Investigation of the Substitution Reaction Pathways of ...

Investigation of the Substitution Reaction Pathways of

Antidiabetic Chromium(III) Supplements

by

Mohammed Kabir Uddin

BSc (Hons.), MSc, MPhil, MS

A Thesis Submitted to the

School of Engineering & Information Technology

in Fulfillment of the Requirements

for the Degree of

Doctor of Philosophy

in Chemical and Metallurgical Engineering and

Chemistry

Murdoch University

Feb 2017

Murdoch Western Australia

ii

Declaration

I declare that the work presented in this thesis is my own account of my

research and contains as its main content work which has not previously

been submitted for a degree at any other university or institution.

......................................................

(Mohammed Kabir Uddin)

iii

Abstract

Chromium(III) complexes are regularly included in nutritional supplements and are

reported to lower blood glucose levels in type 2 diabetics. However, questions have

been raised about potential toxicity. Understanding the reactivity of these complexes is

key to understanding their behaviour in physiological environments. Therefore, the

central aim of this work is to explore the speciation of these complexes under different

conditions using both computational and experimental methods.

The computational studies involved a detailed mechanistic investigation of the

substitution reaction pathways of several chromium(III) nutritional supplements (e.g.

chromium(III) chloride, glycinato-chromium(III) ([Cr(gly)x(H2O)6-2x](3-x)+

where x = 1 ‒

3) and their conjugate bases and chromium(III) picolinate (Cr(pic)3) using state-of-the-

art computational chemistry techniques.

A number of mechanistic aquation pathways were investigated including associative

interchange (Ia), dissociative interchange (Id), associative (A) and dissociative (D), to

predict the overall activation enthalpies of these processes. Our initial computational

investigations have focussed on the mechanistic pathways for aquation of chromium(III)

chloride under acidic (stomach) and mild (duodenum and blood) conditions. DFT

iv

calculations reveal that aquation of CrCl3 is slow in an acidic environment and this

limits speciation of the complex. However, at mild pH, conjugate base complexes easily

form, leading to a significant number of species that may be involved in halide-water

exchange pathways.

The activation enthalpies for aquation of tris-glycinatochromium(III) and its

conjugate base at mild and physiological pHs via the dissociative (D) mechanism are in

consistent agreement with the experimental results. Aquation of [Cr(pic)3]0 initially

proceeds via a dissociative ring-opening step to form the [Cr(pic)2(H2O)(picH2)]+

intermediate in acidic media. The reaction may then proceed via two possible pathways

involving either stepwise (two-step) or concerted (one-step) mechanisms. Throughout

these studies solvent calculations were performed using the polarizable continuum

model (PCM). However, the inclusion of explicit solvation was also found to be

important for many processes.

The experimental component of the study involved the synthesis of chromium(III)

nutritional supplements and characterization (elemental analysis, atomic absorption

spectroscopy, NMR, HPLC, ATR-FTIR, Raman, and ESI−MS spectroscopy).

Investigation of solvent and pH effects using UV−Vis monitoring (pH ~3.0 to ~8.5) and

electron paramagnetic resonance (in both solid and frozen state) complemented the

theoretical studies of the speciation and interaction of Cr(III) metal with amino acid

ligands. The electrochemical properties of [Cr(gly)3]0 were also investigated for the first

time using cyclic voltammetry (CV) and it was found that this complex can undergo

reversible electron transfer, which may play a role in its biological activity.

v

Dedication

To

the Memory of

My father.

I also would like to dedicate this thesis to:

My mother

My wife and my daughters

Tanisha

Maisha

vi

Acknowledgements

I am very grateful to my mother and lovely wife N. Nahar for their enormous

support, inspiration and encouragement during my journey towards achieving my PhD

degree.

I would like to express my deepest gratitude to my principal supervisor Dr. David J.

Henry, for his kind advice, indispensable guidance, encouragement, support, invaluable

suggestions, and careful supervision throughout the progress of my research work. I

would like to thank my secondary supervisors Dr. Leonie Hughes and Dr. David Ralph

(Postgraduate Research Director of Engineering and Information Technology) for their

support, help and encouragements throughout my PhD program.

I would also like to thank my supervisory committee head, Prof. Parisa Arabzadeh

Bahri, for her valuable suggestions and support throughout the progress of my PhD

program. I am indebted and have the great pleasure to thank my collaborator and Master

of Science (MS) supervisor Prof. Raymond A. Poirier FCIC, Department of Chemistry,

Memorial University, Canada for his valuable suggestions and cooperation during the

progress of my research work.

vii

In addition, I would like to acknowledge the School of Engineering and Information

Technology for the Maintenance Funds and the Graduate Research Studies at Murdoch

University for the Conference Travel Award and the Australian Government Research

Training Program Scholarship (i.e. IPRS and APA) and Murdoch top-up for financial

support and all the laboratory instrumental supports during my candidature. I would also

like to thank Prof. Bogdan Dlugogorski, Dean of Engineering and Information

Technology for providing access to the EPR and TGA-DSC facilities. I would like to

acknowledge the following individuals, for their technical assistance during my

laboratory learning experiences for research techniques and expertise in my research:

Andrew Foreman, Dr. Marc Hampton, Dr. Juita, David Zeelenberg, Jacqueline, Saijel

Jani, and Caitlin Sweeney. I am grateful to the Australian National Computational

Infrastructure (NCI) facility and the Atlantic Computational Excellence Network

(ACENET) for computer time.

I would like to give special thanks to my colleagues and friends Dr. Alrawashdeh,

Dr. Delgado, Nick, Lou, Dr. Mahbub, Dr. Kuba, Moktadir, Dastogeer, Ahsan Habib,

Toriqul Islam, and Dr. Michelle, for their support in different aspects during my

candidature especially in the thesis writing.

Finally, I would like to extend a special thanks to my brothers, sisters, and parents-

in-law, for their encouragement and the support of my family. I am very grateful to Dr.

Azad Hasan Yusuf, Dr. Kamrun Nahar and Shamsun Nahar, for their support,

inspiration and encouragement throughout the completion of my PhD candidature. I am

also grateful for my beautiful daughters Tanisha and Maisha, for their chats on Skype,

which made me feel good during my PhD study in Australia.

viii

Table of Contents

Title Page

i

Declaration

ii

Abstract

iii

Dedication

v

Acknowledgements

vi

Table of Contents

viii

List of Symbols and Abbreviations

xii

List of Publications xv

Attribution xvi

Chapter 1 Introduction and Literature Review………………………….. 1

1.1 General Introduction 1

1.2 Literature Review

3

1.2.1 Investigation on the Hydrolysis of Cr(III) Complexes 3

1.2.2 Investigation on the Oxidation of Cr(III) Complexes 5

1.2.3 Formation and Hydrolysis of Cr(III)-Biomolecule Species 7

1.2.4 Theoretical Studies of M(III) Metal Complexes 13

1.3 Theoretical Background 18

1.3.1 Theoretical Methods 18

1.3.1.1 Transition State Theory 19

1.3.1.2 Electronic Structure Theory 20

1.3.1.3 Density Functional Theory 22

ix

1.3.1.4 Solvation Models 26

1.3.1.5 Correlation-Consistent Basis Sets 27

1.4 Aims and Objectives 28

1.5 Significance of the Research 29

1.6 Methodology 30

1.6.1 Computational Procedure 30

1.6.2 Experimental Methods 32

1.7 Outline of the Thesis 33

1.8 References

35

Chapter 2 Mechanistic Investigation of Halopentaaquachromium(III)

Complexes: Comparison of Computational and Experimental Results…...

39

Abstract 39

Introduction 40

Computational Methods 47

Results and Discussion 49

Conclusions 68

Supporting Information Summary 69

Acknowledgements 69

References 70

Chapter 3 Mechanistic Study of the Aquation of Nutritional Supplement

Chromium Chloride and other Chromium(III) Dihalides…………………

73

Abstract 73

Introduction 74

Computational Methods 79


Conclusions 99


Acknowledgements 100

References 101

x

Chapter 4 Further Theoretical Studies of the Aquation of Chromium(III)

Chloride Nutritional Supplement: Effect of pH and Solvation………............ 104

Abstract 104

Introduction 105


Conclusions 137



References 138

Chapter 5 Investigation of the Spectroscopic, Thermal and

Electrochemical Properties of Tris-glycinato)chromium(III)……………… 141

Abstract 141

Introduction 142


Conclusions 161



References 162

Chapter 6 Investigation of Mono‒, Bis‒, and Tris-glycinato

chromium(III): Comparisons of Computational and

Experimental Results…………………………………………………….……

166

Abstract 166

Introduction 167

Experimental 171


Conclusions 196



References 197

xi

Chapter 7 Electron Paramagnetic Resonance Spectroscopy and New

Insights into the Substitution Reaction of Chromium(III) Picolinate……...

200

Abstract 200

Introduction 201

Experimental 206


Conclusions 225



References 226

Chapter 8 Summary and Future Directions………………………….……

231

Appendices……………………………...………………………………………

237

Appendix A 237

Appendix B 239

Appendix C 244

Appendix D 247

xii

List of Symbols and Abbreviations

Cr Chromium

Cr3+

or Cr(III) Trivalent Cr ions

CrCl3 Chromium(III) chloride

L Ligand

Å Ångström, 1010

m

° Degree

ε Dielectric constant of solvent

λ Wavelength

v Scan rate

K Kelvin

k Rate constant

eV Electronvolt

cm‒1 Reciprocal centrimeter

m/z Mass to charge ratio

ppm Parts per million

MHz Megahertz

GHz Gigahertz

nm Nanometer

TMS Tetramethylsilane

g Gauss

M Molar

xiii

mmol or mM Millimole

µM Micromolar

N Normal

Dq Crystal field splitting

Ea or E‡ Activation energy

∆H‡

Activation enthalpy

MAD Mean absolute deviation

dH2O Deionized water

TBAH Tetrabutylammonium hexafluorophosphate

DMSO Dimethyl sulfoxide

TGA Thermogravimetric analysis

DTA Differential thermal analysis

DSC Differential scanning calorimetry

CV Cyclic voltammetry

ATR-FTIR Attenuated total reflection Fourier transform infrared

NMR Nuclear magnetic resonance

ESI-MS Electrospray ionization-mass spectrometry

HPLC High-performance liquid chromatography

EPR Electron paramagnetic resonance

UV-Vis Ultraviolet-visible

AAS Atomic absorption spectrometry

DFT Density functional theory

B3LYP Becke3-Lee-Yang-Parr

cc-pVDZ Correlation-consistent polarized valence double zeta

LANL2DZ Los Alamos National Laboratory 2-double zeta

IRC Intrinsic reaction coordinate

PES Potential energy surface

PCM Polarizable continuum model

MO Molecular orbital

NBO Natural bonding orbital

HOMO Highest occupied molecular orbital

LUMO Lowest occupied molecular orbital

xiv

pic Picolinate

gly Glycinate

PS Product state

RS Reactant state

SCRF Self consistent reaction field

SI Supporting Information

G09 Gaussian09

xv

List of Publications

Journal papers

Uddin, K. M.; Poirier, R. A.; Henry, D. J. Investigation of Mono-, Bis-, and

Tris-glycinatochromium(III): Comparison of Computational and Experimental

Results. Polyhedron (accepted).

Uddin, K. M.; Habib, A. M.; Henry, D. J. Investigation of the Spectroscopic,

Thermal and Electrochemical Properties of Tris-(glycinato)chromium(III).

ChemistrySelect, 2017, 2, 1950‒1958.

Uddin, K. M.; Henry, D. J. Further Theoretical Studies of the Aquation of

Chromium(III) Chloride Nutritional Supplement: Effect of pH and Solvation.

ChemistrySelect, 2016, 1, 5236‒5249.

Uddin, K. M.; Poirier, R. A.; Henry, D. J. Mechanistic Study of the Aquation of

Nutritional Supplement Chromium Chloride and other Chromium(III) Dihalides.

Comp. Theor. Chem., 2016, 1084, 88‒97.

Uddin, K. M.; Ralph, D.; Henry, D. J. Mechanistic Investigation of

Halopentaaquachromium(III) Complexes: Comparison of Computational and

Experimental Results. Comp. Theor. Chem., 2015, 1070, 152‒161.

Conference papers

Uddin, K. M.; Henry, D. J. Investigation of the Biochemical Pathways of

Antidiabetic Chromium(III) Supplements. International Conference on Food

Chemistry and Technology (FCT‒2016) at Las Vegas, USA during 14‒16

November, 2016.

Uddin, K. M.; Hughes, L. Henry, D. J. Kinetics and Mechanistic Investigation

of the Biochemical Pathways of Antidiabetic Chromium(III) Supplements.

International Conference on 16th

Asian Chemical Congress (16ACC) at Dhaka,

Bangladesh during 16‒19 March, 2016.

Uddin, K. M.; Hughes, L. Henry, D. J. Kinetics and Mechanistic Investigation

of the Aquation of Chromium(III) Supplements. Conference on RACI Physical

Chemistry 2016 Meeting at University of Canterbury, New Zealand during 2‒5

February, 2016.

xvi

Attribution

In consultation with my supervisor Dr. David J. Henry, I was responsible for the

development of the innovative concepts, literature review, manuscript writing,

interpretation of results and design of the chapters in the thesis. The thesis contains

published work which has been co-authored. In particular Dr David J. Henry provided

overall supervision of all the research and the final editing of all manuscripts. Dr David

Ralph provided writing support for the manuscript that chapter 2 is based on. Professor

Raymond A. Poirier provided writing support for Chapter 3 and Chapter 5 manuscripts.

Mr Ahsan M. Habib performed the electrochemical measurements reported in Chapter

5, but the analysis and interpretation of this data was conducted by Uddin. All authors

critically reviewed and approved the final version of each manuscript.

......................................................

(Mohammed Kabir Uddin)

Chapter 1: Introduction and Literature Review. 1

Chapter 1

Introduction and Literature Review

1.1 General Introduction

Chromium (Cr) is an element of the first transition series that exhibits a wide

range of oxidation states (Cr(0) – Cr(VI)), which provides a rich chemistry and a

wide range of uses. For example, hexavalent Cr has several industrial applications

including: in the manufacture of stainless steels, in the leather tanning process, in

electro-plating, in pigments, in ceramics and corrosion-resistance industries. The

most important oxidation states of Cr in biological environments are II and III. Cr(II)

and Cr(III) complexes are generally considered kinetically inert and persistent in

many environments. Toepfer et al.1a

measured the relative biological activity of the

extracts of a variety of foods and estimated the total Cr content of several foods.1b

Trivalent chromium is present in many foods, including egg yolks, whole-grain

products, coffee, nuts, green beans, and broccoli. Meat contains appreciable amounts

of Cr (0.80 ppm)1c

and high levels of biologically active Cr (1.00 ppm)1c

are also

found in brewer’s yeast.1a‒1c

Cr(III) was proposed as an essential trace element in the 1950s which has led to a

variety of Cr(III) complex based nutritional supplements that have become available


and increasingly popular, including chromium(III) chloride, chromium(III)

picolinate (Cr(pic)3), and chromium(III) with biological molecules. Specific strong

claims have been made that 50‒200 µg per day of chromium(III) picolinate improves

glycemic control.2 However, a number of scientific studies suggests that these

supplements are of minimal benefit to most people and that Cr(III) is not actually an

essential trace element.3−5

Nevertheless, there is significant evidence that Cr(III)

supplementation can lower blood glucose levels of type II (or type 2) diabetics by

acting as a critical cofactor in the action of insulin.6−8

A recent meta-analysis of a

combination drug using 600 µg Cr(pic)3 and 2 mg biotin, known to enhance

absorption and transport of Cr(III), was found to reduce blood glucose in type II

diabetes patients in two randomized control trials.8 In addition to this it has been

suggested that Cr(III) influences insulin binding, insulin receptor number, and β-cell

sensitivity.9 Anderson et al.

4 observed that Cr(pic)3 complexes may have beneficial

effects on HbA1c, glucose, insulin, and cholesterol levels for patients with type II

diabetes. Hellerstein et al.5 also proposed that chromium(III) picolinate might be

effective in treating type II diabetes or impaired glucose metabolism. However, they

also raised important questions about whether these studies actually represent Cr(III)

supplementation or more appropriately consider as a pharmacologic investigation.

Some recent experimental studies of Cr(III) complexes under physiological

conditions found no adverse effects or no increase in damaging effects.10−11

However, Speetjens et al.6

studied the long term effects of Cr(pic)3 supplementation

and have raised concerns regarding the stability and reactivity of Cr(III)

supplementation and ultimately its potential toxicity under physiological conditions.

Of particular concern is the possibility that Cr(III) could be oxidized up to the

Cr(IV), Cr(V), or Cr(VI) states under physiological conditions. Hexavalent


chromium is 1000 times more toxic than trivalent chromium and is known to cause

asthmatic and allergenic reactions. It is also carcinogenic and even mild exposure to

it can lead to diarrhea and damage to the stomach, liver, and kidney. Therefore, there

is significant interest in understanding the structure and stability of Cr(pic)3 in

aqueous solutions. To fully appreciate the risks of Cr(III) treatment of type II

diabetes, a more fundamental knowledge is required of the stability and reactivity of

Cr(III) complexes and their binding to amino acids, proteins, and carbohydrates.

The focus of this research is investigation of the reaction pathways of Cr(III)

complexes via hydrolysis, ligand exchange, thermal analysis and electrochemistry

reactions using both density functional theory and experimental methods under

different conditions. The aim is to provide a fundamental understanding of

biochemical ligand exchange processes of these complexes with particular emphasis

on chromium(III) chlorides, haloaqua chromium(III) complexes, glycinato-

chromium(III) ([Cr(gly)x(H2O)6-2x](3-x)+

where x = 1 ‒3), their conjugate bases and

Cr(pic)3 species.

1.2 Literature Review

1.2.1 Investigation on the Hydrolysis of Cr(III) Complexes

Although chromium(III)/Cr(III) complexes are generally considered to be

kinetically inert, it has been found that the dark green solution of [Cr(H2O)4Cl2]+

cations gradually turns pale green and then subsequently violet due to the step-wise

replacement of chloride by water to form [Cr(H2O)5Cl]2+

and [Cr(H2O)6]3+

,

respectively. In comparison, the neutral [Cr(H2O)3Cl3]0 complex is only observed

when CrCl3.6H2O is dissolved in concentrated HCl. However, the complex rapidly


returns to the dichlorocomplex, [Cr(H2O)4Cl2]+ in dilute solutions. Experimental

investigation,12−13

including spectroscopic and X-ray absorption EXAFS techniques,

have shown that the octahedral chloroaquachromium complexes,

[Cr(H2O)(6−n)Cln](3−n)+

(n = 1−3) can exist in several different stable conformations

resulting in different geometrical parameters.14−15

The substitution reactions of chloroaquachromium(III) complexes are similar to

those of the equivalent cobalt complexes. As a consequence, a number of

experimental kinetics studies have been carried out on octahedral transition metal

complexes of the types [ML5X]n+

and [ML4YX]n+

(n = 1‒2), where M = Co(III) or

Cr(III), L = H2O or NH3, and X = F‒, Cl

‒, Br

‒, or I

‒ and Y = ‒OH or ‒NH2.

Investigation of aquation processes reveals a strong pH dependence.12,16−17

Several experimental studies12,16−17

have been reported for [MLnX](n-2)+

(n = 5),

where M = Cr(III) and Co(III), L = H2O or NH3, and X = F‒, Cl

‒, Br

‒, or I

‒ using

polarographic techniques to examine the kinetics of the hydrolysis reaction under

neutral, acidic and basic conditions.18−20

Levine et al. investigated14

the first order or

pseudo-first order kinetics of hydrolysis of [Cr(NH3)5X]2+

(X = Cl‒, Br

‒,or I

‒) at pHs

in the range 1 to 10, identifying dependence on the concentration of the complexes.

The associated rate constants at 25°C increased in the order Cl− < Br

− < I

− (5.6 × 10

-4

min1, 6.2 × 10

‒4 min

1, 6.0 × 10

‒4 min

1 respectively). However, above pH ~11.5,

second order kinetics was observed according to the rate law,

rate = kb[complex][OH-] (1.1)

The reported activation energy for hydrolysis of Cr(NH3)5Cl at pH 10 is ~ 93 kJ

mol1

, whereas at pH 11.5 this increases to ~112 kJ mol1

.14

Similarly, Swaddle et al.


proposed15

a first order one-step mechanism for the hydrolysis reaction of

[Cr(OH2)5X]2+

(X= F‒, Cl

‒, Br

‒, or I

‒) ion in neutral, acidic and at physiological pH

(less than 10) over a range of temperatures for the following hydrolysis reactions:

[Cr(OH2)5X]2+

+ H2O [Cr(OH2)6]3+

+ X, (X = F−, Cl

−, Br−, or I

−) (1.2)

The reported enthalpies of activation (∆H‡) for the fluoro-, chloro-, bromo-,

18 and

iodo- complexes are 120 ± 2.5, 102 ± 1, 100 ± 1, and 96 ± 1 kJ mol1

, respectively.15

The corresponding enthalpy of activation values for the hydrolysis reactions of

halopentaamminechromium(III) ions are generally quite similar.14

The

thermodynamic properties for the reverse reactions (equations 1.3 and 1.4) have also

been studied in aqueous media at temperatures ranging from 30oC−95

oC determined

by calorimetric measurements:19−20

[Cr(OH2)6]3+

+ Cl− [CrCl(OH2)5]

2+ + H2O (1.3)

[CrCl(OH2)5]2+

+ Cl− trans-[CrCl2(OH2)4]

+ + H2O (1.4)

Values of reaction enthalpy (ΔH) and reaction entropy (ΔS) for the two processes are

close and differ by less than 7 kJ mol1

and 13 J mol1

, respectively. Thermodynamic

parameters for the hydrolysis reactions were also shown to have dependence upon

the concentration of chloride ion. Related studies include the investigation by Selbin

et al.21

of the kinetics of acid hydrolysis of cis-[Cr(en)2Cl2]+ and cis-

[Cr(en)2Cl(H2O)]2+

and the activation energy (Ea) over the temperature range of

20oC−30oC was found to be 88 kJ mol1

and 86 kJ mol1

, respectively at pH 1.0.

1.2.2 Investigation on the Oxidation of Cr(III) Complexes

Hydrolysis and oxidation by O2 reactions of Cr(III) complexes containing

chloride(Cl‒), fluoride (F

‒), sulphate (SO4

2‒), and carbonate (CO3

2‒) ions play a


minor role in sea-water chemistry and the kinetics of the oxidation of Cr(III) to

Cr(VI) indicate that it is a slow process.22

Furthermore, Schroeder et al.23

demonstrated that the oxidation of Cr(III) in fresh water by oxygen also occurs at a

slow rate with an experimental activation energy of 92 kJ mol1

.

Recently, Levina et al.10

suggested that the oxidation of Cr(III) to Cr(V) and

Cr(VI) might be possible in biological systems for six coordinate octahedral Cr(III)

complexes, with at least two aqua ligands, for example, cis-[Cr(phen)2(OH)2]3+

,

trans-[Cr(salen)(OH2)2]+, and [Cr3O(OCOEt)6(OH2)3]

+.24−25

These complexes are

generally denoted by [CrL4(OH2)2]n+

(A in Scheme 1.1A), where L represents

monodentate (i.e. H2O, OH-, or NH3).

10 Scheme 1.1 shows that the deprotonation of

A leads to an aqua-hydroxo complex (B). Ligand exchange is likely to occur prior to

the oxidation step resulting in an octahedral intermediate (C). Finally, the product

(D), is formed via H2O2, ClO−, ROOH, or oxidized forms of metalloenzymes.

26 The

kinetics of the oxidation of short lived Cr(III) complexes (C) suggest an

intramolecular electron transfer to give Cr(V) oxo species (D). Further oxidation of

Cr(V) to Cr(VI) species with biological oxidants (for example, H2O2 or oxidase

enzymes) in aqueous buffer solutions or blood serum occurs at a slow rate under

physiological conditions.10

Scheme 1.1: Proposed reaction mechanism for the oxidation of Cr(III)

complexes by biological oxo-transfer.

CrL

L L

L

OH2

OH2

n+

H

H

A

CrL

L L

L

OH2

OH

(n 1)

CrL

L L

L

OX

OH

(n 1)

XO

H2O

(slow)

Cr

L

L L

L

O

n+

B C D

XO = H2O2 ClO ROOH Fe(IV) O Mo(VI) (O)2

X = H2O Cl ROH Fe(II) Mo(IV) O

X

OH


1.2.3 Formation and Hydrolysis of Cr(III)-Biomolecule Species

In the chromium chloride supplements, all of the ligands are monodentates.

However, Cr(pic)3 and Cr(III) amino acid complexes are also a common component

of many nutritional supplements and these complexes contain bidentate ligands.

Consquently, chelate (polydentate) effects will have a significant impact on the

reaction kinetics of these species. Therefore, studying hydrolysis reactions of these

Cr(III) species under physiological conditions and in the presence of biological

molecules is important.

Cr

O

O

CO

N

C

O

NCr

OH

HO O

O

CO

N

C

O

NCr

O O

O

CO

N

C

O

N

C

O

N

0

[Cr(pic)3]0

CrH2O

H2O OH2

OH2

OH2

OH2

[Cr(H2O)6]3+

CrOH2

OH2 OH2

OH

OH2

OH2

2+ 4+

pH 72 2

-HCr

H2O

H2OHO

OH

OH2

OH2

CrOH

OH2

OH2

OH2

3+

Polymerization

[Cr(H2O)5OH]2+ [(H2O)4Cr(OH)2Cr(H2O)4]4+

Cr

O

O

CO

N

C

O

N

OH2

OH

[Cr(pic)2(H2O)OH]20

1 2

[Cr(pic)2OH]20

Figure 1.1: Structures of [Cr(H2O)6]3+

undergoes deprotonation at pH 7 forming the

hydroxyl complex, [Cr(H2O)5OH]2+

and condense to give hydroxyl-bridged

polynuclear species, [(H2O)4Cr(OH)2Cr(H2O)4]4+

. Similarly, produced

[Cr(pic)3]0, [Cr(pic)2(H2O)OH]

0 and [Cr(pic)2OH]2

0.

Armstrong et al.27

synthesized the chromium(III) picolinate complex [Cr(pic)3]

where pic = 2-pyridinecarboxylato(−) (Figure 1.1) from chromium(III) chloride and

picolinic acid in aqueous solution. They found the red complex to be the meridional

isomer of [Cr(pic)3]. The facial isomer was not present. They also found a second

product, a purple binuclear species, [Cr(pic)2OH]2 resulting from hydrolysis,

deprotonation, and dimerization. This is analogous to the hexaaquachromium(III),


[Cr(H2O)6]3+

complex that can deprotonate to form [Cr(H2O)5OH]2+

, subsequently

forming the hydroxyl-bridged polynuclear species, [Cr(H2O)4OH]24+

by

polymerization or olation at physiological pH (Figure 1.1).28

Abdel-Messih et al.29

investigated the reaction of hexaaquachromium(III) with

picolinic acid under weakly acidic conditions as shown in Scheme 1.2. In the first

step, an ion-pair is formed between picolinate and the deprotonated [Cr(H2O)6]3+

complex. The carboxylate group then displaces an aqua ligand to coordinate to the

metal ion followed by displacement of a second aqua ligand by the nearby pyridine

ring nitrogen leading to ring closure (see Scheme 1.2). This complex can then

undergo further reaction with picolinate to form [Cr(pic)3]0 .

Scheme 1.2: Mechanism of the formation of [Cr(pic)3]0 via reaction of Cr(III)

chloride with picolinic acid in aqueous solution.

Cr

O O

O

CO

N

C

O

N

C

O

N

0

[Cr(pic)3]0

N

O

O

+ [Cr(H2O)5(OH)]2+

N

O Cr(H2O)5(OH)2+

O

[Cr(H2O)6)]3+

H H

N

O

O

Cr(H2O)4(OH)+

H2O

N

O

O

Cr(H2O)3(OH)+

H2O

N

O

O

+ 2

Recently, Kita et al.30

studied the kinetics and mechanism of hydrolysis of

[Cr(pic)3]0 and [Cr(pic)2OH]2

0 in acidic aqueous HClO4 and identified the formation

of [Cr(pic)2(H2O)2]+ (see Scheme 1.3). They proposed a multi-step mechanism in


which the first step is ring-opening at the Cr−N bond followed by cis-aquation. The

[Cr(pic)2(H2O)4]+ can undergo further exchange but the subsequent aquation to form

[Cr(pic)(H2O)4]2+

is very slow. The Cr(pic)3 is considered stable under these

conditions.

Scheme 1.3: Acid catalysis hydrolysis reaction of [Cr(pic)3]0.

Cr

O O

O

CO

N

C

O

N

C

O

NCr

O O

O

CO

N

C

O

N

C

O

N

k1 OH

H

0 0

Cr

O O

O

CO

N

C

O

N

C

O

N

OH

H

+

H3O+H

Cr

O O

O

H

C

O

N

C

O

N

OH

H

+

H

PicH2

H2O

[Cr(pic)3]0 [Cr(pic)3(OH2)]0

[Cr(pic)2(picH)(OH2)]+[Cr(pic)2(OH2)2]2+

k

H+

1

In comparison, Marai et al.31

found that mer-[Cr(pic)3]0 and [Cr(ox)2(pic)]

2− in

alkaline conditions undergo hydrolysis to give chromate as the final product. These

results clearly indicate the effect of pH on the speciation of chromium complexes in

solution, suggesting the formation of different intermediates in acidic (A) and

alkaline (B) media (Figure 1.2).

NO

Cr

H2O

(pic)2

O H

NO

Cr

OH

(pic)2O

A B

Figure 1.2: Structures of the intermediates in acidic (A) and alkaline (B) media.


Green et al.32

synthesized the first carboxyl-bound nicotinic acid complex, trans-

[Cr(1,3-pn)2(nic-O)2]+ (Figure 1.3) in aqueous solution at physiological pH and

characterized by 1H NMR. They also found that the biologically active form of the

chromium(III)-nicotinic acid complex may be stabilized within a peptide or protein

matrix.

CrNH2

H2N

NH2

O

O

0

trans-[Cr(1,3-pn)2(nic-O)2]+

N

O

NO

H2N

1,3-pn = 1,3-propanediamine nic-O = nicotinate

Figure 1.3: Structures of trans-[Cr(1,3-pn)2(nic-O)2]+.

Maria et al.33

synthesized and characterized three Cr(III) complexes

([Cr(ox)2(AA)]2−

type, where AA = N,O−Asn, N,O−His, or N,N'−His) with

asparagine (Asn) and histidine (His) (Figure 1.4). These complexes undergo acid-

catalysed aquation and the ligand substitution indicates that the amino acid identity

slightly affects the reactivity of the complex. Recently, they synthesized33

and

characterized additional Cr(III) complexes with amino acids ([Cr(ox)2(AA)]2-

where

AA = N,O−Ala, Val, Cys, Asp, and Ser), and they found the substitution reaction of

reactivity order is as follows:

Val < Ala < Asp Ser < His < Asn Cys


CrO

H2N O

O

O

O

O

O

O

OO

H2C

CrNH2

N O

O

O

O

O

O

O

O

CH

H2NO

[Cr(ox)2(Asn)]2

HO

O

N

H2C

[Cr(ox)2(His)]2

CrO

H2OO

O

O

O

O

O

O

O

C

CHH2N

CH2

NH2

O

HO

CrO

H2OO

O

O

O

O

O

O

O

C

CHH2N

CH2

NH2

O

O

21

(a) (b)

Figure 1.4: Structures of asparagine and histidine in the ([Cr(ox)2(AA)]2−

and

intermediates with O-bonded AA of ([Cr(ox)2(N,O-Asns))]2−

in acidic (a) and

neutral (b) solution.

Another example of an amino acid containing Cr(III) complex is tris-(glycinato)

chromium(III) monohydrate, Cr(C3H4NO2)3.H2O, [Cr(gly)3]0 (Figure 1.5). Bryan et

al.34

determined the crystal and molecular structure of this complex by single-crystal

X-ray analysis and characterized the complex by UV-Vis spectroscopy.

Recently, Yang et al.35

synthesized Cr(III) phenylalanine, [Cr(pa)3] (Figure 1.5),

and found that this complex improved insulin–sensitivity, reduced plasma

cholesterol levels and improved whole body glucose tolerance in patients with type

II diabetes and obesity. Importantly [Cr(pa)3] does not generate toxic hydroxyl

radicals that cleave DNA under physiological conditions. Tang et al.37

have

synthesized the chromium methionine complex, [Cr(met)3], using aqueous ethanol

solution and this compound has been used in the food and pharmaceutical industries

(Figure 1.5).


CrO

H2N NH2

O

O

H2N

CH2

O

CH2

O

H2C

O

[Cr(gly)3]0

CrO

H2N NH2

O

O

H2N

CH

O

CH O

CH

O

CH2

CH2

CH2

[Cr(pa)3]0

CrO

H2N NH2

O

O

H2N

CH

O

CH O

CH

O

[Cr(met)3]0

H3CSSCH3

H3CS

Figure 1.5: Structures of the [Cr(gly)3]0, [Cr(pa)3]

0 and [Cr(met)3]

0.

Kortenkamp et al.36

studied the reaction of trans-[Cr(H2O)4Cl2]+, [Cr(H2O)6]

3+,

[Cr(L-Cys)2]−, and Cr(Gly)3 (Gly is glycinate) complexes binding with adesonine

triphosphate (ATP) and found that the formation of these species directly involves

the interaction of the Cr(III) with the triphosphate moiety of ATP, not the adenine

base.

Lugo et al.38

investigated interactions of chromium(III) picolinate and Cr(III)

dipicolinic acids with small blood serum bioligands (e.g., lactic, oxalic, citric and

phosphoric acids (Figure 1.6). They studied these complexes in aqueous solution by

means of electromotive force measurement (emf(H) or E) at 25 °C and with 1.5 mol

dm‒3

KCl as the ionic medium.

Cr

O O

O

CO

N

C

O

N

C

O

N

0

[Cr(pic)3]0

Cr

O O

O

CO

O

C

O

O

C

O

O

0

[Cr(lac)3]0

H3C

CH3

CH3

Cr

O O

O

C

O

O

C

O

O

CO

O

0

[Cr(ox)3]0

O

O

O

Figure 1.6: Structures of the [Cr(pic)3]0, [Cr(lac)3]

0 and [Cr(ox)3]

0.


1.2.4 Theoretical Studies of M(III) Metal Complexes

Accurate determination of the reaction mechanisms is of great importance in

inorganic chemistry, in understanding the role of inorganic species in biological

systems, and fundamental bioinorganic processes. Computational chemistry studies

can provide considerable insight to the specific details of these reaction mechanisms.

For example, Janse van Rensburg et al.39

investigated ethylene trimerization

catalysed by Cr-pyrrolyl complexes with density functional methods and proposed

that a metallacycle mechanism involving coordination of ethylene, hydrogen transfer

and ligand exchange as shown in Scheme 1.4.

Scheme 1.4: Proposed metallacycle mechanism for Cr-pyrrolyl-catalyzed

trimerization of ethylene.

N

CrCl

AlEt3

N

Cr Cl

AlEt3

N

Cr Cl

AlEt3

Cr(EH)3PyrroleEt3AlEt3AlCl

metallacycleformation

ethylenecoordinate

metallacyclegrowth

hydrogentransfer

N

Cr Cl

AlEt3

Recently, Choudhary et al.40

proposed that the conjugate-base of the

chromium(III) hexaaqua ion ([Cr(H2O)5OH]2+

) is the active Cr(III) species for

catalyzing glucose isomerization. Furthermore, Mushrif et al.41

used Car-Parrinello

molecular dynamics with metadynamics to investigate chromium (III) catalyzed

glucose-to-fructose isomerization. They proposed a reaction that [Cr(H2O)5(OH)]2+

binding to glucose occurs prior to ring opening and isomerization (Scheme 1.5). [40‒


41] However, many questions remain surrounding the reaction mechanism of Cr(III)

supplements in physiological environments.40−42

Scheme 1.5: Proposed reaction pathway for the isomerization via

[Cr(H2O)5OH]2+

catalysed aldose−to−ketose in water.

Cr

O

OH2

OH2

OH2

OC

H

C

R

H

O

H

Cr

O

OH2

OH2

OH2

OH2

C

C

R

H

O

H

Cr

O

OH

OH2

OH2

OH2

C

C

R

HO

H

TS1 TS2I1

1CHO

OHH

HOH

OHH

OHH

CH2OH

2

3

D-glucose

Cr

O

OH2

OH2

OH2

OH2

C

C

R

O

H

I2

H

CH2OH

O

HHO

OHH

OHH

CH2OH

D-fructose

Solvent effects are also particularly important when investigating reaction

mechanisms under aqueous conditions. For example, Hanauer et al.43

recently

studied the water exchange mechanism of another trivalent metal cation,

[Al(H2O)6]3+

, and its conjugate base [Al(H2O)5OH]2+

using density functional

theory, both in the gas phase and with the inclusion of a solvation model. Solvation

can take an implicit form by use of a continuum model, an explicit form by inclusion

of additional water molecules in the model or can involve a combination of implicit

and explicit forms. They found that the calculated exchange mechanism can be

directly influenced by explicit involvement of water molecules with the aquated

Al(III) species.


Scheme 1.6: Proposed reaction mechanism for the exchange water of Al(III)

species for associative (A), interchange (I) and dissociation (D) pathways.

Al

H2O

H2O OH2

OH2

OH2

X

Association (A) AlH2O

H2O OH2

OH2

X

H2O*OH2

AlH2O

H2O OH2

OH2

X

H2O *OH2

H2O*+

AlH2O

H2O OH2

OH2

X

H2O*OH2

CrH2O

H2O OH2

OH2

*OH2

X

+ H2OAlH2O

H2O OH2

OH2

X

H2O *OH2

Interchange (I)

AlH2O

H2O OH2

OH2

X

OH2

AlOH2

OH2

OH2

OH2

X

H2O

AlH2O

H2O OH2

OH2

X

H2O*

Dissociative (D)

X = H2O or OH

The water-water exchange reaction mechanism can proceed via three possible

mechanisms: an associative mechanism (A), an interchange mechanism (I), and a

dissociation mechanism (D), presented in Scheme 1.6. The associative water

exchange mechanism proceeds via a seven-coordinate intermediate, whereas the

interchange mechanism involves weak coordination of the two exchanging water

molecules. In comparison, the dissociative mechanism proceeds via a five-coordinate

intermediate. For this main group trivalent metal ion Al(III), the dissociative water-

water exchange mechanism with the five-coordinate aquated Al(III) species is the

most favoured, for both the hexa-aqua and the monohydroxo systems, involving the

interaction with the H-bond network around the complexes. In this case, the

stabilizing effect of the solvent on the transition state decreases the activation energy

(Ea) from 67 to 36 kJ mol1

for the hexa-aqua complex and 43 to 32 kJ mol1

for the

monohydroxo complex.The calculated energies of activation for the Al(III) species

compared well to experimental values. The structures of the reactant complex, for


the reaction of Al(III) species with one water and two water molecules are shown in

Figure 1.7.

A

AlH2O

H2O OH2

OH2

OH2

OH2

OH2

AlOH2

H2O

OH2

OH2

OH2

AlOH2

H2O

H2O

OH2

OH2

OH2

OH2 OH2Al

H2O

H2O OH2

OH2

OH2

OH2

OH2 OH2

B

C D

OH2

OH2

Figure 1.7: Two possible coordination modes (linear and bifurcated) of external

water molecules; B and D structures are more stable than A and C structures

for more hydrogen bonds.

Rotzinger44

carried out a study of the water exchange reactions of aqua amine

and hexaaqua complexes of Cr(III) and Ru(III) in the gas phase and using the

polarizable continuum model (PCM). The dissociative mechanism involving square

pyramidal pentacoordinated intermediates, [Cr(NH3)5]3+

and [Cr(NH2CH3)5]3+

, and

trigonal bipyramidal transition states are found to have higher values (138.8 and

101.5 kJ mol−1

, respectively (Scheme 1.7).

Cr Cr Cr

Scheme 1.7:

A detailed computational investigation45

of the mechanism for the water-water

exchange of transition metal aqua ions Sc(III) through Zn(II) has been conducted

using Hartree-Fock (HF) and Complete Active Space SCF (CAS-SCF) methods. The

hydrolysis mechanisms investigated include associative (A), concerted (IA), and


dissociative (D) pathways (Figure 1.8), however, no transition state was found for

the dissociative interchange (Id) pathway. The preferred pathway for these processes

was very dependent on the electronic configuration of the metal. For example,

transition states for the dissociative mechanism were only located for the high-spin

d8, d

9, and d

10 complexes as shown in Figure 1.8. In contrast, the associative

mechanism was only observed for the transition metal ions that have less than seven

3d electrons (i.e. Sc(III), Ti(III) and V(III)).

