected signal for mononuclear MnII centered at g = 2 (see Figure 5.5 (A), first spectrum). The

lack of the typical six-line pattern (I = 5/2, S = 5/2) is attributed to the formation of different

(B) (A)


148

substituted aqua-species, a mixture of mono- and bis-terpyridine compounds (see Chapter 5.3.1),

and complex-formation with phosphate buffer under the selected experimental conditions. The

second spectrum, taken directly after addition of PAA is shown in Figure 5.5 (B).

Figure 5.5 X-band EPR spectra at 10 K of (A) MnIIterpy (Mn

II:terpy = 1:2) in solution and after mixing with PAA at

different time intervals. (B) Amplification of the first spectrum in the presence of oxidant at approximately 15 s

reaction time. Reaction conditions: 1 · 10-4

M MnII, 2 · 10

-4 M terpy, 0.01 M PAA, 0.05 M phosphate buffer, pH 7.0, 25

°C. EPR conditions: 9.45 GHz, 1 mW microwave power, modulation amplitude 400 mT.

The drastic decrease in the overall spectral features compared to the spectrum in the absence

of oxidant emphasizes the fact that the main species in solution under catalytic conditions is EPR

silent. On the other hand a small amount of a 16-line signal at g = 2 is observed (Figure 5.5 B,

amplification of the second spectrum recorded 15 s after mixing of the reactants). These spectral

features are characteristic for an antiferromagnetically exchange-coupled Mn2III/IV dimer.[21] An

estimated quantification implies that µ-oxo bridged Mn2III/IV dimer, content is approximately ≤ 10

%, whereas the remaining ≈ 90 % are EPR inactive. On the basis of the UV/Vis spectroscopic results

this EPR silent form is assumed to be mostly a Mn2IV/IV dimer which is in fact in very good

agreement with the detailed mechanistic investigations of the intermediates under catalytic

conditions for the [Mn2III/IV(µ-O)2terpy2(H2O)2]3+/ HSO5

- system by Brudvig/Crabtree et al., who

evidenced that the predominant species in the catalytic solution is the EPR silent Mn2IV/IV dimer.[22]

(B) (A)


149

5.3.3 MnIIterpy catalyzed dye degradation with PAA

5.3.3.1 General observations

In order to study the catalytic dye degradation by PAA, a series of measurements were

performed at pH 7.0 and 25 °C. Figures 5.6 (A), (B) and (C) depict the UV/Vis spectral changes

during the catalytic degradation reaction and the concomitant absorbance changes at λmax for the

different model substrates: OII (λmax = 484 nm), TZ (λmax = 420 nm) and PNP (λmax = 400 nm).

Figure 5.6 UV/Vis spectral changes observed for the catalyzed oxidative degradation of OII (A), TZ (B) and PNP (C) by

PAA, insets show the absorbance vs. time plots at the corresponding λmax of the dye. Reaction conditions: (A) and (B) 1

· 10-6

M MnII, 2 · 10

-6 M terpy, 5 · 10

-3 M PAA, 5 · 10

-5 M dye, 0.05 M buffer, pH 7.0, 25 °C; (C) 5 · 10

-6 M Mn

II, 1 · 10

-5 M

terpy, 5 · 10-3

M PAA, 5 · 10-5

M dye, 0.05 M buffer, pH 7.0, 25 °C.

(A) (B)

(C)


150

When the results for TZ are compared to the use of chlorine or hypochlorite in basic media,

which is a well known technology for bleaching dyes,[2b, 23] it is obvious that under these reaction

conditions the bleaching reaction is not only limited to the cleavage of the azo-linkage (420 nm for

TZ) as evidenced by the UV/Vis absorbance after 500 s. Thus, it is concluded that the more stable

aromatic rings and primary dye destruction products which result from the azo-cleavage are also

extensively degraded.[2b, 24] These results highlight the fact that at least in the studied reactions the

absorbance decrease of the dye at λmax provides a satisfactory basis for the evaluation of the

performance of the dye treatment method. In addition, PNP, a highly stable phenolic model

substrate, is efficiently degraded by only minor amounts of oxidant and catalyst (see Figure 5.6).

Based on the excellent catalytic activity of the MnIIterpy/PAA system at lower catalyst and oxidant

concentrations compared to the simple MnII salt,[13] the reaction conditions were refined to

evaluate its putative applicability in the catalytic degradation of various organic dyes with different

structural motives by determining the catalytic turnover frequency. Table 5.1 contains the

calculated turnover frequencies, i.e. mol of dye oxidized by mol of catalyst per hour.

Dye substrate OII TZ PNP

Initial rate (Ms-1) 9,92 · 10-6 5,74 · 10-6 4.43 · 10-7

TOF (h-1) 7142 4133 319

Table 5.1 Summary of initial rates and turn over frequencies for the degradation reactions of OII, TZ and PNP

determined at the corresponding λmax of the dye. Reaction conditions: 5 · 10-6

M MnII, 1 · 10

-5 M terpy, 5 · 10

-5 M dye

substrate, 5 · 10-3

M PAA, 0.05 M phosphate buffer, pH 7.0, 25 °C.

Numerous control experiments were performed (results not shown) to exclude different

influencing factors on the overall dye degradation catalysis, viz. the spontaneous, non-catalyzed

reaction of PAA with the dye substrates, a possible contribution of free radicals, acetate or

phosphate buffer concentration. In organic buffers suitable for this pH range such as MES and Bis-

Tris no reaction with dye substrate was observed (results not shown). This is probably due to the

fact that these organic molecules could serve as substrate to the highly reactive Mn intermediates

formed during the reaction so that the dye degradation is suppressed by the excess content of

buffer required for reliable pH jump experiments.


151

Apart from the fact that the catalytic degradation of the studied organic dyes occurred

efficiently, the optimization of the process by analyzing the effect of several parameters such as

pH, [oxidant] and [catalyst], was investigated in more detail.

5.3.3.2 Kinetics of the MnIIterpy catalyzed dye degradation by PAA

pH dependence. The kinetics of the manganese catalyzed oxidative degradation of the selected

dyes was studied in 0.05 M buffer (phosphate, acetate and bicarbonate), in a pH range from 2.0 to

10.0 at 25 °C. The decomposition of the dyes was followed by monitoring the absorbance change

at the characteristic λmax of each dye substrate. Figure 5.7 shows the changes of the

experimentally determined observed first order rate constants for the MnIIterpy catalyzed

degradation for the different substrates as a function pH.

Figure 5.7 Plots of the observed first order rate constants for the degradation of OII (—), TZ (—) and PNP (—) as a

function of pH determined at the corresponding λmax. Reaction conditions: 5 · 10-6

M MnII, 1 · 10

-5 M terpy, 5 · 10

-3 M

PAA, 5 · 10-5

M dye substrate, 0.05 M acetate or phosphate buffer, 25 °C.

For all systems the rate of oxidation increases with increasing pH and goes through a

maximum at a pH between 6.5 and 7.0, suggesting that the same reactive manganese species and

the same in situ formed oxidizing agent is responsible for the decomposition of the various

substrates. The decrease in oxidation rate at higher pH is partly due to the subsequent formation

of Mn(OH)2 precipitates at higher pH, which negatively affect the stability of the MnIIterpy

precursor and thereby the amount of available catalyst. The deviation observed at a pH of about


152

4.5 is attributet to the pKa of the terpyridine ligand (pKa = 4.7),[25] which interferes in the

equilibrium described in Equation 5.1 and thereby lowers the available precursor concentration

for the catalytic reaction.

Effect of [terpy] on the catalytic oxidative degradation of various dyes. Since the complex-

formation constants of MnII and terpyridine are known to be relatively unfavourable, the dye

degradation reaction was studied as a function of the terpyridine ligand concentration for OII, TZ

and PNP (Figure 5.8).


function of the terpy:Mn ratio determined at the corresponding λmax. Reaction conditions: 5 · 10-6

M MnII, 1 · 10

-3 M

PAA, 5 · 10-5

M dye substrate, 0.05 M phosphate buffer, pH 7.0, 25 °C.

The results clearly show that complexation between the σ-donor and π-acceptor terpyridine

as ligand and the metal, is indispensable for efficient catalytic performance. With increasing ligand

concentration the maximum rate increases and reaches limiting values at higher excess

concentrations of terpyridine ligand for all the studied dye substrates.

For this result two possible explanations are plausible. On the one hand the low binding

constant of terpyridine to manganese may require higher [ligand] to assure complex-formation

between terpyridine and manganese. On the other hand, the UV/Vis spectral changes during the

catalytic degradation reaction imply that partial ligand degradation also occurs, so that a higher


153

[terpy] is required for efficient catalytic performance. In any case, all further measurements were

performed at a MnII:terpy ration of 1:2.

Critical role of [H2O2]eq. Previous results established that the [H2O2]eq plays an important role in

the overall catalytic oxidation reaction by PAA (commercially available PAA always contains

H2O2).[13] It was evidenced that a low [H2O2]eq was required to avoid MnIVO2 precipitaition and

over-oxidation to permanganate. Moreover, in the presence of excess [PAA] the catalytic system

could be reactivated by only minor amounts of [H2O2]. However, larger [H2O2]eq negatively

affected the catalytic performance by delaying the formation of active Mn=O intermediates.

Figure 5.9 Plots of the inverse observed first order rate constants for the degradation of OII (—), TZ (—) and PNP (—)

as a function of the additional [H2O2] determined at the corresponding λmax. Reaction conditions: 5 · 10-6

M MnII, 1 · 10

-

5 M terpy, 5 · 10

-3 M PAA, 5 · 10

-5 M dye substrate, 0.05 M phosphate buffer, pH 7.0, 25 °C.

In the case of the MnIIterpy catalyzed degradation of dyes, no reactivation of the catalytic

system by addition of small aliquots of H2O2 was found. The influence of the [H2O2]eq on the

catalytic degradation reaction was investigated by performing the reaction in the presence of an

additional amount of H2O2 in the concentration range from 0.25 · 10-3 to 5 · 10-3 M (equal to

[PAA]). It shows that a higher [H2O2]eq inhibits the reaction. If the corresponding inverse observed

rate constant is plotted against the additional [H2O2] for OII, TZ and PNP (Figure 5.9), a linear

dependence with no intercept is observed. The slopes of these linear fits were found to be 36.4 ∙

103 s M-1 for OII, 61.2 ∙ 103 s M-1 for TZ and 162.7 ∙ 103 s M-1 for PNP, respectively. To account for

this observation it is suggested that H2O2 reacts with the rapidly formed oxo-bridged dimeric


154

intermediates (Mn2III/IV and Mn2

IV/IV) in an unproductive parallel catalase-like pathway during

which oxygen is formed and the dimers are reduced or partially cleaved. Similar observations have

been reported before.[11c] This represents a kind of back reaction that depending on the [H2O2]eq

lowers the steady state availability of the dimeric, oxo-bridged species, which in turn emphasizes

their role as key intermediates in the MnIIterpy catalyzed dye degradation with PAA.

Effect of [PAA] on the MnIIterpy catalyzed oxidative degradation of various dyes. In order to

investigate the effect of the oxidant concentration on the MnIIterpy catalyzed degradation of the

different dyes, the [PAA] was varied from (0.5 - 15) · 10-3 M at a catalyst concentration of 5 · 10-6

M. Figure 5.10 shows the plot of the observed first order rate constants as a function of [PAA] for

Orange II, TZ and PNP.


function of the [PAA] determined at the corresponding λmax. Reaction conditions: 5 · 10-6

M MnII, 1 · 10

-5 M terpy, 5 ·

10-5

M dye substrate, 0.05 M phosphate buffer, pH 7.0, 25 °C.

Despite of the difference in reactivity towards the different dyes, in all cases the system

reaches limiting values at higher [PAA]. Although this saturation behaviour is in line with Michelis-

Mentin kinetics, which involves a pre-equilibrium reaction prior to the rate-limiting step of the

studied reaction, it has to be considered in light of the findings concerning the effect of [H2O2]eq,

that it might rather be related to the equilibrium content of H2O2 described above. With increasing

[PAA] the omnipresent [H2O2]eq also accelerates the back reaction which in turn slows down the


155

observed dye degradation rates, so that the observed rate constants for the overall degradation

reaction only show an apparent saturation effect.

Effect of [MnIIterpy] on the catalytic oxidative degradation of various dyes. To evaluate the

effect of the catalyst concentration on the oxidative degradation of different organic dyes by PAA,

kinetic studies were performed under variation of the MnIIterpy pre-catalyst concentration.

Surprisingly, in the present case, the catalytic reaction leads to a square dependence of kobs on the

[MnIIterpy], which implies a possible involvement of two equivalents of the MnIIterpy precatalyst

in the overall catalytic reaction as expected when Mn-dimers participate. The plots of the

corresponding observed first order rate constants vs. [catalyst]2 for OII, TZ and PNP (for two

different oxidant concentrations) are presented in Figures 5.11 (A) and (B), respectively. From the

slope of these dependencies the third order rate constants for all substrates were calcultated and

are summarized in Table 5.2.


function of [catalyst]2 determined at the corresponding λmax for (A) 1 · 10

-3 M PAA and (B) 5 · 10

-3 M PAA. Reaction

conditions: MnII:terpy = 1:2, 5 · 10

-5 M dye substrate, 0.05 M phosphate buffer, pH 7.0, 25 °C.

OII TZ PNP

k (M-2s-1) for 1 · 10-3 M PAA 3.74 · 109 2.31 · 109 6.23 · 107

k (M-2s-1) for 5 · 10-3 M PAA 7.25 · 109 4.36 · 109 11.65 · 107

Table 5.2 Third order rate constants for two different [PAA] calculated from the linear fit of the observed first order

rate constants in Figures 5.11 (A) and (B).

(A) (B)


156

Since the observed second order dependence highlights again the perticipation of dimeric

manganese species in the overall catalytic degradation reaction, we decided to additionally repeat

some experiments for the readily prepared oxo-bridged Mn2III/IV and Mn2

IV/IV dimers as catalysts.

5.3.3.3 Readily prepared dimers in the catalytic dye degradation by PAA

A direct comparison of the catalytic reactivity at indentical manganese content revealed a

slightly slower degradation rate when the dimers are used (Figure 5.12 (A)).

Figure 5.12 (A) Comparison of the depedencies of the observed first order rate constants for the degradation of OII on

the [catalyst]2 for monomeric Mn

IIterpy (―), dimeric [Mn2

III/IV(µ-O)2terpy2(H2O)2](NO3)3 (―) and Mn2

IV/IV(µ-

O)2terpy2(SO4)2 (―) at identical [total Mn]. Reaction conditions: 5 · 10-3

M PAA, 5 · 10-5

M OII, 0.05 M phosphate

buffer, pH 7.0, 25 °C. (B) Comparison of the traces recorded at 484 nm that accompany the catalytic OII degradation of

(―)1 · 10-6

M MnII + 2 · 10

-6 M terpy and (―) 0.5 µM [Mn2

III/IV(µ-O)2terpy2(H2O)2](NO3)3 + 1 · 10

-6 M terpy. Reaction


M PAA, 5 · 10-5

M OII, 0.05 M phosphate buffer, pH 7.0, 25 °C.

This observation can be explained by the effect of the two-fold Mn:Ligand ratio described

above. Whereas all prior experiments were performed with a excess of terpyridine ligand to

assure the complexation despite the low binding constant and occurring ligand degradation, the

dimers contain only a 1:1 ratio of Mn:terpy, which accounts for the slightly lower activity. In fact, if

a comparison is made between the monomeric and the dimeric starting compound with the same

amount of terpyridine ligand present, no difference in the catalytic performance is observed

(Figure 5.12 (B)).

(B) (A)


157

In general, oxo-bridged manganese complexes are often involved as key intermediates in

several catalytic oxygenation processes such as photosynthetic water oxidation,[16, 26, 27]

bleaching[28], epoxidation[29] or oxidation of hydrocarbons[30] and alcohols[31]. However, it is known

from literature that these species are not prevalent in catalytic solution when H2O2 is used as

oxygen source.[11c, 32] Moreover, the lability of the [Mn2III/IV(µ-O)2terpy2(H2O)2]3+ compound has

been shown before.[11c, 33] Presumably it is rapidly converted to a mononuclear manganese species

by reduction to lower-valence bridged dimanganese species and eventually concomitant cleavage

of the bis-µ-oxo motive by H2O2.[11c, 19, 20] This could also account for the observation that no

significant difference is found, whether the MnII starting compound or the readily prepared

Mn2III/IV or Mn2

IV/IV µ-oxo dimers are used in the catalytic dye degradation with PAA, since the oxo-

bridged core of the dimers is reduced or partially cleaved by the [H2O2]eq.

Figure 5.13 Comparison of the absorbance vs. time plots at 484 nm in the stoichiometric reaction of 5 · 10-5

M

[Mn2III/IV

(µ-O)2terpy2(H2O)2](NO3)3 (―) and 5 · 10-5

M Mn2IV/IV

(µ-O)2terpy2(SO4)2 (―) with 5 · 10-5

M OII. Reaction

conditions: 0.05 M phosphate buffer, pH 7.0, 25 °C.

However, also in the absence of oxidant the readily prepared Mn2III/IV or Mn2

IV/IV dimers react

with the dye substrate. Figure 5.13 shows the absorbance vs. time plots recordedat the dye’s λmax

upon reaction with a stoichiometric amount of [Mn2III/IV(µ-O)2terpy2(H2O)2](NO3)3 and Mn2

IV/IV(µ-

O)2terpy2(SO4)2. Both dimers show rather sluggish activity towards the dye substrate. Whereas in

the reaction with the Mn2III/IV dimer only minor changes are observed, addition of Mn2

IVIV causes a

rapid decrease of the chacracteristic absorbance. It is clear that a stoichiometric amount of


158

Mn2III/IV and Mn2

IV/V yields incomplete dye degradation, which is reflected by the fact that only 15

to 40 % of the azo-bond absorbance band is decreased.

As a consequence, it has to be considered that both in situ formed dimmer species contribute

at least partially to the observed reactivity in the overall dye degradation. However, this also

implies that there is another, more reactive intermediate formed.

The fact that [Mn2III/IV(µ-O)2terpy2(H2O)2]3+ with its easily exchangeable water ligands is rapidly

formed when an excess of PAA oxidant is present, gives rise to further potential reactive “Mn=O”

intermediates. These could involve the potential dimeric species with a MnV=O or MnIV-O• moiety

as postulated before.[23, 27, 34] Since it is known that the latter reaction steps to even higher valence

species than Mn2III/IV and Mn2

IVIV, ´which afford the labile terminal coordination sites occupied by

two water ligands in the terpy system[35] (see Scheme 5.2), a comparison of the reactivity pattern

of the corresponding bidentate bpy or phen ligand systems could reveal some helpful information

on further putative catalytically relevant species. Both oxo-bridged dimers of the bpy and phen

(metal:bpy/phen ligand ratio in complex = 2:1) lack the essential feature of the terminal water

ligands. These labile coordination sites of the terpyridine compound are believed to result in

enhanced lability of di-µ-oxo dimer and give rise to reaction pathways involving the potential high-

valence MnV=O or MnIV-O• intermediates.[35]

Some preliminary eperiments with the use of phen instead of terpy already indicate that this

motive might be relevant. Figure 5.14 shows the UV/Vis spectral changes at λmax during the

catalytic degradation reaction of Orange II in the presence phen. Although two different

MnII:ligand ratios were used to exlude the interference of the MnII:ligand precursor equilibrium

(related to Equations 5.1 and 5.2), it is evident that the so-formed potential dimeric phen species

are much less reactive.

Similar observations on the influence of the terminal binding site on the rate of oxygen

evolution from the [Mn2III/IVO2]3+/oxidant have been reported before.[35] It is concluded that the

lack of the labile coordination site in the corresponding in situ formed bpy and phen dimers is

responible for the decrease in the catalytic dye degradation activity. Hence, we suggest that the

MnIIterpy + PAA system indeed comprises parallels to the intermediates reported for the water

oxidizing model system [Mn2III/IV(µ-O)2terpy2(H2O)2]3+/HSO5

- and that therefore the postulated


159

MnV=O or MnIV-O• containing intermediates might account for the extraordinarily high activity in

the catalytic dye degradation with PAA.