CoO

O O

O

O

O

H H

H

H

H

H

H H

H

H

H

H

O

H

H

Water adduct: [Co(OH2)6.OH2]2+

V

O

OO

O

O

O

H HH

H

H

H

H H

H

H

H

H

OH

H

Transition state [V(OH2)6...OH2]3+

formed via the A mechanism

V

O

OO

O

O

O

H HH

H

H

H

H H

H

H

H

H

OH

H

Heptacoordinated intermediate

[V(OH2)7]3+

Cr

O

O

O

O

O

O

H H H

H

H

H

H H

H

H

H

H

OH

H

Transition state [Cr(OH2)5...(OH2)2]3+

formed via the Ia mechanism

Pentagonal bipyramidal transition state

[Mn(OH2)5...(OH2)2]3+ formed via

the Ia mechanism

OO

O

OO

Mn

O

O

H H

H

H H

H H

H

HH

H

H

H

H

Cu

O

OO

O

O

H

H

H

HH

H

HH

H

HO

H

H

OH

H

Transition state [Cu(OH2)5...(OH2)2]2+

formed via the D mechanism

Cu

O

OO

O

O

H

H

H

HH

H

HH

H

H

OH

H

Square pyramidal transition state

[Cu(OH2)5...OH2]2+ formed via

the D mechanism

O

Mn

OO

O O

OH H

H

HH

H

H

HH

H

HH

Trigonal bipyramidal transition state

[Mn(OH2)5...OH2]2+ formed via

the D mechanism

Cu

O

OO

O

O

H H

H

H

H

HH

HH

H

O

O

HH

H

H

Pentacoordinated intermediate

[Cu(OH2)5.(OH2)2]2+ formed in

the D mechanism

Figure 1.8: Proposed reaction mechanism for the water-exchange of Sc(III) to Zn(II)

ion with six or seven water molecules.


1.3 Theoretical Background

1.3.1 Theoretical Methods

Computational chemistry is a sub-discipline of chemistry that employs computer

models of chemical systems in order to acquire a better understanding of the

underlying principles that make the system behave or react as it does. Advancements

in computer power and theory compared to the relative expense of chemicals and

analytical equipment, as well as the inherent safety concerns of working with

carcinogenic, explosive, and toxic chemicals have made computational chemistry an

important tool in explaining and rationalizing known chemistry and as a means to

explore new chemistry. In addition, computational models can give fundamental

information about isolated molecules without solvent effects. Computational

modeling techniques have been developed to accurately predict experimental results

of the structure and properties of molecules.46−48

Quantum mechanics (QM) is a widely used approach as it affords the accurate

computation of molecular properties including the geometries, electronic structure,

and reactions of small molecules. QM methods are especially important for studying

processes involving bond breaking, bond making, and electronic reorganization. QM

methods can be divided into roughly three types: ab initio wavefunction methods,

density functional theory (DFT), and semi-empirical approaches. When choosing a

model, it is necessary to consider the computational cost, the accuracy and precision

required, and the applicability.


1.3.1.1 Transition State Theory

A transition state (also called a transition-state complex) of relatively high energy

(Figure 1.9) is formed in equilibrium with the reactants of an elementary reaction as

follows:

A B A...B C D

Reactants Transition State Products Complex

The Arrhenius equation demonstrates the dependence of the rate constant (k) on

the temperature:

(1.5)

where Ea is the activation energy required to form the transition state complex, R is

the gas constant, T is the temperature, and A is the frequency factor (pre-exponential

factor). The rate constant k can also be expressed in terms of the free energy of

activation by the following equations:

(1.6)

Here,

(

) (1.7)

(

)

, where for solution and for

solution gas-phase (1.8)


where H ‡

, G‡, and S

‡ are the standard molar enthalpy, the standard molar Gibbs

energy, and the standard molar entropy of activation, respectively. kB and h are the

Boltzmann and Planck constants, respectively.

Reaction coordinate

Act

iva

tio

n E

nth

alp

y, ∆

H ‡

H ‡ uncat

Activated complex with a lower

energy profile (Catalytic Effect)

A...B

A+B

C+D

Products

Reactants

Transition state

H ‡ cat

slower path

Figure 1.9: Energy diagram of the activation enthalpy versus reaction coordinate.

Reactions can potentially follow more than one pathway. Therefore it is often

necessary to explore a range of transition states and mechanisms to identify the

lowest energy pathway. Exchange reactions via the associative (A), interchange (Ia or

Id) or dissociative (D) mechanism will be calculated to provide activation energies.

1.3.1.2 Electronic Structure Theory

Computational chemistry focuses on approximate solutions to the central equation of

quantum mechanics, the time independent, non-relativistic Schrödinger

equation,4849

, = , (1.9)


where and are the electronic and nuclear coordinates, respectively. is the

Hamiltonian operator, is the wave-function of an ith

state, and is the total

energy of the system. The wave-function is a function of the position of all

fundamental particles (electrons and nuclei) in the system. Equation (1.9) can be

simplified using the Born-Oppenheimer approximation to separate the nuclear and

electronic degrees of freedom, while disregarding relativistic corrections for solving

a many-electron wave-function.

The electronic Hamiltonian operator ( can be written (in atomic units) as,

∑

∑ ∑

∑ ∑ ∑

(1.10)

This equation is the summation of three operators, including the electron kinetic

energy operator ( , electron-nuclei potential energy ( operator (attractive) and

electron-electron potential energy ( operator (repulsive). N is the total number of

electrons, ZA is the nuclear charge on atom A, rij is the distance between electrons i

and j, riA is the distance between electron i and nucleus A.

The Schrödinger equation using the electronic Hamiltonian operator is,

(1.11)

where, is the electronic wave-function that explicitly depends on the electron

coordinates and parametrically on the nuclear coordinates, and is the electronic

energy.

The total energy (Etot ) is the sum of the Eel and nuclear-nuclear repulsion energy

(Enn):

(1.12)


However, the total energy Etot(R) as a function of the nuclear coordinates

provides the potential energy surface (PES) of the system. Thus, the concept of PES

is a result of the Born-Oppenheimer approximation.

1.3.1.3 Density Functional Theory

Density functional theory (DFT) is a quantum mechanical modelling method used in

physics and chemistry to investigate electronic structure properties. The DFT energy

(EDFT) may be written as a sum of the kinetic energy (ET), electron-nuclear potential

energy (EV), Coulomb (EJ), and exchange/correlation energy, EXC:50

(1.13)

All components are expressed as functions of the electron density (r), except

ET. The electron density, (r) = (x,y,z), is a function of only three coordinates. The

mathematical formulation for the energy of the density functional theory is:

(1.14)

where, E[(r)] is the total energy functional. F[(r)] is the universal functional,

which is a sum of the kinetic energy of the electrons, T[(r)], and the electron-

electron interaction energy functional, Vee[(r)]. Vext[(r)] is the external potential,

which is the attraction of nuclei and electrons, Vext = VNe in the absence of an

external field. The foundation of DFT method is based on two theorems provided by

Hohenberg and Kohn (HK).51

The first theorem states that the external potential is

uniquely determined from the ground state electron density 0(r), which

consequently determines the ground state wave function, 0 and energy, E0. The

second theorem states that the energy of the true ground state electron density 0(r),


minimizes the energy functional E[(r)]. The second theorem has the conditions that

the density is non-negative,

(1.15)

and integrates to the number of electrons,

∫ (1.16)

where N is the total number of electrons in the system, (r) is the electron density

function or electron probability distribution function, and (r)dr is the probability of

finding any electron in volume dr at r.

The true ground state electron density minimizes the energy functional of DFT

using the Kohn-Sham approach that involves defining a system of non-interacting

electrons which have the same electron density as the exact electron density KS(r) =

0(r).52

Non-interacting electrons are exactly described by a Slater determinant of

Kohn-Sham orbitals and the electron density. The energy density functional has the

form:

(1.17)

where Ts[(r)] is the kinetic energy functional of the non-interacting electrons,

Vne[(r)] is due to the external potential, or the nucleus-electron attraction potential

energy, Vee[(r)] is the electron-electron potential energy, which is given by the

classical Coulomb energy. Vee[(r)] is the correction to the electron-electron

potential energy and T[(r)] is the correction to the kinetic energy. Vee[(r)] and

T[(r)] are contained in the exchange-correlation energy, Exc[(r)].


(1.18)

and the functional Exc[(r)] can be written as:

∫ (1.19)

(1.20)

where the integral is over all space and

DFT has different functionals, which are defined as the functional derivative of

the exchange-correlation energy (X: exchange functional and C: correlation

functional) with respect to the density. The next sections review in brief the most

common type of the approximate exchange-correlation functionals.

I. Local Density Approximation (LDA)

LDAs are the simplest exchange-correlation functionals that depend only on the ρ(r)

to some power. The ԑx and Ex in LDA are known and given as,53

(1.21)

∫ (1.22)

while the ԑc and Ec have different approximate forms that can be obtained by fitting

to the many-body free electron gas data. For open-shell systems, the alpha and beta

spin densities, ρα(r) and ρβ(r) with ρ(r) = ρα(r) + ρβ(r), must be used in the approxima

-tion, which is called the local spin-density approximation (LSDA).54

The Ex in

LSDA is written as,

[ ]

∫ (1.23)


The Perdew-Wang (PW92)63

and Vosko-Wilk-Nusair (VWN) functionals are

common examples of LDA functionals.

II. Generalized Gradient Approximation (GGA)

In GGA, the Exc is given as a functional of both the density and its gradient at each

point:

∫

(1.24)

The GGA often shows a large improvement over the LDA results when studying

atoms, molecules and solids. Several GGA functionals have been developed and

used by computational chemists, e.g., Becke-88 (B88),54

Perdew-Wang-91

(PW91),55‒56

and Perdew–Burke–Ernzerhof (PBE)57

functionals.

III. The Meta‒GGAs

The meta-GGA functionals use the Laplacian of the density, and the kinetic

energy density,

∑ |

|

, of the occupied Kohn‒Sham orbitals into

the Exc as additional degrees of freedom.

∫

(1.25)

Perdew-Kurth-Zupan-Blaha (PKZB)58

and Lee-Yang-Parr (LYP)59

functionals

are meta-GGA functionals.

IV. Hybrid Functionals

The hybrid functionals incorporate a portion of the HF exchange with a combination


of different levels of exchange and correlation functionals. For instance, the

combination of the LSDA and B88 exchange functionals and the HF exchange,

along with the VWN and LYP correlation energy functionals leads to the B3LYP

functional.59‒60

(1.26)

Where a0, ax, and ac are parameters, and is the HF exchange energy. The

B3LYP, Becke three parameter exchange functional (B3)60

and the Lee‒Yang‒Parr

correlation functional (LYP),59

is the most popular density functional in

computational chemistry.

A few research groups continue to develop new exchange-correlation

functionals.61‒62

The M06 and M06-2X functionals were designed to correct the

shortages of density functional theory by optimizing a number of empirical

parameters.61‒62

These functionals depend on spin density, reduced spin density,

reduced spin density gradient, and spin kinetic energy density. Some other widely

used hybrid density functionals include, Coulomb-attenuating B3LYP (CAM-

B3LYP),63

M05-2X, and M06-2X. The advantages of DFT over alternate theories are

less computer time and efforts and, in general, have better correlation with

experimental results than those obtained from the Hartree-Fock theory.

1.3.1.4 Solvation Models

Solvation system models have been mostly developed in two different ways using

explicit and implicit solvation methods. Explicit solvation involves creating

atomistic representation of the solvent molecule when a system is highly sensitive to


solvent interactions (i.e. hydrogen bonding). A variety of chemical reactions in

nature occur in the presence of solvent, which plays a vital role in the reaction

mechanism. On the other hand, implicit solvation models represent the overall

solvent effect on the solute molecule surrounded by many solvent molecules. The

solvent interactions are sum of the electrostatic and nonelectrostatic contributions.

(1.27)

where is the solvation energy from the electrostatic interaction of the solute

with solvent and solvent with itself, whereas are contributions from other

interactions and can be further broken down:

(1.28)

Where are the cavitation, dispersion, and repulsion

contributions, respectively. Generally, cavitation is the solvent effect arising from

modelling the solvent as a continuous dielectric surrounding the solute cavity.

A broad range of solvent models based on continuum dielectric methods have

been developed, such as Poission-Bolzmann, mutipole expansion (MPE), polarisable

continuum model (PCM), conductor-like screening model (COSMO) and the

solvation model on density (SMD) models.

1.3.1.5 Correlation-Consistent Basis sets

The correlation-consistent (cc) basis sets have been developed by Dunning and co-

workers for high-accuracy calculations with electron-correlation methods.64‒67

The

cc-basis sets are optimized using the CISD energy because the basis sets obtained at

the HF level might not be ideal for calculations of electron-correlation.67

They are


designated cc-pVXZ, where p stands for polarized, V for valence, X (= D, T, Q, 5, 6)

for the number of shells the valence orbitals are split into, and Z for zeta. The cc-

pVXZ can be augmented (aug-cc-pVXZ) by adding diffuse functions, of types f, d,

p, and s on heavy atoms and of types d, p, and s on H and He, to the all types of basis

functions in the set. The cc-pVDZ consists of 3s2p1d = 14 basis functions for Li →

Ne and 2s1p = 5 basis functions for H and He. In general, the cc-pV(X+1)Z contains

for each shell (l) one extra basis function compared to the cc-pVXZ. Thus, the cc-

pVTZ has 4s3p2d1f = 30 basis functions for Li → Ne and 3s2p1d = 14 for H and He,

while the cc-pVQZ has 5s4p3d2f1g = 55 basis functions for Li → Ne and 4s3p2d1f =

30 for H and He. Moreover, results obtained by the correlation consistent basis sets

are systematically improved with increasing their size.69‒70

1.4 Aims and Objectives

The anti-diabetic effects of Cr(III) might be explained in terms of a stimulatory

effect of subtoxic doses of a toxic chemical. However, the current uncertainties are

caused by the lack of understanding of Cr(III) interactions with biological

molecules/environments. There are also increasing concerns regarding potential

toxic side effects. Therefore, a detailed comparative mechanistic study is required for

the hydrolysis, decomposition, electrochemical behaviors, and reactions of Cr(III)

complexes with biological molecules under physiological conditions.

The central aim of this thesis is to investigate the reaction pathways of chromium

complexes under physiological conditions using both state of the art computational

chemistry techniques and experimental methods. The specific focus of this research

is to provide a better understanding of the role of Cr(III) complexes in treating type 2

diabetes and to achieve an understanding of how relevant chemical reactions occur.


The main objective of the research can be summarised as follows:

Quantum Mechanics (QM) modelling of the reaction process:

- To investigate the reaction pathways of Cr(III) compounds under

physiological conditions, thus providing an understanding of the

speciation of Cr(III) complexes that may be relevant in treating type 2

diabetes.

- To characterize the structural, energetic, and electronic properties of

the reactants, transition states and products/intermediates along the

reaction pathway.

- To explore the role of solvent effects including implicit solvation

models (e.g. polarisation continuum model (PCM) and the importance

of explicit solvation in these processes.

Experimental methods:

- To synthesize a selection of Cr(III) complexes and characterize these

by elemental analysis, atomic absorption spectroscopy (AAS), HPLC

1H NMR, UV−Vis, EPR, ATR-FTIR, Raman, and ESI−MS

spectroscopy. In addition, the electrochemical properties and the

stability will be investigated by cyclic voltammetry (CV), and thermal

decomposition reaction by differential scanning calorimetry (DSC).

1.5 Significance of the Research

A wide variety of Cr(III) based nutritional supplements are currently available.

However, a number of scientific studies suggest that these supplements are of

minimal benefit to most people. Nevertheless, there is increasing evidence to suggest


that Cr(III) complexes may have benefits as an insulin enhancing treatment in type 2

diabetics. However, there has been significant debate regarding the safety of these

treatments.

This work is significant because it provides an independent approach to

investigate key aspects of the chemistry of Cr(III) complexes, which are used as

nutritional supplements and treatments in type 2 diabetes patients and may therefore

help resolve experimental uncertainties. By investigating reaction pathways of anti-

diabetic Cr(III) complexes, it may be possible to identify potentially toxic speciation.

1.6 Methodology

1.6.1 Computational procedure

Theoretical calculations have been used in this study to investigate a wide range of

fundamental properties of chromium complexes. The accuracy and reliability of a

number of computational methods on Cr(III) metal in complexes have been

compared to ensure the observed steady states of the geometry optimizations,

energies and composition of the molecular orbitals (MO) of all reactants, transition

states, intermediates, and products are reproducible. Calculations will be performed

using the hybrid density functionals (e.g. B3LYP and PBE0 (=PBE1PBE)), range-

separated hybrid (RSH) density functionals (Coulomb-attenuating B3LYP (CAM-

B3LYP)), and the newly developed meta-hybrid functional M06 methods. All

calculations will be carried out with Gaussian0971

provided by the Australian

National Computational Infrastructure (NCI) and the Atlantic Computational

Excellence Network (ACENET) facility. The optimized structures and vibrational

modes will be visualized using the GaussView and Molden Programs.


A range of basis sets will be used: LanL2DZ (DZ refers double zeta), cc-pVDZ,

and aug-cc-pVDZ. In addition, due to the large number of electrons in heavy atoms

(i.e. Cr), of which the largest contributions to the interactions in these systems is

given by the valence electrons, a combination of an effective core potential (ECP)

for core electrons (i.e. LanL2DZ, CRENBL etc.) and valence basis set can be used.

Effective core potentials were proposed by Hellmann to replace core electrons with

analytical functions that would allow for more accurate calculations.46

On the basis

of these studies compared with experimental data, these methods provide reliable

structural and energetic data at reasonable computational. Frequencies and intensities

will also be calculated for the trivalent Cr(III) complexes. The calculations will be

carried out in the gas phase and with the polarisation continuum model (PCM) using

the effect of solvent, water (ɛ = 78.36) on geometrical parameters, vibrational

frequencies, and the energetics of each pathway, the default in Gaussian09.71

The

calculations for these systems were carried out using hybrid and GGA DFT methods

and energy-consistent relativistic pseudo potentials and correlation consistent basis

sets.

For all the reaction pathways discussed in this study, the transition states will be

analyzed using the intrinsic reaction coordinate (IRC) method. The structures

obtained from each IRC will be further optimized in order to positively identify the

reactant and product complexes to which each transition state is connected.

Vibrational frequencies obtained for all optimized structures will be used to check

for the absence of imaginary frequencies for reactants, intermediate or products and

for the presence of a single imaginary frequency for each transition state. All

activation energies, enthalpies of activation, Gibbs energies of activation, enthalpy

change of reaction and Gibbs energy change of reaction values will be calculated at


298.15 K. Accordingly, the thermal and Gibbs corrections were used to determine

the enthalpy and Gibbs free energy values:

(1.29)

(1.30)

1.6.2 Experimental Methods

In general the preparation of chromium-amino acid complexes (in 1:1 or 1:2 or 1:3)

can be achieved via simple ligand exchange reactions under aqueous conditions

starting from chromium chloride hexahydrate, CrCl3.6H2O or chromium nitrate

nonahydrate, Cr(NO3)3.9H2O. The resulting homogeneous reaction mixtures are

washed with solvent. The solid product was collected by vacuum filtration,

recrystallized in suitable solvent, dried at 100°C in an oven and stored in a

desiccator.

These complexes will be characterized by elemental analysis, atomic absorption

spectroscopy (AAS), 1H NMR, UV−Vis, EPR, HPLC, ATR-FTIR, Raman, and

ESI−MS spectroscopy. Furthermore, the thermal decomposition of glycine ligands

with Cr(III) complex will be explored by differential scanning calorimetry (DSC)

under nonisothermal conditions, at various heating rates from 5, 10, 15 and 20

°C/min over the temperature range from 30−1000 °C. In this study, electrochemical

reactions using cyclic voltammetry was performed with a glassy carbon working

electrode, saturated Ag/AgCl reference electrode, and a Pt-wire auxiliary electrode.

Oxygen was removed by purging the solution with pure Nitrogen and continuous gas

stream passed over the sample. Peak potentials for Epa (anodic peak potentials) and

currents for ipa (anodic peak current) are measured for the negative to positive sweep


(i.e. oxidation) and Epc (cathodic peak potentials) and ipc (cathodic peak current) for

the positive to negative sweep (i.e. reduction). The average of the peak potenatials

(E1/2) values were determined, E1/2 = (Epc + Epa)/2 and the electrochemical

reversibility of a redox couple, ΔEp = Epa – Epc.

Overall, in this study the computed structural properties, kinetics and

thermodynamics will be compared to the experimental values where available. The

focus is on plausible reaction mechanisms and how the results of different models

compare between theory and experiment. A combination of theoretical and

experimental studies will be very helpful to characterize and understand the

behaviour of antidiabetic Cr(III) supplements.

1.7 Outline of the Thesis

The results of this thesis are organized and discussed in eight chapters. An overview

of the research and features are as follows:

Chapter 1: ‘Introduction and Literature Review’ provides an overview of the

current stage of research in the field of biological chemistry of chromium(III)

complexes and also identifies research gaps.

Chapter 2: ‘Mechanistic investigation of halopentaaquachromium(III) complexes:

comparison of computational and experimental results’. In this chapter, we have

applied several computational methods to investigate aquation of [Cr(H2O)5X]2+

species (where X = F, Cl, Br, or I), and comparing calculated results with

experimental values where possible. Our computations were carried out both in

implicit solvation (PCM) and explicit solvation.

[This chapter has been published in Comp. Theor. Chem., 2015, 1070, 152‒161]


Chapter 3: ‘Mechanistic study of the aquation of nutritional supplement chromium

chloride and other chromium(III) dihalides’. In this chapter, we have investigated

detailed new insights of the intimate mechanism of the aquation of chromium

chloride nutritional supplements and related trans-[Cr(H2O)4TX]+ (X or T = Br or I)

complexes.

[This chapter has been published in Comp. Theor. Chem., 2016, 1084, 88‒97]

Chapter 4: ‘Further theoretical studies of the aquation of chromium(III) chloride

nutritional supplement: effect of pH and solvation’. In this chapter, we investigated

the speciation of halochromium(III) complexes under physiological conditions

involving both conjugate-base formation and halide-water exchange mechanism.

[This chapter has been published in ChemistrySelect, 2016, 1, 5236‒5249]

Chapter 5: ‘Investigation of the spectroscopic, thermal and electrochemical

properties of tris-(glycinato)chromium(III)’. In this work, tris-(glycinato)

chromium(III) has been synthesized and characterized by elemental analysis, atomic

absorption spectroscopy (AAS), 1H NMR, UV−Vis, EPR, and ESI−MS spectroscopy

and electrochemical properties that were investigated by cyclic voltammetry (CV).

[This chapter has been published in ChemistrySelect, 2017, 1, 1‒10]

Chapter 6: ‘Investigation of mono-, bis- and tris-glycinatochromium(III):

comparisons of computational and experimental results’. In this chapter, we explore

the water exchange reaction of [Cr(+NH3CH2COO

‒)(H2O)5]3+

and its conjugate base

species including associative interchange (Ia) and dissociative (D) pathways using

computational procedures. This work was complemented by an experimental

investigation of pH effects using UV−Vis monitoring (pH ~3.0 to ~8.5).

[This chapter has been accepted in Polyhedron]


Chapter 7: ‘Electron paramagnetic resonance spectroscopy and new insights into

the substitution reaction of chromium(III) picolinate’. In this chapter, [Cr(pic)3] has

been synthesized, characterized and investigated by electron paramagnetic resonance

(in both room and frozen temperatures). Additionally, computational chemistry was

used to explore the intimate pathways of aquation.

[The manuscript has been revised in the American Chemical Society Journal Inorg. Chem.]

Chapter 8: ‘Summary and Future directions’ provides an overview of the important

finding of the current stage of research in the field of bioinorganic chemistry

involving chromium(III) complexes, as well as expected future work.

1.8 References

(1) (a) Toepfer, E. W.; Mertz, W.; Roginski, E. E.; Polansky, M. M. J. Agric.

Food Chem. 1973, 21, 69‒73. (b) Cefalu, W. T.; Hu, F. B. Diabetes Care

2004, 27, 2741‒2751. (c) Jones, G. B.; Buckley, R. A. J. Sci. Food Agric.

1977, 28, 265‒268.

(2) Anderson, R. A. J. Am. Coll. Nutr. 1998, 17, 548‒555.

(3) Anderson, R. A.; Cheng, N.; Bryden, N. A.; Polansky, M. M.; Cheng, N.;

Chi, J.; Feng, J. Diabetes 1997, 46, 1786‒1791.

(4) Hellerstein, M. K. Nutr. Rev. 1998, 56, 302‒306.

(5) Speetjens, J. K.; Collins, R. A.; Vincent, J. B.; Woski, S. A. Chem. Res.

Toxicol. 1999, 12, 483‒487.

(6) Ghosh, D.; Bhattacharya, B.; Mukherjee, B.; Manna, B.; Sinha, M.;

Chowdhury, J.; Chowdhury, S. J. Nutr. Biochem. 2002, 13, 690‒697.

(7) Vincent, J. The nutritional biochemistry of chromium (III); Elsevier, 2011.

(8) Nahas, R.; Moher, M. Can. Fam. Physician 2009, 55, 591‒596.

(9) Anderson, R. A.; Roussel, A.-M.; Zouari, N.; Mahjoub, S.; Matheau, J.-M.;

Kerkeni, A. J. Am. Coll. Nutr. 2001, 20, 212‒218.

(10) Levina, A.; Lay, P. A. Coord. Chem. Rev. 2005, 249, 281‒298.

(11) Levina, A.; Lay, P. A. Chem. Res. Toxicol. 2008, 21, 563‒571.

(12) Díaz-Moreno, S.; Muñoz-Páez, A.; Martínez, J. M.; Pappalardo, R. R.;

Marcos, E. S. J. Am. Chem. Soc. 1996, 118, 12654‒12664.

(13) Dance, I. G.; Freeman, H. C. Inorg. Chem. 1965, 4, 1555‒1561.

(14) Levine, M. A.; Jones, T.; Harris, W.; Wallace, W. J. Am. Chem. Soc.


1961, 83, 2453‒2457.

(15) Swaddle, T. W.; King, E. L. Inorg. Chem. 1965, 4, 532‒538.

(16) Elving, P. J.; Zemel, B. J. Am. Chem. Soc. 1957, 79, 1281‒1285.


(18) Guthrie, F. A.; King, E. L. Inorg. Chem. 1964, 3, 916‒917.

(19) Gates, H. S.; King, E. L. J. Am. Chem. Soc. 1958, 80, 5011‒5015.

(20) Schug, K.; King, E. L. J. Am. Chem. Soc. 1958, 80, 1089‒1091.

(21) Selbin, J.; Bailar Jr, J. C. J. Am. Chem. Soc. 1957, 79, 4285‒4289.

(22) Elderfield, H. Earth Planet. Sci. Lett. 1970, 9, 10‒16.

(23) Schroeder, D. C.; Lee, G. F. Water Air Soil Pollut. 1975, 4, 355‒365.

(24) Mulyani, I.; Levina, A.; Lay, P. A. Angew. Chem. Int. Ed. Engl. 2004, 43,

4504‒4507.

(25) Dillon, C. T.; Lay, P. A.; Bonin, A. M.; Cholewa, M.; Legge, G. J. Chem.

Res. Toxicol. 2000, 13, 742‒748.

(26) Enemark, J. H.; Cooney, J. J. A.; Wang, J.-J.; Holm, R. Chem. Rev. 2004,

104, 1175‒1200.

(27) Stearns, D. M.; Armstrong, W. H. Inorg. Chem. 1992, 31, 5178‒5184.

(28) Hneihen, A. S.; Standeven, A. M.; Wetterhahn, K. E. Carcinogenesis 1993,

14, 1795‒1803.

(29) Abdel-Messih, M.; Abou-Gamra, Z. Monatsh. Chem. 2012, 143, 211‒216.

(30) Kita, E.; Szabłowicz, M. Transition Met. Chem. 2003, 28, 698‒706.

(31) Marai, H.; Kita, E.; Wiśniewska, J. Transition Met. Chem. 2012, 37, 55‒62.

(32) Green, C. A.; Bianchini, R. J.; Legg, J. I. Inorg. Chem. 1984, 23, 2713‒2715.

(33) Marai, H.; Kita, E.; Kiersikowska, E.; Kuchta, S.; Bajek, A.; Drewa, T.

Transition Met. Chem. 2012, 37, 337‒344.

(34) Bryan, R. F.; Greene, P. T.; Stokely, P. F.; Wilson Jr, E. W. Inorg. Chem.

1971, 10, 1468‒1473.

(35) Yang, X.; Palanichamy, K.; Ontko, A. C.; Rao, M.; Fang, C. X.; Ren, J.;

Sreejayan, N. FEBS Lett. 2005, 579, 1458‒1464.

(36) Kortenkamp, A.; O'Brien, P. Carcinogenesis 1991, 12, 921‒926.

(37) Tang, H.-y.; Xiao, Q.-g.; Xu, H.-b.; Zhang, Y. Org. Process Res. Dev. 2013,

17, 632‒640.

(38) Lugo, M. L.; Lubes, V. R. J. Chem. Eng. Data 2007, 52, 1217‒1222.

(39) Janse van Rensburg, W.; Grové, C.; Steynberg, J. P.; Stark, K. B.; Huyser, J.

J.; Steynberg, P. J. Organometallics 2004, 23, 1207‒1222.

(40) (a) Choudhary, V.; Mushrif, S. H.; Ho, C.; Anderko, A.; Nikolakis, V.;

Marinkovic, N. S.; Frenkel, A. I.; Sandler, S. I.; Vlachos, D. G. J. Am.

Chem. Soc. 2013, 135, 3997-4006. (b). Choudhary, V.; Pinar, A. B.; Lobo,

R. F.; Vlachos, D. G.; Sandler, S. I. ChemSusChem 2013, 6, 2369−2376.


(41) Mushrif, S. H.; Varghese, A.; Vlachos, D. G. Phys. Chem. Chem. Phys. 2014,

16, 19564−19572.

(42) (a) Mulyani, I.; Levina, A.; Lay, P. A. Angew. Chem. Int. Ed. 2004, 43,

4504‒ 4507. (b) Dillon, C. T.; Lay, P. A.; Bonin, A. M.; Cholewa, M.;

Legge, G. J. Chem. Res. Toxicol. 2000, 13, 742‒748. (c) Enemark, J. H.;

Cooney, J. A.; Wang, J.-J.; Holm, R. Chem. Rev. 2004, 104, 1175‒1200.

(43) Hanauer, H.; Puchta, R.; Clark, T.; van Eldik, R. Inorg. Chem. 2007, 46,

1112‒1122.

(44) Rotzinger, F. P. J. Phys. Chem. A 2000, 104, 8787‒8795.

(45) Rotzinger, F. P. J. Am. Chem. Soc. 1997, 119, 5230‒5238.

(46) Cramer, C. J. Essentials of computational chemistry: theories and models;

John Wiley & Sons, 2013.

(47) Young, D. Computational chemistry: a practical guide for applying

techniques to real world problems; John Wiley & Sons, 2004.

(48) Szabo, A.; Ostlund, N. S. Modern quantum chemistry: introduction to

advanced electronic structure theory; Courier Dover Publications, 2012.

(49) Levine, I. N. Quantum Chemistry, Prentice-Hall, Inc., New Jersey, 2000.

(50) Engel, T. Quantum Chemistry & Spectroscopy, 3rd ed., Pearson Education,

Inc., USA, 2013.

(51) Hohenberg, P.; Kohn, W. Phys. Rev. 1964, 136, 864‒876.

(52) Kohn, W.; Sham, L. Phys. Rev. 1965, 140, 1133.

(53) Koch, W.; Holthausen, M. C. A Chemist’s Guide to Density Functional

Theory; Wiley-VCH: second edition, 2001.

(54) Becke, A. D. Phys. Rev. A 1988, 38, 3098.

(55) Perdew, J. P.; Wang, Y. Phys. Rev. B 1992, 45, 13244.

(56) Perdew, J. P.; Chevary, J. A,; Vosko, S. H.; Jackson, K. A.; Pederson, M. R.;

Singh, D. J.; Fiolhais, C. Phys. Rev. B 1993, 48, 4978.

(57) Perdew, J. P.; Burke, K; Ernzerhof, M. Phys. Rev. Lett. 1996, 77, 3865.

(58) Perdew, J. P.; Kurth, S; Zupan, A,; Blaha, P. Phys. Rev. Lett. 1999, 82, 2544.

(59) Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785.

(60) Becke, A. D. J. Chem. Phys. 1993, 98, 5648‒5652.

(61) Zhao, Y.; Truhlar, D. G. Theor. Chim. Acta 2008, 120, 215‒241.

(62) Zhao, Y.; Schultz, N. E.;Truhlar, D. G. J. Chem. Theory Comput. 2006, 2,

364.

(63) Yanai, T.; Tew, D. P.; Handy, N. C. Chem. Phys. Lett. 2004, 393, 51‒57.

(64) Dunning, T. H. J. Chem. Phys. 1989, 90, 1007‒1023.

(65) Woon, D. E.; Dunning, T. H. Jr. J. Chem. Phys. 1995, 103, 4572‒4585.

(66) Peterson, K. A.; Dunning, T. H. Jr. J. Chem. Phys. 2002, 117, 10548‒10560.

(67) Wilson, A. K.; Mourik, T.; Dunning, T. H. Jr. J. Mol. Struct. (Theochem)

1996, 388, 339.

http://www.ncbi.nlm.nih.gov/pubmed/25105840


(68) Jensen, F.; WIREs Comput. Mol. Sci. 2013, 3, 273.

(69) Balabanov, N. B.; Peterson, K. A. J. Chem. Phys. 2005, 123, 064107.

(70) Sinha, P.; Boesch, S. E. Gu, C.; Wheeler, R. A.; Wilson, A. K. J. Phys.

Chem. A 2004, 108, 9213‒9217.

(71) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.;

Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G.

A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.;

Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.;

M.; Nakai, H.; Vreven, T.; Montgomery, J. A.; Jr.; Peralta, J. E.; Ogliaro, F.;

R.; Bearpark, Heyd, J. J.; Brothers, E.; Kudin, K. N.; Staroverov, V. N.; J. C.;

Kobayashi, R; Normand, J.; Raghavachari, K.; Rendell, A.; Burant, Iyengar,

S. S.; Tomasi, J.; Cossi, M.; Rega, N.; Millam, J. M.; Klene, M.; Knox, J. E.;

Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann,

R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.;

Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.;

Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, O.; Foresman, J. B.;

Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Guassian 09, Revision D.01, Gaussian,

Inc., Wallingford CT, 2013.

Chapter 2: * Uddin, K. M.; Ralph, D.; Henry, D. J. Comp. Theor. Chem. 2015, 1070, 152‒161. 39

Chapter 2

Mechanistic Investigation of Halopentaaquachromium(III)

Complexes: Comparison of Computational and

Experimental Results*

Abstract

The mechanism for the substitution reactions of halopentaaquachromium(III)

complexes in the series, [CrX(H2O)5]2+

, where X= F−, Cl

−, Br

−, or I

− have been

investigated using density functional theory. Several different mechanistic pathways

were explored including associative interchange (Ia), dissociative (D) and the

associatively activated dissociation mechanisms (Da). The lowest overall activation

enthalpy (ΔH‡) obtained for the fluoride system is for the Ia pathway, with

immediate proton abstraction leading to the formation of HF and the conjugate base.

For the chloride, bromide and iodide systems the Da pathway has the lowest ΔH‡

values. Activation enthalpies determined at the PBE0/cc-pVDZ level, in aqueous

solution (PCM), are in excellent agreement with the experimental results (MAD is

Chapter 2: *

Uddin, K. M.; Ralph, D.; Henry, D. J. Comp. Theor. Chem. 2015, 1070, 152‒161. 40

1.0 kJ mol1

). Reaction profiles were analyzed in terms of activation volumes to

explain the observed trends.

Introduction

The kinetics of ligand substitution around transition metal complexes is of

fundamental importance in inorganic chemistry and for understanding substitution

reactions in bioinorganic processes.1‒6

Chromium(III) complexes in the form of

chromium chloride, chromium tripicolinate, chromium dinicotinate and chromium

complexes with amino acids are all available as nutritional supplements.7‒16

There is

clinical evidence that chromium(III) complexes can lower blood glucose levels of

type 2 diabetics,17‒22

however, there is little evidence supporting their general

nutritional value and in fact, questions have been raised about potential toxicity.4,22‒

26 Chromium(III) complexes administered orally will encounter a number of

different physiological environments and chemical conditions as they pass through

the body including the high acidity of the stomach and the slightly alkaline

conditions of the bloodstream, before being eliminated in the urine and bile. These

differing conditions provide opportunities for chromium complexes to undergo

ligand substitution raising questions about their reactivity,27‒29

potentially changing

both their absorptivity and toxicity.

A number of mechanisms for ligand substitution around octahedral metal

complexes have been identified. Atkins3 notes that there are two aspects to the

classification of ligand substitution mechanisms. The first of these is the

stoichiometric mechanism that describes each of the steps in the reaction.

Stoichiometric mechanisms include an associative (A) pathway proceeding via an

Chapter 2: *


intermediate of increased coordination number, a dissociative (D) pathway proceeds

via an intermediate of reduced coordination number and an interchange (I) reaction

mechanism can exist where the leaving and entering groups exchange

simultaneously.1 Atkins also considers substitution reactions in terms of an intimate

mechanism that describes the formation of the activated complex in the rate-

determining step. If the rate is sensitive to the entering group then the process is

described as associatively activated (a) and if it is insensitive to the entering group

then it is dissociatively activated (d).3

For example, Langford and Gray1

considered two paths for interchange

mechanisms, one via associatively activated interchange (Ia) indicating a degree of

bond formation between the entering ligand and transition metal and the other

involving dissociatively activated interchange (Id) indicating a degree of bond

breaking as the transition state is formed.30‒31

The substitution reactions at an

octahedral metal center could be expected to show SN1 kinetics if the mechanism is

dissociative or SN2 kinetics if the mechanism is associative.2,32

The substitution of

ligands around Cr(III) under various conditions has been studied by a large number

of workers.33‒55

For example, in the simplest case of substitution around octahedral

chromium(III), Swaddle and Stranks48,56

examined reaction (2.1):

[Cr(H2O)6]3+

+ H218

O → [Cr(H2O)5 (H218

O)]3+

+ H2O (2.1)

and its rate under various pressures. These authors found that the volume of

activation (ΔV‡), the difference between the molar volume of the transition and

initial states, was −9.3 cm3 mol

−1 for reaction (2.1), suggesting that this reaction

proceeds via A or Ia mechanisms.48,56

Chapter 2: *


Rotzinger60

investigated theoretically, the mechanism for water-water exchange

around transition metal aqua ions (Sc(III) through Zn(II)) using Hartree-Fock (HF)

and CAS-SCF methods. The preferred pathway for these processes was dependent

on the electronic configuration of the metal. For example, water-water exchange of

the high-spin d8, d

9, and d

10 complexes was found to proceed by the D mechanism

alone. In contrast, for Sc(III), Ti(III) and V(III) only the A mechanism was found.

However, water-water exchange for transition metals in the middle of the periodic

table (high-spin 3d3

to 3d7 systems) was found to proceed by both the associative (Ia

or A) and dissociative (D) mechanisms. Similarly, Eldik38

found experimentally that

the aquation reactions of pentaammine complexes of cobalt(III) appear to be

characterized by bond breaking (D mechanism), while those of Cr(III) systems

proceed via the bond making (A mechanism).

Nevertheless there has been some debate of the actual mechanism of ligand

substitution at Cr(III). Ardon proposed51

that ligand substitution reactions of

[Cr(OH2)5I]2+

, proceeds via a dissociative mechanism with formation of

[Cr(OH2)5]3+

, as an intermediate. However, tracer studies by Moore et al.54

demonstrated that iodide exerts a strong trans effect and that there is considerable

exchange of H2O trans to the iodide prior to hydrolysis. This is also important for

chlorination of [Cr(OH2)5I]2+

, which is found to proceed with formation of the trans-

[Cr(OH2)4ICl]2+

intermediate. This appears to rule out the formation of [Cr(OH2)5]3+

via a dissociative mechanism. Consequently, the aquation of all hexacoordinated

Cr(III) complex ions was assumed to proceed via a bimolecular or SN2 mechanism

with replacement of ligands (X) involving a simple nucleophilic attack of the

entering group (interchange mechanism) on the Cr(III) metal center.39‒40,61

The study

Chapter 2: *


of Carey et al.55

supported an Ia mechanism for acid-independent substitution and

also ruled out a purely dissociative mechanism. They argue that the Id mechanism is

only possible for Cr(III) substitutions if [Cr(OH2)6]3+

is highly selective in ion-

pairing. Nevertheless, an Id mechanism could be explained in acidic conditions if

Cr─X bond breaking is assisted by ion-pairing. Weekes and Swaddle et al.

determined that the volume of activation (ΔV‡) of hydrolysis of [CrI(H2O)5]

2+

complex in water is ‒5.4 cm3 mol

−1, again suggesting an Ia.

57 These mechanisms

however are strongly influenced by the solvent, steric effects, the size of the

substituent and electronic effects.58‒59

For example King and co-workers obtained

results for [CrI(H2O)5]2+

solvolysis in acidic aqueous methanol and aqueous

dimethylsulfoxide that are consistent with a dissociative mechanism. Experimental

investigations of the closely related halopentaamminechromium(III) complexes have

also contributed to the debate. Mǿnsted assumed33

from the linear free energy

correlation of the activation of enthalpies that hydrolysis of halopentaaqua- and

halopentaammine-chromium(III) systems favour the associative interchange

mechanism (Ia), which is in good agreement with the observation of negative

volumes of activation (ΔV‡).

47,49,57,62

Several other studies have provided evidence that aquation of [Cr(NH3)5L]3+

,

where L = neutral ligand, proceeds via an Ia mechanism, which is consistent with the

studies of [Cr(OH2)6]3+

. However, Lawrence et al.63

note that when aquation

involves charged leaving groups, it is more difficult to identify the mechanistic

details because electrostriction of the solvent influences both the activation volume

and activation entropy. They also suggested that steric factors may play a role in

determining the mechanism, with the [Cr(NH3)5L]3+

, system undergoing aquation via

Chapter 2: *


an Ia mechanism but the bulkier [Cr(NH2Me)5L]3+

, system follows an Id pathway.