Figure 5.14 UV/vis spectral changes at 484 nm that accompany the catalytic degradation of OII by PAA with different

MnIIphen precursors (Mn

II:phen = 1:20 (―) and Mn

IIphen = 1:4 (―)). Reaction conditions: 1 · 10

-6 M Mn

II, 2 · 10

-5 M

phen (―), 4 · 10-6

M phen (―), 5 · 10-5

M OII, 5 · 10-3

M PAA, 0.05 M phosphate buffer, pH 7.0, 25 °C.

Although the exact role of the high-valence Mn2III/IV and Mn2

IV/IV dimers or higher oxidized

dimers with MnV=O or MnIV-O• moieties in this efficient catalytic degradation reaction remains to

be clarified, it has to be considered that there are several catalytically relevant species formed in

the reaction with MnIIterpy and PAA that contribute to the observed overall dye degradation.

5.3.4 Mechanistic implications

Although the existence of free coordination sites in the precursor complex is considered to be

of general importance for the catalytic activity, our experiments on the MnIIterpy catalyzed dye

degradation by PAA point to the necessity of an excess [terpy], which in turn may favour the

formation of a coordinatively saturated bis-terpy compound. This seemingly contradictive

behaviour can be accounted for under consideration of two phenomena: On one hand, the

binding constant of terpy with MnII is not ideal, especially in phosphate containing solution it

probably interferes in the complexation equilibrium of the MnIIterpy precursor by formation of

Mn-phosphate complexes and thereby lowers the available [pre-catalyst]. On the other side,

ligand degradation also occurs during the reaction with oxidant, which results in the formation of


160

free permanganate. Consequently, an excess ligand is required to suppress the effect of the

decreasing ligand concentration.

Since over the applied pH range no catalytic degradation reaction with MnII and PAA is

observed in the absence of the chelate ligand due to insufficient interaction between the

reactants, it is concluded that the complex-formation by terpyridine is indispensable.

Furthermore, it is suggested that a reaction with PAA in its protonated form (pH 7) is in the first

place enabled by the strong π-acceptor properties of the terpyridine ligand and the proximity to

the pKa of PAA.

The UV/Vis spectral changes in the absence of dye imply a rapid formation of µ-oxo-bridged

high-valence Mn2III/IV and Mn2

IV/IV dimers. This particularly complex reaction behaviour (see

absorbance vs. time plots) could be satisfactorily copied by the use of HSO5-/H2O2 under conditions

comparable to those used in the PAA system, which emphasizes the role of the [H2O2]eq

omnipresent in PAA. The [H2O2]eq gives rise to a potential back reaction with the rapidly formed

high-valence Mn-intermediates, which thereby lowers the catalytic dye degradation activity. In

general, oxo-bridged Mn2III/IV and Mn2

IV/IV dimers can be considered as oxidized forms of typical

catalase model compounds and are known to cause H2O2 dismutation.[11c, 19, 20] The involvement of

dimeric manganese intermediates is in addition supported by EPR spectroscopic measurements

and the second-order dependence of the catalytic dye degradation on the [catalyst]. Although the

exact mechanism of action of dimer formation remains to be clarified, it is well known that the

synthesis and isolation of [Mn2III/IV(µ-O)2terpy2(H2O)2]3+ is already achieved by a subequivalent of

oxidant.[16] On the other hand, numerous reports on the reaction between the [Mn2III/IV(µ-

O)2terpy2(H2O)2]3+ and an excess HSO5- oxidant revealed by the use of various spectroscopic

methods that the major species under catalytic conditions is an Mn2IV/IV dimer.[22]

MnIIterpy shows comparable reaction behavior with PAA as well as with HSO5-/H2O2, and our

own EPR spectroscopic measurements of the intermediates formed from MnIIterpy and PAA

confirmed the presence of Mn2III/IV and mainly Mn2

IV/IV dimers. Therefore, it can be concluded that

there are fundamental parallels to the intermediates formed during the well studied reaction of

[Mn2III/IV(µ-O)2terpy2(H2O)2]3+. Brudvig/Crabtree et. al. investigated the conversion of a [MnIII(μ-

O)2MnIV]3+ complex to its one-electron-oxidized product, [MnIV(μ-O)2MnIV]4+, by the two-electron

oxidant HSO5-. According to literature reports, a simplified mechanistic sequence to account for

these putative intermediates is outlined in Scheme 5.3.


161

This oxidation is believed to pass through an intermediate two-electron-oxidized form of the

[Mn2III/IV(µ-O)2terpy2(H2O)2]3+, with either a MnV=O or a MnIV-O• moiety, which through

disproportionation with [MnIII(μ-O)2MnIV]3+ mainly yields the corresponding Mn2IV/IV dimer.[34, 35]

The binding site for this second oxidation step is supposed the MnIII ion since it is expected to

show comparable or even higher substitution rates than the MnIV binding site, due to the

enhanced lability in the Jahn-Teller distorted MnIII ion.[34]

Scheme 5.3 Proposed mechanism for the reaction between [Mn2III/IV

(µ-O)2terpy2(H2O)2]3+

and oxygen-atom transfer

reagents (XO) acting as a water-oxidation catalyst according to references [27] and [34].

Although in the absence of oxidant, both dimers, viz. [Mn2III/IVO2terpy2(H2O)2](NO3)3 and

Mn2IV/IV(µ-O)2terpy2(SO4)2, show slight activity toward the dye substrate, it has to be considered

that the excess of oxidant present allows further oxidation steps as in [Mn2III/IV(µ-

O)2terpy2(H2O)2]3+ / HSO5- (outlined in Scheme 5.3). However, to what extend the differént in situ

formed µ-oxo dimers Mn2III/IV and Mn2

IV/IV, respectively, or the postulated reactive intermediates

and mechanistic ideas of the water oxidizing model system [Mn2III/IV(µ-O)2TERPY2(H2O)2]3+/HSO5

-

apply to the observed dye degradation reactivity, remains to be clarified.


162

5.4 CONCLUSIONS

In summary, we presented an extraordinarily efficient method for the catalytic oxidative

degradation of different highly stable organic dyes, such as Orange II, Tartrazine and p-nitrophenol

as reflected by the high catalytic turnover frequencies. Screening and spectroscopic methods

allowed us to study the catalytic reaction course and to shed light on some key mechanistic

features. The second order dependence of the observed first order rate constants for the overall

dye degradation on the pre-catalyst concentration implies the involvement of dimeric

intermediates in the catalytic reaction. The omnipresent [H2O2]eq in the PAA stock solution was

shown to result in an inverse dependence on the catalytic reaction, most likely by causing a

catalase-like H2O2 decomposition, which also becomes noticeable at higher [PAA]. Consequently,

the saturation kinetics observed for higher oxidant concentrations can be ascribed to the thereby

increased [H2O2]eq.

UV/Vis spectral and EPR spectroscopic measurements in the absence of dye substrate

confirmed that the intermediates formed under catalytic conditions and in the absence of

substrate mainly consist of dimeric species with a [Mn2III/IV(µ-O)2]3+- and [Mn2

IV/IV(µ-O)2]4+-core.

These results strongly resemble earlier findings on the water oxidizing model system [Mn2III/IV(µ-

O)2terpy2(H2O)2]3+/HSO5-. Both these dimers show rather moderate activity towards the dye

substrate. Preliminary experiments were performed with another ligand system than terpy that

does not allow the formation of dimers with a labile coordination site like [Mn2III/IV(µ-

O)2terpy2(H2O)2]3+. These results already imply that the latter dimer might be a pre-stage to even

higher oxidized dimeric species, which could be relevant to the extraordinary performance in the

MnIIterpy catalyzed dye degradation. We are currently trying to clarify to what extend the

different in situ formed µ-oxo dimers and the mechanistic ideas of the water oxidizing model

system [Mn2III/IV(µ-O)2terpy2(H2O)2]3+/HSO5

- and its potential dimeric intermediates with either a

MnV=O or a MnIV-O• moiety, account for the observed reactivity in the MnIIterpy catalyzed dye

degradation with PAA.


163



naphthylazo)benzenesulfonate], 99 % was supplied by Sigma-Aldrich and recrystallised from a

Et2O/H2O mixture at 4 °C. Peracetic acid 39 wt. %, H2O2 30 %, as well as 2,2’:6’,2’’-terpyridine,

1,10-phenanthroline, 2-bis(2-hydroxyethyl)amino-2-(hydroxymethyl)-1,3-propanediol] and 2-(N-

morpholino)ethanesulfonic acid were of analytical grade and provided by Sigma-Aldrich. The

synthesis of [Mn2III/IVO2terpy2(H2O)2](NO3)3 · 6H2O was performed according to literature.[16]

[Mn2IV/IVO2terpy2(SO4)2] · 6H2O was synthesized as described before and crystallographically

characterized (Crystal data: monoclinic, space group: C2/c, a(Å) = 27.060, b(Å) = 9.020, c(Å) =

18.236, V(Å3) = 3539.29, α(°) = 90, β(°) = 127.331, γ(°) = 90, Z = 4.).[26a] All other chemicals were

commercially available (Acros Organics, Sigma-Aldrich) and were used without any further

purification.

GENERAL PROCEDURE AND PH JUMP TECHNIQUE. The manganese terpyridine compound was freshly

produced by dissolving the appropriate ratio of MnII:ligand in water with 10 % CH3CN before use

(MnII:terpy = 1:2). Stock solutions of [Mn2III/IVO2terpy2(H2O)2](NO3)3 and [Mn2

IV/IVO2terpy2(SO4)2]

were prepared in water and used immediately. All solutions were prepared in Millipore Milli-Q

purified water. To a freshly prepared 0.05 M sodium phosphate solution, an adequate amount of

NaOH was added to adjust the pH in a way that the subsequent addition of a specific

concentration of PAA gave the desired pH within less than one minute. The reaction was started

under isothermal conditions by addition of small aliquots of a concentrated manganese stock

solution together with Orange II to the PAA containing buffer solution. The catalytic reaction was

followed by in situ UV/Vis spectroscopy.

INSTRUMENTATION AND EQUIPMENT. All kinetic data were obtained by recording time-resolved UV-Vis

spectra using a Hellma 661.502 – QX quartz Suprasil immersion probe attached via optical cables

to a 150 W Xe lamp and a multi-wavelength J & M detector, which records complete absorption

spectra at constant time intervals. All kinetic measurements were carried out under pseudo-first-

order conditions. The pH of the PAA containing aqueous carbonate solution was carefully

measured and adjusted using a Mettler Delta 350 pH meter previously calibrated with standard

buffer solutions at two different pH values (4 and 10). The kinetics of the degradation reaction was

monitored at 484 nm. First order rate constants, where possible, were calculated using Specfit/32


164

and Origin (version 7.5) software. Perpendicular mode EPR spectra were recorded on an X-band

Joel Jes Fa 200 spectrometer equipped with a cylindrical mode cavity and a liquid helium cryostat.

Samples were taken from the investigated solutions and immediately frozen to quench the

reaction. The EPR measurements were performed in quartz tubes at 10 K (9.45 GHz, 1 mW

microwave power). Data analyses were done with the Jes-Fa Series software package.


[1] (a) Park, H.; Choi, W. J. Photochem. Photobiol. A 2003, 159, 241-247; (b) Brown, D., Laboureur

P. Chemosphere 1987, 12, 397-404; (b) Goszczynski, S.; Paszczynki, A.; Pasti-Grigsby, M. B.;

Crawford, R. L.; Crawford, D. L. J. Bacteriol. 1994, 176, 1339-1347; (c) Chivukula, M.; Spadaro,

J. T.; Renganathan, V. Biochem. 1995, 34, 7765-7772; (d) Spadaro, J. T.; Renganathan, V. Arch.

Biochem. Biophys. 1994, 312, 301-307; (e) Spadaro, J. T.; Isabelle, L.; Renganathan, V. Environ

Sci. Tech. 1994, 28, 1389-1393.

[2] (a) Zollinger, H. Color Chemistry, 3rd ed., Wiley-VCH: Weinheim, Germany, 2003; (b) Beach, E.

S.; Malecky, R. T.; Gil, R. R.; Horwitz, C. P.; Collins; T. J. Catal. Sci. Technol. 2011, 1, 437-443.

[3] Baughman, G. L.; Weber, E. J. Environ. Sci. Technol. 1994, 28, 267-276.

[4] (a) Ellis, W. C.; Tran, C. T.; Roy, R.; Rusten, M.; Fischer, A.; Ryabov, A. D.; Blumberg, B.; Collins,

T. J. J. Am. Chem. Soc. 2010, 132, 9774-9781. (b) Kluson, P.; Drobek, M.; Krejcikova, S.; Krysa,

J.; Kalaji, A.; Cajthaml, T.; Rakusan, J. Appl. Catal. B 2008, 80, 321-326. (c) Wieprecht, T.; Xia,

J.; Heinz, U.; Dannacher, J.; Schlingloff, G. J. Mol. Catal. A 2003, 203, 113-128. (d) Ramirez, J.

H.; Maldonado-Hódar, F. J.; Pérez-Cadenas, A. F.; Moreno-Castilla, C.; Costa, C. A.; Madeira, L.

M. Appl. Catal. B 2007, 75, 312-323. (e) Hage, R.; Iburg, J. E.; Kerschner, J.; Koek, J. H.; M.

Lempers, E. L.; Martens, R. J.; Racherla, U. S.; Russell, S. W.; Swarthoff, T.; van Vliet, M. R. P.;

Warnaar, J. B.; van der Wolf, L.; Krijnen, B. Nature 1994, 369, 637-639.

[5] (a) Oakes, J.; Gratton, P. J. Chem. Soc., Perkin Trans. 2 1998, 1857-1864. (b) Nadtochenko, V.;

Kiwi, J. J. Chem. Soc., Faraday Trans. 1997, 93, 2373-2378. (c) Oakes, J.; Gratton, P; Weil, I. J.

Chem. Soc., Dalton Trans. 1997, 3805-3809. (d) Chahbane, N.; Popescu, D.-L.; Mitchell, D. A.;

Chanda, A.; Lenoir, D.; Ryabov, A. D.; Schramm, K.-W.; Collins, T. J. Green Chem. 2007, 9, 49-


165

57. (e) Gilbert, B. C.; Lindsay-Smith, J. R.; Newton, M. S.; Oakes, J.; Ponts i Parts, R. Org.

Biomol. Chem. 2003, 1, 1568-1577.

[6] (a) Song, Y. J.; Lee, S. H.; Park, H. M.; Kim, S. H.; Goo, H. G.; Eom, G. H.; Lee, J. H.; Lah, M. S.;

Kim,Y.; Kim, S.-J.; Lee, J. E.; Lee, H.-I.; Kim, C. Chem. Eur. J. 2011, 17, 7336-7344. (b) Song, F.;

Wang, C.; Falkowski, J. M.; Ma, L.; Lin, W. J. Am. Chem. Soc. 2010, 132, 15390-15398. (c)

Palucki, M.; Finney, N. S.; Pospisil, P. J.; Guler, M. L.; Ishida, T.; Jacobsen, E. N. J. Am.Chem.

Soc. 1998, 120, 948-954.

[7] (a) Arunkumar, C.; Lee, Y.-M.; Lee, J. Y.; Fukuzumi, S.; Nam, W. Chem. Eur. J. 2009, 15, 11482-

11489. (b) Crestoni, M. E.; Fornarini, S.; Lanucara, F. Chem. Eur. J. 2009, 15, 7863-7866. (c)

Song, W. J.; Seo, M. S.; DeBeer George, S.; Ohta, T.; Song, R.; Kang, M.-J.; Tosha, T.; Kitagawa,

T.; Solomon, E. I.; Nam, W. J. Am. Chem. Soc. 2007, 129, 1268-1277. (d) Rose, E.; Andrioletti,

B.; Zrig, S.; Quelquejeu-Etheve, M. Chem. Soc. Rev. 2005, 34, 573-583.

[8] (a) Wieghardt, K.; Schmidt, W.; Nuber, B.; Weiss, J. Chem. Ber. 1979, 112, 2220-2230; (b)

Romakh, V. B.; Therrien, B.; Süss-Fink, G.; Shul’pin, G. B. Inorg. Chem. 2007, 46, 1315-1331; (c)

Mandelli, D.; Kozlov, Y. N.; Golfeto, C. C.; Shul'pin, G. B. Catal. Lett. 2007, 118, 22-29; (d)

Sibbons, K. F.; Shastri, K.; Watkinson, M.; Dalton Trans., 2006, 645-661; (e) Gilbert, B. C.;

Lindsay-Smith, J. R.; Mairata y Payeras, A.; Oakes, J. Org. Biomol. Chem. 2004, 2, 1176-1180;

(f) Bennur, T. H.; Srinivas, D.; Sivasanker, S.; Puranik, V. G. J. Mol. Cat. A , 2004, 219, 209-216;.

(g) Hage, R.; Kerschner, J. Trends in Inorg. Chem. 1998, 5, 145-159.

[9] De Boer, J. W.; Browne, W. R.; Brinksma, J.; Alsters, P. L.; Hage, R.; Feringa, B. L. Inorg. Chem.

2007, 46, 6353-6372.

[10] (a) Shilov, A. E.; Shul’pin, G. B.; Chem. Rev. 1997, 97, 2879-2932; (b) Sheldon, R. A. J. Chem.

Technol. Biotechnol. 1997, 68, 381-388.


E.; Gazzaz, H. A.; Rothbart, S.; Puchta, R.; van Eldik, R. Appl. Catal. B. 2009, 95, 179-191; (c)

Rothbart, S.; Ember, E.; van Eldik, R. Dalton Trans. 2010, 39, 3264-3272.

[12] (a) Richardson, D. E.; Yao, H.; Frank, K. M.; Bennett, D. A. J. Am. Chem. Soc. 2000, 122, 1729-

1739. (b) Bakhmutova-Albert, E. V.; Yao, H.; Denevan, D. E.; Richardson, D. E. Inorg. Chem.

2010, 49, 11287-11296.

[13] Rothbart, S.; Ember, E. E.; van Eldik, R. New J. Chem. 2012, DOI: 10.1039/C2NJ20852K.


166

[14] Mohr, R.; van Eldik, R. Inorg. Chem. 1985, 24, 3396-3399.

[15] Limburg, J.; Brudvig, G. W.; Crabtree, R. H. J. Am. Chem. Soc. 1997, 119, 2761-2762.

[16] Limburg, J.; Vrettos, J. S.; Liable-Sands, L. M.; Rheingold, A. L.; Crabtree, R. H.; Brudvig, G. W.

Science 1999, 283, 1524-1527.

[17] Baffert, C.; Romain, S.; Richardot, A.; Leprêtre, J.-C.; Lefebvre, B.; Deronzier, A.; Collomb, M.-

N. J. Am. Chem. Soc. 2005, 127, 13694-13704.

[18] Cady, C. W.; Crabtree, R. H.; Brudvig, G. W. Coord. Chem. Rev. 2008, 252, 444-455.

[19] Dubois, L.; Pécaut, J.; Charlot, M.-F.; Baffert, C.; Collomb, M.-N.; Deronzier, A.; Latour J.-M.

Chem. Eur. J. 2008, 14, 3013-3025.

[20] Collomb, M.-N.; Deronzier, A. Eur. J. Inorg. Chem. 2009, 14, 2025-2046.

[21] Thorp, H. H.; Brudvig, G. W. New J. Chem. 1991, 15, 479-490.

[22] Chen, H.; Tagore, R.; Olack, G.; Vrettos, J. S.; Weng, T.-C.; Penner-Hahn, J.; Crabtree, R. H.;

Brudvig, G. W. Inorg. Chem. 2007, 46, 34-43.

[23] Oakes, J; Gratton, P. J. Chem. Soc., Perkin Trans. 2 1998, 2201-2206.

[24] Feng, W.; Nansheng, D.; Helin, H. Chemosphere 2000, 41, 1233-1238.

[25 ] Offenhartz, P. O.; George, P.; Haight, G. P. J. Phys. Chem. 1963, 67,116-118.

[26] (a) Barton, D. H. R.; Choi, S.-Y.; Hu, B.; Smith, J. A.; Tetrahedron 1998, 54, 3367-3378; (b)

Tagore, R.; Chen, H.; Crabtree, R. H.; Brudvig, G. W. J. Am. Chem. Soc. 2006, 128, 9457-9465;

(c) Jacobsen, E. N. in: I. Ojima (Ed.), Catalytic Asymmetric Synthesis, VCH Publishers, New

York, 1993; (d) Katsuki, T. Coord. Chem. Rev. 1995, 140, 189-214; (e) Yachandra, V. K.; Sauer,

K.; Klein, M. P. Chem. Rev. 1996, 96, 2927-2950; (f) Ruttinger, W.; Dismukes, G. C. Chem. Rev.