Lay64‒65

argued on the basis of the ground state structures and similarities in the

entropies of activation that in fact both systems undergo aquation via a common

dissociative interchange mechanism.

The pH of the system also appears to have an impact on the mechanism of

aquation. For example, Swaddle and King measured35

the pseudo-first-order rate

coefficients for aquation of [Cr(OH2)5X]2+

, (X= F−, Cl

−, or I

−) and reported

enthalpies of activation (ΔH‡) that decrease from 120 ± 3 kJ mol

1 for X = F

− to 96 ±

1 kJ mol1

for X = I−:

[Cr(H2O)5X]2+

+ H2O [Cr(H2O)6]3+

+ X− (2.2)

However, the rate of aquation of the fluoro-complex increases with increasing

[H+], whereas the rates for the chloro- and iodo-complexes decrease with increasing

[H+]. Similarly, Guthrie and King observed

46 the aquation of [Cr(OH2)5Br]

2+ ion and

found an activation of enthalpy of 100 ± 1 kJ mol1

.

In comparison to the many experimental studies, there have been only a limited

number of theoretical studies of ligand exchange in these systems.60,66‒67

The

aquation reaction of hexaaquachromium(III) has been investigated via the Ia and D

mechanisms using Hartree-Fock (HF) and CAS-SCF methods.60

The calculated

barriers of 98 kJ mol1

for Ia and 121 kJ mol1

for D mechanisms fall to either side of

the experimental value (109 kJ mol1

).68

Rotzinger reported66

the activation energy

(Ea) for aquation via the Ia mechanism of chloropentaamminechromium(III) complex

to be 113 kJ mol1

at the HF level using the implicit solvation model. This value is

Chapter 2: *


higher than the experimental value42

(93 kJ mol1

) by 20 kJ mol1

. There have also

been attempts to theoretically determine ΔV‡ for comparison with experiment. For

example, the volume of activation, for [Cr(NH3)5Cl]2+

has been estimated at HF level

from the change of the sum of all M−L (L = NH3 or Cl) bond lengths between the

transition state and reactant (∆∑ d(TS−R)).65

The computed ∆∑ d(TS−R) value of

−0.93 Å is consistent with a negative experimental activation volume (ΔV‡ = −10.6

cm3

mol−1

)47

and also supports the associative interchange mechanism.

Scheme 2.1: Associative Interchange (Ia) mechanism for the [CrX(H2O)5]2+

with

H2O (Pathways A (X= F–) to D (X= I

–)).

To date, no computational studies have been reported for the aquation

mechanisms of halopentaaqua-Cr(III) complexes. In this paper, we report a detailed

mechanistic study involving Ia, D and Da mechanisms for the aquation reactions of

[Cr(H2O)5X]2+

complexes; X = F−, Cl

−, Br

−, or I

−. As can been seen in Scheme 2.1,

the aquation process is initiated by outer sphere coordination of a water molecule to

the Cr(III) complex. The incoming H2O hydrogen bonds to inner sphere water

molecules to give the stabilized precursor or reactant complex (RA−D). The Ia

aquation mechanism involves nucleophilic attack by the outer sphere water molecule

on the central Cr3+

metal and simultaneous halide release. This occurs via a

heptacoordinate transition state (TSA−D) and leads to an octahedral product (PA−D),

with outer sphere coordination of the halide, [Cr(H2O)6]3+…X

−.

Chapter 2: *


Scheme 2.2: Dissociative mechanism for the [CrX(H2O)5]2+

(Pathways H (X = F–)

to K (X= I–)).

Cr

O

O

F

O

OH

HH

H

O

H

H

[CrF(H2O)5]2+ (RH) [Cr(H2O)5]2...F (TSH)

H

HH H

Cr

O

O

X

O

O H

HH

H

O

Cr

O

OO

O

X

O

H

H

H

H H

H

H

H

H

H

H

H

H

HH H

Cr

O

O O

OO

H

H

H

H

H

H

X

H

HHH

Cr

O

O O

O

X

O

H

H

H

H

Cr

O

OO

O

F

O

H

H

H

HCr

O

O O

OO

H

H

H

H

H

H

H

H

F

H

[Cr(H2O)5(OH)]2+...FH (PH)

HHH

H

[CrX(H2O)5]2+

H

H

H

H

H

H

HHH

[CrX(H2O)5]2+ (RI-K) [Cr(H2O)5]2+...X (TSI-K) [Cr(H2O)6]3+...X (PI-K)

X = F

X = Cl Br, or I

The dissociative mechanism (D) is a two-step reaction that initially involves

breaking of the Cr‒X bond with formation of a common pentacoordinate

intermediate (Scheme 2.2). The second step (not shown) involves addition of H2O to

this intermediate.

Scheme 2.3: Dissociatively Activated (Da) mechanism for the [CrX(H2O)5]2+

…H2O (Pathways M (X= Cl–) to O (X = I

–)).

Chapter 2: *


The associatively activated dissociation (Da) mechanism is a two-step

mechanism that is generally similar to the dissociative mechanism described in the

previous section but involves activation of the complex via outer sphere coordination

of water. A schematic of the Da mechanism for aquation of [CrX(H2O)5]2+

complexes is shown in Scheme 2.3.

The focus of this study is to firstly establish the reliability of selected methods

for investigating these processes by comparing calculated results with experimental

values, where possible. In addition, the role of solvent is investigated by exploring

both implicit solvation (PCM) and explicit solvation options in the calculations. The

overall objective of this study is to provide a detailed investigation of aquation at

haloaqua-Cr(III) species that will provide new insights to the intimate mechanisms

of this process.

Computational Methods

Standard density functional and hybrid density functional theory calculations

were carried out with Gaussian09.69

The geometries of all reactants, transition states,

intermediates, and products for the associative interchange (Ia) mechanism were

fully optimized in the gas phase and solvent phase (water) using hybrid density

functionals (B3LYP and PBE0 (=PBE1PBE)), long-range corrected hybrid density

functionals (Coulomb-attenuating B3LYP (CAM-B3LYP)), and the newly

developed meta-hybrid functional M06 method. A range of basis sets were also

investigated including LanL2DZ, cc-pVDZ, and aug-cc-pVDZ. A subset of these

methods and basis sets were used to investigate the dissociative (D) and associatively

activated dissociation (Da) mechanisms. The optimized structures and the relative

Chapter 2: *


energies of reactants, intermediates, transition states, and products for all pathways

are shown in Figures SF0‒SF19 and Tables S1‒S11 in the Supporting Information

(SI).

To identify the most reliable methods for calculating structural and energetic

data, theoretical values were compared with experimental data where possible. The

effect of solvent (water) on the structures and energetics of each pathway was

investigated using the polarisable continuum model (PCM). Vibrational frequencies

were obtained for all optimized structures to check for the absence of imaginary

frequencies for reactants and products and for the presence of a single imaginary

frequency for each transition state. Unless otherwise stated, all values given in the

text were obtained at the PBE0/cc-pVDZ level in solution (PCM). Gas phase values

are included in parentheses.

For all the reaction pathways discussed in this study, the transition states were

analyzed using the intrinsic reaction coordinate (IRC) method. The final structures

obtained from each IRC were further optimized in order to positively identify the

reactant and product complexes to which each transition state was connected.

Activation energies, enthalpies of activation and entropies of activation values are

calculated at 298.15 K. All distances and angles shown in the figures are in

angstroms (Å) and degrees (º), respectively.

The difference in partial molar volume between the transition state and the

reactants/intermediates is the volume of activation (∆V‡), which can be used to

identify I or Ia (negative value) and D (positive value). The volume of activation can

Chapter 2: *


be correlated to the change of the sum of all Cr−L bond lengths (∆∑ d(TS−R))

between the transition state and reactant:

∆∑ d(TS−R) = ∑ d(Cr−L)TS − ∑ d(Cr−L)R (2.3)

Similarly, the change in molar volume of the reaction can be estimated from the

change in Cr−L bond lengths (∆∑ d(P−R)), between the product and the reactant

species:

∆∑ d(P−R) = ∑ d(Cr−L)P − ∑ d(Cr−L)R (2.4)

where ∑ d(Cr−L)P is the sum of the Cr−L bond lengths of the product.

Results and discussion

A range of aquation mechanisms of halopentaaquachromium(III) species were

investigated in this work including associative interchange (Ia), dissociative (D) and

associatively activated dissociation (Da) mechanisms. In the first part of this study

we applied a range of computational methods to explore aquation of [Cr(L)nX]2+

complexes, to identify variations with methods and compare with experiment where

possible.

Structural parameter variations with method

Structural parameters for the interchange aquation pathway of [Cr(H2O)5X]2+

complexes; (X = F−, Cl

−, Br

−, or I

−) were determined at several levels of theory to

identify variations with functional and basis sets. Limited experimental gas-phase

and solution phase structural data is available for the [Cr(H2O)5X]2+

complexes.

However, amongst the limited experimental structural data available is an EXAFS

Chapter 2: *


study by Diaz-Moreno et al.70

of chloroaquachromium(III) complexes in aqueous

solution. They obtained bond lengths of 1.98 ± 0.04 Å and 2.26 ± 0.05 Å for the

Cr−Cl and Cr−O bonds of [CrCl(H2O)5]2+

, respectively. Our PBE0/cc-pVDZ values,

2.008 and 2.225 Å, respectively, are in good agreement with these experimental

values. Therefore, the PBE0/cc-pVDZ level was arbitrarily chosen as a reference

point for comparison of complex geometries obtained with other levels of theory.

Table 2.1 provides a summary of the overall variations in chromium-ligand bond

lengths for a range of functionals and basis sets. The full table of geometries and

bond lengths used in the comparison, obtained in the gas phase and with PCM

(water), can be found in Figures SF1 to SF5 and Table S1 of the supporting

information (SI).

Table 2.1: Summary of deviations for Cr−Ligand bond lengths for

[Cr(H2O)5X]2+

complexes; (X = F−, Cl

−, Br

−, and I

−) with functional and basis

set.a,b

Reactant Transition

Structure Overall

Level/basis set MD MAD MD MAD MD MAD

B3LYP/LanL2DZ 0.015 0.038

0.024 0.058

0.019 0.048

CAM-B3LYP/LanL2DZ −0.002 0.046

0.008 0.064

0.003 0.055

PBE0/LanL2DZ −0.006 0.043

−0.002 0.055

−0.004 0.049

M06/LanL2DZ −0.009 0.036

0.004 0.045

−0.002 0.041

PBE0/aug-cc-pVDZ −0.001 0.015 0.003 0.041 0.001 0.028 a Methods are arbitrarily compared against PBE0/cc-pVDZ geometries.

b MAD is the Mean Absolute Deviation and MD is the Mean Deviation.

All levels of theory were found to give similar performance for the bond lengths of

Cr3+

complexes, with MADs from PBE0/cc-pVDZ of 0.015−0.046 Å for reactant

complexes in solution and 0.041−0.064 Å for the transition structures in solution. The

Chapter 2: *


inclusion of diffuse functions in the basis set has minimal effect on the geometries of

both reactants and transition states, as shown by the small MADs for the PBE0/aug-

cc-pVDZ level (Table 2.1 and Table S2 in the SI). The effective core potential basis

set (LanL2DZ) leads to larger deviations from the all-electron cc-pVDZ basis, with an

overall MAD of 0.049 Å. Nevertheless, when LanL2DZ is combined with the other

functionals (B3LYP, CAM-B3LYP and M06), the agreement with PBE0/cc-pVDZ is

reasonable, particularly M06/Lanl2DZ. It was found that M06/LanL2DZ was an

acceptable alternative for obtaining geometries of these systems in a limited number

of cases where PBE0/cc-pVDZ stationary points could not be located.

In the reactant (precursor) complexes (RA−D) it is the Cr…O distance of the outer

sphere water molecule that shows the largest variation with method. In the transition

states, the main contributions to the non-zero MADs come from variations in the

breaking Cr…X bond and the forming Cr…O bond (Tables S1‒S2 and Figures SF1

to SF5 in the SI). However, for most cases there are no major variations in reactant

and transition state geometries.

Comparison of activation enthalpies with experiment

Activation enthalpies (ΔH‡) and entropies (ΔS

‡) of the interchange (Ia) and

associatively activated dissociation (Da) pathways for hydrolysis of [CrX(H2O)5]2+

complexes (X = F−, Cl

−, Br−, or I

−) were calculated at a range of levels and compared

with experiment.35,46

The full range of methods explored were the M06, B3LYP,

CAM-B3LYP and PBE0 levels of theory using the LanL2DZ and cc-pVDZ basis

sets. The ΔH‡ and ΔS

‡ values of selected methods are presented in Table 2.2, with

the full range of data presented in Tables S3‒S5 and S9‒S10 of the SI.

Chapter 2: *


Table 2.2: Enthalpies of Activation (ΔH‡, kJ mol

−1) and entropies (ΔS

‡, J mol

−1)

for aquation of [CrX(H2O)5]2+

complexes via interchange (Ia) and associatively

activated dissociation (Da) in water solvent at 298.15 K.a

M06/

PBE0/

PBE0/

LanL2DZ

LanL2DZ

cc-pVDZ

Exptlb

Cr3+

Species Mecha. ΔH‡ (ΔS

‡)

ΔH

‡ (ΔS

‡)

ΔH

‡ (ΔS

‡)

ΔH

‡ (ΔS

‡)

[CrF(H2O)5]2+

Ia 120 (−30)

129 (−47)

119d

120 ± 3 (−16 ± 3)

D 136 (–26)

119 (–28)

128d

[CrCl(H2O)5]2+

Ia 117 (−17)

116 (−26)

119 (−13)

Da 126 (−14)

102 (−24)

103 (−15)

102 ± 1 (−30 ± 2)

[CrBr(H2O)5]2+

Ia 119 (−47)

115 (−50)

114 (−16)

Da 131 (−22)

105 (−17)

101 (−20)

100 ± 1 (−15 ± 4)c

[CrI(H2O)5]2+

Ia 113 (−9)

121 (−33)

114 (−20)

Da 138 (−8)

106 (−13)

97 (−18)

96 ± 1 (−1 ± 4) a Optimized structures of halopentaaqua Octahedral Cr

3+ complexes defined in Figures SF1

to SF18 and the Da values for overall activation enthalpies (ΔH‡) and activation entropies

(ΔH‡).

b Ref. 36.

c Ref. 47.

d Activation energy calculated by single point using PBE0/cc-

pVDZ//M06/LanL2DZ and scaled thermal correction at M06/LanL2DZ.

Generally, all methods applied to the Ia pathway provide similar values with

differences of no more than ~ 4 kJ mol−1

for ΔH‡ and ~ 10 J K

−1 mol

−1 for ΔS

‡. The

calculated ΔH‡ values for the Ia pathway decrease from X = F through to X = I at all

levels, consistent with the experimental trend. However, activation enthalpies and

activation entropies for the interchange mechanism are overestimated relative to

experiment, by all levels of theory, with MADs of 12.5−16.3 kJ mol−1

for ΔH‡

and

6.0−16.8 J K−1

mol−1

for ΔS‡ in solution (Figure 2.1) (Table S5 in SI). Of the

selected methods, PBE0/cc-pVDZ gives ΔH‡ values for the Ia pathway that are

closest to experiment with MADs of 12.5 kJ mol−1

and 12.3 J K−1

mol−1

for ΔH‡ and

ΔS‡, respectively. The M06/LanL2DZ method gives the next closest agreement with

MADs of 12.8 kJ mol−1

and 16.8 J K−1

mol−1

for ΔH‡ and ΔS

‡ respectively.

Nevertheless, across the series of complexes, the deviations in ΔH‡ from experiment

Chapter 2: *


tend to be small for the fluoride system but become successively larger for the

remaining halides, regardless of the method used. This is a strong indication of a

change in mechanism across the series.

Figure 2.1: Plots of MAD from experiment for activation enthalpy (H, kJ mol1

) and

activation entropy (S, J K−1

mol1

) for aquation of [CrX(H2O)5]2+

(X = F–, Cl

–, Br

–,

or I–) calculated at different levels of theory.

The ΔH‡ values for the Da pathway show a large variation with theoretical

method. At the M06/Lanl2DZ level, ΔH‡ values increase from X = Cl through to X =

I, counter to the experimental trend. In contrast, PBE0 with a range of basis sets

gives values that follow the experimental trend and the PBE0/cc-pVDZ level gives

particularly close agreement with an overall MAD from experiment of 1.0 kJ mol−1

(Figure 2.1). We also note that activation enthalpies calculated in the gas phase are

~60.0 kJ mol1

higher (Table S3 in the SI) indicating the importance of implicit

solvation for the accurate modeling of these systems.

Chapter 2: *


In summary, we have found that activation enthalpies (ΔH‡) calculated at

PBE0/cc-pVDZ for the associative interchange (Ia) mechanism (MAD 12.5 kJ

mol1

) and associatively activated dissociation (Da) mechanism (MAD 1.0 kJ mol1

)

for hydrolysis reactions of halopentaaquochromium(III) species, give the closest

agreement with existing experimental data.1‒2

The PBE0/cc-pVDZ level has

therefore been selected to study in detail each of the mechanistic pathways for

aquation processes of [CrX(H2O)5]2+

complexes and to identify the major

contributing factors for the observed trends in reactivity. However, we note that in a

small number of cases we were unable to locate structures at the PBE0/cc-pVDZ

level. In these limited cases we have obtained PBE0/cc-pVDZ single-point energies

on M06/Lanl2DZ geometries with scaled (0.9) thermal corrections to obtain reliable

energetic data.

Associative interchange pathway (Ia) for aquation of [CrX(H2O)5]2+

As noted in the Introduction, the Associative Interchange (Ia) mechanism is the

generally accepted pathway for substitution of neutral ligands in Cr(III) complexes

and also widely accepted for substitution of charged ligands. A schematic of this

mechanism for halogen-water exchange of these [CrX(H2O)5]2+

species is shown in

Scheme 2.1 and optimized structures can be found in Tables S3 to S6 and Figures

SF1 to SF5 of the SI. The relative energy profiles for the process described in

Scheme 2.1 are shown in Figure 2.2.

Chapter 2: *


Figure 2.2: Energy profiles for Ia mechanism for halide-water exchange of

[CrX(H2O)5]2+

(X = F–, Cl

–, Br

–, or I

–) in water solvent obtained at PBE0/cc-pVDZ.

The stability of the associated reactant complexes (RA‒D, A (X= F–

) to D (X = I–

)) differs

only slightly from X = F through to X = I (Figure 2.2), which reflects the similarity

in the outer sphere coordination of the attacking water molecule. A range of different

precursor complex conformations for each system ([CrX(H2O)5]2+

…H2O) were

explored and the bifurcated arrangement (Figure 2.3a) was found to be the most

stable. We note that the H…O bonds for this conformation exhibit only a very

marginal decrease (~0.01 Å) from X = F through to X = I, indicating that the outer

sphere coordination is largely independent of the halide. The next most stable

conformations were those with a single linear hydrogen bond to an inner sphere

water molecule (Figure 2.3b), which are 3.4, 10.7, 11.5 and 11.2 kJ mol1

higher in

energy for X = F, Cl, Br and I, respectively. The H…O bond lengths for these linear

conformations are also largely independent of the halide. However, in this

conformation there is some interaction with the halide. Not surprisingly, this is

strongest for the fluoride (d(X…H) = 1.93 Å) and weakest for the iodide (d(X…H) =

3.12 Å) (Figure 2.3b). This suggests that the lower stability of the linear

Chapter 2: *


conformations arises largely from the weaker hydrogen bonding to the outer sphere

water molecule rather than a significant destabilization of the metal complex. If these

linear reactant structures are used to calculate ΔH‡ for the Ia pathway then the MAD

from experiment decreases to 5.0 kJ mol−1

(Tables S4‒S5 of the SI). However, IRC

analysis of the transition states leads to the lower energy bifurcated structure rather

than these higher energy linear structures.

Cr

O

OO

O

X

O

H

H

H

H

H

H

H

H

H

H

Cr

O

O O

O

X

O

H

H

H

H

H

H

HHH

H

OH

H

O

H

H

(a) (b)

Figure 2.3: Conformations and relative energies (kJ mol–1

) of precursor complexes:

(a) bifurcated and (b) linear for X = F−, Cl

−, Br

−, or I

−.

As noted earlier, the experimental enthalpy of activation for the four systems

decreases in the order, [CrF(H2O)5]2+

[CrCl(H2O)5]2+

[CrBr(H2O)5]2+

[CrI(H2O)5]2+

(Table 2.2). The ΔH‡ is very high for X = F because F

− ion is more

tightly bound to Cr3+

than the other halide ions due to the strong hard acid (Cr3+

) −

hard base (F− ion) interaction i.e. this is largely an electrostatic interaction (Cr) =

+2.472 e, F) = ‒0.798 e). The decrease in ΔH‡ for the other halides corresponds

with the softer nucleophilicities and decreases in electronegativity of these ions, so

that the leaving group ability follows the order I− Br

− Cl

− F

−. In particular, the

valence electrons of the I− ion are more loosely bound, so the iodide ion is more

polarizable than fluoride ion. The greater polarizability leads to a distortion of the

Chapter 2: *


electron cloud of the halide that stabilizes the transition state. The importance of

polarizability of the halide is also reflected in the Cr…X distances in the TS, which

increase from 2.22 Å for the fluoride system through to 3.24 Å for the iodide system.

The Cr…OH2 distances in the transition states exhibit much less variation with X

and range from 2.34 to 2.37 Å. Hydrogen bonding is also important in stabilising the

transition states for these processes and in each system the departing halide interacts

with the three closest water molecules, which includes the incoming water molecule

and two inner-sphere water molecules on the departing face of the complex.

However, the strength of these interactions varies across the series. In the case of the

fluoride system the X...H distance between fluoride and the incoming water

molecule is 1.69 Å and the remaining H-bonds are both ~1.82 Å, which is indicative

of the high basicity of F‒. However, the corresponding distances for the other

systems systematically increase to values of 2.64 Å, 2.49 Å and 2.54 Å, respectively,

indicating that H-bonding becomes progressively less important in stabilising these

transition states. In conjunction with this, the calculated ∆∑ d(Cr−L)TS‒R values

range from ‒0.63 Å to ‒0.94 Å (Table S11 of the SI), which is consistent with a

negative activation volume.

As noted earlier, the successor complexes (PA‒D) of this process involve outer

sphere coordination of the released halide ion. The structures of these are generally

similar but the relative stabilities differ with the identity of the halide. In particular,

the successor complex for the fluoride system is substantially less stable than the

corresponding species for the other halides. Our calculations reveal that there is only

a very small barrier for this species to undergo intramolecular proton transfer to the

fluoride leading to formation of HF and [Cr(OH)(H2O)5]2+

. This result is consistent

Chapter 2: *


with experimental observations for the fluoride system. The higher stability of the

other successor complexes, coupled with the lower basicities of these halides means

that these species do not exhibit proton transfer. Not surprisingly, the stability of the

ion pair successor complexes PA‒D is substantially greater than the isolated species.

However, explicit solvation of the free ions (X‒(H2O)6) substantially lowers the

energy of these free ions (Figure SF18 of the SI). When the solvation energy of the

free ions (−37, −55, −58 and −59 kJ mol1

, for X = F, Cl, Br and I, respectively), is

included, the overall reaction enthalpies (H) are substantially reduced and follow

the experimental trend.71‒72

To further explore the importance of explicit solvation we investigated the

effects of inclusion of a second outer sphere water molecule on the reaction profile

for [CrF(H2O)5]2+

, [CrCl(H2O)5]2+

and [CrBr(H2O)5]2+

. The additional explicit

solvation of the TSs achieved with inclusion of the second water molecule lead to

differences in activation enthalpies of no more than 3 kJ mol1

and therefore was not

considered further.

Dissociation pathway (D) for Aquation of [CrX(H2O)5]2+

The dissociative mechanism is a two-step reaction that initially involves breaking

of the Cr‒X bond with formation of a common pentacoordinate intermediate

(Scheme 2.2). For the chloride, bromide and iodide systems, the released halide ions

are coordinated to the outer sphere of the complex to give ion-pair complexes,

[Cr(H2O)5]2+

…X−. The fluoride system is similar but the strong interaction with the

neighbouring water molecule leads to proton abstraction and the formation of the

conjugate base. The structure of this intermediate is described in more detail below.

Chapter 2: *


The activation enthalpies and activation entropies for the Cr‒X dissociation of

[CrX(H2O)5]2+

(X = F−, Cl

−, Br

− or I

−) are given in Table 2.3. The optimized

structures of the reactants, products, and transition states are shown in Figures SF10

to SF13 in the SI, respectively.

Table 2.3: Enthalpies of Activation (ΔH‡, kJ mol

−1) and entropies (ΔS

‡, J mol

−1)

for aquation of [CrX(H2O)5]2+

complexes via dissociative (D) and associatively

activated dissociation (Da) in water solvent at 298.15 K.a

D Pathway Da Pathway Exptlb

Cr3+

Species ΔH‡ (ΔS

‡) ΔH

‡ (ΔS

‡) ΔH

‡ (ΔS

‡)

e[CrF(H2O)5]

2+ TS1 128

d – 120 ± 3 (−16 ± 3)

[CrCl(H2O)5]2+

TS1 103 (−38) 97 (−20)

TS2 –

34 (−7)

Overall 103 (−38) 103 (−15) 102 ± 1 (−30 ± 2)

[CrBr(H2O)5]2+

TS1 99 (−18) 93 (−29)

TS2 – 20 (−18)

Overall 99 (−18) 101 (−20) 100 ± 1 (−15 ± 4)c

[CrI(H2O)5]2+

TS1 86 (−16) 83 (−17)

TS2 – 34 (21)

Overall 86 (−16) 97 (−18) 96 ± 1 (−1 ± 4)

a Optimized structures of halopentaaquo Octahedral Cr

3+ complexes defined in Figures SF10

to SF13. b

Ref. 36.c Ref. 47.

d The values are for activation energy (TS) calculated by single

point using PBE0/cc-pVDZ//M06/LanL2DZ and scaled thermal correction at

M06/LanL2DZ. e The value is for the reaction: [CrF(H2O)5]

2+ → [Cr(H2O)4(OH)]

2+…HF.

The activation enthalpy for the dissociation pathways of the F−, Cl

−, Br

− and I

−

systems at PBE0/cc-pVDZ level are 128, 103, 99 and 86 kJ mol1

, respectively

(Table 2.3 and Figure 2.4). Although the chloride and bromide values are close to the

corresponding experimental values, the value for the fluoride system is significantly

higher than experiment and the iodide value is slightly lower than the experimental

value (Table 2.3).

Chapter 2: *


Figure 2.4: Energy profiles for D mechanism for halide-water exchange of

[CrX(H2O)5]2+

(X = F–, Cl

–, Br

–, or I

–) in water solvent obtained at PBE0/cc-pVDZ.

The leaving group ability follows the order I− Br

− Cl

− F

−. The high

polarizability and low electronegativity of the soft base I− ion again plays a role in

the lowering of the activation enthalpy of the dissociation step, whereas the hard

base F− ion is more tightly bound in [CrF(H2O)5]

2+ and has a high dissociation

enthalpy.

The Cr…X distances of the transition structures range from 2.63 Å for the

fluoride system through to 3.70 Å for the iodide system. These TS distances are

longer (0.41– 0.46 Å) than the corresponding distances for the TSs of the Ia pathway,

with fluoride showing the largest increase and iodide the least, consistent with a

dissociative (D) process. As noted above Cr‒F bond breaking from [CrF(H2O)5]2+

occurs with formation of F‒H in the intermediate complex, ([Cr(H2O)4(OH)]+…F‒

H). This complex is stabilized due to the delocalization of the negative charge as

shown in the following manner:

Chapter 2: *


This structure is consistent with the observation of Arshadi et al.71

for the

hydration of halide ions. They observed that in the F‒ system, the bonding in the

monohydrate (F‒.H2O) is not well approximated by the electrostatic model and that

F‒H bond formation occurs and the electronic structure of the monohydrate is

probably best described by mesomeric structures.

This intramolecular process comprises ligand dissociation with partial formation

of the covalent bond and partial proton transfer in the complex (F−…H

+…OH

−). The

interaction between F−

ion and H2O ligand involves charge dispersal and formation

of the semicovalent bond with electron transfer that leads to the hydroxide ion being

formed, as a conjugate base, in the transition state (TSH), resulting in a high

activation enthalpy (128 kJ mol–1

). The mesomeric structures of conjugate bases of

F−

ion (I and II canonical forms) do not occur with the other halides and this effect

correlates well with the high bond energy of H−F (565 kJ mol–1

) , compared with

H−Cl (431 kJ mol–1

), H−Br (363 kJ mol–1

), and H−I (297 kJ mol–1

).71‒72

The ∆∑

d(Cr−L) values of the first step of this pathway are all positive (0.65 – 0.96 Å),

which is consistent with a dissociative mechanism (Table S10 of the SI).

Chapter 2: *


Associatively activated dissociation (Da) for Aquation of [CrX(H2O)5]2+

A schematic of the Da mechanism for aquation of [CrX(H2O)5]2+

complexes is

shown in Scheme 2.3 and optimized structures of the reactants, products, and

transition states can be found in Figures SF14 to SF17 of the SI. The Da pathway

begins from the same associated precursor (R) complexes as described for the

associative interchange mechanism (Ia). However, the Da mechanism differs from

the Ia mechanism because it is a two-step process. Chromium–halide bond breaking

occurs in the first step to form a five-coordinated intermediate (I1(M−O)). The released

halide ion is coordinated to the outer sphere of the complex through two X…H

hydrogen bonds, while at the same time the outer sphere water molecule adopts a

position to facilitate attack on the metal centre. Nucleophilic attack by the water

molecule on the Cr3+

metal ion occurs in the second step and results in the octahedral

product ion pair (P(M−O)). The formation of this product occurs via a hexacoordinate

transition state (TS2(M−O)). Following this pathway, the fluoride system proceeds with

hydrogen abstraction in the first step (TS1(L)), resulting in formation of HF and

[Cr(OH)(H2O)5]+ rather than the hexaaqua complex and is therefore not considered

further (Figure SF14 of the SI).

The activation enthalpies and activation entropies for each of the steps of the Da

pathway are given in Table 2.3 and the relative energy profiles for the individual

reactions are shown in Figure 2.5.

Chapter 2: *


Figure 2.5: Energy profiles for associatively activated dissociation (Da) mechanim

of [CrX(H2O)5]2+

…H2O (X = Cl–, Br

–, or I

–) in the water solvent. Relative energies

(in kJ mol1

) are at the PBE0/cc-pVDZ.

The activation enthalpies for the Cr‒X dissociation of the first step decrease from

chloride (97 kJ mol1

) through to iodide (83 kJ mol1

), which reflects the differences

in the Cr−X bond strengths. These values are lower than the corresponding values for

the D mechanism, indicating the outer-sphere coordination of water helps to stabilise

the transition states. The Cr…X distances of the Da transition states (TS1(M‒O)) are

only marginally longer than the corresponding distances for the D transitions

structures. Consistent with the D mechanism, there is outer sphere coordination of

the halides in the intermediates. However, in these species the halides are

coordinated between two of the cis-inner sphere water molecules rather than along

an extension of the Cr‒X bond.

The barriers for the second step are substantially smaller than the first indicating

that nucleophilic attack by the outer sphere water molecule on the pentacoordinate

metal centre of the intermediate is the fast step of the reaction. As noted earlier, the

Chapter 2: *


overall activation enthalpies at PBE0/cc-pVDZ, are in excellent agreement with the

experimental values (Table 2.3).35,46

These results provide strong evidence that

aquation of [CrX(H2O)5]2+

complexes (X = Cl, Br and I) proceeds via a associatively

activated dissociation mechanism rather than an associative interchange (Ia)

mechanism. The calculated ∆∑d(TS−R) and ∆∑ d(Cr−L)P‒R values range from 0.82

Å to 1.00 Å and ‒0.31 Å to ‒0.42 Å, respectively, which is consistent with a D-type

mechanism (Table S10 and Figure SF0 of the SI).

Discussion

In this study we have carried out a detailed investigation of three different

mechanisms (Ia, D and Da) for the aquation of [CrX(H2O)5]2+

complexes. The very

close agreement between experimental activation enthalpies and these calculations

suggest that aquation of the fluoride system proceeds via an associative interchange

(Ia) mechanism with immediate intramolecular proton transfer to form HF and the

conjugate base [Cr(OH)(H2O)5]2+

. Aquation of the other systems (X = Cl, Br and I)

proceeds via SN1 mechanism of associatively activated dissociation (Da). No

computational studies of Da mechanism have previously been reported for the

hydrolysis reactions of these complexes. However, given the extensive debate that

has occurred on this topic it is worthwhile comparing the details of our mechanisms

with previous theoretical and experimental studies.

Rotzinger60

carried out an extensive study of the water-water exchange

mechanism for hexaaqua metal ions of the first transition series. For Cr(III) he was

able to locate both Ia and D transitions structures. Although the calculated Ia barrier

(98 kJ mol‒1

) was substantially lower than the barrier for the D pathway (121 kJ ml‒

Chapter 2: *


1) neither value was particularly close to the experimental value (109 kJ mol

‒1.68

In

this study the ∆H‡ values for the Da pathway (X = Cl, Br and I) are not only

substantially lower than the Ia values, they are also in excellent agreement with

experiment (MAD = 1.0 kJ mol1

).

Our Ia transition structures for aquation of [CrX(H2O)5]2+

generally resemble

Rotzinger’s water-water exchange TS for [Cr(H2O)6]3+

. In particular, the incoming

H2O is in a cis position relative to the leaving X. However, in the water-water

exchange process, the incoming and leaving Cr…O bonds are equivalent, whereas in

the systems studied here the metal-halide bonds (M…X) and metal-water bonds

(M…O) are not equivalent. Although we find that the Cr…O bonds of the incoming

water molecule are almost constant across the series, the Cr…X bonds vary with the

identity of the halide X, being slightly shorter than the Cr…O bond for the fluoride

system and substantially longer for the iodide system. The sum of all Cr−L bond

lengths (∆∑ d(Cr‒L)) between the transition state and reactant/intermediate is often

correlated with the activation volume (∆V‡). Negative ∆∑ d(Cr−L) values are

generally correlated with negative activation volumes which implies an associative

mechanism. Rotzinger reports a ∆∑ d(Cr−L) value for water-water exchange of

Cr(III) of ‒1.20 Å and the experimental ∆V‡ is ‒9.6 cm

3 mol

‒1.68

The ∆∑ d(Cr−L)

values for our systems along the Ia pathway are all negative (‒0.63 to ‒0.94 Å) but

smaller in magnitude than the value for water-water exchange. However, there are

important differences between water-water exchange and halide-water exchange.

Firstly, the leaving X‒ groups have different polarizabilities that can be linked to

stabilization of charges in the TSs. Secondly, electrostriction should be relatively

small in the water-water exchange process but will vary for halide-water exchange

Chapter 2: *


across the four systems. This is qualitatively demonstrated in our calculations by the

hydrogen bonding of the leaving halides to inner and outer sphere water molecules in

the TSs and successor complexes.

Rotzinger also identified a TS corresponding to the dissociative (D) pathway for

water-water exchange of Cr(III), with a reaction barrier of 121 kJ mol‒1

, which is ~

12 kJ mol‒1

higher in energy than the experimental value. The TS structure and the

corresponding intermediate compare well with those obtained in our study with

square pyramidal geometries. It was found that the dissociative pathway lead to

barriers that were higher than experiment (Table 2.3). In this study, the activation

enthalpies for the Cr‒X dissociation decrease from chloride (103 kJ mol1

) through

to iodide (86 kJ mol1

), which reflects the differences in the Cr−X bond strengths.

These values are higher than the corresponding values for the Da mechanism,

indicating the outer-sphere coordination of water helps to stabilise the transition

states.

The associatively activated dissociation pathway is closely related to the

dissociative interchange mechanism. Interestingly, Rotzinger did not consider the Da

pathway for water-water exchange and was unable to locate a TS for the dissociative

interchange pathway. This is in contrast to our study where we located Da TSs for

the chloride, bromide and iodide systems and as noted earlier the activation

enthalpies for this pathway are in excellent agreement with experiment. The Da

pathway passes through a square pyramidal intermediate in which the dissociated

halide and the incoming water molecule are located in the outer-sphere of the

complex, approximately trans to one another. The ∆∑ d(Cr−L) values for the

dissociative first step (X ligands leaving) are positive and for the second step (water

Chapter 2: *


ligand entering) are negative as shown in Table S10 and Figure SF0 of the SI. For

[CrI(H2O)5]2+

…H2O process, the calculated ∆∑ d(Cr−L) values (Table S10 of the

SI) for the overall process are negative, which is consistent with the experimental

value.57

We also note that the overall ∆S‡ is negative and that there is relatively little

variation across the three systems of this study. However, there are variations in ∆S‡

for the individual steps, particularly for step 2 that can be related to the size of the

leaving group and changes in the coordination structure of the inner and outer sphere

of the complexes for formation of the TSs. A key feature of the Da mechanism is that

the complex becomes associatively activated through outer sphere coordination that

lowers the activation enthalpy for dissociation of the Cr‒X bond. Previously,41

the

substitution reaction of [Co(NH3)5(H2O)]3+

…X− was designated “accidental

bimolecular” to indicate that the substitution process is a rearrangement between

inner and outer coordination spheres, which in consistent with our observations for

the Da mechanism. Likewise, Bushey and Espenson73

investigated anation of

[Cr(H2O)5CH2Cl]2+

and [Cr(H2O)5CHCl2]2+

with NCS‒ and obtained results that

were consistent with both a limiting SN1 and an ion-pairing mechanism.

A key feature of the experimental investigations of the aquation halopenta-

aquachromium(III) complexes is the large difference in experimental parameters for

the fluoride system compared to the other systems which are largely attributed to the

higher basicity of fluoride ion. Our calculations are completely consistent with this

observation. In particular, the departing fluoride ion is observed to abstract a proton

from a coordinated water molecule in both the D and Da TSs and has particularly

close contacts with the hydrogen atoms of neighboring water molecules in the TS of

the Ia pathway. We propose that the most likely mechanism for aquation of

Chapter 2: *


fluoropentaaquachromium(III) complex proceeds via an Ia pathway (∆H‡ = 119 kJ

mol‒1

) in excellent agreement with the reported experimental value (120 ± 3 kJ mol‒

1),

35 followed by immediate formation of the conjugate base through intramolecular

proton transfer.

Conclusions

A comprehensive computational investigation was conducted to explore three

possible mechanisms of aquation reactions of halopentaaqua Cr3+

species: associative

interchange (Ia), dissociative (D) and associatively activated dissociation (Da)

mechanisms. A range of density functional theory methods were investigated for

determination of structures and reaction energetics. All levels of theory were found

to give similar performance for the bond lengths of Cr3+

complexes, with MADs

from PBE0/cc-pVDZ of 0.015−0.046 Å for reactant complexes and 0.041−0.064 Å

for the transition structures (TS) in solution. The inclusion of diffuse functions in the

basis set has minimal effect on the geometries of both reactants and transitions states.

Activation enthalpies (ΔH‡) and entropies (ΔS

‡) for the associative interchange (Ia)

and associatively activated dissociation (Da) pathways for hydrolysis of

[CrX(H2O)5]2+

complexes (X = F−, Cl

−, Br−, or I

−) were calculated at a range of

levels (M06, B3LYP, CAM-B3LYP, and PBE0 using the LanL2DZ and cc-pVDZ

basis sets) and compared with experimental data.35,46

Activation enthalpies

calculated at PBE0/cc-pVDZ for the associative interchange (Ia) mechanism (MAD

12.5 kJ mol1

) and associatively activated dissociation (Da) mechanism (MAD 1.0 kJ

mol1

) for hydrolysis reactions of halopentaaquachromium(III) species, give the

closest agreement with existing experimental data. Use of implicit solvent models

Chapter 2: *


has a minimal effect on the structures of reactants, products and transition states but

is essential for energetic agreement between theoretical and experimental values.

Calculations reveal that the mechanism for aquation of the fluoride system

differs from the other halides. The lowest calculated overall activation enthalpy for

the fluoride system was obtained for the Ia mechanism and was in close agreement

with the experimental value.35,46

This is followed by immediate proton abstraction

leading to the formation of HF and [Cr(OH)(H2O)5]2+

. In comparison, for the X =

Cl−, Br

− or I

− the lowest H

‡ values are found for the Da mechanism which are in

excellent agreement with the experimental values. Nevertheless, the identification of

a Da mechanism is in conflict with the negative experimental activation volumes,

obtained in totally aqueous conditions, which imply an associative interchange

mechanism. We propose that electrostriction/ion-pairing and the stability of the

hydration spheres around Cr3+

complexes may be factors that contribute to the

anomaly and may therefore warrant further experimental investigation of these

processes in mixed solvents.

Supporting Information Summary

Energies of all pertinent structures are reported in the accompanying supporting

information.

Acknowledgements

We are grateful to the Australian National Computational Infrastructure (NCI)

facility for computer time.

Chapter 2: *


References

(1) Langford, C. H. G., H. B. Ligand substitution processes W. A. Benjamin,

Inc., New York, N. Y., 1966.

(2) Basolo, F. P., R. G. Mechanisms of Inorganic Reactions John Wiley and

Sons, Inc., New York, N. Y., 1967.