1997, 97, 1-24.

[27] Limburg, J.; Vrettos, J. S.; Chen, H.; de Paula, J. C.; Crabtree, R. H.; Brudvig, G. W. J. Am. Chem.

Soc. 2001, 123, 423-430.

[28] (a) Hage, R.; Iburg, J. E.; Kerschner, J.; Koek, J. H.; Lempers, E. L. M.; Martens, R. J.; Racherla,

U. S.; Russel, S. W.; Swarthoff, T.; van Vliet, M. R. P.; Warnaar, J. B.; van der Wolf, L.; Krijnen,

B. Nature, 1994, 369, 637-639; (b) de Boer, J. W.; Browne, W. R.; Brinksma, J.; Alsters, P. L.;

Hage, R.; Feringa, B. L. Inorg. Chem. 2007, 46, 6353-6372.


167

[29] (a) Groves, J. T.; Stern, M. K. J. Am. Chem. Soc. 1987, 109, 3812-3814; (b) Groves, J. T. Stern,

M. K. J. Am. Chem. Soc. 1988, 110, 8628-8638.

[30] (a) Tetard, D.; Rabion, A.; Verlhac, J.-B.; Guilhem, J. J. Chem. Soc., Chem. Commun., 1995, 531-

532; (b) Menage, S.; Collomb-Dunand-Suathier, M.-N.; Lambeaux, C.; Fontecave, M. J. Chem.

Soc., Chem. Commun. 1994, 16, 1885-1886; (c) Vincent, J. M.; Menage, S.; Lambeaux, C.;

Fontecave, M. Tetrahedron Lett. 1994, 34, 6287-6290.

[31] Handervan, C.; Hage, R.; Feringa, B. L. Chem. Commun. 1997, 5, 419-420.

[32] Stamatis, A.; Doutsia, P.; Vartzouma, C.; Christoforidis, K. C.; Deligiannakis, Y.; Louloudi, M. J.

Mol. Cat. A 2009, 297, 44-53.

[33] (a) Limburg, J.; Vrettos, J. S.; Crabtree, R. H.; Brudvig, G. W.; de Paula, J. C.; Hassan, A.; Barra,

A.-L.; Duboc-Toia, C.; Collomb, M.-N. Inorg. Chem. 2001, 40, 1698-1703; (b) Wei, Z.; Cady, C.

W.; Brudvig, G. W.; Hou, H. J. M. J. Photochem. Photobio. B 2011, 104, 118125; (c) Zhang, F.;

Cady, C. W.; Brudvig, G. W.; Hou, H. J. M. Inorg. Chim. Acta 2011, 366, 128-133, (d) Lieb, D.;

Zahl, A.; Shubina, T. E.; Ivanović -Burmazović, I. J. Am. Chem. Soc. 2010, 132, 7282-7284.

[34] Tagore, R.; Crabtree, R. H.; Brudvig, G. W. Inorg. Chem. 2008, 47, 1815-1823.

[35] Tagore, R.; Crabtree, R. H.; Brudvig, G. W. Inorg. Chem. 2007, 46, 2193-2203.


168

6. Summary

169

6 SUMMARY

Although during the last decades a wide variety of manganese compounds has been

investigated towards their ability to activate peroxide for numerous homogeneous and

heterogeneous applications (summarized in Chapter 1.1), there are several aspects that require

further clarification. In fact, there is still an ongoing controversy among scientists on many

mechanistic issues in this area of research. In this context, the general goal of this thesis was to

contribute to a more detailed mechanistic understanding of these processes in aqueous solution

with the help of kinetic methods and spectroscopic techniques, such as UV/Vis- and EPR-

spectroscopy.

In Chapter 2 a fast and environmentally benign method for the oxidative degradation of a

typical industrial dye waste compound was introduced using H2O2 in conjunction with catalytic

amounts of simple manganese salts as catalyst precursors in a carbonate containing aqueous

solution under mild reaction conditions. The results provide an innovative, viable and simplistic

solution for efficient and clean oxidative degradation of highly stable organic dyes and shed light

on some key mechanistic features of this intriguing catalytic system.

The choice of the right buffer turned out to be a decisive factor. It was shown that the

oxidative degradation of the model substrate Orange II is only catalytic in carbonate containing

aqueous solution, since no other buffer solution showed comparable degradation rates. In the

absence of bicarbonate buffer MnIIaq is not able to activate H2O2 for the studied reaction under

identical reaction conditions (pH, ionic strength, concentrations etc.). By the use of excess

bicarbonate buffer a rapid pre-equilibrium between HCO3- and H2O2 leads to the formation of the

deprotonated peracid, HCO4-, as a key molecular entity and the actual oxidizing agent. The

generation of peroxymonocarbonate was further supported by 13C-NMR measurements on the

reaction of H2O2 with H13CO3-, which confirmed the known equilibrium constant for this reaction in

purely aquatic solution. However, the second order dependence of the observed overall dye

degradation rate on the total carbonate concentration in solution implies the involvement of

HCO3- ions in two catalytically relevant reaction steps. According to CV and UV/Vis spectroscopic

measurements the requirement of a second equivalent of HCO3- can be attributed to its potential

6. Summary

170

use as stabilizing ligand. Its electron donating abilities facilitate the formation and stabilization of

catalytically relevant, high-valent Mn-oxo species.

Kinetic investigations performed at different pH could provide relevant information about the

nature of the species involved in the catalytic reaction, viz. the formation of peroxycarbonate and

the deprotonation of the aquated MnII starting catalyst. On the basis of the systematic

investigation presented in Chapter 2, a reaction sequence for the underlying reaction mechanism

was proposed: In the first step a labile manganese-hydroxo precursor reacts with the in situ

formed nucleophilic peroxycarbonate in a rapid pre-equilibrium reaction. The so formed transient

intermediate, most likely a MnII-η2-peroxycarbonate complex, undergoes cleavage of the peroxo

bond to yield a high valent manganese-oxo compound. As a consequence of a heterolytic O-O-

bond scission a MnIV=O species is formed, as indicated by UV/Vis spectroscopic measurements and

confirmed by complementary EPR spectroscopy. Radical degradation pathways which are

expected for one-electron reactions could be excluded. Hence, a key feature of the proposed

reaction mechanism is that the overall oxidation of Orange II with the MnII precursor occurs in a

two-electron oxidation step and leads to the formation of a MnIV=O intermediate. This high valent

MnIV=O species represents a potent key catalytic intermediate which subsequently transfers the

oxo group to the substrate to reform the MnII catalyst and thereby close the catalytic cycle.

Another interesting aspect of a metal ion catalyst is the possible direct interaction of the latter

with the dye molecule itself. In titration experiments the formation of complexes with different

stoichiometries between the azo dye Orange II and MnII was depicted. This precedented

simultaneous σ,π - coordination of the organic dye was underlined by the presented additional

DFT studies and may significantly contribute to the stabilization of the MnII ion pre-catalyst in

slightly basic solution (Chapter 2).

In Chapter 3 the speciation of the catalytic hydrogen peroxide activation for homogeneous dye

degradation by MnII compounds was studied to answer the question to what extent µ-oxo-bridged

manganese dimers of mixed, higher valence nature contribute as reactive intermediates in the

catalytic cycle, since they represent frequently postulated intermediates for this type of reaction.

To have a closer look at the nature of putative intermediates, a comparative study on the

reactivity of a [Mn2III/IV(µ-O)2]3+ dimer and its readily accessible mononuclear analogous MnII

complex in the hydrogen peroxide assisted catalyzed oxidation of Orange II was performed

(Chapter 3). As the study progressed, verification of the obtained results for other dyes as model

6. Summary

171

substrates became increasingly relevant. Consequently p-nitrophenol, an extremely stable

aromatic dye, and Morin as polyphenolic representative were chosen to yield a broader spectrum

of potential substrates.

In summary, it was demonstrated by a well studied example that elaborated catalysts in higher

oxidation states are not always required for efficient catalysis. Instead, the same results are

achieved by the application of [MnII(bpy)2Cl2] as catalyst to oxidize various organic substrates.

Moreover, detailed kinetic investigations revealed that both catalysts show identical reactivity,

implying that the same reactive intermediate is formed regardless of the starting material used.

This was further confirmed by EPR-spectroscopic measurements, which disclosed the generation

of a high-valent MnIV=O intermediate along with a MnII precursor for the [Mn2III/IV(µ-O)2]3+-dimer

and the analogous mononuclear MnII complex upon addition of H2O2 in carbonate containing

solution. The large excess concentrations of H2O2, indispensable for the in situ formation of

peroxycarbonate, is responsible for the rapid reduction and cleavage of the dimer core. Thus, it is

concluded that both catalysts, i.e. [Mn2III/IV(µ-O)2(bpy)4]3+ and the analogous MnII monomer,

exhibit the same two-electron oxidation from a MnII precursor to a high-valent MnIV=O

intermediate.

In addition, it was possible to attain the same catalytic reactivity by simple in situ preparation

from a 1:2 ratio of MnII salt and the bipyridine ligand within the reaction mixture. A comparison

with the results in Chapter 2 confirmed that complexation of the MnII ion by a chelating ligand

generally benefits the reactivity. On the other hand, the stronger coordination at higher bipyridine

concentrations favors the undesired parallel decomposition of H2O2 above substrate oxidation.

Through UV/Vis-titration experiments it was determined that the dominating complex

stoichiometry reflects an 1:2 Mn:bpy compound under the applied reaction conditions.

Furthermore, the detailed investigations revealed very similar kinetics and intermediates as for

the simple MnII ion catalysis in the absence of a ligand, as described above for Chapter 2. Hence,

the general rate enhancement by the addition of the bipyridine ligand is attributed to a potential

stabilization of the reactive high-valent MnIV=O intermediate. However, this implies that the

observed second-order dependence of the observed dye oxidation rate on the total carbonate

concentration is not necessarily the result of bicarbonate coordination to the Mn II-center. In fact,

another plausible explanation might be the occurence of general acid-catalysis of the heterolytic

O-O bond cleavage in which the proton is provided by bicarbonate.