(3) Atkins, P. Shriver and Atkins' Inorganic Chemistry; Oxford University Press,

2010.

(4) Kaim, W.; Schwederski, B.; Klein, A. Bioinorganic Chemistry-Inorganic

Elements in the Chemistry of Life: An Introduction and Guide; John Wiley &

Sons, 2013.

(5) Hay, R. W. Reaction mechanisms of metal complexes; Elsevier, 2000.

(6) Basolo, F.; Pearson, R. G. Mechanisms of inorganic reactions: a study of

metal complexes in solution, 1967.

(7) Plaza, S. M. Alternative Med. Rev. 2002, 7, 218‒235.

(8) Anderson, R. A. Regul. Toxicol. Pharmacol. 1997, 26, S35‒S41.

(9) Zhitkovich, A.; Voitkun, V.; Costa, M. Biochemistry 1996, 35, 7275‒7282.

(10) Sreejayan, N.; Dong, F.; Kandadi, M. R.; Yang, X.; Ren, J. Obesity 2008, 16,

1331‒337.

(11) Vincent, J. B. Biol. Trace Elem. Res. 2004, 99, 1‒16.

(12) Vincent, J. B. Proc. Nutr. Soc. 2004, 63, 41‒47.


Toxicol. 1999, 12, 483‒487.

(14) Mertz, W. J. Nutr. 1993, 123, 626‒633.

(15) Vincent, J. B. Acc. Chem. Res. 2000, 33, 503‒510.

(16) Vincent, J. B. J. Nutr. 2000, 130, 715‒718.

(17) Ghosh, D.; Bhattacharya, B.; Mukherjee, B.; Manna, B.; Sinha, M.;

Chowdhury, J.; Chowdhury, S. J. Nutr. Biochem. 2002, 13, 690‒697.

(18) Cefalu, W. T., in The Nutritional Biochemistry of Chromium(III), J. B.

Vincent (ed.), Chapter 8, pp163-181, 2007.

(19) Nahas, R.; Moher, M. Can. Fam. Physician 2009, 55, 591‒596.

(20) Anderson, R. A. J. Am. Coll. Nutr. 1998, 17, 548‒555.

(21) Press, R. I.; Geller, J.; Evans, G. W. West. J. Med. 1990, 152, 41.


(23) Wilkins, P. C.; Wilkins, R. G. Inorganic chemistry in biology; Oxford

University Press: Oxford, 1997.

(24) Silva, J.; Williams, R. The Biological Chemistry of the Elements: the

Inorganic Chemistry of Life, 2001.

(25) Cefalu, W. T.; Hu, F. B. Diabetes Care 2004, 27, 2741‒2751.

Chapter 2: *



(27) Mulyani, I.; Levina, A.; Lay, P. A. Angew Chem. Int. Ed. 2004, 43, 4504‒ 4507.

(28) Dillon, C. T.; Lay, P. A.; Bonin, A. M.; Cholewa, M.; Legge, G. J. Chem.

Res. Toxicol. 2000, 13, 742‒748.

(29) Enemark, J. H.; Cooney, J. J. A.; Wang, J.-J.; Holm, R. Chem. Rev. 2004,

104, 1175‒1200.

(30) Merbach, A. E. Pure Appl. Chem. 1982, 54, 1479‒1493.

(31) Merbach, A. E. Pure Appl. Chem. 1987, 59, 161‒172.

(32) Pearson, R. G.; Boston, C. R.; Basolo, F. J. Phys. Chem.

1955, 59, 304‒308.

(33) Monsted, O. Acta Chem. Scand. 1978, 32, 297‒302.

(34) Duffy, N. V.; Earley, J. E. J. Am. Chem. Soc. 1967, 89, 272‒278.

(35) Swaddle, T. W.; King, E. L. Inorg. Chem. 1965, 4, 532‒538.

(36) Swaddle, T. W. J. Am. Chem. Soc. 1967, 89, 4338‒4344.

(37) Swaddle, T.; Jones, W. Can. J. Chem. 1970, 48, 1054‒1058.

(38) Eldik, R. J. Chem. Soc., Chem. Commun. 1987, 1105‒1106.

(39) Swaddle, T.; Guastalla, G. Inorg. Chem. 1968, 7, 1915‒1920.

(40) Espenson, J. H. Inorg. Chem. 1969, 8, 1554‒1556.

(41) Langford, C. H.; Muir, W. R. J. Am. Chem. Soc. 1967, 89, 3141‒3144.

(42) Levine, M. A.; Jones, T.; Harris, W.; Wallace, W. J. Am. Chem. Soc. 1961,

83, 2453‒2457.

(43) Espenson, J. H.; King, E. L. J. Phy. Chem. 1960, 64, 380‒381.

(44) Ardon, M. Inorg. Chem. 1965, 4, 372‒375.

(45) Gates, H. S.; King, E. L. J. Am. Chem. Soc. 1958, 80, 5011‒5015.

(46) Guthrie, F. A.; King, E. L. Inorg. Chem. 1964, 3, 916‒917.

(47) Guastalla, G.; Swaddle, T. Can. J. Chem. 1973, 51, 821‒827.

(48) Swaddle, T.; Stranks, D. J. Am. Chem. Soc. 1972, 94, 8357‒8360.

(49) Angermann, K.; Van Eldik, R.; Kelm, H.; Wasgestian, F. Inorg. Chem. 1981,

20, 955‒959.

(50) Jones, W.; Carey, L.; Swaddle, T. Can. J. Chem. 1972, 50, 2739‒2746.

(51) Ardon, M.; Sutin, N. Inorg. Chem. 1967, 6, 2268‒2269.

(52) Ferraris, S. P.; King, E. L. J. Am. Chem. Soc. 1970, 92, 1215‒1218.

(53) Bracken, D. E.; Baldwin, H. W. Inorg. Chem. 1974, 13, 1325‒1329.

(54) Moore, P.; Basolo, F.; Pearson, R. G. Inorg. Chem. 1966, 5, 223‒228.

(55) Carey, L.; Jones, W.; Swaddle, T. Inorg. Chem. 1971, 10, 1566‒1570.

Chapter 2: *


(56) Swaddle, T.; Stranks, D. J. Am. Chem. Soc. 1971, 93, 2783‒2784.

(57) Weekes, M. C.; Swaddle, T. W. Can. J. Chem. 1975, 53, 3697‒3701.

(58) Anslyn, E. V.; Dougherty, D. A. Modern physical organic chemistry;

University Science Books, 2006.

(59) Helm, L.; Merbach, A. E. Chemical reviews 2005, 105, 1923‒1960.

(60) Rotzinger, F. P. J. Am. Chem. Soc. 1997, 119, 5230‒5238.

(61) Chan, S. J. Chem. Soc. 1964, 447, 2375‒2379.

(62) Swaddle, T. W. Coord. Chem. Rev. 1974, 14, 217‒268.

(63) Lawrance, G. A.; Schneider, K.; Van Eldik, R. Inorg. Chem. 1984, 23, 3922‒ 3925.

(64) Lay, P. A. Comments Inorg. Chem. 1991, 11, 235‒284.

(65) Lay, P. A. Inorg. Chem. 1987, 26, 2144‒2149.

(66) Rotzinger, F. P. Inorg. Chem. 1999, 38, 5730‒5733.


(68) Lincoln, S. F.; Merbach, A. E. Adv. Inorg. Chem. 1995, 42, 1‒88.

(69) Frisch, M. et. al. Guassian 09, Revision D.01, Gaussian,Inc., Wallingford CT,

2013.



(71) Arshadi, M.; Yamdagni, R.; Kebarle, P. J. Phys. Chem. 1970, 74, 1475‒1482.

(72) Hiraoka, K.; Mizuse, S.; Yamabe, S. J. Phys. Chem. 1988, 92, 3943‒3952.

(73) Bushey, W. R.; Espenson, J. H. Inorg. Chem. 1977, 16, 2772‒2776.

Chapter 3: *

Uddin, K. M.; Poirier, R. A.; Henry, D. J. Comp. Theor. Chem. 2016, 1084, 88‒97. 73

Chapter 3

Mechanistic Study of the Aquation of Nutritional

Supplement Chromium Chloride and other

Chromium(III) Dihalides*

Abstract

The mechanism for aquation of dihalotetraaquachromium(III) complexes (trans-

[Cr(H2O)4TX] +

, where

X or T = Cl, Br, or I), including a common component of

many nutritional supplements (X, T = Cl) has been investigated using density

functional theory. A number of mechanistic pathways were explored including

associative interchange (Ia), and dissociative (D) mechanisms. The overall activation

enthalpy for the D pathway of the dichloro (trans-[Cr (H2O)4Cl2]) system calculated

at the PBE0/cc-pVDZ level, with inclusion of an explicit outer sphere water

molecule and in aqueous solution (PCM), is in excellent agreement with the

experimental result. The results provide a detailed understanding of the mechanism

Chapter 3: *


for the hydrolysis of trans-Cr(III) complexes, which could be useful in

understanding the speciation of Cr(III) complexes in physiological environments.

Introduction

Chromium(III) complexes are regularly incorporated into nutritional

supplements. However, there is conflicting evidence supporting their nutritional

value and in fact questions have been raised about their potential toxicity.1‒3

Consequently there has been some interest in determining the fate of Cr(III) in

physiological environments.2‒3

One of the common chromium containing

components of nutritional supplements is chromium(III) chloride (CrCl3∙6H2O),

which actually exists as the dichlorotetraaquachromium complex, trans-

[Cr(H2O)4Cl2]Cl∙2H2O.4 Although chromium(III) may be of little nutritional benefit

to most people, there is evidence to suggest Cr(III) supplements may be beneficial in

improving glucose metabolism, maintaining blood sugar and cholesterol levels,

weight loss and building muscles in subjects with type II diabetes.1,5‒6

Glinsmann

and Mertz reported7

that oral supplementation with amounts of chromium(III)

chloride ranging from 150 to 1000 μg per day for periods of 15 to 120 days led to

improved glucose tolerance in type II diabetics. Chromium chloride has been

claimed to pharmacologically influence blood sugar levels in type II diabetics when

used at dosages less than the commonly accepted minimum nutritional level of 200

μg/day.7‒9

This claim is largely based on the results of clinical trials and animal

studies.10‒15

Anderson et al.5,16

found that a three month course of supplemental

chromium(III) chloride at a dose of 50 to 200 μg/day significantly improved glucose

tolerance and blood sugar levels in diabetics. Similar results were found with a dose

of 125 μg/day of yeast-based chromium(III).17

There are also reports suggesting that

Chapter 3: *


chromium chloride has potential as an anti-depressant and alternatively to reduce

lipid deposits.10‒11

The aquation of [Cr(H2O)4Cl2]+, is likely to play an important role

in determining the speciation of Cr(III) complexes, prior to binding to biomolecules.

Chromium(III) complexes are generally considered to be kinetically inert which

might explain the relatively low absorption (~2−5%) of orally administered

chromium(III) supplements. Nevertheless, a deep green solution of [Cr(H2O)4Cl2]+

ion will gradually turn to pale green [Cr(H2O)5Cl]2+

and then subsequently to violet

[Cr(H2O)6]3+

due to the step-wise replacement of chloride by water.4,18‒24

Therefore

there may be significant speciation of chromium(III) in a physiological environment

and it is not clear which of these species may have an impact on physiological

processes in the body. For example, it has been suggested that chromium may form

polymeric species with biomolecules such as lipids, nucleic acids and proteins.2‒3,25‒

26 Understanding the stability and reactivity of dichlorotetraaquachromium(III) and

other dihalotetraaquachromium(III) complexes will provide a basis for exploring the

biochemical pathways of chromium supplements.

Substitution and aquation reaction mechanisms of metal complexes can be

classified according to whether they are associative (A), interchange (I or Ia or Id) or

dissociative (D).27‒28

The substitution reactions at an octahedral metal center could

be expected to show SN1 (unimolecular) kinetics if the mechanism is dissociative or

SN2 (bimolecular) kinetics if the mechanism is associative or interchange.29‒30

Measurement of the volume of activation (ΔV‡) is often used experimentally to

assign a mechanism, with negative values associated with A, I, Ia or Id pathways and

positive values linked to D pathways. Although this approach works well for

substitution of neutral ligands, the interpretation of ΔV‡ values for charged leaving

groups is less straightforward.

Chapter 3: *


Ardon proposed31

that ligand substitution reactions of [Cr(H2O)5I]2+

proceeds via

a dissociative mechanism with formation of [Cr(H2O)5]3+

as an intermediate.

However, tracer studies by Moore et al.32

demonstrated that iodide exerts a strong

trans-effect leading to considerable exchange of H2O trans to the iodide prior to

hydrolysis. This is also important for chlorination, which is found to proceed with

formation of the trans-[Cr(H2O)4ICl]+ intermediate.

A number of experimental studies have investigated the trans-effects in

chromium(III) species including dihaloaqua-,33‒35

dihaloammine-,36‒38

and dihalo-

bis-(ethylenediammine) -chromium(III) complexes.35,38‒42

Mǿnsted and co-workers34

measured the pseudo-first-order rate coefficients for aquation of cis- and trans-

[Cr(H2O)4Cl2]+ and reported ΔH

‡ values of 101 ± 5 kJ mol

1 for cis and 96 ± 1 kJ

mol1

for trans systems. This indicates increased reactivity of the trans-isomer.

Similarly, Williams and Garner33

investigated the aquation of the cis- and trans-

[Cr(NH3)(H2O)3Cl2]2+

ion and found activation energies (Ea) of 97 ± 2 and 85 ± 1 kJ

mol1

, respectively. Hoppenjans et al. investigated36

the aquation reactions of the

trans-dichloro-, trans-chloroaqua-, trans-bromochloro- and trans-chloroiodo-

tetraammine chromium(III)43

complexes and determined that the dihalo complexes

aquate much more rapidly than the haloaqua complexes. They also found36

that Br‒

leaves at least 8.7 times faster than Cl‒ from trans-[Cr(NH3)4BrCl]

+ and the loss of

Br‒

from bromochloro complex is about 10 times faster than Cl‒ from trans-

[Cr(NH3)4Cl2]+. The situation is more complicated for the chloroiodo complex.

Analysis of the product ratios suggests the rate of loss of iodide is 30 – 40 times

greater than the loss of chloride from this complex. This reflects the better leaving

group ability of I‒ versus Cl

‒. However, if the ratio is solely due to leaving group

ability, the ratio should in fact be much higher. The experimentally observed ratio of

Chapter 3: *


products can be explained if the trans-directing influence of the iodide is taken into

account, which increases the rate of chloride loss.

The mechanistic details of these reactions have also been investigated in several

studies. Bushey and Espenson proposed44

a large trans labilizing effect of alkyl

groups (–CH2Cl or –CHCl2) for the 1:1 substitution of H2O with NCS‒ in

[Cr(H2O)5CH2Cl]2+

and [Cr(H2O)5CHCl2]2+

. The kinetic data indicate that this

reaction proceeds via either an SN1 (dissociative) or an ion-pairing mechanism.

Analysis of experimental data for the trans-[Cr(AA)Cl2]+, (AA = en, pn, tmd)

species, combined with the observed stereochemical changes, indicate formation of a

square-pyramidal intermediate, which is indicative of a dissociative mechanism.45

Similarly, Macdonald et al.35

investigated aquation for both cis- and trans-isomers of

[Cr(en)2Cl2]+. Aquation of trans-[Cr(en)2Cl2]

+ led to the formation of trans-

[Cr(en)2Cl(H2O)]2+

, cis-[Cr(en)2Cl(H2O)]2+

and trans-[Cr(en)Cl2(H2O)2]2+

, whereas

aquation of cis-[Cr(en)2Cl2]+ led to only cis-[Cr(en)2Cl(H2O)]

2+. The reactions were

proposed to proceed via a dissociative mechanism (SN1) that may include trans-to-

cis isomerisation but this could not be confirmed. These results highlight the

significant speciation that can occur in these reactions and the need for further

investigation.

There have been only a limited number of theoretical studies of the trans effect

in chromium(III) species under aqueous conditions. Rotzinger46

carried out an

extensive study of the water-water exchange mechanism for a number of complexes

including trans-[Cr(NH3)4(H2O)2]3+

and trans-[Cr(NH2CH3)5H2O]3+

. He identified

an Ia pathway for the former and an Id mechanism for the latter due to steric effects.

Chapter 3: *


Recently we investigated47

the aquation of halopentaaquachromium(III)

complexes in the series, [Cr(H2O)5X]+ , (X = F, Cl, Br, or I) using density functional

theory. On the basis of excellent agreement with experimental activation enthalpies,

we proposed that aquation of the fluoropentaaquachromium(III) complex proceeds

via an Ia pathway, followed by immediate formation of the conjugate base through

proton transfer. In comparison, the most likely mechanism for aquation of the X =

Cl, Br or I complexes is a variation of the dissociative mechanism. In our earlier

study we found that coordination of a water molecule to the outer sphere of the

complex had a significant effect on lowering ΔH‡ for the dissociation step.

Consequently, we referred to dissociative aquation from these [Cr(H2O)4TX]+…H2O

precursor complexes as an associatively activated dissociation process (Da).

Unfortunately this leads to a mixing of the labeling convention of Langford and Gray

(D)28

with the labeling convention of Shriver and Atkins (Da)48

. Additionally, the

inclusion of the outer sphere water molecule in these systems might best be

considered an improvement in the model rather than initiating an alternate reaction

pathway. Therefore, to eliminate confusion in this study we use only the D label of

Langford and Gray to refer to reaction mechanisms in which the first step involves

dissociation of the Cr‒X bond.

This current work represents and extension of our previous study to include the

aquation reactions of trans-[Cr(H2O)4TX] +

complexes, where T and X are chloride,

bromide or iodide ions (See Scheme 3.1). To the best of our knowledge no

computational studies have been reported for the potential energy surfaces for the

aquation mechanism of these systems.

Chapter 3: *


Scheme 3.1: Hydrolysis reaction of trans-[Cr(H2O)4TX] +

(X or T = Cl, Br, or I).

The major objective of this work is to provide new insights of the intimate

mechanisms of the aquation of the chromium chloride nutritional supplement and

related trans-[Cr(H2O)4TX]+ (X or T = Br or I) complexes. This will lead to the

development of a better understanding of the stability and speciation of these species

in aqueous and physiological environments.


Standard density functional and hybrid density functional theory calculations

were carried out with Gaussian09.49

The geometries of all reactants, transition states,

intermediates, and products in this study were fully optimized in the gas phase and

solvent (water) at the PBE0/cc-pVDZ level of theory, which we have previously

shown to give excellent agreement with experiment for the activation enthalpies of

related species.47

The small-core relativistic pseudopotential and correlation

consistent basis set (cc-pVDZ-PP) of Peterson and co-workers was used for iodine

throughout these calculations.50

The effect of solvent (water) on the structures and energetics of each pathway

was investigated using the polarisable continuum model (PCM). Vibrational

frequencies were obtained for all optimized structures to check for the absence of

imaginary frequencies for reactants, intermediates and products and for the presence

Chapter 3: *


of a single imaginary frequency for each transition state. For all the reaction

pathways discussed in this study, the transition states were also analyzed using the

intrinsic reaction coordinate (IRC) method. The final structures obtained from each

IRC were further optimized in order to positively identify the reactant and product

complexes to which each transition state is connected. Activation energies,

enthalpies of activation and entropies of activation were calculated at 298.15 K. In

the figures, all distances are in angstroms (Å), angles in degrees (º) and the energies

in kJ mol‒1

. Unless otherwise stated, all values given in the text were obtained at the

PBE0/cc-pVDZ level in solution (PCM). The optimized structures and the relative

energies of reactants, intermediates, transition states, and products for all pathways

are shown in Tables ST1 to ST4 and Figures SF01 to SF65 of the Supporting

Information (SI).

The difference in partial molar volume between the key transition state and the

reactant is the volume of activation (∆V‡), which can be used experimentally to

determine if the mechanism is Ia (negative value) or D (positive value). The volume

of activation can be correlated to the change in the Cr−L bond lengths between the

TS and reactant (eqn 3.1).

∆∑ d(TS−R) = ∑ d(Cr−L)TS − ∑ d(Cr−L)R (3.1)

Similarly, the reaction volume (∆V) can be estimated from the change in Cr−L

bond lengths from product to the reactant species (eqn 3.2).

∆∑ d(P−R) = ∑ d(Cr−L)P − ∑ d(Cr−L)R (3.2)

where ∑ d(Cr−L)TS, ∑ d(Cr−L)R and ∑ d(Cr−L)P are the sums of the Cr−L bond

lengths in the transition state, reactant and product, respectively.

Chapter 3: *



Several different aquation pathways of trans-dihalotetraaquachromium(III)

complexes, [Cr(H2O)4TX]+, where T or X = Cl, Br or I, were investigated including

associative interchange (Ia, SN2) and dissociative (D, SN1) mechanisms. In the

following discussion of trans-dihalotetraaquachromium(III) complexes, we define

the leaving group as X (Cl, Br or I) and the trans-directing group as T (H2O,47

Cl, Br

or I).

Structures of trans-[Cr(H2O)4TX]+ complexes

The only experimental structural data available for the dihaloaquachromium(III)

complexes considered in this work is an EXAFS study by Díaz-Moreno et al.51

for

trans-[Cr(H2O)4Cl2]+, in aqueous solution. The PBE0/cc-pVDZ calculated Cr−Cl

and Cr−O bond lengths for trans-[Cr(H2O)4Cl2]+ (2.28 Å and 2.01 Å, respectively)

are in good agreement with the experimental values, (2.30 ± 0.03 Å and 1.99 ± 0.03

Å, respectively). On the basis of this and the previous results47

we have confidence

that PBE0/cc-pVDZ is a suitable level of theory for determining the structural

parameters for the [Cr(H2O)4TX]+ complexes (Figure SF1 of the SI).

Inspection of the chromium-halide bond lengths reveals that, with the exception

of the di-iodo species, there is little evidence of a structural trans-effect.

Consistently, across the series of complexes, Cr‒Cl bonds are 2.28 Å and Cr‒Br

bonds are 2.44 Å. The Cr‒I bond is 2.67 Å in both the chloro/iodo and bromo/iodo

complexes but increases to 2.70 Å in the di-iodo complex, which may indicate a

marginal structural trans-effect.

Chapter 3: *


Associative interchange (Ia) mechanism

The structures of reactants, transition states and products involved in the Ia

mechanism for halogen-water exchange of [Cr(H2O)4TX]+

are shown in Scheme 3.2

and optimized structures can be found in Figures SF2 to SF10 of the SI.

Scheme 3.2: Associative Interchange (Ia) mechanism for [Cr(H2O)4TX]+ (X, T =

Cl–, Br

–, or I

–) with H2O.

The reaction is initiated by outer sphere coordination of a water molecule to the

Cr(III) complex as shown in Scheme 3.2. In this reaction, the incoming H2O forms a

linear hydrogen bond to an inner sphere water molecule and has a weak interaction

with the halide X, to give the stabilized precursor or reactant complex (R). The

stability of the associated reactant complexes (R), relative to the isolated complex

and water molecule, are shown by the depths of the first potential energy wells in

Figure 3.1.

IRC calculations confirm the linear H-bonded structures (Figure 3.2a) as the

precursor complexes for the Ia pathway. However, we also identified the bridging

bifurcated arrangement (Figure 3.2b) and found that this structure is lower in energy

than the linear structure by 7 ‒ 12 kJ mol1

for T = Cl and X = Cl, Br and I.

Chapter 3: *


Figure 3.1: Energy profiles (in kJ mol1

) for the Ia pathway of [Cr(H2O)4TX]+ (X = Cl, Br,

or I) with H2O for (a) T =Cl, (b) T = Br and (c) T = I in solution phase obtained at PBE0/cc-

pVDZ.

The X…HO-H distances for the linear conformations (Figure 3.2a) show some

variation with halide X (2.56 Å, X= Cl to 3.66 Å, X = I). Additionally, for any given

X, the X…HOH distances exhibit a marginal increase as T changes from Cl, to Br

and I. Generally, coordination of the H2O to the outer sphere of the complex leads to

a small increase in Cr‒X (~ 0.02 – 0.03 Å) and a marginal increase in Cr‒T (~0.01 –

0.02 Å) bond lengths compared to the isolated species. Furthermore, the trans-

X−Cr−T bond angles in the precursor complexes (R) are all quite similar (173.7 –

177.5) and differ only slightly from the 180.0 found in the isolated species. This

suggests that the lower stability of the linear conformations arises largely from the

weaker hydrogen bonding to the outer sphere water molecule rather than a

Chapter 3: *


significant destabilization of the metal complex. On the other hand, there is

essentially no change in the H…O bond lengths for these bifurcated conformations

from X or T = Cl through to X or T = I, indicating that the outer sphere coordination

is largely independent of the halide.

Figure 3.2: Conformations of precursor complexes: (a) linear and (b) bifurcated for X or T

= Cl, Br, or I.

The Ia mechanism involves nucleophilic attack by the outer sphere water

molecule on the central Cr(III) metal, simultaneous halide release via a

heptacoordinate transition state (TS) and leads to an octahedral product (P), with

outer sphere coordination of the halide ([Cr(H2O)4T]2+

…X−) (Scheme 3.2). The

energy profiles for hydrolysis of [Cr(H2O)4TX]+

are shown in Figure 3.1 and the

enthalpies of activation (ΔH‡) and entropies of activation (ΔS

‡) are given in Table

3.1. Also included are values for T = H2O from our previous study.47

Table 3.1: Enthalpies of activation (ΔH‡, kJ mol

‒1) and entropies (ΔS

‡, J mol

‒

1K

‒1) for Ia pathways of trans-[Cr(H2O)4TX]

+ with H2O in the water solvent at

298.15 K.a‒c

[Cr(H2O)4TCl]+

[Cr(H2O)4TBr]+

[Cr(H2O)4TI]+

T ΔH‡ ΔS

‡

ΔH‡ ΔS

‡ ΔH

‡ ΔS

‡

H2Od 109 (119) ‒27 (‒13) 102 (114) ‒32 (‒16)

103 (114) ‒41 (‒20)

Cl 90 (107) ‒37 (‒49) 83 (97) ‒41 (‒36) 74 (100) ‒16 (‒39)

Br 85 (105) ‒44 (‒20)

81 (96) ‒15 (‒32)

72 (98) ‒29 (‒42)

I 81 (101) ‒28 (‒31) 72 (90) ‒29 (‒28) 63 (83) ‒37 (‒47) a Optimized structures of trans-Cr(III)

complexes defined in Figures SF2 to SF10 of

the SI.

b The experimental value for trans-[Cr(H2O)4Cl2]

+ is 96 ± 1 kJ mol

1 of ref. 34.

c The values in parentheses are calculated relative to bifurcated reactant structures.

d Ref. 47.

Chapter 3: *


The calculated ΔH‡ for aquation of the chromium nutritional supplement trans-

[Cr(H2O)4Cl2]+ is 90 kJ mol

‒1, using the linear reactant structure, which

underestimates the experimental value by 6 kJ mol‒1

. If the reaction enthalpy is

calculated relative to the bifurcated reactant structures then ΔH‡ increases by 17 kJ

mol‒1

, to 107 kJ mol‒1

, which overestimates the experimental value by 11 kJ mol‒1

.

Similarly, if the bifurcated reactant structures (Figure 3.2b) are used for T = Cl and

X = Br and I, ΔH‡ increases by 20 kJ mol

‒1. Although the IRC links to the linear

structures, it is likely that these arise from rearrangement of the lower energy

bifurcated structures. However, we have not located a TS for this intramolecular

process.

Clear trends can be seen in the activation enthalpies for aquation of the

complexes studied. Generally, for any series of complexes with a constant trans

ligand (T), the barriers decrease with X changing from chloride through to iodide.

We also note that for any given leaving group (X‒) the activation enthalpies decrease

with T changing from water and chloride to iodide. Consequently, the mono- and di-

chloride systems have the highest activation enthalpies for aquation and the di-iodide

system has the lowest. This is evidence of a trans-effect due to the high polarisability

of the valence electrons of iodide which leads to stabilization of the transition state in

the substitution reaction. The influence of the trans-effect is also evident in the Cr‒T

distances in the transition structures that decrease by ~ 0.07 – 0.09 Å for T = I but

only ~ 0.04 – 0.06 Å for T = Cl. The Cr…O distances in the TSs for the incoming

H2O range from 2.346 Å and 2.599 Å in the mono- and di-chloride systems through

to 2.797 Å for the di-iodide system.

As the hydrolysis reactions proceed via the Ia pathway, there is a distortion of the

complex geometry including the trans-(X−Cr−T) bond angles and Cr−X bond

Chapter 3: *


lengths, which contributes to ΔH‡. The T‒Cr‒X angles in the transition structures are

generally quite similar for T = Cl and Br systems (145.0 – 147.0). However, in the

trans-iodide systems (T = I), the T‒Cr‒X angle increases slightly to ~ 150.0 for X =

Cl, Br but decreases to 135.0 for X = I. To explore the importance of atomic

charges on the hydrolysis reactions, natural bonding orbital (NBO)52

analysis was

performed and the calculated charges were computed and are given in Table ST1 in

SI. The changes in geometry are accompanied by changes in the charges on several

key atoms in the complex. The positive charge on Cr is greater in the TS than in the

reactant complex. In comparison the magnitude of the negative charge on T

decreases on formation of the TS. The largest decrease is observed for T = I and X =

Cl, Br, reflecting significant charge transfer from the trans-ligand to stabilise the TS

(Figure 3.3 and Table ST1 of SI). The oxygen atom of the incoming water molecule

exhibits relatively little change in charge across the various systems as shown in the

selected maps of electrostatic potential for the Ia systems (Figure 3.3). Of the various

leaving groups, iodide develops the smallest absolute charge in the TS.

Consequently, the more polarizable I‒ leads to less charge-charge repulsion between

the halide and the incoming water, compared to the other halides. However, these

iodide systems also show the largest increase in charge on X, relative to the reactant

complex. This corresponds with these being later TSs, due to the polarisability of the

iodide.

We have determined that H-bonding is also important in stabilising the TSs for

these processes and in each system the departing halide interacts with the three

closest water molecules, which includes the incoming water molecule and two inner-

sphere water molecules on the departing face of the complex (Scheme 3.2). This is

evident by the shorter X…H‒OH distances in the TSs that range from 2.129 Å in the

Chapter 3: *


di-chloride through to 2.758 Å in the di-iodide. In comparison, there are only

marginal differences in the outer sphere coordination in the successor complexes (P)

(~0.01 Å).

Figure 3.3: Electrostatic potential (0.02 electrons hartree‒3

electron density

isosurface) for selected TSs.

The ΔH‡ values reveal the leaving group ability follows the expected order, I

‒ >

Br‒ > Cl

‒. In general, bond breaking of the leaving groups makes a significant

contribution to ΔH‡. Although this might also be attributed to changes in steric

effects and decrease in HOMO‒LUMO gap (Table ST2 of SI), it was found that

interaction between X– and Cr(III) was weaker for the iodide system alone, which

correlates with the lower ΔH‡. Changes in (∆∑ d(Cr−L)TS‒R) (Table 3.2) range from

−0.33 to −0.93 Å, consistent with a negative ΔV‡ and an Ia mechanism. The changes

in Cr−L bond lengths from product to the reactant species range from −0.46 to −0.59

Å.

The stability of the ion pair products P is substantially greater than the isolated

species. However, the calculated single ion enthalpies of hydration (X‒ + 6H2O ⇄

[X(H2O)6]‒: ΔH = −42, −34 and −8 kJ mol

1, for X = Cl, Br and I, respectively

(Figure SF17 of the SI),) are all exothermic and when these are taken into account

the overall reaction enthalpies are substantially reduced and in better agreement with

Chapter 3: *


the experimental53

trend. In our previous work,47

we have shown that inclusion of a

second outer sphere water molecule for hydrolysis of [Cr(H2O)5F]2+

, [Cr(H2O)5Cl]2+

and [Cr(H2O)5Br]2+

does not have a significant effect, with differences of no more

than 3 kJ mol‒1

in the enthalpies of activation. Therefore we have not considered

inclusion of additional explicit solvation in these systems.

Table 3.2: Calculated selected parameters for the change of the sum of the

Cr(III)−Ligand bond lengths (in Å) in the water solvent at 298.15 K.a‒b

∑ d (Cr‒L) TS-R (∆∑ d (Cr‒L)P-R)

T [Cr(H2O)4TCl]+

[Cr(H2O)4TBr]+

[Cr(H2O)4TI]+

trans-[Cr(H2O)4TX]+

+ H2O [Cr(H2O)5T]

2+ +

X

- (Ia) Cl

‒0.51 (‒0.46)

‒0.93 (‒0.54)

‒0.67 (‒0.59)

Br

‒0.63 (‒0.47)

‒0.61 (‒0.56)

‒0.51 (‒0.53)

I

‒0.37 (‒0.48)

‒0.33 (‒0.53)

‒0.49 (‒0.47)

trans-[Cr(H2O)4TX]+...H2O [Cr(H2O)5T]

2+ +

X

- (D) Cl

0.90 (1.50)

0.77 (1.26)

0.85 (1.47)

Br 0.88 (1.43)

0.89 (1.40)

0.90 (1.59)

I

0.37 (1.46)

0.81 (1.37)

0.81 (1.44) a The ∆∑ d (Cr‒L)TS-R were calculated at PBE0/cc-pVDZ using eqn 3.1.

b The ∆∑ d (Cr‒L)P-R values in parentheses are for using eqn 3.2.

The results show that the complexes with a hard ion (X or T = Cl) have a large

HOMO‒LUMO gap (Table ST2 in SI). On the other hand, complexes with a soft ion

(X or T = I) have a small HOMO‒LUMO gap due to the greater polarizability

(Figure 3.4) of the valence electrons of the I ion. This facilitates distortion of the

electron cloud of the halide that stabilizes the TS. Both a σ-trans effect

(destabilization of reactant) and a π-acid influence (back-donation to stabilize the

TS) play an important role and account for the shorter Cr‒T bond. Figure 3.4 shows

HOMO‒LUMO energy gaps for reactants, TSs and products along the Ia pathways.

The isodensity surface of these orbitals for TS, R and P structures (Figures SF11 to

Chapter 3: *


SF16 of the SI) indicates the main contributions to the frontier molecular orbitals

come from the Cr(III) d atomic orbitals.

Figure 3.4: HOMO‒LUMO energy gap (eV) for R, TS and P of [Cr(H2O)4TX]2+

(X

= Cl, Br or I) for (a) T = Cl, (b) T = Br and (c) T = I via Ia pathway in solution phase

obtained at PBE0/cc-pVDZ.

Analysis of the frontier MO energies shows there is a correlation between the

energy of the HOMO and LUMO for the transition state structures and ΔH‡. In fact,

these data suggest that the large HOMO-LUMO gap for the T = Cl and X = Cl,

compared to the other systems is due to the strong hard acid (Cr(III)) − hard base

(Cl− ion) interaction (Figure 3.4a). Similarly, Figure 3.5a shows the overall

Chapter 3: *


activation enthalpy versus the HOMO‒LUMO gaps of TSs of the Cr(III) species for

the Ia, mechanism. For any given trans group T, there is a good correlation between

ΔH‡ and Egap. On that basis, the decrease in ΔH

‡ corresponds with the

electronegativity of the leaving group influencing the electron density at the Cr(III)

metal centre and the leaving group ability follows the order I− Br

− Cl

− (Table ST2

in the SI).

Figure 3.5: ∆H‡ (kJ mol

‒1) versus HOMO‒LUMO energy gap (eV) for TS species

of [Cr(H2O)4TX]2+

(X = Cl, Br or I) for T = Cl, T = Br and T = I in solution phase

obtained at PBE0/cc-pVDZ.

Dissociative (D) mechanism

Aquation via the dissociative mechanism is a two-step reaction. The first step

involves breaking of the Cr‒X bond with formation of a pentacoordinated

Chapter 3: *


intermediate (Scheme 3.3). The released halide ion is coordinated to the outer sphere

of the complex to give the ion-pair, [Cr(H2O)4T]+…X

− (X or T= Cl, Br or I).

Scheme 3.3: Cr-X dissociation for the [Cr(H2O)4TX]+ (X or T = Cl

–, Br

–, or I

–)

complexes.

The enthalpies and entropies of activation for the dissociation step of trans-

[Cr(H2O)4TX]+ (X or T = Cl, Br

or I) are given in Table 3.3. The optimized


to SF50 in the SI, respectively.



‡, J mol

‒1

K‒1

) for Cr‒X bond dissociation of trans-[Cr(H2O)4TX]+ in the water solvent at

298.15 K.a

ΔH‡ (ΔS

‡)

T [Cr(H2O)4TCl]+

[Cr(H2O)4TBr]+ [Cr(H2O)4TI]

+

H2Ob 103 (‒38) 99 (‒18)

86 (‒16)

Cl 85 (‒1) 76 (‒17)

63 (‒24) Br 77 (‒9)

71 (‒17)

57 (‒30) I 69 (0)

65 (‒10)

49 (‒2)

a Optimized structures of trans-Cr(III)

complexes defined in Figures SF36 to SF50 of the SI.

b Ref. 47.

The ΔH‡

values for the Cr‒X dissociation gradually decrease from X= Cl−, Br

−

and I−

systems and for T = Cl through T = I. The [Cr(H2O)5Cl]2+

complex barrier is

significantly higher (Table 3.3) than the other systems.47

The leaving group ability

Chapter 3: *


follows the order I− Br

− Cl

− and the trans-effect follows the order I

− Br

− Cl

−

H2O.

As noted in the Introduction, we have previously found that coordination of a

water molecule to the outer sphere of the complex had a significant effect on

lowering ΔH‡ for the dissociation step. In these systems, the dissociative pathway

begins from precursor complexes with bifurcated H-bonding to an outer sphere water

molecule. In the first step, chromium–halide bond breaking occurs via transition

state (TS1) to form a pentacoordinated intermediate (I1), an ion-pair complex

(Scheme 3.4). The halide ion becomes coordinated to the outer sphere of the

complex through two X…H hydrogen bonds, while at the same time the outer sphere

water molecule adopts a position to facilitate attack on the Cr(III) metal centre.

Nucleophilic attack by the water molecule on the Cr(III) metal ion occurs in the

second step and results in the octahedral product ion-pair, [Cr(H2O)5T]2+

…X− (P).

The formation of this product occurs via a hexacoordinate transition state (TS2).

Optimized structures of the reactants, products, and transition states can be found in

Figures SF18 to SF26 of the SI.

Scheme 3.4: Dissociative (D) mechanism for aquation of [Cr(H2O)4TX]+… H2O

(X or T = Cl–, Br

–, or I

–).

Chapter 3: *


The ΔH‡ and ΔS

‡ for each of the steps of the D pathway for the associated

complex are given in Table 3.4 and the relative energy profiles for the individual

reactions are shown in Figure 3.6.



‡, J mol

‒1

K‒1

) for D pathways of trans-[Cr(H2O)4TX]+ with H2O in the water solvent

at 298.15 K.a

ΔH‡ (ΔS

‡)

T [Cr(H2O)4TCl]+

[Cr(H2O)4TBr]+ [Cr(H2O)4TI]

+

H2Ob TS1 97 (‒20) 93 (‒29)

83 (‒17)

TS2 34 (‒7)

20 (‒18)

34 (‒21)

Overall 103 (‒15)

101 (‒20)

97(‒18)

Cl TS1 72 (‒13) 71 (‒21) 67 (‒26)

TS2 36 (‒8) 39 (‒23) 39 (‒9)

Overall 96 (‒3) 95 (‒21) 89 (‒20)

Br TS1 64 (‒12)

59 (‒11)

61 (‒22)

TS2 39 (‒39)

50 (‒46)

39 (‒11)

Overall 91 (‒35)

89 (‒29)

86 (‒5)

I TS1 55 (‒30)

53 (‒29)

50 (‒29)

TS2 38 (‒9)

34 (‒25)

38 (‒3)

Overall 85 (‒19) 82 (‒42) 78 (‒4) a

Optimized structures of trans-Cr(III) complexes defined in Figures SF18 to SF26 of the SI.

b Ref. 47.

The enthalpies of activation for dissociation of the Cr‒X bond decrease from 97

kJ mol1

for T = H2O and X = Cl to 50 kJ mol1

for T = I and X = I. Comparison of

Tables 3.3 and 3.4 reveals that inclusion of the outer sphere water leads to a lowering

of ΔH‡ by 6 ‒ 15 kJ mol

‒1 for X = Cl and Br. However, for X = I, the inclusion of the

outer sphere water molecule leads to a slight increase (1 – 4 kJ mol‒1

) in ΔH‡ for T

= Cl, Br and I. Nevertheless, the ΔH‡ values again indicate that the leaving group

ability follows the order I− Br

− Cl

−. The high polarizability and low

electronegativity of the soft base I− ion once again plays a role in lowering of the

Chapter 3: *


activation enthalpy of the dissociation step, whereas the hard base Cl− ion is more

tightly bound and has a high dissociation enthalpy.

Figure 3.6: Energy profiles (in kJ mol1

) for the D pathway of

[Cr(H2O)4TX]+…H2O (X = Cl, Br, or I) for (a) T =Cl, (b) T = Br and (c) T = I in

solution phase obtained at PBE0/cc-pVDZ.

The ΔH‡ values for the second step, nucleophilic attack by the outer sphere water

molecule on the pentacoordinate metal centre, are substantially lower than the values

for the first step and are generally very similar across all systems. The overall ΔH‡

for the hydrolysis of X and T = I is lower than that for X and T = Cl by 18 kJ mol1

,

reflecting the combined contributions of leaving group ability and trans-effect. The

Chapter 3: *


calculated overall ΔH‡ and ΔS

‡ for aquation of trans-[Cr(H2O)4Cl2]

+ (Table 3.3), are

in excellent agreement with the experimental34

values (96 ± 1 kJ mol1

and ‒7 ± 4 J

mol1

, respectively).

These results provide evidence that the hydrolysis of trans‒[Cr(H2O)4TX]+

complexes (X or T = Cl, Br and I) proceeds via a D mechanism. The calculated

∆∑d(Cr‒L)TS‒R and ∆∑d(Cr‒L)P‒R values are positive for the rate-determining step

(TS1) of the reaction, consistent with a D-type mechanism (Table 3.2). The Cr…X

distances in the transition structures for this pathway are somewhat longer (+0.2 –

0.4 Å) than the corresponding values in the Ia pathway. In conjunction with this there

is a shortening of the Cr‒T distances for T = Cl, Br. However, for the trans-iodide

systems (T = I), the Cr‒T distances are either the same as the reactant complexes (X

= Cl, Br) or slightly longer (X = I).

Generally the magnitudes of the positive charge on Cr and negative charge on X

increase from reactant to TS. In comparison, the charge on T decreases, while there

is little change in the charge on O of the nucleophilic water molecule. These results

demonstrate the shift in charge from the chromium and the trans-ligand to the

leaving group. Analysis of the TSs charges reveals that the magnitude of the charge

on Cr decreases with change of T from Cl to I and also with change of X from Cl to I

(Figure 3.3). However, the charge on X does not very systematically across the

systems. For example, when T = Cl, the charge on X increases for X = Cl to X = I.

However, for the corresponding T = Br and T = I systems, there is a decrease in the

charge on X. For any given leaving group X, there is generally a decrease of ~ 18 kJ

mol‒1

in ΔH‡

for the first step from T = Cl through T = I, indicating a trans-effect in

these systems. Analysis of the frontier MO energies (HOMO and LUMO) (Figures

Chapter 3: *


SF27 to SF35 and Table ST2 of the SI) along with the corresponding Egap reveals a

gradual decrease (Figures 3.5b and Figures 3.7a, 3.7b and 3.7c) from the hard ion (T

= Cl‒) through soft ion (T = I

‒). Thus it is likely that dissociation of the Cr‒X bond

and migration of the halide ions into the outer coordination sphere via dissociation

pathways is stimulated by electronic and charge neutralizing effects.

Figure 3.7: HOMO‒LUMO energy gap (eV) for R, TS and P species of

[Cr(H2O)4TX]+…H2O(X = Cl, Br or I) for (a) T = Cl, (b) T = Br and (c) T = I via D

pathway in solution phase obtained at PBE0/cc-pVDZ.

Discussion

In this study, we have carried out a detailed investigation of the Ia and D

mechanisms for the aquation of trans-[Cr(H2O)4TX]+ complexes. For nutritional

supplement trans-[Cr(H2O)4Cl2]+, the ΔH

‡ value for the D pathway is in excellent

agreement with the experimental enthalpy of activation.34

The ΔH‡ values for the Ia

Chapter 3: *


pathway, starting from the lower energy bifurcated reaction complex, is 11 kJ mol‒1

higher than the experimental value. As noted previously, additional explicit solvation

can lower values by up to 3 kJ mol‒1

, which is insufficent to bring the Ia value in line

with experiment.

More broadly, if reaction energies are referenced to the lower energy bifurcated

reactant complex then, the calculated ∆H‡ for TS1 via D pathways (X = Cl and T =

Cl, Br or I) are substantially lower (~18 kJ mol‒1

) than the Ia pathway from X or T =

Cl through to X = Cl and T = I. In comparison, if the linear reactant complex is used

the overall ∆H‡ values for the Ia pathway drops below that of the D pathway.

However, this linear structure must arise from an endothermic intramolecular

rearrangement of the lower energy bifurcated structure, which is not captured in the

IRC analysis. We propose that the most likely mechanism for the aquation of trans-

[Cr(H2O)4TX]+ complexes proceeds via D mechanism.

A key feature of the D mechanism is that the complex becomes associatively

activated through outer sphere coordination that lowers the activation enthalpy for

dissociation of the Cr‒X bond due to H-bonding. Previously, the substitution

reaction of [Co(NH3)5(H2O)]3+

…X− was designated “accidental bimolecular” to

indicate that the substitution process is a rearrangement between inner and outer

coordination spheres,54

which in consistent with our observations for the D

mechanism in this study. Trends in ΔH‡ for the D mechanism, can be accounted for

by the trans group influence, stabilization of the electron-deficient TSs structures

and leaving group (X) ability. The frontier MO energies (HOMO‒LUMO gap) show

that complexes with the soft ion ligands (X or T = I) have a small HOMO‒LUMO

gap due to greater polarizability and electron density.

Chapter 3: *


Chloride is a harder base than bromide and iodide and therefore forms a stronger

bond with the hard acid Cr(III). Therefore, the chloride is a poorer leaving group

than the bromide or the iodide, which explains the higher ΔH‡. The decrease in ΔH

‡

for the other halides corresponds with the softer nucleophilicities and decrease in

electronegativity of these ions (X = I), so that the leaving group ability follows the

order I− Br

− Cl

−. In general, the valence electrons of the I

− ion are more loosely

bound, so the iodide is more polarizable than chloride (X or T = Cl). The greater

polarizability leads to a distortion of the electron cloud of the halide that stabilizes

the transition state. This is consistent with the analysis of the transition structures for

aquation via Ia, which reveals the Cr…O distances for the incoming water molecule

is almost constant across the series but the Cr…X distances vary from chloride

through to iodide.

The enthalpies of activation of the aquation reactions also depend on the nature

of the trans group. The greater polarizability of the trans iodide also leads to a

distortion of the electron cloud of the halide that stabilizes the transition state.

Furthermore, electronegativity influences the electron density at the Cr(III) center,

which effects the activation barrier and bond formation with the entering water

molecule. In these octahedral complexes the labilizing strength of the trans ligand

follows the expected order: I− Br

− Cl

− > H2O. Additionally, the iodide forms ion-

pairs with a more favourable ΔH‡ than the chloride ion.

Our identification of the activated D mechanism for aquation of chromium(III)

chloride has some interesting implications for the speciation of chromium(III) in

physiological environments. Activation via outer-sphere coordination is expected in

all aqueous environments including the stomach and gut. Furthermore, in addition to

Chapter 3: *


water a range of biological molecules could potentially act as nucleophile and

replace chloride after dissociation. Nevertheless, the activation enthalpy for the di-

chloride system is 72 kJ mol‒1

, so this will be a relatively slow process under

physiological conditions and the amount of aquated or substituted product could be

low. In comparison, the activation enthalpy for the corresponding di-iodide system is

only 50 kJ mol‒1

and therefore it will undergo aquation or substitution much faster.

The activation enthalpies for aquation of the mono-halide product complexes are

higher than for the dihalides due to the poor trans-directing effect of water.

Therefore, formation of the hexaaquachromium(III) complex will be slow regardless

of the starting complex. Unless the incoming nucleophile has a strong trans-directing

effect, it seems likely that the active complex in chromium(III) chloride treatments is

either [Cr(H2O)4Cl2]+ or [Cr(H2O)5Cl]

2+. However, our results suggest that there may

be some value to explore the efficacy of other dihalotetraaquachromium(III)

complexes for treatment of Type II diabetes and other physiological disorders due to

their different aquation/substitution rates.

Conclusions

A comprehensive DFT investigation was conducted to explore possible

mechanisms for aquation of trans-[Cr(H2O)4TX]+. These include the associative

interchange (Ia) and dissociative (D) mechanisms. Structural parameters and energy

barriers were calculated at PBE0/cc-pVDZ and compared with the available

experimental data. Overall, the activation enthalpy and entropy of activation

calculated at PBE0/cc-pVDZ for aquation of the chromium chloride nutritional

supplement (trans-[Cr(H2O)4Cl2]+ ) via the D mechanism, is in excellent agreement

with existing experimental data. Use of implicit solvent models has a minimal effect

Chapter 3: *


on the structures of reactants, products and transition states but is essential for

agreement between theoretical and experimental ΔH‡ values.

Our calculations indicate that the D pathway is favoured over the Ia pathway,

when referenced to the same low energy bifurcated reactant complex. Clearly steric

and electronic effects influence the enthalpies of activation for the substitution

reactions of trans-[Cr(H2O)4TX]+

complexes. The leaving group ability follows the

order: I− Br

− Cl

− and the trans-effect influence follows the order: I

− Br

− Cl

−

H2O.

We conclude that the [Cr(H2O)4Cl2]+ and [Cr(H2O)5Cl]

2+ complexes are key

active components in chromium chloride treatment of physiological disorders such

as Type II diabetes and depression. However, our calculations imply that the other

dihalides have lower activation enthalpies for aquation and therefore may offer

greater efficacy in treating these disorders.



information.

Acknowledgements


facility and the Atlantic Computational Excellence Network (ACEnet) for computer

time.

Chapter 3: *


References

(1) Vincent, J. The nutritional biochemistry of chromium (III); Elsevier,

2011.




(5) Anderson, R. A.; Cheng, N.; Bryden, N. A.; Polansky, M. M.; Cheng,

N.; Chi, J.; Feng, J. Diabetes 1997, 46, 1786‒1791.

(6) Hellerstein, M. K. Nutr. Rev. 1998, 56, 302‒306.

(7) Glinsmann, W. H.; Mertz, W. Metabolism 1966, 15, 510‒520.

(8) Nath, R.; Minocha, J.; Lyall, V. S.; Kumar, V.; Kapoor, S.; Dhar, K.

Develop. Nutr. Metabol. 1979.

(9) Mossop, R. Centr. Afric. J. Med. 1983, 29, 80‒82.

(10) Piotrowska, A.; Siwek, A.; Wolak, M.; Pochwat, B.; Szewczyk, B.;

Opoka, W.; Poleszak, E.; Nowak, G. J. Physiol. Pharmacol. 2013, 64,

493‒498.

(11) Evans, D. A. P.; Tariq, M.; Dafterdar, R.; Al Hussaini, H.; Sobki, S.

H. Biol. Trace Elem. Res. 2009, 130, 262‒272.

(12) Plaza, S. M. Alter. Med. Rev. 2002, 7, 218‒235.

(13) Anderson, R. A.; Polansky, M. M.; Bryden, N. A.; Canary, J. J. Am. J.

Clin. Nutr. 1991, 54, 909‒916.

(14) Tang, H.-y.; Xiao, Q.-g.; Xu, H.-b.; Zhang, Y. J. Trace Elem. Med.

Biol. 2015, 29, 136‒144.

(15) Haas, M. J.; Sawaf, R.; Horani, M. H.; Gobal, F.; Wong, N. C.;

Mooradian, A. D. Nutrition 2003, 19, 353‒357.

(16) Anderson, R. A. Regul. Toxicol. Pharmacol. 1997, 26, S35‒S41.

(17) Clausen, J. Biol. Trace Elem. Res. 1988, 17, 229‒236.

(18) Bjerrum, N. Collegium 1927, 691, 505.

(19) Lamb, A. B.; Fonda, G. R. J. Am. Chem. Soc. 1921, 43, 1154‒1178.

(20) King, E. L.; Woods, S. M. J. M.; Gates, O.; Gates, H. S. J. Am. Chem.

Soc. 1958, 80, 5015‒5018.

(21) Gates, H. S.; King, E. L. J. Am. Chem. Soc. 1958, 80, 5011.

(22) Elving, P. J.; Zemel, B. Can. J. Chem. 1959, 37, 247‒252.

(23) Hamm, R. E.; Shull Jr, C. M. J. Am. Chem. Soc. 1951, 73, 1240.

(24) Salzman, J. D.; King, E. L. Inorg. Chem. 1967, 6, 426‒427.

(25) Pham, T. N.; Aitken, J. B.; Levina, A.; Lay, P. A. Inorg. Chem. 2014,

53, 10685‒10694.

(26) Cooper, J. A.; Anderson, B. F.; Buckley, P. D.; Blackwell, L. F.

Chapter 3: *


Inorg. Chim. Acta 1984, 91, 1‒9.

(27) Mønsted, L.; Mønsted, O. Coord. Chem. Rev. 1989, 94, 109‒150.

(28) Langford, C. H.; Gray, H. B. Ligand Subst. Process. 1966.

(29) Basolo, F.; Pearson, R. G. Mechanisms of inorganic reactions: a study

of metal complexes in solution, 1967.

(30) Pearson, R. G.; Boston, C. R.; Basolo, F. J. Phys. Chem. 1955, 59,

304‒308.

(31) Ardon, M. Inorg. Chem. 1965, 4, 372‒375.

(32) Moore, P.; Basolo, F.; Pearson, R. G. Inorg. Chem. 1966, 5, 223‒228.

(33) Williams, T. J.; Garner, C. S. Inorg. Chem. 1970, 9, 52‒57.

(34) Monsted, L.; Monsted, O. Acta. Chem. Scand. A 1978, 32, 19‒24.

(35) MacDonald, D. J.; Garner, C. S. J. Am. Chem. Soc. 1961, 83, 4152‒

4157.

(36) Hoppenjans, D.; Hunt, J. B.; Gregoire, C. Inorg. Chem. 1968, 7,

2506‒2510.

(37) Glerup, J.; Schaeffer, C. E. Inorg. Chem. 1976, 15, 1408‒1411.

(38) Dubicki, L.; Hitchman, M.; Day, P. Inorg. Chem. 1970, 9, 188‒190.

(39) Keeton, M.; Chou, B. F.-C.; Lever, A. Can. J. Chem. 1971, 49, 192‒

198.

(40) Wirth, G.; Bifano, C.; Walters, R. T.; Linck, R. Inorg. Chem. 1973,

12, 1955‒1957.

(41) Pyke, S. C.; Linck, R. Inorg. Chem. 1971, 10, 2445‒2449.

(42) Vaughn, J. W.; Stvan Jr, O. J.; Magnuson, V. E. Inorg. Chem. 1968, 7,

736‒741.

(43) Gordon, G.; Hoppenjans, D. W.; Hunt, J. B. Inorg. Chem. 1971, 10,

754‒760.

(44) Bushey, W. R.; Espenson, J. H. Inorg. Chem. 1977, 16, 2772‒2776.

(45) Couldwell, M.; House, D.; Powell, H. Inorg. Chem. 1973, 12, 627‒

634.


(47) Uddin, K. M.; Ralph, D.; Henry, D. J. Comput. Theor. Chem. 2015,

1070, 152‒161.

(48) Atkins, P. Shriver and Atkins' inorganic chemistry; Oxford University

Press, USA, 2010.


2013.

(50) Peterson, K. A.; Figgen, D.; Goll, E.; Stoll, H.; Dolg, M. J. Chem.

Phys. 2003, 119, 11113‒11123.



(52) Reed, A. E.; Curtiss, L. A.; Weinhold, F. Chem. Rev. 1988, 88, 899‒

Chapter 3: *


926.

(53) Arshadi, M.; Yamdagni, R.; Kebarle, P. J. Phys. Chem. 1970, 74,

1475‒1482.

(54) Langford, C. H.; Muir, W. R. J. Am. Chem. Soc. 1967, 89, 3141‒3144.

Chapter 4: *

Uddin, K. M.; Henry, D. J. ChemistrySelect 2016, 1, 5236‒5249. 104

Chapter 4

Further Theoretical Studies of the Aquation of

Chromium(III) Chloride Nutritional Supplement:

Effect of pH and Solvation*

Abstract

We have explored speciation of halochromium(III) complexes under physiological

conditions involving both conjugate-base formation and halide-water exchange using

DFT. Conjugate base formation is spontaneous under neutral and alkaline

conditions. Several different pathways were explored for the water exchange

reaction of cis- or trans-Cr(III) complexes in the series trans-[Cr(H2O)3(OH)X2]0 (X

= Cl− or Br

−) and cis- or trans-[Cr(H2O)4(OH)X]

+, (X= F

−, Cl

−, Br

−, or I

−) including,

associative interchange (Ia), dissociative interchange (Id) and dissociative (D)

mechanisms. We have also investigated nucelophilic attack by OH‒ on [Cr(H2O)5X]

+

species, (X = Cl− or Br

−). The lowest overall activation enthalpies (ΔH

‡) obtained for

the F− and Cl

− systems of [Cr(H2O)4(OH)X]

+, are from the Ia pathway, whereas for

the Br− and I

− systems the Id pathway was found, consistent with the experiment. The

Chapter 4: *


importance of outer sphere solvation was investigated with the addition of up to five

explicit water molecules in the second coordination sphere. The ΔH‡ values for the

aquation of cis-[Cr(H2O)4(OH)X]+…3H2O, are lower than the corresponding values

for [Cr(H2O)5X]2+

. This indicates that there will be greater speciation of

chromium(III) halides under physiological conditions. These values also imply that

chromium(III) derivatives that may be active in the treatment of physiological

disorders such as Type II diabetes will be formed at different rates depending on the

identity of the starting halide.

Introduction

There is significant debate whether chromium(III) is of nutritional benefit for

healthy humans. However, there is clinical evidence that Cr(III) supplements can

lower blood sugar levels, improve glucose metabolism, and cholesterol levels in

people with type II diabetes.1

Nevertheless, questions remain about the potential

toxicity of chromium(III).1‒2

A key factor in investigating the biochemical pathways

of chromium(III) compounds is identification of the speciation of these compounds

in aqueous and/or physiological environments. One common chromium containing

component of nutritional supplements is chromium(III) chloride (CrCl3∙6H2O),

which actually exists as the dichlorotetraaquachromium(III) complex, trans-

[Cr(H2O)4Cl2]Cl∙2H2O.3 Although aquation of chromium(III) dihalide complexes has

been proposed to occur via either an interchange or dissociative process at low pHs,

the conjugate base pathway may be important at moderate pH and under neutral

conditions.4

For example, hexaaquachromium(III) has a pKa of ~ 4 so removal of a

proton from a coordinated water molecule is relatively facile.5

The relevance of this

pathway is suggested by Choudhary et al.6 who proposed that the conjugate-base for

Chapter 4: *


the chromium(III) hexaaqua ion ([Cr(H2O)5OH]2+

) is the most active Cr(III) species

for catalyzed glucose isomerization. Recently, Mushrif et al.7

used Car-Parrinello

molecular dynamics with metadynamics to investigate chromium (III) catalyzed

glucose-to-fructose isomerization. They proposed that the conjugate base,

[Cr(H2O)5(OH)]2+

,binds to glucose occurs prior to ring opening and isomerisation.6‒7

The conjugate base of chromium(III) complexes can also undergo hydrolysis or

coordinate to a second metal centre to form bridged complexes and polymeric

species.2,8

At higher pHs, the process can continue through to formation of hydrated

chromium(III) oxide,2

which has been implicated in a number of pathological

disorders, such as dermatitis, liver or kidney damage, lung or throat cancer, and is

therefore toxic in large doses in humans.9

Consequently there is interest in studying

base hydrolysis processes and determining the speciation of Cr(III) complexes and

their conjugate-bases.

Several experimental studies have investigated the mechanistic details of

conjugate-base hydrolysis of halo-hydroxochromium(III) species.4,10‒12

Xu et al.4a

reported that the conjugate-base, [Cr(H2O)5(OH)]2+

undergoes water exchange ~75

times faster than [Cr(H2O)6]3+

. However, they obtained ΔH‡ values of 108.6 ± 2.7 kJ

mol1

and 111.0 ± 2.5 kJ mol1

for water exchange at [Cr(H2O)6]3+

and

[Cr(H2O)5(OH)]2+

respectively, using oxygen-18 tracer studies. They suggest that

the difference in exchange rates therefore mainly depends on a large positive entropy

of activation and is nearly independent of temperature. The experimental volumes of

activation (∆V‡) suggest that the reactions proceeds via an associative mode for

[Cr(H2O)6]3+

and a dissociative mode, for [Cr(H2O)5(OH)]2+

due to stabilisation of

the transition state by the hydroxide.4c,13

Espenson investigated13

the reactions of

Chapter 4: *


[Cr(H2O)6]3+

and [Cr(H2O)5(OH)]2+

with different anionic and neutral ligands and

proposed that the reaction of [Cr(H2O)5(OH)]2+

and HX was far more important than

that of [Cr(H2O)6]3+

and X‒ systems. Similarly, Thusius found

4c ΔH

‡ by 17‒21 kJ

mol1

for aquation of the same reaction series13

of [Cr(H2O)5X]2+

(X = Cl‒, Br

‒, I

‒,

ONO2‒, SCN

‒, and NCS

‒) compared to [Cr(H2O)6]

3+ differ. Thusius observed

4c the

aquation of [Cr(OH2)5OH]2+

with X (X = HF, Cl, Br, and I) and found the activation

of enthalpy (ΔH‡) of 79, 109, 95 and 113 kJ mol

1, respectively. However, the lower

ΔH‡ of the [Cr(H2O)5(OH)]

2+ reactions is of dissociative nature and may be due to

the stabilization of the transition state by coordinated hydroxide.13‒14

Parris and co-

workers[12b]

prepared a series of [Cr(RNH2)5Cl]2+

complexes (where R = H, CH3,

C2H5, n-C3H7, and n-C4H9) and investigated the aquation reactions. They also

determined12b

from the corresponding trends of ΔS‡ values for acid and base

hydrolysis that the rate-determining step proceeds via a dissociative mechanism in

both cases. Similarly, Carey et al. proposed4d

that the substitution reactions of the

conjugate-bases [Cr(H2O)4(OH)X]+ (X = Br or I) occurs via a dissociative

interchange (Id) mechanism. Ardon proposed[12a]

that ligand substitution reactions of

[Cr(H2O)5I]2+

proceeds via a dissociative mechanism with formation of [Cr(H2O)6]3+

and I‒.4f,11

Many workers investigated4,10‒11

that the changes of green chromic

chloride solution into the blue (violet) isomer convert through the intermediate

formation of an hydrolysis product of the green chloride solution,

[Cr(H2O)3(OH)Cl2] of the change in significant dependence upon pH.

In comparison to the many experimental studies, there have been only a limited

number of theoretical studies of the hydrolysis of conjugate-bases of Cr(III)

complexes. Rotzinger et al.12d

carried out an extensive study of the base hydrolysis of

the octahedral second and third-row transition metal amine complexes and proposed

Chapter 4: *


the conjugated base, trans-[Cr(NH3)4(NH2)Cl]+ complex proceed via a concerted

substitution process (i.e. an Id mechanism). Rotzinger et al.15

found no structural

difference between the conjugate base, cis-[Cr(NH3)4(NH2)Cl]+ and the trigonal

bipyramidal pentacoordinated intermediates, cis-eq-tbp-Cr(NH3)4(NH2)∙Cl+.

Recently we investigated16

the aquation of halopenta aquachromium(III) complexes

in the series [Cr(H2O)5X] +

, (X = F, Cl, Br, or I) using density functional theory. On

the basis of excellent agreement with experimental activation enthalpies, we

proposed that aquation of the fluoropentaaquachromium(III) complex proceeds via

an Ia pathway, followed by immediate formation of the conjugate base through

proton transfer. In comparison, the most likely mechanism for aquation of the X =

Cl, Br or I complexes is the dissociative pathway from an activated precursor

([Cr(H2O)5X]2+…H2O). Furthermore, we investigated

17 aquation of chromium(III)

chloride (trans-[Cr(H2O)4Cl2]) a common component of many nutritional

supplements and some related trans-dihalotetraaquachromium(III) complexes using

density functional theory. The overall activation enthalpy for the D pathway

calculated at the PBE0/cc-pVDZ level, in aqueous solution (PCM), is in excellent

agreement with the experimental result.18

Our previous study of chromium(III) chloride and other dihalides, focussed on

acidic conditions as might be found in the stomach after oral ingestion of these

species. The main finding of this study was that aquation of the dichloride complex

has a high activation enthalpy and therefore the most likely species present are the

dichloride or the dihalide. If these are indeed the key species that survive the highly

acidic environment of the stomach then they will pass on to the duodenum which has

a milder pH and then may eventually enter the blood stream, which has a

physiological pH of ~ 7.4. Conjugate-base formation is favourable under these

Chapter 4: *


conditions. Therefore, in this study we extend our previous work of chromium(III)

hydrolysis to conjugate-base systems through investigation of the aquation reactions

of trans-[Cr(H2O)3(OH)X2]0 (X = Cl or Br) and cis or trans-[Cr(H2O)4(OH)X]

+ (X =

F, Cl, Br or I) species and [Cr(H2O)5X] +

, (X = Cl or Br) with OH‒. This will lead to

the development of a better understanding of the stability and speciation of these

species under physiological conditions.


Effect of solvation on structural parameters

Our previous studies16‒17

have shown that PBE0/cc-pVDZ combined with PCM

implicit solvation is a suitable level of theory for determining the structural

parameters of chromium(III) complexes. However, given the important role that

solvent plays in conjugate-base formation, we investigate the effects of full explicit

hydration using the supermolecular representation, with 12 or 13 second-shell waters in

the second coordination sphere using a supermolecular-PCM (SM-PCM) model.

Structural parameters for the lowest energy conformations of several

haloaquachromium(III) species are presented in Table 4.1 obtained using

quasichemical (PCM) and supramolecular representations of the systems the SM-

PCM model (Figure 4.1), with experimental values where available.19

Also included

are values for cis- and trans- isomers of the corresponding conjugate-base species

obtained with SM-PCM (cis- or trans-[Cr(H2O)4(OH)X]+∙mH2O (m = 13) and

[Cr(H2O)5(OH)]2+

∙mH2O (m = 12)) (Figures 4.1 and SF1 to SF7 of the SI).

Chapter 4: *


Table 4.1: Selected distances (Å) for conjugate bases cis or trans-[Cr(H2O)4(OH)X

and [Cr(H2O)5X]2+

in the water solvent at 298.15 K.

T Bond cis-

[Cr(H2O)4

(OH)X]+a

trans-

[Cr(H2O)4

(OH)X]+a

[Cr(H2O)5X]2+

(PCM)b

[Cr(H2O)5X]2+

(SM-PCM)a

Exptlc

F Cr‒X 1.910 1.968 1.806 1.873

Cr‒OH2d 2.038 2.006 2.003 2.002

Cr‒OHe 1.847 1.888 ‒ ‒

Cr‒Of 4.027 3.973 ‒ 3.95

Cl Cr‒X 2.397 2.444 2.230 2.302

2.26 ± 0.05

Cr‒OH2d 2.023 2.001 2.010 1.986

1.98 ± 0.04

Cr‒OHe 1.861 1.882 ‒ ‒

Cr‒Of 4.095 4.089 ‒ 4.068 3.97 ± 0.02

Br Cr‒X 2.565 2.615 2.385 2.515

Cr‒OH2d 2.030 1.999 2.011 1.986

Cr‒OHe 1.839 1.881 ‒ ‒

Cr‒Of 4.142 3.961 ‒ 4.076

I Cr‒X 2.826 2.888 2.619 2.745

Cr‒OH2d 2.023 2.000 2.016 1.988

Cr‒OHe 1.843 1.865 ‒ ‒

Cr‒Of 4.168 4.109 ‒ 4.078

H2Og

Cr‒OH2d 2.016 ‒ 1.998 1.978 2.00 ± 0.01

Cr‒OHe 1.857 ‒ ‒ ‒

Cr‒Of 3.918 ‒ ‒ 3.887 4.02 ± 0.08

a Optimized structures of cis or trans-[Cr(H2O)4(OH)X]

+∙13H2O; [Cr(H2O)5X]

2+ and [Cr(H2O)5X]

2+∙

13H2O complexes defined in Figures 4.1 and SF0 to SF7 in in water of the SI. b Ref. 16.

c Ref. 19.

d The average Cr‒O distances to the water molecules in the inner shell.

e The average Cr‒O distances to the hydroxyl ligand in the first hydration shell.

f The average Cr‒O distances to the water molecules in the second hydration shell.

g The average distances for [Cr(H2O)6]

3+∙12H2O and cis or trans-[Cr(H2O)5(OH)]

2+∙12H2O.

Inclusion of the explicit second solvation shell leads to a significant network of

hydrogen bonds around the complex but only a slight increase in the Cr−X distance.

The average distance to water molecules in the second hydration shell is slightly

overestimated with SM-PCM. Nevertheless, both the PCM and SM-PCM structures

are in good agreement with experiment.

Chapter 4: *


Figure 4.1: Structures of the precursor complexes: (a) cis-[Cr(H2O)5(OH)Cl]+; (b)

cis-[Cr(H2O)4(OH)Cl]+∙13H2O;(c) trans-[Cr(H2O)5(OH)Cl]

+;(d) trans-[Cr(H2O)4

(OH)Cl]+∙3H2O; (e) [Cr(H2O)5(OH)]

2+ and (f) [Cr(H2O)5(OH)]

2+∙12H2O in solution

phase obtained at PBE0/cc-pVDZ.

Focussing on the SM-PCM structures, the Cr‒X bonds of the cis- and trans-

isomers of [Cr(H2O)4(OH)X]+ are longer (~0.12 Å) than the corresponding bonds in

the [Cr(H2O)2X]2+

complexes, particularly for the trans isomers where the hydroxide

exerts a structural trans-effect (~0.05 Å) (see Table S0 of the SI). There is generally

only a slight increase in the Cr−O distance to inner shell water molecules of

[Cr(H2O)4(OH)X]+ relative to [Cr(H2O)5X]

2+, except again for bonds trans to OH.

Chapter 4: *


The SM-PCM value for Cr‒OH bond lengths vary between 1.839 Å and 1.888 Å

with the larger values arising from the trans-[Cr(H2O)4(OH)X]+ complexes.

Our calculations indicate that formation of the conjugate base has a larger effect

on the structural parameters than the explicit outer-sphere solvation of the complex.

Gibbs free energies for proton transfer (ΔGdp) from haloaqua chromium(III)

complexes

In all cases, the aqueous deprotonation of [Cr(H2O)5X]2+

was investigated using

reactions (4.1) and (4.2) and the SM-PCM model with explicit inclusion of water

molecules in the outer shell of the complexes (Tables 4.2 and ST11 in the SI).

However, as noted in previous section, solvation of H+ and OH

‒ is also important in

determining the thermodynamics of these processes. Therefore, we have calculated

aqueous Gibbs free energies for deprotonation from the cis-position of

halohydroxoaqua Cr(III) species with different values of n, the number of water

solvent molecules around the H+ or OH

‒ ions (i.e. H

+(H2O)n and OH

‒( H2O)n).

Table 4.2: Calculated aqueous Gibbs free energies (kJ mol‒1

) for deprotonation

(ΔGdp) of [Cr(H2O)5X]2+

∙mH2O at 298.15 K.a‒b

[Cr(H2O)5X]2+∙13H2O + nH2O cis- or trans-[Cr(H2O)4(OH)X]+∙13H2O + H+( H2O)n (4.1)

([Cr(H2O)5X]2+∙12H2O + OH‒( H2O)n cis- or trans-[Cr(H2O)4(OH)X]+∙13H2O + nH2O) (4.2)

n F Cl Br I

1 59 (‒383) 48 (‒394) 46 (‒397) 38 (‒404)

2 ‒15 (‒272) ‒27 (‒284) ‒29 (‒286) ‒36 (‒293)

3 ‒43 (‒232) ‒55 (‒244) ‒57 (‒246) ‒64 (‒253)

4 ‒45 (‒216) ‒56 (‒227) ‒58 (‒230) ‒66 (‒237) a The proton transfer reaction Gibbs free energies in neutral media are calculated using

reaction 4.1.b The parenthesis values are for aqueous reaction Gibbs free energies in alkaline

media calculated using reaction 4.2.

The ΔGdp values decrease for both cis- and trans- species from X = F through to

X = I. A variation of ~ 21 kJ mol‒1

is observed in ΔGdp across the series of halides

Chapter 4: *


for each value of n (Tables 4.2 and ST11). Deprotonation from the trans-position

also leads to a systematic change in ΔGdp with X but the variation is larger (~ 61 kJ

mol‒1

) (Table ST11 in the SI). We also note that for deprotonation from both the cis-

and trans-positions, ΔGdp values systematically decrease by 104 kJ mol‒1

for each

halide as n increases from 1 to 4. The cis-Cr(III) conjugate bases are more stable

than the corresponding trans-species by 12.4, 2.5, 8.1 and 3.5 kJ mol‒1

for X = F, Cl,

Br and I, respectively. Deprotonation of all haloaqua Cr(III) complexes was found to

be exergonic except when calculated using equation 1 and hydronium ion (n = 1).

The largest change in ΔGdp values with solvation occurs from n = 1 to n = 2 (~ 75 kJ

mol‒1

). The subsequent change from n = 2 to n = 3 is ~ 28 kJ mol‒1

and this

decreases to ~1.5 kJ mol‒1

from n = 3 to n = 4. These changes emphasise the

importance of explicit solvation of the proton in these processes and imply that at

least three water molecules are required to adequately stabilise this ion product. The

ΔGdp values are quite different between H2O and OH‒ as base. Although the same

systematic variation in ΔGdp is observed with change of halide, the opposite trend

with n is observed. This occurs because the smaller ion (OH‒) is on the reactant side

of the process. Therefore, increased stabilization of this ion through explicit

solvation reduces the magnitude of ΔGdp. We note that the changes in ΔGdp with n

are larger for reaction 4.2 than reaction 4.1 and suggest that at least 4 water

molecules are required to stabilize OH‒.

Interchange mechanism

As shown in the previous section, deprotonation of the haloaquachromium

complexes is energetically favourable under neutral and alkaline conditions. The

resulting conjugate base species can subsequently undergo aquation in a similar

Chapter 4: *


manner to the parent conjugate acids. Aquation reactions at chromium(III) halide

complexes can potentially occur via several different mechanisms including

interchange and dissociative pathways. To simplify the process of locating

transitions states for these processes we use a quasichemical representation. We

begin by considering, the interchange mechanism, a concerted bimolecular process,

for which Langford and Gray20

proposed two variations: associative interchange (Ia),

indicating a significant degree of bond formation between the entering ligand and

metal centre in the TS (i.e. ∆V‡ is positive) and dissociative interchange (Id),

involving a degree of bond rupture as the transition state is formed (i.e. ∆V‡ is

negative).

Scheme 4.1: Interchange mechanism for the aqaution of trans-

[Cr(H2O)3(OH)X2]0 (X = Cl

– or Br

–) and cis or trans-[Cr(H2O)4(OH)X]

+ (X = F

–,

Cl–, Br

–, or I

–).

H2O

Cr

L1

O O

O

X

L

H

HH

H

cis- or trans-[Cr(H2O)4(OH)X]+

Cr

O

OO

X

O

OH

HH

H

O

H

H

Cr

O

O O

OO

H

H

H

H

H

H

OX

H

H

H

H

H

H

H

H

CrO

O O

O

X

O

H

H

H

H

H

H

H

H

OH

H

Cr

O

OO

X

O

OH

HH

H

L

H

H

Cr

O

O O

OL

H

HH

H

OX

H

H

H

H

H

H

CrO

O O

O

X

L

H

H

H

H

H

H

OH

H

H

H

(R) (TS) (P)

(R) (TS) (P)

HH

H

H H H

cis: L = H2O

trans-[Cr(H2O)3(OH)XL]0

X or L = Cl or Br

The interchange pathway for aquation of trans-[Cr(H2O)3(OH)X2]0 and cis- or

trans-[Cr(H2O)4(OH)X]+ with H2O is initiated by outer sphere coordination of a

water molecule to the Cr(III) species as shown in Scheme 4.1. For both the cis- and

trans-complex, the incoming H2O forms a hydrogen bond with an inner sphere water

Chapter 4: *


molecule and has a weak interaction with the halide X, to give the stabilized

precursor/reactant complex (R).

Figure 4.2: Energy profiles (kJ mol1

) of cis- and trans-[Cr(H2O)4(OH)X]+ (X = F

–,

Cl– Br

–, or I

–) for (a−b) aquation via the interchange pathway; for (c−d) the Cr‒X

dissociation (D) pathways; and for (e−f) the D pathway of

[Cr(H2O)4(OH)X]+…H2O

–) in solution phase obtained at PBE0/cc-pVDZ.

As noted previously, outer-sphere coordination of the attacking nucleophilic

water molecule can occur either with linear hydrogen bonds to X and H2O(inner) or via

bifurcated H-bonding to H2O(inner). The precursor bifurcated structures (F−, Cl

− Br

−or

I−) with bridging hydrogen bonds (Figures SF8 and SF9 of the SI) are energetically

more stable than those with linear hydrogen bonds by 1.6, 9.6, 16.2 and 14.6 kJ

Chapter 4: *


mol1

, respectively. However, IRC analysis of the transition states confirms the

linear H-bonded structures (Scheme 4.1) as the precursor of cis or trans-Cr(III)

complexes for the interchange pathway (Figures SF10 to SF15). The associated

reactant complexes (R) all show a similar level of stability (~72 kJ mol‒1

) relative to

the isolated reactants, except for the cis-fluoride system that is stabilised by only 38

kJ mol‒1

(Figures 4.2a‒4.2b).

For the trans-[Cr(H2O)3(OH)X2]0 reactant complexes (X = Cl or Br) the HO‒

H…X distances for the linear conformations differ from X = Cl (2.32 Å) through to

X = Br (2.47 Å) as shown in Figures SF14 to SF15. The HO‒H…X distances for the

linear conformations of cis-[Cr(H2O)4(OH)X]+ (Figures SF10 to SF13 of the SI)

show variation with halide X from X = F (1.69 Å) through to X= I (2.84 Å). These

variations reflect differences in the electronegativity and polarisability of the halides.

The corresponding values for the trans-[Cr(H2O)4(OH)X]+ and trans-

[Cr(H2O)3(OH)X2] complexes are marginally shorter (0 – 0.09 Å). Coordination of

the H2O to the outer sphere of the reactant complex leads to a marginal increase in

cis-Cr‒OH (~0.01 – 0.02 Å) bond lengths compared to the isolated species. The

geometry around the metal centre of the associated cis-[Cr(H2O)3(OH)X]+ reactant

complexes remains close to octahedral with trans-X−Cr−OH2 bond angles of 175.2

– 177.1) and cis-X−Cr−OH2 bond angles of 97.2 – 98.9. Likewise, for trans-

[Cr(H2O)3(OH)X]+ reactant complexes the geometry remains close to octahedral

with trans-X−Cr−OH bond angles of 172.1 – 173.6) and cis-X−Cr−OH2 bond

angles of 87.4 – 94.7 (Figures SF16 to SF19). The lower stability of the linear

conformations (cis- or trans-) arises largely from a weaker hydrogen bonding to the

outer sphere water molecule rather than a significant destabilization of the metal

complex.

Chapter 4: *


The energy profiles for hydrolysis of cis- and trans-[Cr(H2O)4(OH)X]+

are

shown in Figures 4.2a‒4.2b and the enthalpies of activation (ΔH‡), free energies of

activation (ΔG‡) and entropies of activation (ΔS

‡) are given in Table 4.3. For the cis-

series of complexes, the barriers decrease with X changing from chloride through to

iodide, which is consistent with our previous study of aquation of

halopentaaquachromium(III) complexes and is explained in terms of the changes in

electronegativity and polarisability of the leaving halide. The variations in the ΔH‡

values for the trans- complexes reflect competing electronic effects between the

trans halide and hydroxide ligands. This leads to an increase in ΔH‡ for the trans-F

system relative to the cis-F system but decreases for the other systems. As the

hydrolysis reactions for conjugate bases proceed via the interchange pathway there is

a distortion of the complex geometry including the cis-(X−Cr−OH) or trans-

(X−Cr−OH) bond angles and Cr−X bond lengths, which contributes to ΔH‡.


‒1), Gibbs free energies of

activation (ΔG‡, kJ mol

‒1), and entropies (ΔS

‡, J mol

‒1K

‒1) for interchange

pathways of cis- or trans-[Cr(H2O)4(OH)X]+ and trans-[Cr(H2O)3(OH)X2]

0 with

H2O in the water solvent at 298.15 K.a

cis-

[Cr(H2O)4(OH)X]+

trans-

[Cr(H2O)4(OH)X]+

trans-

[Cr(H2O)3(OH)X2]0

X ΔH‡ ΔG

‡ ΔS

‡ ΔH

‡ ΔG

‡ ΔS

‡ ΔH

‡ ΔG

‡ ΔS

‡

F 77 90 ‒43

98 107 ‒28

‒ ‒ ‒

Cl 105 107 ‒6

70 77 ‒25

41 50 −31

Br 82 88 ‒20

51 59 ‒24

36 47 −39

I 79 82 ‒2 71 82 ‒36 ‒ ‒ ‒ a Optimized structures of cis- or trans-Cr(III)

complexes defined in Figures SF8 to SF19 of

the SI.

The Cr(III)…X bond distances in the transition structures range from 2.20 Å in

the X = F through to 3.87 Å in the X = I for cis systems (Figures SF10 to SF13 of the

SI). The Cr(III)…OH2 distances in the fluoride and chloride are 2.41 Å and 2.52 Å,

Chapter 4: *


respectively, whereas the corresponding value for the bromide and iodide systems is

2.99 Å. The (Cr(III)…X) distances in the transition structures for trans-Cr(III)

systems are slightly longer and range from 2.38 Å through to 4.06 Å. The

Cr(III)…OH2 distances are slightly longer in the trans-Cr(III) systems compared to

the cis-Cr(III) systems but follow the same general trend. Therefore, the bromide and

iodide ions may be mechanistically dissociative interchange in character due to

steric-crowding.