6. Summary

172

Chapter 4 describes the extension of the MnII ion system in Chapter 2 by another peroxide -

peracetic acid or peracetate (PAA), which in contrast to HCO4- is readily accessible. Preliminary

NMR measurements with 13C-labelled acetic acid permitted in situ monitoring of the formation

and dissociation of PAA and in particular of its decomposition at higher pH. The latter is often

assumed to intervene in sensible reproducibility of kinetic data, but the time scale of this

decomposition turned out to be negligible compared to the much faster reaction between the

MnII salt and PAA. In conclusion, the results led to the suggestion that a highly reactive mixture is

formed upon addition of PAA to a weakly basic solution of MnII and that free radical pathways do

not account for the observed performance. Thereby the application of the readily prepared

peroxide that is structurally related to peroxycarbonate, results in an enhanced reactivity towards

the dye substrate Orange II compared to H2O2 (HCO4-). This is a rare example of an investigation on

peracetic acid activation and most notably the first to specifically address the role of the

equilibrium amount of H2O2 omnipresent in PAA.

In the absence of dye substrate the reaction between MnII and PAA shows a biphasic reaction

behavior which is very sensitive to the applied experimental conditions. By careful choice of the

reaction parameters, it was possible to closely investigate the different reaction stages, the

composition of intermediates and their relevance to the catalytic dye degradation reaction by

means of UV/Vis and EPR spectroscopy. It was shown that during the initial stage a MnIV=O

compound is formed, whereas the second phase of the reaction is characterized by the sudden

formation of permanganate and colloidal manganese dioxide. A direct comparison of the catalytic

reactivity of the different intermediates towards the oxidative dye degradation provided more

detailed information on the nature of possible reactive species. Especially the initial reaction

phase shows extraordinary reactivity, whereas a tremendous loss of reactivity is observed

beginning with the formation of MnVIIO4-/MnIVO2. Further assignment of potential reactive species

was achieved by a comparison of the oxidative reactivity between Orange II and several high-

valent Mn-oxo anions, viz. manganate(VI) and hypomanganate(V), in the presence and absence of

PAA, since these oxo-anions often represent postulated intermediates in permanganate

oxidations. In addition, measurements with readily prepared solutions of colloidal MnIVO2 of

known particle size confirmed that heterogeneous pathways do not account for the catalytic

course.

6. Summary

173

However, the omnipresent equilibrium content of H2O2 in the PAA oxidant turned out to play a

decisive role as reductive species in this biphasic reaction. During the initial reaction stage it avoids

accumulation of the less reactive Mn-species, viz. MnVIIO4- and MnIVO2, and thus keeps a favorable

steady state concentration of the suggested reactive MnIV=O intermediate as long as H2O2 is

present in the reaction mixture. The corresponding reactivity tests and EPR measurements

evidenced the unequivocally higher reactivity in the first reaction phase, while a tremendous drop

in the catalytic performance was observed at the stage of MnVIIO4-. Moreover, selective reduction

of the less reactive intermediates in the second phase was achieved by application of small

amounts of additional H2O2 as reductant which thereby restored the catalytic activity. In this way

the lifetime of the catalytic system can be tuned by avoiding two undesired secondary processes,

namely precipitation of inactive MnIVO2 and the over-oxidation of the catalyst to MnVIIO4-. On the

other hand, a large initial excess content of H2O2 was shown to interfere with the in situ formation

of the MnIV=O intermediate, as well as the reactivity towards the dye substrate, by shifting the

complex “oxidation by PAA vs. reduction by H2O2” sequence back to the side of the MnII precursor.

Detailed kinetic studies of the oxidative catalytic degradation of the model substrate Orange II

in conjunction with the results described above provided the basis for a simplified underlying

reaction mechanism. Hence, for the MnII ion catalyzed peroxide activation a major mechanistic

conclusion can be drawn: For the MnII/H2O2/HCO3- system (Chapter 2) the excess H2O2 content,

which is required for more efficient HCO4- formation, enhances the back reaction of the in situ

formed MnIV=O to MnII, thus only a minor amount of catalytically active MnIV=O is constantly

present in the catalytic reaction mixture. The MnII catalyzed activation of PAA (Chapter 4) benefits

from the application of the prefabricated peroxide, viz. peracetate, so that more of the reactive

oxidant is present in solution. This results in a more efficient formation of the active Mn IV=O while

the minor equilibrium content of reducing H2O2 keeps it at a higher steady-state concentration as

compared to the combination of MnII/H2O2/HCO3-. In the case of PAA, a further oxidation to

higher-valent species along with enhanced deactivation of the transient MnIV=O by MnIVO2

precipitation is observed. For this reason small amounts of H2O2 are required to avoid these

secondary processes and thereby extend the catalytic lifetime. Consequently, both reaction types,

viz. MnII/H2O2/HCO3- vs. MnII/PAA, are characterized by reaction sequences of rapid oxidation and

reduction processes which basically lead to the same reactive MnIV=O intermediate. The striking

differences in the catalytic degradation reactivity apparently reflect the differing steady-state

availability of the reactive MnIV=O species.

6. Summary

174

Chapter 5 reports preliminary investigations on an efficient method for the catalytic oxidative

degradation of different highly stable organic dyes, such as Orange II, Tartrazine and p-nitrophenol

by using PAA and a MnIIterpy starting compound. Contrary to our former studies, it shows the

highest activity at pH ≈ 7 and the extraordinary efficiency is reflected by the catalytic TOFs which

were determined for the different substrates. The excess ligand concentration required for

catalysis is needed for a more efficient precursor formation. Moreover, it can be assumed that also

a partial degradation of the terpy ligand occurs during the catalytic reaction. Screening and

spectroscopic methods allowed us to study the reaction course and to shed light on some

mechanistic features. The omnipresent [H2O2]eq in PAA stock solution was shown to result in an

inverse dependence on the catalytic reaction, by causing a catalase-like H2O2 decomposition,

which also becomes noticeable at higher [PAA]. Consequently, the saturation kinetics observed for

higher oxidant concentrations can be ascribed to the thereby increased [H2O2]eq. The second order

dependence of the observed first order rate constants for the overall dye degradation on the

catalyst concentration implies the involvement of dimeric intermediates in the catalytic reaction.

This is confirmed by the UV/Vis and EPR measurements of the species formed in the absence

of substrate which strongly resemble earlier findings concerning the water oxidizing model system

[Mn2III/IV(µ-O)2terpy2(H2O)2]3+/HSO5

-. In analogy, it was found that the main species during the

catalytic phase is a µ-oxo bridged Mn2IV/IV dimer with only minor (≤ 10 %) amounts of the

corresponding µ-oxo Mn2III/IV dimer. Unfortunately it was not possible to resolve the complex

reaction behavior which leads to the mixture of different dimers by variation of the numerous

reaction parameters, i.e. [catalyst], [PAA], [H2O2], pH or temperature. However, very similar

UV/Vis spectral characteristics were observed when a combination of HSO5- and H2O2 at similar

conditions as those of the PAA system was used, which already implies parallels in the use of PAA

and HSO5- if the [H2O2]eq is not considered.

Both readily prepared dimer species with a [Mn2III/IV(µ-O)]3+ and [Mn2

IV/IV(µ-O)2]4+ core showed

rather sluggish reactivity towards the dye substrate, which leads to the suggestion that the excess

oxidant might result in the formation of even higher oxidized intermediates as known for the

[Mn2III/IV(µ-O)2terpy2(H2O)2]3+/HSO5

- system. There is evidence from preliminary experiments with

phen as ligand system that there are indeed parallels to the intermediates previously postulated

for [Mn2III/IV(µ-O)2terpy2(H2O)2]3+/HSO5

-. Contrary to terpy, the in situ formed µ-oxo bridged phen

dimers lack the essential feature of the terminal water ligands such that further oxidation is

6. Summary

175

hindered. However, to what extent the different in situ formed µ-oxo dimers and mechanistic

suggestions from the water oxidizing model system [Mn2III/IV(µ-O)2terpy2(H2O)2]3+/HSO5

- (i.e. its

potential dimeric intermediates with either a MnV=O or a MnIV-O• moiety) account for the

extraordinary reactivity in the MnIIterpy catalyzed dye degradation with PAA, remains to be

clarified.

In summary, this work provides further insights into the mechanism of action, intermediates

and characteristic features of the activation of peroxides by manganese compounds with and

without a defined ligand system which serve as simplest representatives for the numerous

manganese-based bleaching and oxygenation catalysts.

6. Summary

176

7. Zusammenfassung

177

7 ZUSAMMENFASSUNG

Obwohl in den letzen Jahrzehnten eine Vielzahl verschiedenster Manganverbindungen auf ihre

Fähigkeiten zur Aktivierung von Peroxiden für zahlreiche homogene und heterogene

Anwendungen untersucht worden sind, sind viele Einzelheiten bisher noch nicht geklärt. Vielmehr

bietet dieser Bereich nach wie vor Raum für kontroverse Diskussionen innerhalb der

wissenschaftlichen Gemeinde bezüglich verschiedenster mechanistischer Details. In diesem

Zusammenhang war das allgemeine Ziel der vorliegenden Arbeit mittels kinetischer und

spektroskopischer Methoden, wie UV/Vis- und EPR-Spektroskopie, zu einem tieferen

mechanistischen Verständnis dieser katalytischen Prozesse, im Besonderen in wässriger Lösung,

beizutragen.

In Kapitel 2 wurde eine schnelle und umweltfreundliche Methode zum oxidativen Abbau eines

beispielhaften Abfallprodukts der Farbenindustrie unter milden Reaktionsbedingungen mittels

H2O2 und katalytischen Mengen des simplen MnII-Salzes vorgestellt. Die Ergebnisse stellen eine

innovative, praktikable Lösung zum oxidativen Abbau stabiler Farbstoffe dar und geben Aufschluss

über wesentliche mechanistische Merkmale dieser verblüffenden Reaktion.