Figure 4.3: Maps of electrostatic potential (0.02 electrons Bohr‒3

) for selected

transition structures of cis- or trans-[Cr(H2O)4(OH)X]+ (X = F, Cl, Br or I) and cis-

[Cr(H2O)3(OH)X2]0 (X = Cl or Br).

Hydrogen bonding is important in stabilising the TSs for these processes and in

each system the departing halide interacts with the three closest water molecules,

which includes the incoming water molecule and two inner-sphere water molecules

Chapter 4: *


on the departing face of the complex (Scheme 4.1 and Figure 4.2a−4.2b). The

particularly, low activation enthalpy for X = F is due to O‒H…X, H-bond formation,

whereas for the other systems there is formation of HO‒H…X H-bonds. There are

only marginal differences in the outer sphere coordination in the successor

complexes (P).

The changes in geometry are accompanied by changes in the charges on several

key atoms in the complex. To explore the importance of atomic charges on the

hydrolysis reactions, natural bonding orbital (NBO)21

analysis was performed and

the calculated charges are given in Table ST1 of the SI. The positive charge on

Cr(III) and the magnitude of the negative charge on halide increase on formation of

the TS. However, the charges on Cr (+0.756) and I (‒0.302) are smallest in

magnitude in the iodide system due to the high polarisability of the valence electrons

of iodide, which stabilises the TS (Figure 4.3 and Table ST1). The oxygen atom of

the incoming water molecule exhibits relatively little change in charge across the

various systems as shown in the selected maps of electrostatic potential for the

interchange mechanism systems (Figures 4.3 and SF27 to SF29 and Table ST1 of the

SI). Consequently, the more polarizable iodide ion leads to less charge-charge

repulsion between the halide and the incoming water, compared to the other halides

(Figure SF30 of the SI). The variation of H-bond and change in charge for cis-

[Cr(H2O)4(OH)]2+

…X‒…H2O (X = F or Cl) observed in the conjugate bases implies

more association of water in the departure of fluoride or chloride ion systems due to

solvent electrostriction around the charge centres in the transition states. We note,

the successor complex of the fluoride system is substantially less stable than the

corresponding species for the other halides. However, there is only a very small

Chapter 4: *


barrier for this species to undergo intramolecular proton transfer to the fluoride

leading to formation of HF (Figures 4.2a−4.2b and SF10 of the SI).

To further explore the importance of outer-sphere explicit solvation, we

investigated the effects of inclusion of two, three, four and five outer sphere water

molecules on the aquation of cis-[Cr(H2O)4(OH)Cl]+ via the interchange pathway

(Figure SF22 of the SI). To begin with, several conformations were explored for the

precursor reactant complexes, cis-[Cr(H2O)4(OH)Cl]+…mH2O (m = 2), as shown in

Figures 4.4 and SF20 of the SI.

Figure 4.4: Conformations of precursor cis-[Cr(H2O)4(OH)Cl]+…2H2O complexes

involving both linear and bifurcated structures from (a) to (f).

Conformations for cis-[Cr(H2O)4(OH)Cl]+…2H2O complexes with bridging

bifurcated hydrogen bonds (Figure 4.4a) are energetically more stable than those

with a combination of linear and bifurcated hydrogen bonds (Figures 4.4b‒4.4f) by

2.4, 3.4, 7.7, 16.7 and 30.8 kJ mol1

, respectively. Similarly, a range of different

precursor complex conformations with three outer sphere water molecules were

investigated (Figure SF21 of the SI). Again, the lowest energy precursor

conformation has the outer sphere water molecules linked to the inner sphere through

Chapter 4: *


bridging bifurcated H-bonds (Figure SF21). Other conformations were found to be

2.5‒45.1 kJ mol1

higher in energy. Figure 4.5 reveals that the transition structures

for cis-[Cr(H2O)4(OH)Cl]+…mH2O (m = 2−5) are generally similar to Scheme 4.1

but have additional water molecules coordinated to the outer sphere of the complex

(Figures S20 to S22 of the SI).

Figure 4.5: Optimized transition structures and maps of electrostatic potential on the

0.02 electrons Bohr‒3

total electron density isosurface for selected transition

structures of cis-[Cr(H2O)4(OH)Cl]+…mH2O for (a) 2H2O, (b) 3H2O, (c) 4H2O, (d)

5H2O, and (e) activation enthalpies (∆H‡) in different conformers (m = 1‒5) via

interchange pathway in solution phase obtained at PBE0/cc-pVDZ.

Hydrogen bonding between the departing halide and the neighbouring water

molecules again plays a role, assisting in the ligand transfer in cis-

[Cr(H2O)4(OH)Cl]+.mH2O (m =1‒5) systems (Figure 4.5a−4.5e). This is clearly

demonstrated by the changes in overall activation enthalpies for the interchange

pathway of cis-[Cr(H2O)4(OH)Cl]+…mH2O complexes (105, 86, 73, 73 and 72 kJ

Chapter 4: *


mol1

, for m =1‒5, respectively). These results indicate that three water molecules

are required to stabilise the transition states for this process with additional explicit

solvation having a minimal effect. The oxygen atom of the incoming water molecule

and chloride atom of the leaving group exhibits relatively little change in charge

across the two, three, four and five water systems as shown in the selected maps of

electrostatic potential (Figure 4.5) for the interchange mechanism. To further

explore the importance of explicit solvation of the TS, additional conformations were

explored. Figure 4.5e provides a plot of activation enthalpy versus conformation for

different numbers of explicit outer sphere water molecules. Although changes in

conformation can have a significant effect on ∆H‡, it is again found that three outer

sphere water molecules are generally sufficient to stabilise the TS. Interestingly, the

most stable TS conformers, have a combination of linear and bifurcated hydrogen

bonds. This can be understood from their interaction with the inner sphere water

molecules in the TS as the ion migrates to the outer sphere of the complex. When

more than three water molecules are included the additional water molecules have

the weakest interaction with the cis-water molecules and therefore there is a small

difference (1 kJ mol1

) in ΔH‡ between three to five waters. Therefore, the lowest

energy conformation of the cis-[Cr(H2O)4(OH)Cl]2+

…3H2O TS was used as a

starting point for location of the interchange transition states of the other halides.

The activation energies for the hydrolysis of cis-[Cr(H2O)4(OH)X]+∙3H2O

complexes (X = F, Cl, Br or I) with three waters (Figure 4.6) are 85, 59, 47, and 32

kJ mol1

, respectively, indicating the additional solvation leads to a decrease in ΔH‡

for X = Cl, Br and I but an increase for the fluoride system. Clearly, the hydrogen

bonding network around the complexes makes a relatively significant contribution to

ΔH‡, as shown by the decreases in the activation enthalpies (35‒47 kJ mol

1) for X =

Chapter 4: *


Cl, Br and I, compared to the corresponding one water systems (Figures 4.6 and

SF10‒SF13 and SF22A of the SI). It was found that the interaction between X– and

H2O was strongly repulsive for the fluoride system alone, which correlates with the

higher activation enthalpy. The Cr…X distances of the transition states of [Cr

(H2O)4(OH)X]+…3H2O

are longer than the m = 1 values (0.19 – 0.35

Å) and show a

gradual increase from fluoride through to iodide.

Figure 4.6: Optimized transition structures of cis-[Cr(H2O)4(OH)X]+∙3H2O (X = F,

Cl, Br or I) via interchange pathway in solution phase obtained at PBE0/cc-pVDZ.

Dissociative (D) mechanism

Aquation of the conjugate base complexes [Cr(H2O)4(OH)X]+ via the

dissociative (D) mechanism is a two-step reaction. The first step involves breaking

of the Cr‒X bond with formation of a pentacoordinated intermediate (Scheme 4.2).

The released halide ion is coordinated to the outer sphere of the intermediate

Chapter 4: *


complexes (I) to give the ion-pairs, cis-[Cr(H2O)3(OH)]+…FH or cis-

[Cr(H2O)4(OH)]2+

…X− (X = Cl, Br or I).

Scheme 4.2: Dissociative mechanism (D) for the cis-[Cr(H2O)4(OH)X]+ (X = F,

Cl, Br, or I).

The barrier for Cr‒X dissociation of trans-[Cr(H2O)3(OH)Cl2]0

is quite low (46

kJ mol‒1

), which indicates that release of the halide will be quite fast (Table 4.4 and

Figures SF36 to SF37 of the SI). Consistent with our previous study, the barrier is

even lower for the tran-bromide species (40 kJ mol‒1

). Aquation of both of these

complexes will lead to formation of cis-[Cr(H2O)4(OH)X]+ species. The enthalpies

and entropies of activation for the dissociation step of cis- or trans-

[Cr(H2O)4(OH)X]+ (X = F, Cl, Br

or I) are given in Table 4.4. The optimized


to SF41 of the SI, respectively.

Chapter 4: *




‡, J mol

‒1

K‒1

) for Cr‒X bond dissociation pathways of cis- or trans-[Cr(H2O)4(OH)X]+

and trans-[Cr(H2O)3(OH)X2]0 in the water solvent at 298.15 K.

a

ΔH‡ (ΔS

‡ )

X

cis-

[Cr(H2O)4(OH)X]+

trans-

[Cr(H2O)4(OH)X]+

trans-

[Cr(H2O)3(OH)X2]0

FH 93 (‒11)

70 (‒22)

‒

Cl 72 (‒34)

55 (‒13)

46 (‒21)

Br 70 (‒30)

44 (‒19)

40 (‒31)

I 64 (‒29) 41 (‒15) ‒ a Optimized structures of cis or trans-Cr(III)

complexes defined in Scheme 4.2 and Figures

4.2c−4.2d and SF32 to SF41 of the SI.

The ΔH‡

values for the Cr‒X dissociation of the monohalides are higher than for

the dihalides and gradually decrease from X = F‒, Cl

−, Br

− and I

− for both the cis-

and trans-systems (Figure 4.2c−4.2d). However, the values for the trans-systems are

17 – 26 kJ mol‒1

lower than the corresponding values for the cis-systems. The

importance of polarizability of the halide is reflected in the Cr…X distances in the

TS for the cis-[Cr(H2O)4(OH)X]+ species, which increase from 2.55 Å for the

fluoride system through to 3.61 Å for the iodide system, (Figures SF32 to SF35 of

the SI). The trans-[Cr(H2O)4(OH)X]+ species exhibit a similar variation with Cr…X

distances that range from 2.53 to 3.64 Å (Figures SF38 to SF41). The corresponding

Cr…X distances (Cl or Br) for the cis-[Cr(H2O)3(OH)X2]0

(X = Cl or Br) systems

are 3.18 and 3.41 Å, respectively. The calculated ∆∑ d(Cr−L)TS‒R values for the all

cis- (trans-) systems, range from 0.66 (0.61) Å to 0.87 (0.86) Å, which is consistent

with a positive activation volume and a D-type mechanism (see Table ST8 of the SI).

In the second step, there is nucleophilic attack by a water molecule on the

pentacoordinate species (Scheme 4.3) (Figures SF42 to SF43 of the SI) that results in

the octahedral product complex, (P). The activation enthalpy for H2O addition to the

Chapter 4: *


trans-[Cr(H2O)5(OH)]2+

complex is 12 kJ mol‒1

lower than for addition to the cis-

complex (Table ST7 of the SI).

Scheme 4.3: Associative mechanism (A) for the cis-[Cr(H2O)4(OH)]2+

with H2O.

We have previously shown an improved model of the dissociation process

includes water coordination to the outer sphere of the complex, which can have an

influence on lowering ΔH‡ for the dissociation step (Table 4.5 and Figure 4.2e−4.2f).

In these systems, the dissociative pathway begins from precursor complexes with

bifurcated H-bonding to an outer sphere water molecule. The other features of the

process are generally similar to Scheme 4.4. However, when the halide ion becomes

coordinated to the outer sphere of the complex through two X…H hydrogen bonds,

the outer sphere water molecule adopts a cis position to facilitate attack on the

Cr(III) metal centre (Scheme 4.4). Nucleophilic attack by the water molecule on the

pentacoordinate Cr(III) metal ion occurs in the second step and results in the

octahedral ion-pair product, (P). The formation of this product occurs via a

Chapter 4: *


hexacoordinate transition state (TS2). Optimized structures of the reactants, products,

and transition states can be found in Figures SF44 to SF51 of the SI.

Scheme 4.4: Dissociative (D) mechanism for aquation of cis-[Cr(H2O)4

(OH)X]+… H2O (X = Cl, Br, I).

The ΔH‡ and ΔS

‡ for each of the steps of the D pathway from the associated

complex with H2O are given in Table 4.5 and the relative energy profiles for the

individual reactions are shown in Figure 4.2e−4.2f.

Analysis of the charges in the reactant complexes (R) reveals that the magnitude

of the charge on Cr decreases as the halide atom changes from F (1.081) to I (0.735)

and the charge on X decreases from F (‒0.537) to I (‒0.298) for the cis- systems

(Table ST10 and Figure SF57 of the SI). Similarly, for the trans-systems the R

charges on Cr decreases from F (1.079) to I (0.794). However, analysis of the TSs

charges reveals that the charge on X changes from F (‒0.596) to I (‒0.722) for the cis

systems. This suggests that Cr‒X dissociation for the fluoride system passes through

an early TS while the iodide system passes through a late TS.

Chapter 4: *




‡, J mol

‒1

K‒1

) for D pathways of cis- or trans-[Cr(H2O)4(OH)X]+ with H2O in the water

solvent at 298.15 K.a

ΔH‡ (ΔS

‡)

X cis-[Cr(H2O)4(OH)X]+ trans-[Cr(H2O)4(OH)X]

+

FH TS1 93 (‒15) 63 (‒18)

TS2 ‒ ‒

Overall 93 (‒15) 63 (‒18)

Cl TS1 72 (‒5)

55 (‒35)

TS2 5 (‒9)

47 (‒15)

Overall 72 (‒27)

81 (‒35) Br TS1 67 (‒20)

43 (‒5)

TS2 8 (‒35)

24 (4)

Overall 67 (‒20)

61 (2) I TS1 58 (‒25)

40 (‒15)

TS2 15 (17)

24 (5) Overall 62 (‒13) 53(1) a

Optimized structures of cis or trans-Cr(III) complexes defined in Figures SF44 to

SF52 in the SI.

To further investigate the effect of outer sphere coordination on the D pathway

we investigated Cr‒X dissociation with three explicit outer sphere water molecules

([Cr(H2O)4(OH)X]+…mH2O, m = 3) (Figure SF53 of the SI). The dissociation of the

Cr‒X bond (TS1) (Figure 4.7) for cis-[Cr(H2O)4(OH)X]+…3H2O (X = Cl or Br)

complexes with three waters is 68 and 58 kJ mol1

. This represents a decrease of 4

and 9 kJ mol1

, respectively, as result of the additional solvation. The overall

activation enthalpies are 78 and 67 kJ mol1

, respectively. This corresponds to

differences of +6 and 0 kJ mol1

, respectively from the corresponding values with a

single outer sphere water molecule.

Chapter 4: *


Figure 4.7: Optimized transition structure (TS1) and map of electrostatic potential

(0.02 electrons Bohr‒3

) for cis-[Cr(H2O)4(OH)Cl]+… 3H2O via D pathway in


Direct nucleophilic attack by hydroxide

The previous sections have considered deprotonation of the complexes followed

by aquation via either an interchange or dissociative pathway. For completeness we

now consider, direct attack by the nucleophile OH‒ on the Cr(III) metal species as

shown in Scheme 4.5. The trans-[Cr(H2O)4)X2]+…OH

‒ and [Cr(H2O)5)X]

2+…OH

‒

(X = Cl or Br) associated reactant complexes are difficult to locate on the potential

energy surface because deprotonation of Cr(III) complexes by OH‒ occurs easily,

forming the conjugate bases, trans-[Cr(H2O)3(OH)X2]0…H2O and cis-[Cr(H2O)4

(OH) X]+

…H2O (R) complexes (Scheme 4.5 and SF23 to SF26 of the SI).

Furthermore, the separated species of trans-[Cr(H2O)3(OH)X2]0 and cis-

[Cr(H2O)4(OH)X]+ with H2O (cis-[Cr(H2O)4(OH)X]

+ + H2O) are energetically more

stable than precursor reactants (R) complexes by ~3.0 kJ mol1

(Figure F2 of the SI).

Chapter 4: *


The enthalpies of activation for trans-[Cr(H2O)4X2]+ and [Cr(H2O)5X]

2+ with

OH‒ pathways are shown in Table ST3 and Figures F2−F3 of the SI.

The activation enthalpy for the interchange pathways of the trans-[Cr(H2O)4X2]+

and [Cr(H2O)5X]2+

(X = Cl–, or Br

–) systems are 259, 228, 257 and 237 kJ mol

1,

respectively (Table ST3 of the SI). However, the deprotonation of [Cr(H2O)5Cl]2+

by the hydroxide ion to give conjugate base, cis-[Cr(H2O)4(OH)Cl]+…H2O (R) is

very exothermic (Figures F2 to F3). Therefore, nucleophilic attack by OH‒ on

[Cr(H2O)5X]2+

has significantly higher activation enthalpies (Table ST3 in the SI)

than the corresponding hydrolysis by H2O for the interchange mechanism.16

NBO charge distribution analysis of the reactants (Rs) and transition states (TSs)

for OH- nucleophilic attack on Cr(III) are given in Table ST4 of the SI. On

formation of the TSs, the positive charge on Cr(III) and the magnitude of the

negative charge on halide increase compared to R. The oxygen atom of the OH‒

molecule exhibits relatively large changes in charges from R (‒0.919 to ‒0.929) to

TSs (‒0.661 to ‒0.746) across the various systems (Table ST4 and Figure SF30 of

the SI.

Discussion

In the highly acidic environment of the stomach, orally ingested chromium(III)

halides such as chromium(III) chloride nutritional supplement will exist most likely

as either the dihalotetraaqua Cr(III) complex, the monohalopentaaqua Cr(III)

complex or a mixture of both. However, when these complexes enter the duodenum

they will encounter an environment with a much milder pH. Our calculations show

that proton transfer from theses complexes is spontaneous under neutral and alkaline

Chapter 4: *


conditions, in agreement with experiment.8a

Therefore speciation in the duodenum of

chromium(III) chloride in the duodenum will include [Cr(H2O)3(OH)X2]0 and

[Cr(H2O)4(OH)X]+ species. Aquation mechanisms were also investigated for these

conjugate-base species including, [Cr(H2O)3(OH)X2]0

, X = Cl or Br and the four

mono-halohydroxoaquachromium(III) complexes ([Cr(H2O)4(OH)X]+ , X = F, Cl, Br

or I). Pathways investigated include associative interchange mechanisms (Ia or Id,

SN2), and a variation of the dissociative mechanism (D, SN1). For the

[Cr(H2O)4(OH)X]+ species two structures were identified coinciding with the cis

and trans orientations of the OH‒ ligand.

Mechanistic investigations were carried out both without and with explicit water

molecules in the outer sphere of the complexes. For conjugate base species, cis-

[Cr(H2O)4(OH)X]+, the ΔH

‡ value for the Ia or Id pathways are in consistent

agreement with the experimental results.4c‒4d

Solvation is an important factor in the mechanism of this process. This is

demonstrated clearly by explicit solvation for the interchange pathway of cis-

[Cr(H2O)4(OH)Cl]+…mH2O complexes (m = 1‒5). In particular ΔH

‡ values for one

specific conformation were found to decrease from 105 for m = 1 to 72 kJ mol1

, for

m = 5 (Figures 4.5 and SF22a of the SI). However, the solvation with three explicit

water molecules appears to be sufficient to stabilize the transition state.

Nevertheless, the position of these three water molecules is also important and the

lowest energy m = 3 TS for the chloride system has ΔH‡ = 59 kJ mol

1. The

corresponding values for the fluoride, bromide and iodide systems with additional

three external water molecules are 85, 47 and 32 kJ mol1

, receptively (Figure 4.6).

Chapter 4: *


Table 4.6: Selected Cr(III)−Ligand bond lengths sums (Å) in the water solvent

at 298.15 K.a

∑ d (Cr‒L)TS‒R

X

cis-

[Cr(H2O)4(OH)X]+

pathway trans-

[Cr(H2O)4(OH)X]+

pathway

[Cr(H2O)4(OH)X]+

+ H2O [Cr(H2O)5(OH)]

2+ + X

‒

F ‒0.85 Ia

‒0.55 Ia

Cl ‒0.80 Ia

‒0.41 Ia

Br +0.46 Id

+0.40 Id

I +0.31 Id

+0.2 Id

ClClb ‒1.41 Ia ‒ ‒

BrBrb +0.36 Id ‒ ‒

[Cr(H2O)5X]

+ + OH

‒ [Cr(H2O)5(OH)]

2+ +

X

‒

Cl ‒1.31 Ia

‒ ‒

Br ‒1.38 Ia

‒ ‒

ClClb ‒1.32 Ia

‒ ‒

BrBrb ‒0.31 Ia

‒ ‒

a The ∆∑ d (Cr‒L)TS-R were calculated at PBE0/cc-pVDZ.

b The Structures of trans-[Cr(H2O)3(OH)X2]

0.

The ∆H‡ value for dissociation of the Cr‒X bond (TS1) for cis-[Cr(H2O)4

(OH)Cl]+…3H2O complex with three waters is 68 kJ mol

1 (Figures F2−F3 of the

SI) and the overall activation enthalpy is 78 kJ mol1

. Therefore, we propose that cis-

or trans-[Cr(H2O)4(OH)X]+ complexes proceeds via an interchange pathway, which

is good agreement with experimental results.4c‒4d

Nevertheless, there are variations

across the series of complexes. The calculated ∆∑ d(Cr−L)TS‒R values for the

fluoride and chloride cis- or trans-[Cr(H2O)4(OH)X]+ systems are consistent with a

negative activation volume and therefore an Ia mechanism. On the other hand, for the

hydrolysis reaction on cis- or trans-[Cr(H2O)4(OH)X]+ (X = Br or I), the ∆∑

d(Cr−L) values are positive (Table 4.6), indicating a dissociative interchange (Id)

mechanism, which is consistent with experimental results.4d

Chapter 4: *


The frontier MO energies (HOMO‒LUMO gap) show (Figures F4a‒F4b of the

SI) that complexes with the soft ion ligands (X = I) have a small HOMO‒LUMO gap

due to greater polarizability and electron density. This facilitates distortion of the

electron cloud of the halide that stabilizes the transition state. The fluoride is a poor

leaving group, which explains the high ΔH‡. The decrease in ΔH

‡ for the other

halides corresponds with the softer nucleophilicities and decrease in

electronegativity of these ions (X = I), so that the leaving group ability follows the

order I− Br

− Cl

− F

−. In general, the valence electrons of the I

− ion are more

polarizable. A distortion of the electron cloud of the halide can therefore stabilize the

transition state. Furthermore, electronegativity influences the electron density at the

Cr(III) centre, which effects the activation barrier and bond formation with the

entering water molecule. In these conjugated base systems, the cis- or trans- OH‒

group can also have a labilising effect. The trends might also be attributed to an

increase in the steric repulsion with the increase in the ionic radii of the leaving

halides.22

However, it was found that interaction between X– and H2O was strongly

repulsive for the fluoride system alone, which correlates with the higher activation

enthalpy.

Analysis of the frontier MO energies (HOMO and LUMO) (Figure SF58 of the

SI) along with the corresponding Egap for TS1 of the dissociation pathway reveals a

gradual decrease from the hard ion (X = F‒) TS1 through soft ion (X = I

‒). Figure 4.8

shows a correlation between the activation enthalpy (ΔH‡) versus the HOMO-

LUMO gaps of TS1 for the conjugate bases, cis- or trans-Cr(III) species for the D

mechanism. Thus it is likely that dissociation of the Cr‒X bond and migration of the

halide ions into the outer coordination sphere via dissociation pathways is stimulated

Chapter 4: *


by electronic and charge neutralizing effects.

Figure 4.8: ∆H‡ (kJ mol

‒1) versus HOMO-LUMO energy gap (eV) for TS1 species

of cis- or trans-[Cr(H2O)4(OH)X]+…H2O (X = F, Cl, Br or I) via D pathway in


An alternative reaction pathway under alkaline conditions is direct nucleophilic

attack. However, this is found to be a very high energy pathway. Selected maps of

electrostatic potential for nucleophilic attack by OH‒ on [Cr(H2O)5X]

2+ and trans-

[Cr(H2O)4X2]+ (X = Cl or Br) are shown in Figure F5 in the SI. The more polarizable

bromide ion leads to less charge-charge repulsion between the halide and the

incoming OH‒, compared to the chloride systems. However, all systems show that

the complexes with a hard ion (X or T = Cl) have a large HOMO‒LUMO gap (Table

ST4 and Figures SFT1 and SF31 of the SI). Oppositely, the complexes with a

borderline ion (X or T = Br) have a small gap due to the greater polarizability of the

valence electrons, which enables distortion of the electron cloud to stabilise the

transition states (see Figure F5 of the SI).

The aquation of haloaquachromium(III) species via the conjugate base pathways

identified here has interesting implications by providing a link to the formation of

Chapter 4: *


[Cr(H2O)5(OH)]2+

in environments with a pH ~ 4.6.23

As noted in the introduction,

the [Cr(H2O)5(OH)]2+

complex is the active catalytic species in the isomerization

reaction of glucose.6‒7

The OH‒ ligand and the Cr(III) centre, may act as a

bifunctional Brønsted-basic/Lewis-acidic site to bind glucose or biological molecules

in aqueous media. However, the rate determining step is the dissociation of the Cr‒X

bond and the formation of the conjugate base species, [Cr(H2O)5(OH)]2+

, which

replaces two water molecules by glucose in the active site and resulting in the

complex, [Cr(C6H11O6)(H2O)4]2+

.6‒7

Consequently, CrCl3 in a range of biological

solutions could potentially lead to formation of the active species such as

[Cr(H2O)5(OH)]2+

, which in turn could catalyse conversion of glucose. Nguyen et

al.24

used XANES spectroscopy to investigate the reactivity of trans-[Cr(H2O)4Cl2]+

under biologically relevant conditions. The trans-[Cr(H2O)4Cl2]+ complex was

found to react to form both aqua/hydroxo and amino acid complexes in a cell culture

medium. The trans-[Cr(H2O)4Cl2]+ was also found to undergo binding to cell

membranes through complexation with macromolecules. Moreover, the stabilities of

the relatively inert trans-[Cr(H2O)4Cl2]+ and [Cr(H2O)5Cl]

2+ in aqueous solutions

have a dependence upon pH.23,25

Therefore at some pHs these halides will be in

equilibrium with [Cr(H2O)6]3+

and [Cr(H2O)5(OH)]2+

(Scheme 4.6). Merideth et al.

reported11a

that the aquation reaction of trans-[Cr(H2O)4Cl2]+ occurs in the pH range

of 1.0 to 3.7. Bjerum26

and Lamb and Fonda investigated4b

Cr(III) chloride or nitrate

solutions and found that the addition of NaOH changed the color from green to blue

(violet) and concluded that Cr(III) solutions convert to [Cr(H2O)6]3+

at a pH 2.28

and to [Cr(H2O)5(OH)]2+

at a pH ~ 4.6.23

We note that the calculated ∆H‡ value for

the aqauation of trans-[Cr(H2O)3(OH)Cl2]0 complex on the Ia (D) pathways are

lower than those of the trans-[Cr(H2O)4Cl2]+ system

17 by 49 (39) kJ mol

1 (Scheme

Chapter 4: *


4.6). Furthermore, the calculated overall ∆H‡ values for the cis-[Cr(H2O)4

(OH)Cl]+…H2O convert to [Cr(H2O)5(OH)]

2+ complex on the Ia (D) pathways drops

below that of the [Cr(H2O)5Cl]2+

…H2O system by 14 (31) kJ mol1

.16

This suggests

that the green chloride solution reacts slowly with water to form a monochloro

hydroxide, [Cr(H2O)4(OH)Cl]H2O+. Finally, this product reacts slowly to form the

hexa-aqua hydroxide, [Cr(H2O)5(OH)]H2O2+

due to the step-wise replacement of

chloride by water (see Scheme 4.6) at higher pH. The hydroxides can either undergo

binding to bio-molecules or can form hydrolytic polymers. For example, Blankert et

al.27

note that there is a strong pH dependence in the interaction of trans-

[Cr(H2O)4Cl2]+ with DNA. They found that trans-[Cr(H2O)4Cl2]

+ causes unwinding

of supercoiled DNA which reaches a maximum at pH 6.50. However, at higher pHs

there was significant precipitation of Cr(III) hydroxide oligomers that led to a

decrease in reactivity with DNA.

trans-[Cr(H2O)4Cl2]+H2O

Cl[Cr(H2O)5Cl]2+ H2O

Cl[Cr(H2O)6]3+

trans-[Cr(H2O)3(OH)Cl2]0cis-[Cr(H2O)4(OH)Cl]+

OHH+

H2O

Cl

H2O

Cl[Cr(H2O)5(OH)]2+

H+

Scheme 4.6:

Our aquation studies of cis- or trans-[Cr(H2O)4(OH)X]+ and other systems,

provides mechanistic understanding of the speciation of these species under different

conditions. This will provide the basis for further investigation of reactions with

biomolecules including binding to glucose which can initiate isomerization and other

reactions as proposed by Choudhary et al.6

and Mushrif et al.7

Chapter 4: *


Conclusions

We have carried out an extensive study of the deprotonation and aquation of

haloaquachromium(III) complexes using hybrid density functional theory. Our

calculations indicate that formation of the conjugate base has a larger effect on the

structural parameters than the explicit outer-sphere solvation of the complex.

Deprotonation of the halopentaaquachromium(III) complexes is calculated to be

exogonic under both neutral and alkaline conditions. However, under neutral

conditions the conjugate base of the iodide is most easily formed, whereas under

alkaline conditions proton transfer from the fluoride is energetically more

favourable. Regardless of the halide, the cis-[Cr(H2O)4(OH)X] complexes are more

stable than the corresponding trans-[Cr(H2O)4(OH)X] species.

We investigated aquation of cis- or trans-[Cr(H2O)4(OH)X]+ and other related

systems via the associative interchange (Ia), the dissociative interchange (Id) and

dissociative (D) mechanisms. Use of implicit solvent models has a minimal effect on

the structures of reactants, products and transition states but it is essential for an

agreement between theoretical and available experimental ΔH‡ values. We found

that the activation enthalpy and entropy of activation calculated at PBE0/cc-pVDZ

for aquation of the cis- or trans-[Cr(H2O)4(OH)X]+ via the Ia to Id mechanisms

decrease with change of the ligand from F‒ to I

‒. Our calculations indicate that the Ia

or Id pathways is favoured over the D pathway in agreement with experiment. At

least three additional water molecules are required in the second coordination sphere

to stabilize the transition states of Ia and D pathways, reducing ΔH‡. It is clearly

indicated that steric and electronic effects influence ΔH‡ for the aquation reactions of

cis- or trans-[Cr(H2O)4(OH)X]+

and other systems and the leaving group ability

follows the order: I− Br

− Cl

− F

−. Therefore, we conclude that cis- or trans-

Chapter 4: *


[Cr(H2O)4(OH)Cl]+ complexes as well as [Cr(H2O)5(OH)]

2+ are key derivatives of

chromium chloride nutritional supplement at mild and physiological pHs. These may

be the active site for the treatment of physiological disorders such as type II diabetes

and depression.



information and Appendix A.

Acknowledgements


facility for computer time.

References

(1) (a) Vincent, J. The nutritional biochemistry of chromium(III), Elsevier, 2011.

(b) Anderson, R. A.; Cheng, N.; Bryden, N. A.; Polansky, M. M.; Cheng, N.;

Chi, J.; Feng, J. Diabetes 1997, 46, 1786‒1791. (c) Hellerstein, M. K. Nutr.

Rev. 1998, 56, 302-306. d) Mertz,W. Med. Clin. North Am. 1976, 60, 739‒744. (e) Levina, A.; Lay, P. A. Chem. Res. Toxicol. 2008, 21, 563‒571. (f)

Levina, A.; Lay, P. A. Coord. Chem. Rev. 2005, 249, 281‒298.

(2) Torapava, N.; Radkevich, A.; Davydov, D.; Titov, A.; Persson, I. Inorg.

Chem. 2009, 48, 10383‒10388.


(4) (a) Xu, F. C.; Krouse, H. R.; Swaddle, T. W. Inorg. Chem. 1985, 24, 267‒270. (b) Lamb, A. B.; Fonda, G. R. J. Am. Chem. Soc., 1921, 43,1154‒1178.

(c) Thusius, D. Inorg. Chem., 1971, 10, 1106‒1108. (d) Carey, L.; Jones, W.;

Swaddle, T.; Inorg. Chem. 1971, 10, 1566‒1570. (e) Weekes, M. C.;

Swaddle, T. W. Can. J. Chem. 1975, 53, 3697‒3701. (f) Hamm, R. E.; Shull

Jr, C. M. J. Am. Chem. Soc. 1951, 73, 1240‒1243.

(5) Swaddle, T.; Kong, P. C. Can. J. Chem. 1970, 48, 3223‒3228.

(6) Choudhary, V.; Mushrif, S. H.; Ho, C.; Anderko, A.; Nikolakis, V.;

Marinkovic, N. S.; Frenkel, A. I; Sandler, S. I.; Vlachos, D. G. J. Am. Chem.

Soc. 2013, 135, 3997‒4006.

Chapter 4: *


(7) Mushrif, S. H.; Varghese, J. J.; Vlachos, D. G. Phys. Chem. Chem. Phys.

2014, 16, 19564‒19572.

(8) (a) Stuenzi, H.; Marty, W. Inorg. Chem. 1983, 22, 2145‒2150. (b) Crimp, S.

J.; Spiccia, L.; Krouse, H. R.; Swaddle, T. W. Inorg. Chem. 1994, 33, 465‒470. (c) Rotzinger, F. P.; Stuenzi, H.; Marty, W. Inorg. Chem. 1986, 25, 489‒495. (d) Stuenzi, H.; Spiccia, L.; Rotzinger, E. P.; Marty, W. Inorg. Chem.

1989, 28, 66‒71. (e) Rao, J. R.; Gayatri, R.; Rajaram, R.; Nair, B. U.;

Ramasami, T. Biochim. Biophys. Acta 1999, 1472, 595‒602. (f) Merakis, T.;

Spiccia, L. Aust. J. Chem. 1989, 42, 1579‒1589; (g) Thompson, M.; Connick,

R. E. Inorg. Chem. 1981, 20, 2279‒2285; (h) Spiccia, L.; Marty, W.

Polyhedron 1991, 10, 619‒628.

(9) (a) Wahlberg, J. E. ; Wennersten, G.; Br. J. Dermatol, Br. J. 1977, 97, 411‒416. (b) Murphy, V.; Tofail, S. A.; Hughes, H.; McLoughlin, P. Chem. Eng.

J. 2009, 148, 425‒433. (c) Goldoni, M. ; Caglieri, A.; Corradi, M.; Poli, D.;

Rusca, M.; Carbognani, P.; Mutti, A. Int. Arch. Occup. Environ. Health

2008, 81, 487‒493.

(10) Lamb, A. B. J. Am. Chem. Soc. 1906, 28, 1710‒1715.

(11) (a) Merideth, C. W.; Mathews, W. D.; Orlemann, F. F. Inorg. Chem. 1964, 3,

320‒322. (b) Bjerrum, J. Chem. Rev. 1950, 46, 381‒401; (c) Spiccia, L.;

Stoeckli-Evans, H.; Marty, W.; Giovanoli, R. Inorg. Chem. 1987, 26, 474‒482.

(12) (a) Ardon, M. Inorg. Chem. 1965, 4, 372‒375. (b) Parris, M.; Wallace, W.

Can. J. Chem. 1969, 47, 2257‒2262. (c) Guastalla, G.; Swaddle, T. W. Can.

J. Chem. 1974, 52, 527‒535. (d) Rotzinger, F. P.; Weber, J.; Daul, C. Helv.

Chim. Acta 1991, 74, 1247‒1263.

(13) Espenson, J. H. Inorg. Chem., 1969, 8, 1554‒1556.

(14) Basolo F.; Pearson, R. G. Mechanisms of inorganic reactions: a study of

metal complexes in solution. 2nd

edn. Wiley & Sons, Inc., New York, pp 701,

1967.

(15) Curchod, B. F.; Rotzinger, F. P. Inorg. Chem. 2011, 50, 8728‒8740.

(16) Uddin, K. M.; Ralph, D.; Henry, D. J. Comp. Theor. Chem. 2015, 1070, 152‒161.

(17) Uddin, K. M.; Poirier, R. A.; Henry, D. J. Comp. Theor. Chem. 2016, 1084,

88‒97.

(18) Monsted, L.; Monsted, O. Acta Chem. Scand. A 1978, 32, 19‒24.


Marcos, E. S. J. Am. Chem. Soc., 1966, 118, 12654‒12664.

(20) Langford, C. H.; Gray, H. B. Ligand substitution processes. W. A. Benjamin,

Inc., New York-Amsterdam, 1965.

(21) Reed, A. E.; Curtiss, L. A.; Weinhold, F. Chem. Rev. 1988, 88, 899‒26.

(22) Hodgman, C. D.; Veazey, W. R. CRC Handbook of chemistry and physics.

Chemical Rubber Publishing Company, 1977.

(23) Emerson, K.; Graven, W. J. Inorg. Nucl. Chem. 1969, 11, 309‒313.

http://www.ncbi.nlm.nih.gov/pubmed/25105840

Chapter 4: *


(24) Nguyen, A.; Mulyani, I.; Levina, A.; Lay, P. A. Inorg. Chem. 2008, 47,

4299‒4309.

(25) (a) Gates, H. S.; King, E. L. J. Am. Chem. Soc. 1958, 80, 5011‒5015. (b)

Johnson, D. A.; Reid, M. A. J. Electrochem. Soc. 1985, 132, 1058‒1062.

(26) Bjerrum, N. Z. Physik. Chem. 1907, 59, 336.

(27) Blankert, S. A.; Coryell, V. H.; Picard, B. T.; Wolf, K. K.; Lomas, R. E.;

Stearns, D. M. Chem. Res. Toxicol. 2003, 16, 847‒854.

Chapter 5: * Uddin, K. M.; Habib, A. M.; Henry, D. J. ChemistrySelect 2017, 2, 1950−1958. 141

Chapter 5

Investigation of the Spectroscopic, Thermal and

Electrochemical Properties of

Tris-(glycinato)chromium(III)*

Abstract

The tris-glycinatochromium(III) complex ([Cr(gly)3]) has been proposed as a

nutritional supplement and treatment for Type 2 diabetes. The exact mode of action

of this and other Cr(III) supplements is not known but may involve redox reactions

with biomolecules. In this study fac-[Cr(gly)3] was synthesized and the

electrochemical properties investigated by cyclic voltammetry. The complex

undergoes a reversible one electron reduction with E1/2 = −0.66 to −0.69 V that is

essentially independent of scan rate. Thermal decomposition was explored by

differential scanning calorimetry under nonisothermal conditions from 30−1000 °C.

Structural parameters obtained from DFT calculations are in good agreement with

crystallographic data. The reduced form of the complex has a high-spin quintet

electronic configuration due to the small crystal field splitting from the weak -


donor ligands. Both the theoretical calculations and the electrochemical

measurements demonstrate that the complex is stable in the anionic form. These

results indicate that [Cr(gly)3] can undergo reversible electron transfer, which may

play a role in its biological activity.

Introduction

A number of chromium(III) complexes with amino acid (AA) ligands have been

proposed as nutritional supplements and treatments in glucose metabolism.1−4

However, several studies5−9

suggest that Cr(III)-AA supplements are of minimal

benefit to most people. Nevertheless, there is some evidence to suggest that Cr(III)-

AA systems may have benefits as an insulin enhancing treatment in type 2

diabetics,6,10−14

as an anti-depressant and alternatively to reduce lipid deposits.15−16

For example, Tsave et al.12

investigated the potential of Cr(III)-citrate to induce pre-

adipocyte cell differentiation, which is a factor in obesity and may be linked to type

2 diabetes. They found that Cr(III)-citrate was able to induce adipogenesis and did

not appear to be toxic to the pre- or mature adipocytes. Saiyed and Lugo13

found that

treatment of type 2 diabetic subjects with chromium(III) dinicocysteinate led to a

decrease in both insulin resistance and protein carbonyl and also out-performed the

commonly used supplement, chromium picolinate. Recently, Refat et al.14

prepared a

new niacinamide based chromium(III) complex, that was found to significantly

decrease blood glucose levels and HbA1c in diabetic rats. The complex was also

effective at lowering glycogen levels in the liver while also supporting anti-oxidant

defences.

Amino acids in the anionic form can coordinate to Cr(III) through the primary

amine and the carboxyate group to form facial or meridional isomers, Cr(III)-AA


(fac- (I) or mer- (II)) or in the Zwitterionic forms (Cr(gly)3 (III) or Cr(gly)3 (IV))

(see Figure 1).

Cr3+

N

ON

O

O

N

O

N

O

Cr3+

O

NO

Cr3+

O

NN

O

O

N

O

H

H

H

H

O

HH

H

HO

H

H

HH

fac-[Cr(gly)3]0 (I)

O

H

H

H H O

H H

O

H

H

H

H

H

H

mer-[Cr(gly)3]0 (II)

N

HH

H

H

HH

H

H

O

H H

OH

H

O

O

N

Cr3+

O

O

NN

HH

H

H

O

H

H

H

H

O

H

H

H

H

[Cr(gly)3]0 (III) [Cr(gly)3]0 (IV)

Figure 5.1: Structures for conformers of [Cr(gly)3].

The molecular structures of the red facial isomer of tris(glycinato) chromium(III)

(fac-[Cr(gly)3]), and the binuclear complex [Cr(gly)2OH]2 have been identified from

single-crystal XRD by Bryan et al.17

and Veal et al.,18

respectively, along with a

series of oxo-centered trinuclear glycinate-bonded complexes.19

Wallace and

Hoggard20

identified the fac-[Cr(gly)3] complex using electronic excitation

spectroscopy and an angular-overlap-model analysis. The preparation of the

meridional isomer is reported to be more difficult.20−23

Nevertheless, the purple

meridional isomer of α-[Cr(gly)3] has been prepared by Israily.24

Replication of this

synthesis by Gillard et al.25

yielded a purple material in a hydrated form. However it

is believed to be the dihydroxo-bridged dimer, ( [Cr2(gly)4(OH)2].6H2O). They also

prepared the analogous L-alanine dimer complex, ([Cr2(L-ala)4(OH)2].7H2O). Kita et

al.26

synthesized, fac-[Cr(gly)3] and [Cr2(gly)4(OH)2] and carried out kinetic studies


for aquation of these compounds under acidic and alkaline conditions. A range of

other Cr(III)-AA complexes have also been synthesized and characterized including,

complexes with methionine (Cr-Met and Cr2(met)4Cl2.2H2O∙KCl),5

anthranilic acid

(Cr(ant)3∙H2O), valine (Cr(val)3∙H2O∙KCl), tryptophan (Cr2(trp)4Cl2∙2H2O), glutamic

acid (Cr2(glu)3(OH)2∙2H2O∙KCl), cysteine (K[Cr(cys)2]∙4H2O), serine ([Cr(ser)3Cl3]

H2O∙3KCl) and histidine (Cr(his)3Cl3∙3H2O).7, 27−32

The efficacy of tris-glycinatochromium(III) complexes in treating disorders such

as type 2 diabetes may involve binding to a receptor or ligand exchange processes.

Alternatively, redox processes have been proposed as mechanisms by which

chromium(III) complexes might modulate biochemical processes. Reports suggest

that in the presence of weak-field ligands Cr(III/II) couples undergo a slow electron

transfer.33−36

Thus, an understanding of the electrochemical reactions and redox

processes involving Cr(AA)3 complexes is also of biological interest. A number of

investigators29,33,37−58

have studied the electrochemical oxidation-reduction reactions

of chromium(III) complexes. However, of particular interest is the redox chemistry

of those compounds that have been linked to nutritional supplements or

physiological studies including, tris-picolinatechromium(III) [Cr(pic)3],59

the dimeric

picolinate complex [Cr(pic)2OH]2,59

as well as Na[Cr(dipic)2].2H2O,52

[Cr(dipic)(phen)Cl].1/2H2O,52

tris-(8-hydroxyquinolinate)chromium(III),60

and

chromium complexes with a series61

of tri- and hexadentate amino carboxylate

ligands. Parajόn-Costa et al. observed59

that both [Cr(pic)3] and [Cr(pic)2(OH)]2

complexes undergo a one-electron reversible reduction process. However, the half

wave potential (E1/2) for the binuclear [Cr(pic)2(OH)]2 complex is more negative

(E1/2 = −1.49 V) than the mononuclear [Cr(pic)3] complex (E1/2 = −1.23 V). Freitas et

al. investigated60

mer-tris(8-hydroxyquinolonato)-chromium(III) and identified two


pairs of reversible peaks in the cyclic voltamogram (E1/2 of −0.76 and −1.40 V vs

Ag/AgCl).60

Hecht et al. investigated61

the electrochemical behaviour of 10 cis-

amino carboxylateCr(III) complexes and observed reversible one-electron reductions

at formal potentials between ca. −1.4 and ca. −1.2 V vs SCF. Bond et al.57‒58

investigated the redox behaviour of solid cis- and trans-Cr(CO)2(dpe)2 (dpe =

Ph2PCH2CH2PPh2) mechanically attached to carbon electrodes. In this system the

redox chemistry occurs at the solid‒solution‒electrode interface. Sugden et al.49

determined the reduction potentials of five mutagenic Cr(III) complexes

([Cr(bpy)2Cl2]+, [Cr(phen)2Cl2]

+, [Cr(bpy)2(OH)2]

+, [Cr(phen)2(OH)2]

+, and

[Cr(en)2Cl2]+) and proposed that the biological activity of chromium(III) complexes

was linked to generation of oxygen radical in a Fenton-like reaction. The potential

mutagenesis of the Cr(III) complexes was identified by salmonella revision assay

and the redox kinetics for mutagenic and nonmutagenic Cr(III) species were

investigated by cyclic voltammetry. The mutagenic complexes had less negative

reduction potentials than their non-mutagenic counterparts and displayed

reversibility in the reduction process. These results suggest that the electrochemical

reactivity of other chromium(III) species may be significant and warrants further

investigation.

In this paper, we report on the synthesis of fac-[Cr(gly)3] complex and the

characterization by elemental analysis, AAS, 1H NMR, UV−Vis, EPR, and ESI−MS

spectroscopy. However, the major objective of this work is to investigate the

electrochemical behavior of this complex, which will lead to the development of a

better understanding of the Cr(III)/Cr(II) electron transfer process. The redox

behavior of the compound and the changes in structural parameters with change in

oxidation state are discussed. We have also performed thermogravimetric-


differential scanning calorimetry (TGA/DSC) analysis to calculate the thermal

decomposition reaction kinetics. The experimental study is complemented by DFT

calculations with determination of the molecular structures, electronic density,

natural bond orbital charges, frontier molecular orbitals and electrostatic potentials.


Synthesis and characterization of fac-[Cr(gly)3] complex

Tris-glycinatochromium(III) ([Cr(gly)3]) was synthesized using the procedure of

Bryan et al.17

Elemental analysis (% C, H, N, and Cr) and electrospray mass

spectrometry (ESI-MS) data confirm the formation of the [Cr(gly)3] complex. In

particular, the ESI mass spectrum of [Cr(gly)3] complex (Figure 5.2 and SF0 of the

SI) shows a peak at m/z = 275.2205 (calc. m/z = 275.0210), corresponding to

[Cr(C6H12N3O6) + H+] for the Cr-52 isotope. The peaks at 297.2037 and 245.2768

corresponds to [Cr(C6H12N3O6) + H+ + Na

+] and [Cr(C4H8N2O4) + CO2 + H

+],

respectively. Additional peaks at m/z = 200.1770 and m/z = 98.1765, correspond to

[Cr(C4H8N2O4) + H+] and a Gly sodium salt hydrate fragment [H2NCH2CO2Na.H2O

+H+], respectively (see Figure 5.2 and SF0 of the SI).

The UV/Vis spectrum of fac-[Cr(gly)3] complex obtained in the solid state and in

DMSO solution are presented in Figure 5.3. Bryan et al. 17

previously measured the

UV/Vis spectrum for fac-[Cr(gly)3] as a mineral oil mull. They reported two main

peaks assigned to the spin allowed d-d transitions. The 503 nm transition was

assigned to 4A2g →

4T2g (ν1) and the 384 nm transition to

4A2g →

4T1g (ν2). They also

noted three weak bands (> 630 nm) but were only able to assign one of these (689

nm) to the spin forbidden 4A2g →

2E transition. In the solid-state spectrum obtained


from the current study the spin-allowed transitions are clearly observed at 510 nm

and 388 nm, respectively. However, the spin forbidden transitions are not clearly

evident at the resolution of our experiment. The ligand field splitting (∆o or 10Dq =

ν1) in the solid state is 19.9 × 103 cm

−1.

Figure 5.2: ESI mass spectrum of fac-[Cr(gly)3] for (a) simulation and (b)

experimental.

In the UV-Vis spectrum of fac-[Cr(gly)3]0 in DMSO (Figure 5.3b) the two spin-

allowed peaks are shifted to longer wavelengths, 518 nm (ν1) and 393 nm (ν2),

respectively. The weak spin-forbidden transition 4A2g →

2Eg is observed at 694 nm,

which is close to the value reported by Bryan et al.17

The ligand field splitting and

Racah parameter (B = (2ν12 + ν2

2 ‒3ν1ν2)/(15ν2 ‒27ν1))

62‒65, obtained from the

spectrum in DMSO are 19.3 × 103 cm

−1 and 575 cm

‒1, respectively. Although the

ligand field splitting value in solution is consistent with an (RCO2)3(RNH2)3


environment around Cr(III), there is clearly a decrease in ∆o from solid to solution.

This might suggest a slight change in the coordination environment around the

metal. Given that the spectrochemical series for Cr(III) has the order RCO2− < H2O <

RNH2, this shift might be due to Cr−N ring opening, with coordination of H2O.

However, this change may also be due to solvation effects. In dimethylformamide

(DMF) the spin-allowed bands are located at 528 nm (∆o = 18.9 × 103 cm

‒1)22

and

393 nm and Cooper et al.2 also note that in 2M HNO3, these bands shift to 535 (∆o =

18.7 × 103 cm

‒1) and 398 nm, respectively.

Figure 5.3: UV-Vis spectrum of fac-[Cr(gly)3]0 (a) solid-state and (b) DMSO

solution (8.9 mM).

In order to obtain some insight into the coordination geometry of [Cr(gly)3] in

different environments we have performed EPR on the solid state powder sample, in

0.0

0.4

0.8

1.2

1.6

2.0

300 400 500 600 700

Ab

s.

λmax, (nm)

DMSO

0.03

0.04

0.05

680 690 700 710

0.78

0.78

0.79

0.79

680 690 700 710

(a)

(b)

0.8

0.9

1.0

1.1

1.2

300 400 500 600 700

Ab

s.

λmax, (nm)

Solid

4A2g → 4T2g (ν1)

4A2g → 4T1g (ν2)

4A2g → 2Eg

393

518

694

694

688


H2O and in DMSO frozen state solution samples. Chromium(III) complexes are

paramagnetic and can be identified by two different X-band EPR signals (g = 2 and g

= 3.5−5.5).63,66

In this study, the broad solid state powder EPR signal of the fac-

[Cr(gly)3] complex has a g value of ≈ 1.9758−1.9865 (Figure 5.4a). This signal

corresponds with the (CO2)3(NH2)3 coordination geometry identified in the

molecular structure determined from single-crystal XRD by Bryan et al.17

Very

similar values are obtained in the frozen state at 100 K in DMSO (1.9747) and in

H2O (1.9787) (Figure 5.4b), which suggests that there is no change in coordination

environment and is consistent with other six-coordinate octahedral Cr(III)

complexes.62‒63,66

For example, Nguyen Pham et al.62

recently reported the EPR

spectrum of the closely related Cr(III) nicotinato complex, which shows a broad

signal with g ≈ 1.97−1.99, also consistent with octahedral geometry.

Figure 5.4: Analysis of X-band EPR spectra for fac-[Cr(gly)3] in solid state (a)

frozen state (100 K) (b) and solution 12.1 mM DMSO (blue), and 14.4 mM in H2O

(violet).


The paramagnetic nature of Cr(III) means that there have been only a limited

number of NMR studies of these species.53,67−68

Green et al. 67

characterized a series of

bis(ethylenediamine) Cr(III)−amino acid complexes by 2H NMR spectroscopy and

assigned the resonances located between +19 and +37 ppm to monodentate,

carboxylate-bound amino acid ligands. Recently, Freitas et al.60

reported the 1H

NMR spectra of both 8-hydroxyquinoline (8-Hq) and tris-(8-hydroxyquinoline)-

chromium(III), Crq3 in DMSO-d6.

The 1H NMR spectrum of glycine (Gly) in D2O (Figures S1a−S1b of the SI)

exhibits a sharp singlet peak at 3.56 ppm, assigned to co-resonant methylene

protons.70

However, the fac-[Cr(gly)3] complex has only limited solubility in D2O,

therefore, the 1H NMR spectrum of fac-[Cr(gly)3] complex was obtained in DMSO.

The methylene protons of the glycine are shifted up-field in the complex to 2.08

ppm, a broad signal at 3.34 ppm arises from water and the DMSO solvent peak

appears at 2.50 ppm (Figures S2a−S2b of the SI). In order to confirm that the

chemical shift at 3.34 ppm is due to H2O, additions of Millipore water (1, 5, 25 and

125 µL) were made to the fac-[Cr(gly)3] complex in DMSO. Consequently, the

water peak shifted from 3.34 to 3.71, 3.80, 3.81, and 4.09 ppm, respectively.

However, the complexed glycine methylene protons only shifted from 2.08 to 2.11

ppm (Figures S3 to S6 of the SI).

The up-field shift of the methylene proton resonances in the complex can be

related to the contact mechanism of paramagnetic interaction with the unpaired

electrons in the t2g* orbitals of the metal. Eaton69

and Freitas et al.60

have previously

attributed the contact shift to delocalisation of the d3 electrons into the ligand p-

orbitals, which in this case has a shielding effect on the methylene protons. We

discuss this in more detail in the DFT description of the electronic structure.


Thermal Decomposition Kinetic Analysis

The stability of the Cr(III) complex was also analysed using thermogravimetry

(TG) under argon. The differential scanning calorimetry (DSC) curves (Figures 5.5

and S7 to S9 of the SI) show three endothermic stages of mass loss in the

temperature ranges (a) 50 °C to 250 °C, (b) 380 °C to 450 °C, and (c) 740 °C to 810

°C. The first stage shows a gradual weight loss around 250 °C indicating the release

of lattice water, a second mass loss around 417 °C corresponds to the decomposition

of the glycine ligand, and the third stage occurs around 780 °C due to the thermal

decomposition of the complex, which is consistent with previous studies14,71−79

of

other Cr(III) complexes. For example, Freitas et al.60

recently reported that for tris-

(8-hydroxy quinolinate)chromium(III), the second stage shows the greatest fraction

of weight loss, with a corresponding sharp endothermic peak at around 472 °C.

Thermal analysis by thermogravimetry (TG), differential thermal analysis (DTA)

and differential scanning calorimetry (DSC) can also be used for the measurement of

kinetics and Arrhenius parameters of the decomposition of metal complexes, in the

solid state using several methods.76−79

DSC and TG experiments were carried out at four different heating rates (5, 10,

15, and 20 °C/min) to evaluate the kinetics of thermal decomposition of fac-

tris(glycinato)chromium(III) (Table 5.1, Figures 5.5 and S7−S9 of the SI). The

second endothermic peak occurs at a temperature of 417.3 °C and is very sharp

compared to the other two endothermic processes.


Figure 5.5: TG-DSC curve of the fac-[Cr(gly)3] complex at heating rate 10 °C/min.

The activation energy (E) and pre-exponential factor (A) of this endothermic

decomposition process were determined from non-isothermal DSC studies using

Kissinger’s (5.1)78

and Ozawa-Doyle’s (5.2)79

methods as follows:

ln(β/Tp2) = ln(AR/Ea) − (Ea /RTp) (Kissinger’s method) (5.1)

log β + 0.4567(Ea/RT) = C (Ozawa-Doyle’s method) (5.2)

where β is the linear heating rate (⁰C/min), Tp is the peak temperature (⁰C), R is the

gas constant (8.314 J mol−1

K−1

), and C is a constant. The peak temperatures of the

endothermic process obtained at various heating rates are given in Table 5.1 (Figures

S7 to S9 of the SI). The value of Ea was calculated from the slope of the fitted

straight line from a plot of ln(β/Tp2) vs. 1/Tp (Figure S9 of the SI).


Table 5.1: Peak temeraptures for thermal decomposition of fac-[Cr(gly)3]0

at

different heating rates.a

Heating rates Peak temp.

β (⁰C/min) Tp (⁰C)

5

409.9

10

417.3

15

422.5

20 424.0

a The activation energy determination from the equations 5.1−5.2.

The activation energy for thermal decomposition is calculated to be 360.4 kJ

mol−1

by the Kissinger method and 353. 6 kJ mol−1

by Ozawa’s method.

Electrochemistry Analysis

The redox behaviour of the solid tris-glycinatochromium(III) complex was

studied by cyclic voltammetry in acetonitrile using 0.1M TBAH as the supporting

electrolyte and a saturated Ag/AgCl reference electrode (Table 5.2, Figures 5.6 and

5.7).

The cyclic voltammograms indicate a reversible response of the

[Cr(gly)3]/[Cr(gly)3]−1

redox couple at the solid‒solvent (CH3CN)‒electrode

interface that corresponds to the following electrode reaction:

fac-[Cr(gly)3]0 fac-[Cr(gly)3] e

e

The reduction of [Cr(gly)3] from the 0 to the −1 state is shown by a cathodic peak

(Epc = −0.79 V vs Ag/AgCl, v = 100 mV/s and E1/2 = −0.69 V). The separation

between anodic and cathodic peak potentials, ∆EP, is 189 mV at v = 100 mV/s.

However, ∆EP for this electron-transfer process depends on the scan rate and ∆EP is


greater than 58/n mV for a one electron Nernstian process, where n is the number of

electrons transferred.49

Furthermore, the redox activity of fac-[Cr(gly)3] complex in

DMSO, also examined by CV (see Figure SF10 of the SI), featured two weak anodic

peaks ((Epa = −0.71 and Epa = 0.83 vs Ag/AgCl, v = 100 mV/s) assigned to the

oxidation of the glycinate ion group.

Table 5.2: Selected electrochemical data of fac-[Cr(gly)3]0 at different scan

rates.a

scan rate cyclic voltammetric data

v (mV/s) Epc/V Epa/V E1/2/V ∆Ep/mV ipc/µA ipa/µA ipa/ipc

10 −0.74 −0.59 −0.66 150 −1.099E-05 3.605E-06 0.33

100 −0.79 −0.60 −0.69 189 −2.965E-05 1.020E-05 0.34

200 −0.77 −0.59 −0.68 175 −4.250E-05 1.209E-05 0.28

300 −0.78 −0.57 −0.67 205 −5.102E-05 1.886E-05 0.37

400 −0.79 −0.56 −0.67 236 −6.547E-05 2.641E-05 0.40

500 −0.80 −0.54 −0.67 256 −6.981E-05 3.053E-05 0.43

600 −0.81 −0.53 −0.67 280 −7.586E-05 3.464E-05 0.45

700 −0.82 −0.52 −0.67 297 −8.466E-05 3.903E-05 0.46

800 −0.82 −0.51 −0.67 311 −9.122E-05 4.330E-05 0.47

900 −0.83 −0.50 −0.67 329 −9.942E-05 4.756E-05 0.48

1000 −0.84 −0.50 −0.67 347 −1.091E-04 4.204E-05 0.39

a E1/2 = 0.5(Epa + Epc) and ∆E1/2 = 0.5(Epa − Epc),where the Epa and Epc are anodic and

cathodic peak potentials, respectively.

Figure 5.6a‒5.6c shows voltammograms for different numbers of cycles,

obtained for the same scan rate (v = 100 mV/s) demonstrating only minor variation

in the peak potentials with scan number, indicating chemical stability of the solid

fac-[Cr(gly)3] complex for the electron-transfer process.


Figure 5.6: Cyclic voltammograms of solid fac-[Cr(gly)3] complex mechanically

attached to a glassy carbon electrode in 0.1 M TBAH/CH3CN at scan rate , v = 100

mV/s for (a) scan 1, (b) scan 1−5, and (c) scan 1−10.

To further explore the electrochemistry of this complex, a range of different scan

rates were investigated (Table 5.2). We note that the scan rate causes a gradual

increase in the peak separation ∆Ep from 150 to 347 mV (Table 5.2 and Figure 5.7a),

suggesting a diffusion controlled process for the reduction of tris-glycinato Cr(III)

complex. Figure 5.6a shows the effect of various scan rates between 10 to 1000 mV.

These results are consistent with Nicholas and Shai80

for a reversible charge transfer

process.

Plots of cathodic peak current, ipc, vs square root of the scan rate, v1/2

for the

range 10 mV/s ≤ v ≤ 1000 mV/s were linear with y (intercept = 0) (Figure 5.7b),

indicating a diffusion controlled process for reversible electron transfer.81

The

dependence of the current peak (ipc) on the v1/2

value relates to the strength of the

metal-ligand binding in the complex. The ratio of anodic and cathodic peak currents,

ipa/ipc was around 0.3−0.50 at all v, indicating that Cr(III) and Cr(II) species are

stable during the CV measurement and may be weakly absorbed at the glassy carbon

electrode, under these conditions (ipa/ipc <1).82

In addition, also found there were no

significant changes in the current as the potential was cycled up to ten times at v =

100 mV/s (Figures 5.6a−5.6c).


Figure 5.7: (a) Cyclic voltammograms of solid fac-[Cr(gly)3] mechanically attached

to a glassy carbon electrode in 0.1 M TBAH/CH3CN at different scan rates (v = 10 –

1000 mV/s) and (b) plot of cathodic peak current (ipc) versus the square root of scan

rate (v1/2

).

The solid fac-[Cr(gly)3] complex consists of three bidentate glycine ligands that

bind through amine and carboxylate donors and undergoes reduction at E1/2 = −0.66

to −0.69 V at different scan rates, v (10 mV/s ≤ v ≤ 1000 mV/s). Previously, Hecht et

al.61

reported redox potentials of E1°′ = −1.35 to −1.46 V and E2

°′ = −1.19 to −1.24 V

for Cr(III) complexes with cis-N2O4 and trans-N2O4 coordination environments.

These values are somewhat higher than the value for our complex, which has an

N3O3 coordination environment.61

Similarly, Ogino et al.55

obtained Cr(III)/Cr(II)

redox potentials of E1°′ = −1.36 ± 0.5 V and E2

°′ = −1.22 V for hexadentate and


(aquo)pentadentate amino carboxylate complexes, respectively. They attribute the

difference in formal potentials to differing electron donor strength of aquo and

carboxylate ligands. Parajón-Costa et al.59

reported Cr(pic)3 and [Cr(pic)2(OH)2]2

one-electron redox potentials (Cr(III)/Cr(II) and Cr(III)Cr(III)/Cr(III)Cr(II)) of E1/2 =

−1.23 V and E1/2 = −1.49 V, respectively. Sugden et al. reported49

that the

mutagenic bis- and tris-bipyridyl and bis-phenanthroline Cr(III) complexes have

half-wave potentials ranging from −0.73 to −0.93 V versus the standard hydrogen

electrode (SHE), which is slightly more negative than our value and suggests that

further investigation of the toxicity of fac-[Cr(gly)3] may be warranted.

Description of the Structure of fac-[Cr(gly)3] complex

The X-ray crystal structure of the facial tris(glycinato)chromium(III) complex,

has been reported by Bryan et al.17

and the central Cr(III) atom is octahedrally

coordinated by bidentate chelate rings of glycinate residues, through their nitrogen

and oxygen atoms.

Table 5.3: Selected experimental and computed bond distances (Å) and angles

(°) for fac-[Cr(gly)3]−1

and fac-[Cr(gly)3] complexes.a

Bond Distances (Å) [Cr(gly)3]−1

[Cr(gly)3] Exptlb

Cr‒N 2.310 2.095 2.068

Cr‒O 2.138 1.949 1.965

N‒C 1.467 1.476 1.479

C‒C 1.534 1.524 1.517

C‒O(coordinate) 1.283 1.305 1.290

C‒O(carbonyl) 1.234 1.219 1.223

Bond Angles (°)

N1‒Cr−O1 90.6 80.8 81.7

N1‒Cr−O3 174.1 174.9 174.2

N1‒Cr−N3 90.9 95.5 95.5

N1‒Cr−N2 95.1 96.3 96.4 a Optimized structures of facial

complexes at PBE1PBE/cc-pVDZ using the PCM


solvation model. Parameters defined in Figure 5.8. b Ref. 17.

Our previous results,85−87

indicate that PBE0/cc-pVDZ is a suitable theory for

determining the structural parameters of chromium(III) complexes. The calculated

structural parameters for the neutral (fac-[Cr(gly)3]) and reduced form (fac-

[Cr(gly)3]−1

) are reported in Table 5.3 (Figure 5.8) and compared to the X-ray data17

of fac-[Cr(gly)3]. The agreement between calculated and experimental structural

parameters is good. The calculated values for the N−Cr−N and N−Cr−O angles

indicate near octahedral geometry around the Cr3+

ion in the neutral form of the

complex (Table 5.3).

Figure 5.8: Structures of (a) fac-[Cr(gly)3]0 and fac-[Cr(gly)3]

−1 at PBE0/cc-pVDZ;

(b) maps of electrostatic potential (0.02 electrons Bohr−3

) (red = electron-rich, blue =

electron-deficient).


Reduction of the complex leads to differences in some of the key geometric

parameters. For example, there is a significant increase (~0.2 Å) in both the average

Cr‒N and Cr‒O bond lengths due to weaker binding to the metal centre. This also

appears to alleviate some strain within the chelate rings, such that the N1‒Cr‒O1 and

N1‒Cr‒N3 angles adjust to 90° (Figure 5.8 and Table 5.3). We note minor

differences were observed in the structural parameters of both species obtained in the

gas-phase and using the PCM solvation model.

The highest occupied molecular orbital (HOMO) and the lowest unoccupied

molecular orbital (LUMO) surfaces were determined for fac-[Cr(gly)3] using DFT

calculations (Figure 5.9). The HOMO is triply degenerate for the fac-[Cr(gly)3] and

consists of t2g* orbitals that are antibonding between the metal d-orbitals and the

carboxylate groups, consistent with weak -donor interactions. The LUMO consists

of the eg* orbitals that are anti-bonding -donor interactions between the metal and

the ligands.

Figure 5.9: Frontier orbitals for fac-[Cr(gly)3]0 at PBE0/cc-pVDZ (0.02 electrons

Bohr−3

surface).

eg*

t2g*


As noted earlier, the 1H NMR resonances for the glycine methylene protons

exhibit an up-field shift when coordinated to the metal, which is attributed to

paramagnetic contact between the d3 electrons and the ligand. It is evident from

Figure 5.9 that there is significant delocalisation of the electron density in the t2g*

orbitals between the metal and the -system of the ligands. Eaton69

and Freitas et

al.60

have suggested that this process is facilitated by donation from the ligand

orbitals into the e orbitals. The plots of the eg* orbitals of fac-[Cr(gly)3] clearly

indicate a contribution from the ligand orbitals, which is consistent with this

explanation of the paramagnetic NMR contact shift.

The weak field environment around the metal means that reduction of the

complex results in fac-[Cr(gly)3]−1

having a high-spin quintet electronic state with

one of the eg* orbitals being singly occupied. Occupation of this anti-bonding orbital

weakens the ligand binding and contributes to the increased Cr‒N and Cr‒O bond

lengths noted above. These results suggest that the reduced form of the complex may

be more susceptible to aquation and ligand exchange. However, the consistency in

the cyclic voltamograms over multiple cycles implies that ligand exchange is slow

on the time scale of the experiment.

Atomic charges were determined using natural bond orbital (NBOs) analysis

88

(see Figure 5.8). The positive charge on Cr decreases from +1.020 in fac-[Cr(gly)3]0

to +0.703 in fac-[Cr(gly)3]−1

(Figure 5.8a). However, the nitrogen and oxygen atoms

of the complex exhibit relatively little change in charge after the reduction of the

complex (Figure 5.8b).


Conclusions

We have prepared the fac-[Cr(gly)3] complex and fully characterized it by elemental

analysis, AAS, ESI-mass spectrometry, thermogravimetry, UV-Vis, EPR and 1H

NMR spectroscopy. The activation energy for thermal decomposition of fac-

[Cr(gly)3] complex is 360.4 kJ mol−1

obtained by the Kissinger method and 353.6 kJ

mol−1

by Ozawa method. DFT calculations of the geometrical parameters (bond

lengths and bond angles) are in good agreement with the X-ray crystal data.17

Our

calculations indicate one-electron reduction of the d3 [Cr(gly)3] complex leads to

[Cr(gly)3]−1

with a high-spin d4 electronic configuration. The changes in geometric

parameters upon reduction of the complex can be understood in terms of the partial

occupation of an eg* orbital, which is anti-bonding between Cr(III) and the ligands.

Cyclic voltammetry analysis identified a reversible redox process for fac-[Cr(gly)3]

with a half-wave potential of ‒0.66 to ‒0.69 V. Our results are consistent with a

reversible one-electron transfer process that is diffusion controlled at a three-phase

boundary (electrode‒solid‒solvent). It is therefore possible that the biological

activity of the [Cr(gly)3] complex may involve slow reversible electron transfer. We

note that the reduction potential of fac-[Cr(gly)3] is slightly more positive than the

values Sugden et al.49

obtained for a series of mutagenic bipyridyl and o-

phenanthroline Cr(III) complexes and might therefore warrant further investigation.


Full experimental details and characterization data for fac-[Cr(gly)3], including NMR

(1H), ESI-mass spectra, TG-DSC curves, cyclic voltammograms are available in the

Supporting Information and Appendix B. Also included are Cartesian atomic

coordinates for fac-[Cr(gly)3] from the DFT calculations.


Acknowledgements

We thank to Professor Bogdan Dlugogorski for providing access to the EPR

facilities. In addition, we are grateful to the Australian National Computational

Infrastructure (NCI) facility for computer time.

References

(1) Barrett, J.; Brien, P.; De Jesus, J. P. Polyhedron 1985, 4, 1‒14.

(2) Cooper, J. A.; Blackwell, L. F.; Buckley, P. D. Inorg. Chim. Acta 1984, 92,

23‒31.

(3) Vincent, J. The nutritional biochemistry of chromium (III), Elsevier, 2011.

(4) Glinsmann, W. H.; Mertz, W. Metabolism 1966, 15, 510‒520.

(5) Tang, H.-y.; Xiao, Q.-g.; Xu; H.-b.; Zhang, Y. Org. Process Res. Dev. 2013,

17, 632‒640.

(6) Broadhurst, C. L.; Domenico, P. Diabetes Technol. Ther. 2006, 8, 677‒687.

(7) Gielzak, K.; Wojciechowski, W. Chem. Pap. 1994, 48, 380‒385.

(8) Kallen, T. W.; Hamm, R. E. Inorg. Chem. 1977, 16, 1147‒1153.

(9) Ramasami, T.; Wharton, R. K.; Sykes, A. G. Inorg. Chem. 1975, 14, 359‒365.


Chi, J.; Feng, J. Diabetes 1997, 46, 1786‒1791.

(11) Anderson, R. A. Regul. Toxicol. Pharm. 1997, 26, S35‒S41.

(12) Tsave,O.; Yavropoulou, M. P.; Kafantari, M.; Gabriel, C.; Yovos, J. G.;

Salifoglou, A. J. Inorg. Biochem. 2016, 163, 323‒331.

(13) Saiyed, Z. M.; Lugo, J. P. Food Nutr. Res. 2016, 60, 31762.

(14) Refat, M. S.; El-Megharbel, S. M.; Hussien, M. A.; Hamza, R. Z.; Al-Omar,

M. A.; Naglah, A. M.; Afifi, W. M.; Kobeasy, M. I. Spectrochim. Acta. A

2017, 173, 122‒131.

(15) Evans, D. A. P.; Tariq, M.; Dafterdar, R.; Al Hussaini, H.; Sobki, S. H. Biol.

Trace Elem. Res. 2009, 130, 262‒272.

(16) Piotrowska, A.; Siwek, A.; Wolak, M.; Pochwat, B.; Szewczyk, B.; Opoka,

W.; Poleszak, E.; Nowak, G. J. Physiol. Pharmacol. 2013, 64, 493‒498.


1971, 10, 1468‒1473.

(18) Veal, J. T.; Hatfield, W. E.; Jeter, D. Y.; Hempel, J. C.; Hodgson, D. J. Inorg.

Chem. 1973, 12, 342‒346.


(19) Straughan, B. P.; Lam, O. M.; Earnshaw, A. J. Chem. Soc., Dalton Trans.

1987, 97‒99.

(20) Wallace, W. M.; Hoggard, P. E. Inorg. Chem. 1983, 22, 491‒496.

(21) Coleman, W. F. J. Lumin. 1980, 22, 17‒22.

(22) Yuen, G.; Heaster, H.; Hoggard, P. E. Inorg. Chim. Acta 1983, 73, 231‒234.

(23) Baca, A. G.; Coleman, W. F. J. Lumin. 1980, 22, 29‒35.

(24) Israily, N. Compt. Rend. 1966, 262, 1426.

(25) Gillard, R.; Laurie, S.; Price, D.; Phipps, D.; Weick, C. J. Chem. Soc., Dalton

Trans. 1974, 1385‒1396.

(26) Kita, E.; Marai, H.; Muziol, T.; Lenart, K. Transition Met Chem 2011, 36,

35−44

(27) Oki, H.; Yoneda, H. Inorg. Chem. 1981, 20, 3875‒3879.

(28) Vicens, M.; Fiol, J.; Terron, A.; Moreno, V. Inorg. Chim. Acta 1989, 165,

131‒137.

(29) Hoggard, P. E. Inorg. Chem. 1981, 20, 415‒420.

(30) Oki, H.; Otsuka, K. Bull. Chem. Soc. Jpn. 1976, 49, 1841‒1844.

(31) Oki, H. Bull. Chem. Soc. Jpn. 1977, 50, 680‒684.

(32) Perumareddi, J. R. Coord. Chem. Rev. 1969, 4, 73‒105.

(33) Ogard, A. E.; Taube, H. J. Am. Chem. Soc. 1958, 80, 1084‒1089.

(34) Wilkinson, G.; Gillard, R.; McCleverty, J. Eds.; Vol. 2, Ch. 15.; Pergamon

Press: Oxford, U. K., 1987.

(35) Larkworthy, L. F.; Nolan, K. B.; O,Brien. P. In Comprehensive Coordiation

Chemistry; G. Wilkinson, R. D. Gillard, J. A. McCleverty, Eds.; Vol. 3, pp

699−969; Pergamon Press: Oxford, U. K., 1987.

(36) Wilkins, R. G. Kinetics and Mechanism of Reaction of Transition Metal

Complexes, 2nd

ed.; VCH Publishers: New York, 1984.

(37) Szynkarczuk, J.; Drela, I.; Kubicki, J. Electrochim. Acta 1989, 34, 399‒403.

(38) Vandiver, M. S.; Bridges, E. P.; Koon, R. L.; Kinnaird, A. N.; Glaeser, J. W.;

Campbell, J. F.; Priedemann, C. J.; Rosenblatt, W. T.; Herbert, B. J.;

Wheeler, S. K. Inorg. Chem. 2009, 49, 839‒848.

(39) Barker, K. D.; Barnett, K. A.; Connell, S. M.; Glaeser, J. W.; Wallace, A. J.;

Wildsmith, J.; Herbert, B. J.; Wheeler, J. F.; Kane-Maguire, N. A. Inorg.

Chim. Acta 2001, 316, 41‒49.

(40) Tziouris, P. A.; Tsiafoulis, C. G.; Vlasiou, M.; Miras, H. N.; Sigalas, M. P.;

Keramidas, A. D.; Kabanos, T. A. Inorg. Chem. 2014, 53, 11404‒11414.

(41) Song, Y.; Chin, D.-T. Electrochim. Acta 2002, 48, 349‒356.

(42) Serpone, N.; Jamieson, M.; Henry, M.; Hoffman, M.; Bolletta; F.; Maestri,

M. J. Am. Chem. Soc. 1979, 101, 2907‒2916.

(43) Goforth, S. K.; Gill, T. W.; Weisbruch, A. E.; Kane-Maguire, K. A.; Helsel,

M. E.; Sun, K. W.; Rodgers, H. D.; Stanley, F. E.; Goudy, S. R.; Wheeler, S.


K. Inorg. Chem. 2016, 55, 1516‒1526.

(44) McDaniel, A. M.; Tseng, H.-W.; Hill, E. A.; Damrauer, N. H. ; Rapp , A. K.;

Shores, M. P. Inorg. Chem. 2013, 52, 1368‒1378.

(45) Cottica, S. M.; Nozaki, J.; Nakatani, H. S.; Oliveira, C. C.; Souza, N. E. d.;

Visentainer, J. V. J. Braz. Chem. Soc 2009, 20, 496‒501.

(46) Gagné, R. R.; Koval, C. A.; Lisensky, G. C. Inorg. Chem. 1980, 19, 2854‒2855.

(47) Isaacs, M.; Sykes, A. G.; Ronco, S. Inorg. Chim. Acta 2006, 359, 3847‒3854.

(48) Labuda, J.; Bustin, D.; Mocák, J. Chem. Pap. 1983, 37, 273‒283.

(49) Sugden, K.; Geer, R.; Rogers, S. Biochemistry 1992, 31, 11626‒11631.

(50) Protsenko, V.; Danilov, F. Electrochim. Acta 2009, 54, 5666‒5672.

(51) Stern, K. H.; Rolison, D. R. J. Electrochem. Soc. 1990, 137, 178‒183.

(52) González-Baró, A. C.; Pis-Diez, R.; Piro, O. E.; Parajon-Costa, B. S.

Polyhedron 2008, 27, 502‒512.

(53) S. Macgregor, E. McInnes, R. Sorbie and L. Yellowlees, Molecular

Electrochemistry of Inorganic, Bioinorganic and Organometallic

Compounds. Springer Netherlands, 1993. 503-517.

(54) Johnson, D. A.; Reid, M. A. J. Electrochem. Soc. 1985, 132, 1058‒1062.

(55) Ogino, H.; Nagata, T.; Ogino, K. Inorg. Chem. 1989, 28, 3656‒3659.

(56) Marangoni, D. G.; Smith, R. S.; Roscoe, S. G. Can. J. Chem. 1989, 67, 921‒926.

(57) Bond, A. M.; Colton, R.; Marken, F.; Walter, J. N. Organometallics 1994,

13, 5122‒5131.

(58) A. M. Bond, R. Colton, F. Daniels, D. R. Fernando, F. Marken, Y. Nagaosa,

R. F. M. Van Steveninck, J. N. Walter, J. Am. Chem. Soc. 1993, 115, 9556-

9562.

(59) Parajón‐Costa, B. S.; Wagner, C. C.; Baran, E. J. Z. Anorg. Allg. Chem. 2003,

629, 1085‒1090.

(60) Freitas, A. R.; Silva, M.; Ramos, M. L.; Justino, L. L.; Fonseca, S. M.;

Barsan, M. M. ; Brett, C. M.; Silva, M. R.; Burrows, H. D. Dalton Trans.

2015, 44, 11491‒11503.

(61) Hecht, M.; Schultz, F. A.; Speiser, B. Inorg. Chem. 1996, 35, 5555‒5563.

(62) Nguyen Pham, T. H.; Aitken, J. B.; Levina, A.; Lay, P. A. Inorg. Chem.

2014, 53, 10685‒10694.


Keramidas, A. D.; Kabanos, T. A. Inorg. Chem. 2014, 53, 11404‒11414.

(64) Lever, A. B. P. J. Chem. Educ. 1968, 45, 711‒712.

(65) Dou, Y. J. J. Chem. Educ. 1990, 67, 134.

(66) Weckhuysen,; B. M.; Schoonheydt, R. A.; Mabbs, F. E.; Collison, D. J.

Chem. Soc., Faraday Trans. 1996, 92, 2431‒2436.


(67) Green, C. A.; Bianchini, R. J.; Legg, J. I. Inorg. Chem. 1984, 23, 2713‒2715.

(68) Green, C. A.; Place, H.; Willett, R. D.; Legg, J. I. Inorg. Chem. 1986, 25,

4672‒4677.

(69) D. R. Eaton J. Am. Chem. Soc. 1965, 87, 3097 – 3102.

(70) Govindaraju, V.; Young, K.; Maudsley, A. A. NMR Biomed. 2000, 13, 129‒153

(71) Mateescu, C.; Gabriel, C.; Raptopoulou, C.; Terzis, A.; Tangoulis, V.;

Salifoglou, A. Polyhedron 2013, 52, 598‒609.

(72) (a) Spiccia, L.; Marty, W.; Giovanoli, R. Inorg. Chem. 1988, 27, 2660‒2666.

(b) Kanagaraj, J.; Panda, R. C.; Sumathi, V. RSC Adv. 2015, 5, 45300‒45319.

(73) López Muñoz, B. E.; Rivera Robles, R.; Iturbe García, J. L.; Olguín

Gutiérrez, M. T. J. Mex. Chem. Soc. 2011, 55, 137‒141.

(74) Tang, Y.; Michel, F. M.; Zhang, L.; Harrington, R.; Parise, J. B.; Reeder, R.

J. Chem. Mater. 2010, 22, 3589‒3598.

(75) Gabriel, C.; Raptopoulou, C.; Terzis, A.; Tangoulis, V.; Mateescu, C.;

Salifoglou, A. Inorg. Chem. 2007, 46, 2998‒3009.

(76) Lesnikovich, A.; Ivashkevich, O.; Lyutsko, V.; Printsev, G.; Kovalenko, K.;

Gaponik, P.; Levchik, S. Thermochim. Acta 1989, 145, 195‒202.

(77) Friedman, H. L. Physica 1964, 30, 537.

(78) Kissinger, H. E. Anal. Chem. 1957, 29, 1702‒1706.

(79) Ozawa, T. Bull. Chem. Soc. Jpn. 1965, 38, 1881‒1886.

(80) Nicholas, R. S.; Shain, I. Anal. Chem. 1964, 36, 706.