Die Verwendung des richtigen Puffers erwies sich als ausschlaggebend in diesem katalytischen

System. Es konnte gezeigt werden, dass die oxidative Zersetzung des Modellsubstrates Orange II

nur in wässriger HCO3-/CO3

2--Lösung katalytisch verläuft, da kein anderes Puffersystem

vergleichbare Zersetzungsgeschwindigkeiten zeigte und aquatisierte MnII-Ionen alleine nicht in der

Lage sind, H2O2 unter vergleichbaren Bedingungen (pH, Ionenstärke, Konzentrationen) für die zu

untersuchende Reaktion zu aktivieren. Durch die Verwendung eines Überschusses an HCO3-/CO3

2-

hingegen wird die rasche Einstellung eines vorgelagerten Gleichgewichts zwischen HCO3- und H2O2

ermöglicht, was zur Bildung von HCO4-, der einfach deprotonierten Form der Perkohlensäure, als

eigentlichem Oxidationsmittel führt. Die Peroxycarbonat-Bildung wurde außerdem mittels 13C-

NMR-Messungen der Reaktion zwischen H2O2 und H13CO3- untersucht, wobei die Ergebnisse die

bekannte Gleichgewichtskonstante der Reaktion für eine rein wässrige Lösung bestätigen. Jedoch

impliziert eine Abhängigkeit zweiter Ordnung der beobachteten Geschwindigkeit der katalytischen

Bleichreaktion vom Gesamtgehalt an HCO3-/CO3

2- eine Beteiligung des Hydrogencarbonat an zwei

katalytisch relevanten Reaktionsschritten. Cyclovoltammetrische sowie UV/Vis-spetroskopische

7. Zusammenfassung

178

Messungen deuten auf HCO3- als potentiellen Liganden hin, da es durch seine elektronischen

Eigenschaften die Bildung und Stabilisierung katalytisch relevanter, hochvalenter Mn-oxo-

Intermediate erleichtert.

Durch detaillierte kinetische Untersuchungen bei verschiedenen pH-Werten konnten

grundlegende Informationen über die Natur der oxidierenden Spezies gewonnen werden. Auf

Basis der in Kapitel 2 beschriebenen Ergebnisse wurde ein Reaktionsschema vorgeschlagen, das

den zugrundeliegenden Mechanismus reflektiert: Im ersten Schritt reagiert eine labile MnII-

Hydroxo-Spezies mit dem in situ gebildeten nukleophilen Peroxycarbonat in einem raschen

vorgelagerten Gleichgewicht. Die dadurch gebildete Verbindung, wahrscheinlich ein MnII-η2-

Peroxycarbonat-Komplex, ergibt nach der Spaltung der Peroxo-Bindung ein hochvalentes Mn-oxo-

Intermediat. Als Konsequenz eines heterolytischen O-O-Bindungsbruchs ergibt sich eine MnIV=O-

Spezies, wie aus den UV/Vis-spektroskopischen Messungen und den zusätzlichen EPR-Ergebnissen

zu schließen ist. Reaktionswege, welche auf die Beteiligung freier Radikale hindeuten, wie sie im

Falle einer 1e--Reaktion zu erwarten sind, konnten ausgeschlossen werden. Folglich ist der

Schlüsselschritt des katalytischen, oxidativen Abbaus von Orange II die 2e--Oxidation einer MnII-

Vorstufe zu einem MnIV=O-Intermediat. Dieses kurzlebige MnIV=O-Intermediat stellt eine

potentiell reaktive Spezies dar, welche in der Lage ist, den Sauerstoff auf das Substrat zu

übertragen und somit den katalytischen Zyklus unter Rückbildung der MnII-Vorstufe zu schließen.

Ein weiterer interessanter Aspekt eines Metallionenkatalysators ist die mögliche

Wechselwirkung des freien Ions mit dem Farbstoffmolekül selbst. Durch UV/Vis-

spektrophotometrische Titrationen konnte die Bildung von MnII-(Orange II)-Komplexen

verschiedener Stöchiometrien gezeigt werden. Die wohlbekannte, gleichzeitige σ,π –Bindung des

organischen Farbstoffes wurde zusätzlich durch DFT-Rechnungen einer derartigen

Koordinationsform, welche auch eine mögliche Rolle bei der Stabilisierung der freien MnII-Ionen in

schwach basischem Medium spielt, bestärkt.

Kapitel 3 schildert die weitere Ausführung der homogenen Aktivierung von Wasserstoffperoxid

durch MnII-Verbindungen, wobei die Frage geklärt werden sollte, inwieweit höhervalente, µ-oxo

verbrückte Mn-Verbindungen als reaktive Spezies in den katalytischen Zyklus involviert sind, da

diese häufig postulierte Intermediate in derartigen Reaktionen darstellen. Hierfür wurde eine

vergleichende Untersuchung der Reaktivität eines [Mn2III/IV(µ-O)2]3+-Dimers und des

entsprechenden mononuklearen MnII-Analogons als Katalysatoren für den oxidativen Abbau von

7. Zusammenfassung

179

Orange II durchgeführt. Im Verlauf der Studie gewann es zunehmend an Bedeutung, die

erhaltenen Ergebnisse auch für andere Farbstoffarten als Modellsubstrate zu verifizieren. Mit der

Wahl von p-Nitrophenol, einem hoch stabilen aromatischen Farbstoff, und Morin, exemplarisch

für polyphenolische Farbstoffe, ergab sich ein breites Spektrum potentieller Substrate.

Insgesamt war es möglich, anhand dieser wohlbekannten Verbindungen zu zeigen, dass

effiziente Katalyse nicht immer derartig komplexer Katalysatoren bedarf. So können

verschiedenste organische Substrate ebenso mittels des einfachen [MnII(bpy)2Cl2]-Katalysators

oxidativ zersetzt werden. Die detaillierten kinetischen Untersuchungen ließen vielmehr erkennen,

dass bei gleichem Mangangehalt beide Katalysatoren identische Reaktivität aufweisen, was

wiederum auf die vom Ausgangsmaterial unabhängige Bildung der gleichen reaktiven Spezies

hindeutete. Dies konnte mittels EPR-Spektroskopie bestätigt werden, welche die Bildung eines

MnIV=O-Intermediates neben der Präsenz einer mononuklearen MnII-Verbindung unter den

entsprechenden Reaktionsbedingungen für beide Systeme ([Mn2III/IV(µ-O)2]3+-Dimer und der

analogen MnII-Verbindung) offenbarte. Der für die in situ Bildung des Peroxycarbonats

unverzichtbare große Überschuss an H2O2 zeichnet verantwortlich für die rasche Reduktion und

Spaltung des Dimer-Kerns. Durch die Verwendung von tBuOH als Radikalfänger konnte die

Beteiligung freier Radikale ausgeschlossen werden. Daraus folgt, dass beide Katalysatoren, also

[Mn2III/IV(µ-O)2(bpy)4]3+ und das analoge MnII-Monomer, das selbe katalytische Hauptmerkmal

aufweisen, nämlich eine 2e--Oxidation der MnII-Vorstufe zu einem hochvalenten MnIV=O-

Intermediat.

Des Weiteren wurde eine identische katalytische Aktivität durch simple in situ Herstellung des

Katalysators aus einem 1:2 Verhältnis von MnII-Salz und Bipyridin-Ligand erzielt. Der Vergleich mit

den Ergebnissen für unkomplexiertes MnII zeigt, dass die Komplexierung des MnII-Ions durch einen

Chelatliganden allgemein von Vorteil ist, da höhere katalytische Zersetzungsgeschwindigkeiten

beobachtet werden. Andererseits resultiert aus einem größeren bpy-Gehalt eine höhere

Komplexierung, und damit einhergehend eine Verstärkung der H2O2-Zersetzung als unerwünschte

Nebenreaktion zu Lasten des oxidativen Farbstoffabbaus. Durch UV/Vis-Titration konnte gezeigt

werden, dass die dominierende Komplexstöchiometrie unter den Reaktionsbedingungen der

Studie, die einer 1:2 Mn:bpy-Verbindung ist.

Die Untersuchungen lassen weiterhin auf sehr ähnliche Intermediate und ein ähnliches

mechanistisches Verhalten wie im Falle der in Kapitel 2 geschilderten einfachen MnII-Ionen ohne

7. Zusammenfassung

180

Chelatliganden schließen. Als Folge wird die allgemeine Steigerung der katalytischen Aktivität

durch Zusatz des idealen Verhältnisses an Bipyridin als Ligand der dadurch ermöglichten

Stabilisierung des reaktiven MnIV-Intermediates zugeschrieben. Somit ergibt sich jedoch, dass die

Abhängigkeit zweiter Ordnung der beobachteten Geschwindigkeitskonstanten der katalytischen

Farbstoffzersetzung nicht zwingend das Resultat der Koordination von HCO3- an das MnII-Zentrum

ist. Tatsächlich stellt die Möglichkeit der allgemeinen Säurekatalyse, welche durch Bicarbonat als

Protonenquelle die heterolytische Spaltung der Peroxo-Bindung begünstigt, einen weiteren

Erklärungsansatz dar.

Kapitel 4 schildert die Erweiterung des MnII-Ionen-Systems aus Kapitel 2 um ein weiteres

Peroxid ― Peroxyessigsäure bzw. Peroxyacetat (PAA) ― welches im Gegensatz zu HCO4-

kommerziell zugänglich ist. Vorausgehende NMR-Messungen mit 13C-markierter Essigsäure

ermöglichten zusätzlich eine in situ Analyse der Bildung und Dissoziation, aber vor allem der

Zersetzung von PAA bei höherem pH. Letztere wird häufig als Grund für die schlechte

Reproduzierbarkeit kinetischer Messdaten genannt, doch der zeitliche Rahmen dieser Zersetzung

erwies sich als vernachlässigbar gegenüber der Geschwindigkeit der untersuchten Reaktion von

PAA mit dem MnII-Salz. Die Ergebnisse lassen auf eine nicht-radikalische Bildung eines

hochreaktiven Gemisches durch Zugabe eines MnII-Salzes zu einer schwach basischen Lösung von

PAA schließen. Die Verwendung eines bereits zugänglichen Peroxids, welches Peroxycarbonat

strukturell ähnelt, ermöglicht eine um mehrere Größenordnungen gesteigerte Reaktivität

gegenüber H2O2 (bzw. HCO4-) unter idealen Reaktionsbedingungen. Die Ergebnisse stellen eines

der seltenen Beispiele einer Untersuchung der Aktivierung von PAA dar und behandeln erstmals

ausführlich die Rolle des in PAA allgegenwärtigen H2O2, so dass ein tieferes mechanistisches

Verständnis der komplexen Wechselwirkung mit MnII erhalten wurde.

Die Reaktion zwischen MnII und PAA zeigt in Abwesenheit eines Substrates einen zweiphasigen

Reaktionsverlauf, der sich als äußerst empfindlich bezüglich der verwendeten

Reaktionsbedingungen und Konzentrationen erwies. Durch eine vorsichtige Wahl dieser

Bedingungen war es möglich, mittels UV/Vis- und EPR-Spektroskopie die verschiedenen

Reaktionsphasen, die entsprechende Intermediatzusammensetzung sowie deren Relevanz für die

katalytische Farbstoffzersetzung genauer zu bestimmen. Die Analyse zeigte die Bildung einer

MnIV=O-Spezies innerhalb der ersten Reaktionsphase, während die zweite Reaktionsphase durch

abrupte Bildung von kolloidalem Braunstein und Permanganat gekennzeichnet ist. Eine gezielte

7. Zusammenfassung

181

Untersuchung der Reaktivität bezüglich des oxidativen Farbstoffabbaus der verschiedenen

Reaktionsintermediate erlaubte eine genauere Eingrenzung möglicher reaktiver Formen. Hierbei

zeigt im Besonderen die erste Phase sehr hohe katalytische Aktivität, wohingegen ein erheblicher

Reaktivitätsverlust mit Beginn der Permanganat/MnIVO2-Bildung einhergeht. Ein Vergleich der

Reaktivität mit und ohne PAA gegenüber dem Modellsubstrat für die häufig als Intermediate in

Permanganat Oxidationen postulierten, hochvalenten Mangan-oxo-Anionen Manganat(VI) und

Hypomanganat(V) lieferte eine weitere Eingrenzung der reaktiven Spezies. Zusätzlich konnte ein

heterogener katalytischer Verlauf durch Experimente mit kolloidalen MnIVO2-Lösungen

ausgeschlossen werden.

Als für den zweiphasigen Reaktionsverlauf von MnII und PAA ausschlaggebende Komponente

offenbarte sich die in PAA-Lösungen enthaltene Gleichgewichtskonzentration an H2O2. Diese

verhindert als Reduktionsmittel während der ersten Reaktionsphase nachweislich die Bildung von

Permanganat und kolloidalem MnIVO2, wodurch ein stabiler stationärer Zustand der vermutlich

reaktiven MnIV=O-Spezies erreicht wird, solange H2O2 in der Reaktionsmischung zugegen ist. Die

entsprechenden Reaktivitätsstudien und EPR-Untersuchungen zeigten eindeutig die Korrelation

zwischen den intermediären Spezies und der anfänglich katalytisch aktiven Phase des MnII/PAA-

Systems auf, während ein erheblicher Aktivitätsverlust mit dem Beginn der zweiten Phase

einhergeht. Darüber hinaus war es möglich, die weniger aktiven Spezies der zweiten

Reaktionsphase durch Zugabe kleinerer Mengen an H2O2 selektiv zu reduzieren und somit bei

entsprechend großem PAA-Überschuss die katalytische Aktivität des Systems wieder herzustellen.

Auf diese Art und Weise konnten die unerwünschten Sekundärprozesse, nämlich Desaktivierung

durch MnIVO2-Bildung und „Überoxidation“ des Katalysators zu MnVIIO4-, verhindert werden, was

eine verbesserte Feineinstellung der Lebensdauer des katalytischen Systems ermöglicht.

Andererseits konnte unter Beweis gestellt werden, dass größere Mengen an zusätzlichem H2O2

sowohl die Bildung des vermutlich reaktiven MnIV=O als auch die katalytische Abbaureaktion

behindern, indem die komplexe Reaktionssequenz „Oxidation durch PAA vs. Reduktion durch

H2O2“ weiter auf die Seite der MnII-Ausgangsverbindung geschoben wird.

Detaillierte kinetische Untersuchungen der katalytischen, oxidativen Abbaureaktion des

Modellsubstrats Orange II lieferten in Verbindung mit bereits geschilderten Ergebnissen die Basis

für eine vereinheitlichte mechanistische Vorstellung der zugrundeliegenden Reaktionen. Somit

ergibt sich folgender größerer mechanistischer Zusammenhang für die Peroxidaktivierung durch

7. Zusammenfassung

182

MnII-Salze: Im Falle des MnII/H2O2/HCO3--Systems begünstigt der für die Peroxycarbonat-Bildung

benötigte große Überschuss an H2O2 die Rückreaktion des in situ gebildeten MnIV=O zu MnII, so

dass nur sehr geringe Mengen des katalytisch aktiven MnIV=O dauerhaft in der Reaktionsmischung

verfügbar sind. Dagegen profitiert die MnII-katalysierte Aktivierung von PAA (Kapitel 4) von der

Verwendung des vorgefertigten Peroxides PAA, indem durch den geringeren H2O2-Gehalt die

stationäre Verfügbarkeit des MnIV=O erhöht ist. Andererseits erfolgt im Falle des PAA auch eine

merkliche Weiteroxidation zu höhervalenten Spezies sowie eine Desaktivierung des intermediären

MnIV=O durch Bildung von Braunstein, weshalb geringe Mengen H2O2 benötigt werden, um diese

Sekundärprozesse zu unterbinden und somit die katalytische Lebensdauer zu verlängern.

Zusammengefasst handelt es sich daher bei beiden Reaktionstypen, also MnII/H2O2/HCO3- und

MnII/PAA, um eine komplexe Reaktionsfolge rascher Oxidations- und Reduktionsprozesse, welche

zu demselben aktiven Intermediat führt und in denen lediglich die unterschiedliche stationäre

Verfügbarkeit dieser Spezies für die unterschiedliche Reaktivität verantwortlich zeichnet.

Kapitel 5 beschreibt die anfänglichen Untersuchungen einer neuen effizienten Methode des

katalytischen oxidativen Abbaus verschiedener sehr stabiler organischer Farbstoffe unter

Verwendung von PAA und einer MnIIterpy-Verbindung. Im Gegensatz zu unseren früheren Studien

zeigt das System höchste Aktivität bei pH ≈ 7, und die außergewöhnliche Effizienz wird durch die

für die verschiedenen Substrate bestimmten Wechselzahlen untermauert. Die

Überschusskonzentration an terpy-Ligand, welche für die Abbaukatalyse benötigte wird, lässt sich

der damit einhergehenden effizienteren Bildung der mono-terpy-Katalysatorvorstufe zuschreiben.

Darüber hinaus wird angenommen, dass während des katalytischen Reaktionsverlaufs ebenfalls

eine oxidative Zersetzung des Liganden stattfindet. Verschiedene Arten der spektroskopischen

Reaktionsverfolgung ermöglichten es einige mechanistische Teilaspekte näher zu beleuchten. Es

konnte gezeigt werden, dass die allgegenwärtige [H2O2]eq des Oxidationsmittels PAA in einer

inversen Abhängigkeit der katalytischen Abbaureaktion resultiert. Dies erfolgt durch eine

katalaseähnliche H2O2-Zersetzung, welche sich auch bei höheren [PAA] bemerkbar macht. Damit

ergibt sich auch, dass die festgestellte Sättigungskinetik bei höheren Konzentrationen des

Oxidationsmittels eigentlich der damit erhöhten [H2O2]eq zuzuschreiben ist. Eine mögliche

Teilnahme dimerer Intermediate an entscheidenden Reaktionschritten des katalytischen

Farbstoffabbaus wird durch die Abhängigkeit zweiter Ordnung der beobachteten

Geschwindigkeitskonstanten nahegelegt.

7. Zusammenfassung

183

Diese Vermutung wurde durch UV/Vis- und EPR-Messungen bestätigt, welche sehr stark an die

Ergebnisse des [Mn2III/IV(µ-O)2terpy2(H2O)2]3+/HSO5

--Modellsystems zur katalytischen

Wasseroxidation erinnern. Analog hierzu fanden wir, dass die Hauptspezies während der

katalytisch aktiven Phase ein µ-oxo verbrücktes Mn2IV/IV-Dimer ist. Des Weiteren wurde ein

geringerer Prozentsatz (≤ 10 %) des entsprechenden µ-oxo-verbrückten Mn2III/IV-Dimers gefunden.

Leider gelang es nicht durch eine Variation der unterschiedlichen Reaktionsparameter, wie [Kat],

[PAA], [H2O2], pH oder Temperatur, das komplexe Reaktionsverhalten, welches zu Bildung der

Mischung beider Dimere führt, besser aufzulösen. Jedoch zeigte eine Kombination aus HSO5- und

H2O2 bei vergleichbaren Konzentrationen sehr ähnliche UV/Vis-Charakteristika, was auf Parallelen

in der Verwendung von PAA und HSO5- schließen lässt, sofern die [H2O2]eq außer Betracht gelassen

wird.

Die bereits zuvor hergestellten dimeren Spezies mit einem [Mn2III/IV(µ-O)]3+- bzw. [Mn2

IV/IV(µ-

O)2]4+-Kern zeigten beide eine eher geringe Aktivität gegenüber dem Farbstoffsubstrat, was die

Vermutung nahelegt, dass der Überschuss an Oxidationsmittel die Bildung weiterer hochvalenter

Intermediate, wie vom [Mn2III/IV(µ-O)2terpy2(H2O)2]3+/HSO5

- System bekannt, ermöglicht.

Tatsächlich lassen einige vorläufige Experimente mit phen als Ligandensystem darauf schließen,

dass Parallelen zu den hierfür postulierten Intermediaten bestehen. Im Gegensatz zu den terpy-

Verbindungen verfügen die entsprechenden in situ gebildeten phen-Dimere nicht über die

ausschlaggebenden terminalen H2O-Liganden, so dass ein weiterer Oxidationsschritt in ihrem Falle

erschwert ist. Inwieweit genau die in situ gebildeten µ-oxo-Dimere bzw. die mechanistischen

Vorstellungen des Modellsystems zur katalytischen Wasseroxidation, d.h. insbesondere die

potentiellen intermediären Dimere mit einer MnV=O-oder MnIV-O•-Einheit, für die ausgesprochen

hohe Reaktivität im MnIIterpy-katalysierten Farbstoffabbau mit PAA verantwortlich zeichnen, muss

hierbei noch geklärt werden.

Zusammengenommen liefert diese Arbeit weitere Erkenntnisse über Intermediate, besondere

Merkmale und den zugrundeliegenden Reaktionsmechanismus der Peroxidaktivierung durch

Manganverbindungen mit und ohne definiertes Ligandensystem, welche als einfache Modelle der

zahlreichen Mangan basierten Bleich- und Oxidationskatalysatoren dienen.

mechanistic investigation on the activation of …€¦ · sabine rothbart, erika ember and rudi...

Documents