(81) Bard, A. J.; Faulkner, L. R. Electrochemical Methods, 2nd ed.; Wiley: New

York, 2001.

(82) Wopschall, R. H.; Shain, I. Anal. Chem. 1967, 39, 1514‒1527.

(83) Serpone, N.; Jamieson, M. A.; Henry, M. S.; Hoffman, M. Z.; Bolletta, F.;

Maestri, M. J. Am. Chem. Soc. 1979, 101, 2907‒2916.

(84) S. Chandrasekhar, D. G. Fortier, A. McAuley, Inorg. Chem. 1993, 32, 1424‒1429.



88‒97.

(87) Uddin, K. M.; Henry, D. J. ChemistrySelect, 2016, 1, 5236‒5249.

(88) Reed, A. E.; Curtiss, L. A.; Weinhold, F.; Chem. Rev. 1988, 88, 899‒926.


2013

(90) Miertuš, S.; Scrocco, E.; Tomasi, J. Chem. Phys 1981, 55, 117‒129.

(91) Tomasi, J.; Mennucci, B.; Cammi, R. Chem. Rev. 2005, 105, 2999‒3094.

Chapter 6: *

Uddin, K. M.; Poirier, R. A.; Henry, D. J. Polyhedron (accepted) 166

Chapter 6

Investigation of Mono-, Bis- and Tris-glycinatochromium

(III): Comparisons of Computational and

Experimental Results*

Abstract

The synthesis and characterization of mono-, bis- and tris-glycinatochromium(III)

complexes by elemental analysis, AAS, UV-Vis, electron paramagnetic resonance

(EPR), ATR-FTIR and Raman spectroscopy has been performed in this work. IR

stretching bands obtained from DFT calculations of the mono-, bis- and tris-

glycinatochromium(III) complexes are in good agreement with experimental data.

Different mechanistic pathways were explored for the water exchange reactions of

[Cr(+NH3CH2COO

‒)(H2O)5]3+

and its conjugate base species including, associative

interchange (Ia), and dissociative (D) mechanisms The lowest activation enthalpies

are obtained for the Ia pathways of with explicit outer sphere solvation (88 and 76 kJ

mol1

), which are in good agreement with the experimental values (87 and 75 kJ

mol1

). In comparison, tris-glycinatochromium(III) undergoes aquation via the

Chapter 6: *


dissociative (D) mechanism. Investigation of these systems in the pH range ~3.0 to

~8.5 by UV-Vis monitoring, helps identify the speciation of these complexes in

physiological environments.

Introduction

Chromium(III)−amino acid (AA) complexes are commonly found in nutritional

supplements and their physiological activity is of biological importance.1−6

Several

studies7−10

suggest that Cr(III) in nutritional supplements is of minimal benefit to the

function of healthy biological systems. However, there has been significant evidence

showing that daily treatment of alloxan-induced diabetic (AID) mice with 500 and

1000 µg Cr/kg body weight (BW) of a Cr(III)−AA complex such as chromium

methionine (CrMet),10

has significant beneficial effects as an insulin enhancing

treatment and is more effective than chromium trichloride hexahydrate (CrCl3.6H2O)

and chromium nicotinate (CrNic). There are also reports suggesting that CrMet has a

beneficial impact on body weight, triglyceride, total cholesterol and liver glycogen

levels, while also reducing lipid metabolism.10

However, the details of the bio-

inorganic pathways for these complexes are not known. Trivalent chromium

complexes administered orally will encounter a number of different physiological

environments (pH and chemical conditions) as they pass through the body including

the high acidity of the stomach and the slightly alkaline conditions of the

bloodstream. These differing conditions provide opportunities for chromium(III)

complexes to undergo ligand substitution. Furthermore, olation and precipitation can

occur at neutral or alkaline pH. It also appears that the bioinorganic mechanism

depends upon the ligands coordinated to chromium, with a very slow rate of

exchange with the active sites of enzymes.1−2

Chapter 6: *


It is well-known that the amino acids (AA) can adopt several forms including

anion (AA‒), zwitterion (HAA), or cation (H2AA

+) (see Figure 6.1a) depending on

the pH of the solution.

HH3N

O

O

R

H2N

O

O

R H

H

H

H3N

O

OH

R

AA HAA H2AA

R = Amino acid (AA) side chains

(a)

(b)

H3N

O

O . . . Cr3+

R

A

Figure 6.1: (a) Protolytic forms of glycine and amino acids (R is the amino acid side

chain) and (b) Cr(III) metal interaction with the amino acid zwitterion to form

monodentate Cr(III) complex (A).

Characterization studies suggests that coordination of Cr(III) with amino acid

ligands (e.g. glycine, serine and methionine) occurs through the glycinate nitrogen

and carboxylate oxygen to form a chelate ring. However, side chain functional

groups can also play a prominent role in metal ion binding at physiological pH.

There have been a number of experimental investigations of Cr(III)-AA

complexes.11−18

For example, Abdullah et al.12

measured the kinetics for the

formation of tetra-aquaglycinechromium(III) (pH = 3.0‒3.8, temperature = 40‒45 °C

and ionic strength = 0.4 mol dm‒3

). They proposed an associative interchange (Ia)

pathway with outer-sphere complexation involving glycine-hexa-aquachromium(III)

(ΔH‡ = 87 kJ mol

1) or its conjugate base (ΔH

‡ = 75 kJ mol

1).

12 By contrast, the

Chapter 6: *


dissociative mechanism has been reported for reaction of [Cr(NH3)5(H2O)]3+

with

glycine (ΔH‡ = 106 kJ mol

1 and ΔS

‡ = +23.5 J mol

1K

‒1) [13]. In the pH range 3.0

to 4.5, chromium(III) will interact with amino acids in the zwitterion (HAA) and

cationic (H2AA+) forms.

11−12,16 Therefore, the formation of the chromium(III)−AA

complex (A), is coupled with the acid dissociation equilibrium of the amino acid as

shown in Figure 6.1b.

Cr(III) can coordinate up to 3 glycinate ligands to give the mono-, bis-, and tris-

glycinatochromium(III) species. In addition to formation studies, there have been

several experimental aquation studies of Cr(III)-AA complexes.17−22

Kita et al.23

obtained ΔH‡ and ΔS

‡ values of 91.4 ± 2 kJ mol

1 and 17 ± 7 J mol

1 K

‒1 for acid-

catalysed and 66.7 ± 16 kJ mol1

and ‒71 ± 53 J mol1

K‒1

for base-catalysed

aquation of [Cr(gly)3]. There is evidence that the acid-catalysed aquation of fac-

[Cr(gly)3]0 is initiated by dissociation of the Cr‒N bond.

23 They also determined that

[Cr(gly)2(OH)2]‒ undergoes base-catalysed aquation ~4 times faster than

[Cr(gly)2(O‒gly)(OH)]‒. Furthermore, Kita and Lisiak reported

24 that the acid-

catalysed aquation of the [Cr(ox)2(AA)]2‒

complexes (where AA = alanine (Ala‒),

valine (Val‒) or cysteine (Cys

‒)) involves formation of a metastable intermediate

with a monodentate O-bonded ligand. Kiersikowska et al. also investigated25−26

the

acid- and base-catalysed aquation of fac-[Cr(Aa)3] (Aa = Gly, Ala, Asn) and its aqua

derivatives. However, to date there have not been any theoretical studies of the

intimate details of ligand exchange on Cr(III)-AA complexes.

Recently we reported27−29

theoretical studies of aquation of chromium(III)

chloride and other dihalides under acidic conditions, as might be found in the

stomach after oral ingestion of these species. A key finding in these studies was the

Chapter 6: *


importance of outer-sphere solvation on the reaction mechanism. We have also

studied29

the aquation of conjugate-base systems of chromium(III), which is

favourable under physiological conditions (pH ~ 7.4), where once again outer sphere

solvation is involved in the reaction mechanism. Therefore, the study of these

systems leads to a better understanding of the stability and speciation under

physiological conditions, prior to binding to enzymes or peptides. Investigation of

the aquation of mono-, bis-, and tris-glycinatochromium(III) complexes may also

provide insight into the stability of chromium(III) protein complexes. Recently,

Uddin et al.30

studied the synthesis of the fac-[Cr(gly)3] complex, which was fully

characterized by elemental analysis, AAS, ESI-mass spectrometry, UV-Vis, EPR

and 1H NMR spectroscopy. Furthermore,

the electrochemical properties of fac-

[Cr(gly)3] were investigated by cyclic voltammetry and the thermal decomposition

determined using differential scanning calorimetry (DSC).

To date, no computational mechanistic studies have been reported for the aquation of

open-type monodentate and closed-type bidentate systems of glycinato-

chromium(III) ([Cr(gly)x(H2O)6-2x](3-x)+

where x = 1 ‒3) and their conjugate bases

species. Therefore the major objective of this study is to provide a detailed

investigation of the pathways for these processes using DFT calculations.

Theoretical calculations are complemented by experimental characterisation of

Cr(gly), Cr(gly)2, and Cr(gly)3 by elemental analysis, thermal analysis, AAS,

UV−Vis, EPR, ATR-IR, and Raman spectroscopy.

Chapter 6: *


Experimental

Reagents

Chromium(III) chloride hexahydrate, CrCl3.6H2O (Sigma-Aldrich, ≥98.0%),

glycine (Sigma-Aldrich, ≥98.5%), ethanol (LabServ, 99.8%) and sodium hydroxide

(Sigma-Aldrich, 98.5%) were used without further purification. Millipore filtered

deionized water was used throughout the experimental work.

Analytical Methods

Thermal analysis was conducted using a PerkinElmer STA 8000 TGA-DSC

instrument. Before the heating routine program was activated, the entire system was

purged with argon for 10 min at a rate of 20 mL/min, to ensure that the desired

environment was established. Thermogravimetric analysis (TGA) experiments were

performed with a heating rate of 20 °C min−1

from 30 to 1000 °C. High-purity argon

gas was used at a constant flow rate of 20 mL/min. Solution ultraviolet-visible (UV-

Vis) spectra were recorded on a HP 8453 UV-Vis spectrophotometer over the range

190−1100 nm. A Bruker EMX EPR spectrometer running the Xenon software with a

Bruker ER 036TM NMR tesla meter was used to measure the X-band spectra of the

solid Cr(III) complexes with EP parameters: center field, 3200 G; sweep width,

6000 G; width TM, 200 G; Frequency Mon, 9.79 GHz; microwave power, 2.0 mW;

microwave attenuation, 20.0 dB; conversion time, 4.0 ms; gain, 30dB; modulation

amplitude, 4.0 G; modulation frequency, 100 kHz; resolution, 15000 and sweep

time, 60 s. For frozen solution samples, the EPR tube was flushed with with nitrogen

gas for 2 min before analysis and the samples frozen by slow immersion in a liquid

nitrogen bath. EPR parameters for frozen samples were gain, 1.0 × 104; modulation

frequency, 100 kHz; modulation amplitude, 1 G; conversion time, 0.41 ms; time

Chapter 6: *


constant, 81.92 ms; sweep time, 41.98 s; field center, 3350 G; sweep width, 6400 G;

frequency, 9.531415 GHz; and power, 2.0 mW.

Infrared analysis was carried out using a Perkin Elmer FT-IR with a universal

ATR sampling accessory. ATR spectra were generated using 4 scans with a

resolution of 4 cm‒1

in a range of 4000 ‒ 350 cm‒1

. A constant pressure between the

ATR foot, sample and ATR crystal was achieved using an in-built pressure gauge

and software monitor. Raman analysis was conducted using a Nicolet 6700 FT-IR

with NXR FT-Raman module. Raman spectra were collected with: 1064 nm

excitation wavelength, 90o detection angle, CaF2 beam splitter, InGaAs detector,

gain of 1, optical velocity of 0.3165, aperture of 150, focus and side-to-side settings

optimised, laser power of 1.5 W and 256 scans with resolution of 8 cm‒1

in the range

4000 ‒ 200 cm‒1

.

The pH of solutions was measured using LabCHEM-pH meter with a PBFB

electrode and calibration of the pH combination electrode was carried out.

Preparation of [Cr(gly)(H2O)4]2+

and [Cr(gly)2(H2O)2]+

The fac-[Cr(gly)3] complex was synthesized and fully characterized as reported

previously.30

The mono- and bis-glycinatochromium(III) complexes were

synthesised in the current study using 1:1 and 1:2 ratios of chromium(III) chloride

hexahydrate, CrCl3.6H2O and glycine, prepared according to a similar method used

for the tris-system.19,30

The solid product was collected by vacuum filtration, washed

with ethanol and dried in a desiccator. A greenish (light blue) product was obtained

for mono-glycinatochromium(III) ([Cr(gly)(H2O)4]2+

) and a red (light pink) solid

was isolated for bis-glycinatochromium(III) ([Cr(gly)2(H2O)2]+). The yield of the

reactions were 1.35 g (57%) for mono- and 1.37 g (69%) for bis-dentate

Chapter 6: *


chromium(III) glycinate complexes. These complexes were characterised by

thermogravimetry, UV-Vis, and EPR spectroscopy in the solid state (room temp.)

and frozen solution state (100 K). Anal. Calcd for C2H12Cl2CrNO6: C, 8.93; H, 4.50;

Cl, 26.36; N, 5.21; Cr, 19.33. Found: C, 8.91; H, 4.93; Cl, 26.43; N, 4.85; Cr,

18.87%. Anal. Calcd for C4H12ClCrN2O6: C, 17.69; H, 4.45; Cl, 13.06; N, 10.31;

Cr, 19.15. Found: C, 16.67; H, 4.46; Cl, 12.97; N, 10.28; Cr, 18.71%.


Standard hybrid density functional theory calculations were carried out with

Gaussian 09.31

The geometries of all complexes in this study were fully optimized in

the gas phase and solvent (water) at the PBE0/cc-pVDZ level of theory, which we

have previously shown to give excellent agreement with experiment for the

activation enthalpies for aquation of related species.27−29

The effect of solvent (water)

on the structures and energetics of each complex was investigated using the

polarisable continuum model (PCM) of Tomasi and co-workers.32−33

Vibrational

frequencies were obtained for all optimized structures to check for the absence of

imaginary frequencies for reactants, intermediates and products and for the presence

of a single imaginary frequency for each transition state. For all the reaction

pathways discussed in this study, the transition states were also analyzed using the

intrinsic reaction coordinate (IRC) method. The final structures obtained from each

IRC were further optimized in order to positively identify the reactant and product

complexes to which each transition state is connected.

Natural bonding orbital (NBO)34

analysis was performed and the charges were

calculated. The HOMO and LUMO energies were determined from the ground state

quartet geometries. Enthalpies of activation and entropies of activation were

Chapter 6: *


calculated at 298.15 K. In the figures, all distances are in angstroms (Å), angles in

degrees (º) and the energies in kJ mol‒1

. Unless otherwise stated, all values given in

the text were obtained at the PBE0/cc-pVDZ level in solution (PCM). The optimized

structures and the relative energies of reactants, intermediates, transition states, and

products for all pathways are shown in Tables ST1 to ST7 and Figures S1 to S49 of

the Supporting Information (SI). The volume of activation (∆V‡) and reaction

volume (∆V) were correlated to the change in the Cr−L bond lengths of the TS,

reactant and product as reported previously.27−29

Figure 6.2: (a) Optimized structures of the conformers for [Cr(gly)]3 at PBE0/cc-

pVDZ; (b) maps of electrostatic potential (0.02 electrons Bohr−3

) (red = electron-

rich, blue = electron-deficient).

The [Cr(gly)3]0

complex can occur as either a facial or meridional isomer. The

XRD crystal structure for [Cr(gly)3]0 reported by Bryan et al.

19 shows only the facial

isomer, which implies that this isomer is the most stable in the solid state. However,

our DFT calculations of the complex with PCM solvation indicate that the energies

of the fac- (I) and mer- (II) isomers are almost identical with the meridional structure

very slightly favoured (~ 0.5 kJ mol1

) at PBE0/cc-pVDZ (Figure 6.2). There have

Chapter 6: *


been some reports that suggest that glycine can also form chelate rings via

coordination of the two carboxylate oxygen atoms. Therefore, for completeness the

stability of two additional structures of the form [Cr(gly)2(O2CCH2NH2)] (III and

IV) were investigated and found to be less energetically stable than those with facial

configuration by 30 and 40 kJ mol1

.

As reported recently30

, the calculated Cr−N and Cr−O bond distances for fac-

[Cr(gly)3] are in good agreement with the experimental X-ray crystal data.19

The

calculated average values for the Cr−N and Cr−O bond lengths of mer-[Cr(gly)3] are

2.107 and 1.944 Å, respectively (Table ST1 of the SI).

Figure 6.3: Optimized conformations of precursor [Cr(gly)2(H2O)2]+ for (a) cis-

conformers (V‒VII forms) and (b) trans-conformers (VIII‒IX forms) at PBE0/cc-

pVDZ.

Three different cis-conformations of [Cr(gly)2(H2O)2]+ were investigated with

structure (V) the most stable (Figure 6.3). Isomers, VI and VII were higher in energy

by 1.5 and 6.3 kJ mol1

, respectively. Two trans isomers of [Cr(gly)2(H2O)2]+ (VIII

and IX) were also investigated and were found to be 4.1 and 19.1 kJ mol−1

higher in

Chapter 6: *


energy than (V). Isomers of the mono-glycinatochromium(III) complex were also

investigated including closed-type bi-dentate (X) and open-type mono-dentate via

intramolecular Cr−N ring opening (Figure 6.4).

Figure 6.4: Optimized structures of [Cr(gly)(H2O)4]2+

and [Cr(+NH3CH2COO

‒

)(H2O)5]3+

at PBE0/cc-pVDZ.


Synthesis and characterization of [Cr(gly)(H2O)4]2+

, [Cr(gly)2(H2O)2]+ and

[Cr(gly)3]0

In our earlier work30

, the solid-state UV-Vis spectrum of the synthesised fac-

[Cr((gly)3]0 complex revealed two spin-allowed d‒d transitions with absorption

bands at 388 nm and 510 nm. However, several studies have shown that the

absorption spectra of Cr(III) complexes can have significant solvent and pH

dependence. For example, Emerson and Graven35

measured the UV/Vis spectra of

chromium(III) perchlorate solutions within a pH range of 2‒5 and noted a transition

in the absorbance, corresponding to the conversion of the chromium(III) perchlorate

solution to [Cr(H2O)5OH]2+

.36

Consequently they were able to report the equilibrium

constant for the acid dissociation reaction of [Cr(H2O)6]3+

to [Cr(H2O)5OH]2+

.

Therefore the solvent dependence of UV/Vis spectra of the complexes in this study

(CrCl3.6H2O (or [Cr(H2O)4Cl2]+), [Cr(gly)(H2O)4]

2+, [Cr(gly)2(H2O)2]

+ and fac-

Chapter 6: *


[Cr(gly)3]0) was investigated. The two major spin-allowed d-d bands for the

chromium(III) chloride [Cr(H2O)4Cl2]+

starting material were observed at 443 nm

(20.6 M‒1

cm‒1

) and 634 nm (21.0 M‒1

cm‒1

) in water at pH 3.86 (Figures 6.5a and

S1 to S2 of the SI). Previously, Elving and Zemel examined, [Cr(H2O)4Cl2]+ in 12

M HCl, reporting bands at 450 nm (27.9 M‒1

cm‒1

) and 635 nm (23.9 M‒1

cm‒1

)

over the range of 350 to 800 nm.37

Figure 6.5: UV-visible (UV-Vis) spectrum of CrCl3.6H2O, [Cr(gly)(H2O)4]2+

,

[Cr(gly)2(H2O)2]+ and fac-[Cr(gly)3]

0 in the range 350‒800 nm for (a) at pH ~3.0 to

~6.0 and (b) at pH ~5.0 to ~8.3, in water. The concentration of the species are c = 11,

21, 14, and 9 mM, respectively.

During formation of mono-glycinatochromium(III) complex at pH 3.07, these

two main bands are observed to shift to 433 nm (18.5 M‒1

cm‒1

) and 591 nm (16.6

M‒1

cm‒1

), respectively (Figure 6.5). With the addition of a second glycine ligand to

form bis-glycinatochromium(III) at pH 3.76 there is a further shift of the bands to

409 nm (23.6 M‒1

cm‒1

) and 553 nm (27.2 M‒1

cm‒1

). Finally, the two bands are

located at 395 nm (23.6 M‒1

cm‒1

) and 527 nm (44.2 M‒1

cm‒1

) for tris-

Chapter 6: *


glycinatochromium(III) at pH 6.02 (Figure 6.5). Bryan et al.19

reported the solid

UV/Vis spectrum of fac-[Cr(gly)3]0

and assigned 4A2g →

4T2g to the 503 nm transition

and 4A2g →

4T1g (F) to the 384 nm transition. On this basis the longer wavelength

peaks above can be tentatively attributed to the 4A2g →

4T2g transition and the shorter

wavelength peaks to the 4A2g →

4T1g (F) transition in the octahedral (Oh)

approximation. Mono- and bis-glycinatochromium(III) complexes are intermediates

in both the step-wise formation and aquation of tris-glycinatochromium(III).38−39

Therefore, the progress of these reactions can be monitored by shifts in these two

peaks. As noted above, raising the pH of solutions of chromium(III) species can lead

to the formation of the corresponding conjugate base species, which also leads to

changes in the absorption spectra. UV-Vis absorption spectra of solutions of

CrCl3.6H2O, [Cr(gly)(H2O)4]2+

, [Cr(gly)2(H2O)2]+ and fac-[Cr(gly)3]

0 at elevated pH

are shown in Figure 6.5b. Notably, when the pH of CrCl3.6H2O and

[Cr(gly)(H2O)4]2+

solutions was raised to ~ 8.1, the samples became cloudy and

turned purple indicating the formation of the corresponding conjugate base

complexes and subsequent formation of oligomeric species. However,

[Cr(gly)2(H2O)2]+ and fac-[Cr(gly)3]

0 complexes were less susceptible to this change

in pH and gave identical spectra under acidic and alkaline conditions as shown in

Figure 6.5b. The aqueous deprotonation of [Cr(gly)x(H2O)6-2x](3-x)+

(where x = 1 ‒ 3),

and their conjugate bases was investigated by DFT using reactions of the following

form:

[Cr(H2O)5 (OglyH)]3+

[Cr(H2O)4(gly)]2+

+ H3O+ (6.1)

[Cr(H2O)4(gly)]2+

+ H2O [Cr(H2O)4(OH)(gly)]+ + H3O

+ (6.2)

Consequently, we have calculated aqueous Gibbs free energies for deprotonation

(ΔGdp) using reactions (6.1) and (6.2), for the mono-, bis-, and tris-species. The

Chapter 6: *


ΔGdp values decrease for these two reactions from tris- through to the mono-species

(see Figs. S2A−S2B in the SI), with a variation of ~ 35 and ~ 111 kJ mol‒1

,

respectively, across the series. The large positive ΔGdp values for the tris- and bis-

species indicates that deprotonation of these complexes is not favourable and

corresponds with the experimentally observed inertness towards mildly alkaline

conditions. The lower calculated ΔGdp for mono-glycinatochromium(III) represents a

greater propensity for formation of the corresponding conjugate base species and is

consistent with the experimentally observed colour change and precipitate formation.

Figure 6.6: (a) Thermogravimetric (TG) analysis and analysis of X-band EPR

spectra for fac-[Cr(gly)3], [Cr(gly)2]+ and [Cr(gly)]

2+ species, (b) in solid state

(room temp.) (c) in frozen state (100 K) and 15.3 mM (red), 14.4 mM (violet) and

20.9 mM (dark blue) in H2O, respectively.

Chapter 6: *


Thermogravimetric (TG) analysis under argon was carried out to determine the

stability of the [Cr(gly)(H2O)4]2+

and [Cr(gly)2(H2O)2]+ complexes (Figures 6.6a and

S3 in the SI). In this work, the differential scanning calorimetry (DSC) curves show

three endothermic stages of mass loss in the temperature ranges (a) (50 °C to 280

°C), (b) (350 °C to 480 °C), and (c) (750 °C to 920 °C) for mono- and bis-

complexes with a heating rate of 20 °C min−1

from 30 to 1000 °C. The first stage

shows a gradual weight loss indicating the release of lattice water for both mono-

and bis-complexes. A second mass loss occurs around 383 °C for mono-, 413 °C for

bis- and 424 °C for tris-glycinatochromium(III)30

(Figure 6.6a) corresponding to the

decomposition of the glycine ligand(s). Notably, the transition temperature increases

across the series from mono, bis and tris-glycinato complexes, reflecting the

increased stability with increased chelation. However, these transition temperatures

are lower than the value recently reported by Freitas et al.40

(472 °C) for tris(8-

hydroxyquinolinate)chromium(III).

The paramagnetic behaviour of these octahedral chromium(III) complexes

should be sensitive to these changes in coordination environment and can be

characterized by two different X-band EPR signals. The g = ~2 β‒signal is a broad

isotropic signal and the g = 3.5−5.5 δ‒signal is a positive lobe in the region.41−43

In

this study, the broad β‒signal of the solid EPR gradually decreases from 1.9924 to

1.9886 for mono-, bis-, and tris-glycinatochromium(III) complexes, respectively,

(Figure 6.6b). For the mono-, bis-, and tris-glycinatochromium(III) series of

complexes, the Cr(III) coordination environment changes from (CO2)(NH2)(H2O)4 to

(CO2)2(NH2)2(H2O)2 to (CO2)3(NH2)3, respectively. Values obtained in the frozen

solution state at 100 K in H2O are slightly lower (g = 1.9710, 1.9737, and 1.978730

,

Chapter 6: *


for mono-, bis- and tris-glycinatochromium(III) species, respectively) (see Figure

6.6c), which may be the result of solvent effects.30,43

Vibrational Spectra of [Cr(gly)(H2O)4]2+

, [Cr(gly)2(H2O)2]+ and [Cr(gly)3]

0

The IR and Raman spectra of solid glycine, [Cr(gly)3]0, [Cr(gly)2(H2O)2]

+ and

[Cr(gly)(H2O)4]2+

complexes are presented in Figure 6.7 (see Figure S4‒S13 and

Table ST2 of the SI). These spectra indicate that glycine coordinates to Cr(III) in the

mono-, bis-, and tris-systems via the carboxylic and amino groups (see Figure S4 of

the SI). For example, the [Cr(gly)3]0 complex has six IR(Raman) frequencies

attributed to NH2 motions including, asymmetric (~3222 (~3248) cm‒1

) and

symmetric (~3134 (~3184) cm‒1

) N‒H stretches, and the NH2 scissoring (~1590

(~1593) cm‒1

), twisting (~1311 (~1325) cm‒1

), wagging (~1144 (~1161) cm‒1

) and

rocking (~752 (~746) cm‒1

) vibrations (Figures 6.7 and S4‒S5 and Table S1).

Furthermore, the −COO− stretching bands of the free glycine acid, ν(C꞊O) = 1586

(1564) cm‒1

and ν(C−O) = 1412 (1410) cm‒1

, are transformed in the complexes via

νas(COO−) and νs(COO

−) stretching vibrations of the carboxylate group. In the

spectrum of [Cr(gly)3]0 the antisymmetric COO

‒ vibrations appeared in the range

1690‒1630 (1658‒1629) cm‒1

, whereas the symmetric COO‒ vibrations for the C‒O

stretching bands are in the range 1425‒1379 (1430‒1405) cm‒1

(Figure 6.7 and

Table S1). As can be seen from these ranges, the variance between the νas(COO−)

and νs(COO−) stretching vibrations of the carboxylate group is larger than 200 cm

‒1.

Generally, differences of this magnitude between νas(COO−) and νs(COO

−) are

attributed to resonance, which arises when the carboxylate group is deprotonated or

coordinated to a metal.44−46

Guindy et al.14

reported the IR spectrum of solid Cr(leu)2,

which showed a shift of the carboxyl group (∆ν = νas − νs) of 254 cm‒1

. Likewise

Chapter 6: *


for solid CrMet, the observed9

difference is 219 cm‒1

. There is reasonable agreement

between the experimental and theoretical data (Table ST2 and Figures S4‒S13 of the

SI), for all three complexes.

Figure 6.7: ATR-IR (a‒d) and Raman (e‒h) spectra (1800‒350 cm‒1

) for glycine,

fac-[Cr(gly)3], [Cr(gly)2(H2O)]+ and [Cr(gly)(H2O)4]

2+ complexes.

In the spectrum of fac-[Cr(gly)3]0, [Cr(gly)2(H2O)2]

+ and [Cr(gly)(H2O)4]

2+

complexes, the C‒N stretching bands are observed at 1032 (1043), 1038 (1040) and

1039 (1041) cm‒1

, respectively, whereas the free glycine band was observed at 1033

(1036) cm‒1

(see Table ST2). The C‒C‒N bending vibration of glycine with Cr(III)

metal is assigned to the bands at 917 (894), 912 (923) and 898 (924) cm‒1

for fac-

[Cr(gly)3]0, [Cr(gly)2(H2O)2]

+ and [Cr(gly)(H2O)4]

2+ species, respectively, whereas

Chapter 6: *


in the free glycine the band appears at 924 (893) cm‒1

(Figure 6.7). The O‒C‒O

wagging vibration of these complexes is assigned to the bands observed at 696

(686), 695 (698) and 699 (707) cm‒1

, respectively. The corresponding vibration of

glycine is assigned to the band at 707 (697) cm‒1

. The Cr‒N stretching modes appear

at 581 (578), 576 (577) and 515 (552) cm‒1

, and the Cr‒O bands appear at 468 (470),

469 (475) and 416 (415) cm‒1

, respectively. Guindy et al.14

reported Cr‒O stretching

bands in the 600 cm‒1

region for chromium(III) coordinated to DL-leucine but they

did not report a value for the Cr−N stretching band. The X-ray crystal structure of

the facial tris(glycinate)chromium(III) complex has been reported by Bryan et al.19

,

showing that the Cr‒N and Cr‒O bond distances are 2.068 and 1.965 Å, respectively,

suggesting the Cr−N bond is weaker than the Cr−O bond and should have a higher

frequency vibration. Robert et al.47

measured the IR spectra of bis(glycine)

complexes of Pt(II), Pd(II), Cu(II), and Ni(II) species, finding that the metal‒

nitrogen stretching bands (550‒430 cm‒1

) appear at higher frequencies than the

metal‒oxygen stretching bands (420‒290 cm‒1

), which is consistent with our

systems. However, the effects of the molecular symmetry and H-bonding may also

impact the ‒NH2 and ‒COO‒ modes.

48−49 Nevertheless, the IR stretching bands of

the tris-, bis- and mono-glycinatochromium(III) complexes in this study are in good

agreement with theoretical values, which were obtained without explicit outer sphere

solvation of the complexes (Table ST2 and Figures S4‒S13 of the SI).

DFT Studies of glycinato-chromium(III) complexes

The experimental characterization reveals that the tris-glycinatochromium(III)

complex is quite stable in both solution and solid forms. In comparison, the TGA

results imply that the mono-glycinatochromium(III) complex is the least stable and

Chapter 6: *


therefore likely to be the most reactive. Therefore we have investigated the

mechanisms for aquation of these complexes.

Aquation mechanisms for open-type chromium(III) systems

Under acidic conditions, the [Cr(H2O)4(gly)]2+

complex is expected to convert to

[Cr(H2O)5(+NH3CH2COO

−)]

3+ via a Cr−N ring opening with the two structures in

equilibrium (see Figures 6.3−6.4). This is supported by the observation of higher

frequency vibrations for the Cr−N bonds compared with Cr−O bonds. In fact the

Cr−N vibration of the tris complex has the highest frequency value in the series,

suggesting it has the weakest Cr−N bond(s) and therefore will be most susceptible to

ring-opening. Water exerts a weaker trans effect than carboxylate and hence there is

an increase in the Cr−N bond strengths with aquation. All our attempts to locate a

transition state for this ring opening process failed across the three bidentate systems.

Therefore, we use the open-type structures ([Cr(H2O)5(+NH3CH2COO

−)]

3+) as the

starting point for investigation of aquation reactions. The structures of the reactants,

transition states and products involved in the interchange mechanism of the aquation

of [Cr(+NH3CH2COO

‒)(H2O)5]

3+ are presented in Scheme 6.1 and Figures S14 to

S16 of the SI.

As shown in Scheme 6.1, deprotonation of an inner sphere water molecule can

also occur to give the conjugate base ([Cr(+NH3CH2COO

‒)(H2O)4(OH)]

2+), which

can subsequently undergo aquation in a similar manner to the parent complex

(Figures S17 to S20). Aquation reactions of [Cr(+NH3CH2COO

‒)(H2O)5]3+

can

potentially occur via several different mechanisms including interchange and

dissociative pathways.

Chapter 6: *


Scheme 6.1: Associative Interchange (Ia) mechanism for the aquation of

[Cr(+NH3CH2COO

‒)(H2O)5]3+

and [Cr(+NH3CH2COO

‒)(H2O)4(OH)]2+

with

H2O.

H2O

Cr3+

O

O L

O

O

O

H

H

[Cr(NH3CH2COO)(H2O)5]3+

Cr

O

OO

O

O

OH

HH

H

O

HH

H

H

H

CrO

OO

O

O

OH

HH

H

O

H

H

Cr

O

O O

OO

H

HH

H

OO

H

H

H

H

H

H

(R) (TS) (P)

(P)

H

H

H

L = H2O or OH

HH

H

H

N

H

H

H

H

H

O

H

Cr3+

O

O O

O

O

O

H

HH

H

HH

H

H

N

H

H

H

H

H

O

H

H

OH

HH

N

H

H

HH

H

O

HH

2.238

2.364

HHH

N

O

H

HH

H

H

Cr

O

O O

OO

H

H

H

OO

H

H

H

H

(R) (TS)

H

Cr3+

O

O O

O

O

O

H

H

H

HH

H

H

N

H

H

H

H

H

O

H

HH

N

H

H

HH

H

O

HH

2.330

2.392

H

H

O

H H

O

N

H

H

H H

H

H

Parent system

Conjugate base system

The interchange pathway for aquation of both [Cr(+NH3CH2COO

‒)(H2O)5]

3+ and

[Cr(+NH3CH2COO

‒)(H2O)4(OH)]

2+ is initiated by outer sphere coordination of a

water molecule to the Cr(III)-glycinate species as shown in Scheme 6.1 (Figures S14

to S20 of the SI). For both the neutral and conjugate base complexes, the incoming

H2O forms a hydrogen bond with an inner sphere water molecule, to give the

stabilized precursor/reactant complex (R). The precursor structures (R) with outer-

sphere coordination of the attacking nucleophilic water molecule involve bifurcated

H-bonding to an inner sphere water molecule (H2O(inner)). Previously27−29

, we found

that the precursor haloaqua Cr(III) structures with bifurcated bridging hydrogen

bonds are energetically more stable than those with linear hydrogen bonds. IRC

analysis of the transition states also confirms the bifurcated H-bonded structures

(Scheme 6.1) as the precursors for aquation of both of [Cr(+NH3CH2COO

‒)(H2O)5]

3+

Chapter 6: *


and [Cr(+NH3CH2COO

‒)(H2O)4(OH)]

2+ along the interchange pathway (Figures S14

to S20). For these precursor species the bifurcated HO‒H…O distances for H-

bonding of Cr−H2Oinner with the outer sphere H2O molecule are 1.61 and 1.63 Å for

the parent system (Figure S14) and 1.45 and 1.71 Å for the conjugate base systems

(Figure S17 of the SI). These variations reflect differences in the electronegativity

and distortion of geometry of these species.


) for aquation via the interchange pathway of

(a) [Cr(+NH3CH2COO

‒)(H2O)5]3+

and (b) [Cr(+NH3CH2COO

‒)(H2O)4(OH)]2+

with

nH2O (n =1‒3) in solution phase obtained at PBE0/cc-pVDZ.

The energy profiles for aquation of [Cr(+NH3CH2COO

‒)(H2O)5]3+

and

[Cr(+NH3CH2COO

‒)(H2O)4(OH)]2+

are shown in Figures 8a‒8b and the enthalpies of

activation (ΔH‡), Gibbs energies of activation (ΔG

‡) and entropies of activation

(ΔS‡) are given in Table 6.1.

Chapter 6: *


Table 6.1: Enthalpies (ΔH‡, kJ mol

‒1), gibbs energies (ΔG

‡, kJ mol

‒1), and

entropies (ΔS‡, J mol

‒1K

‒1) of activation for interchange pathways of


−)]

3+ and [Cr(H2O)4(OH)(

+NH3CH2COO

−)]

2+ with

H2O in the solution phase at 298.15 K.a

[Cr(H2O)5(glyH)]3+

[Cr(H2O)4(OH)(glyH)]2+

nH2O ΔH‡ ΔG

‡ ΔS

‡ ΔH

‡ ΔG

‡ ΔS

‡

1 114 128 ‒46

97 109 ‒42

2 110 113 ‒11

91 95 ‒14

3 88 102 ‒47

76 86 ‒35

a Optimized structures defined in Figures S14 to S20 of the SI.

Figure 6.9: Optimized transition structures of [Cr(H2O)5(+NH3CH2COO

−)]

3+ and

[Cr(H2O)4(OH) (+NH3CH2COO

−)]

2+ for (a) H2O, (b) 2H2O, and (c) 3H2O via

interchange pathway and maps of electrostatic potential (0.02 electrons Bohr−3

) (red

= electron-rich, blue = electron-deficient).

The aquation barrier for the conjugate base system is lower due to changes in the

electronic environment and distortion of the complex geometry in the TSs (Table 6.1

and Figures 8a‒8b and 9). For example, the conjugate base system is less strained

Chapter 6: *


with a COO-…Cr…OH angle of 80.6 ° compared with a COO-…Cr…OH2 angle of

69.7 ° in the parent complex. Likewise, the Cr−O(inner) bond lengths are 2.01 and 1.82

Å for the parent and conjugate base systems, respectively. In comparison, the

transition Cr…OOC distances are very similar in both systems, (2.35 Å and 2.39 Å)

(Figures S14 to S20 of the SI). Likewise, the Cr(III)…OH2 distances in the neutral

and conjugate base are 2.36 Å and 2.39 Å, respectively.

In our previous study29

of cis-[Cr(H2O)4(OH)Cl]+ we found explicit outer-sphere

water molecules are required to stabilise the transition states. Therefore, we also

investigated the effects of inclusion additional outer sphere water molecules on the

aquation of [Cr(H2O)5(+NH3CH2COO

−)]

3+ and its conjugate base species, via the

interchange pathway (Figures S14‒S20 of the SI). In this work, three outer sphere

water molecules were also found to be necessary to lower ∆H‡ (Table 6.2 and

Figures S14‒S20 of the SI). Hydrogen bonding between the departing glycinate

ligand and the neighbouring water molecules assists the ligand transfer in these

systems (Figures 6.9). This is clearly demonstrated by the changes in activation

enthalpies for the interchange pathway of [Cr(H2O)5(+NH3CH2COO

−)]

3+…3H2O and

its conjugate base complexes (88 and 76 kJ mol1

, which are in good agreement

with experimental12

results (87 and 75 kJ mol1

). The hydrogen bonding network

around these complexes with three water molecules makes a relatively significant

contribution to ΔH‡, as shown by the decreases in the activation enthalpies (26 and

21 kJ mol1

, respectively), compared to the corresponding one water systems

(Figures S14‒S20 of the SI). The Cr…‒OOCH2

+NH3 distances in the TSs of


−)]

3+…3H2O complex and its conjugate base species are

slightly longer than in the single H2O systems (0.03 Å and 0.14 Å, respectively). The

oxygen atom of the incoming water molecule and glycinate ion of the leaving group

Chapter 6: *


exhibits relatively little change in charge across the one, two and three water systems

as shown in the selected maps of electrostatic potential (Figure 6.9) for the

interchange mechanism. We have determined the changes in (∆∑ d(Cr−L)TS‒R) are ‒

0.71 Å and ‒0.65 Å for [Cr(H2O)5(+NH3CH2COO

−)]

3+ and it conjugate base species

with H2O, which corresponds with a negative activation volume (ΔV‡) associated

with an Ia mechanism, and is consistent with experimental results.12

For comparison, we have also investigated the dissociative (D) mechanism,

which involves breaking the Cr(III)‒‒OOCH2

+NH3 bond with formation of a

pentaqau- or tetraaquahydroxo-coordinated intermediates (Scheme 6.2). The released

‒OOCH2

+NH3 ion is coordinated to the outer sphere of the intermediate complexes

(I) to give an ion-pair as shown in Figures S21 to S22 of the SI.

Scheme 6.2: Dissociative (D) mechanism for the reaction of [Cr(+NH3CH2COO

‒

)(H2O)5]3+

.

Cr

O

O O

OO

H

H

H

H

H

O

H

H

Cr

O

OO

O

O

O

H

H

H

H

H

Hslow

Cr

O

O O

OO

H

H

H

H

H

H

H

1st step

HH

O

H H H HH

NH

H

HH H

O

N

H

H H

H

H

O

2.692

R TS I

N

H

HH

H

HO

H HH

The enthalpy of activation, Gibbs energy of activation and entropy of activation

for Cr(III)-−OOCH2

+NH3 bond dissociation of [Cr(

+NH3CH2COO

‒)(H2O)5]

3+ are 138

kJ mol‒1

, 144 kJ mol‒1

and ‒22 J mol‒1

K‒1

, respectively (see Table ST3 and Figures

Chapter 6: *


S21−S22 of the SI), indicating that this pathway is energetically less favourable than

the interchange pathway. However, we have previously

28−29 shown an improved

model of the dissociation process is achieved when water is coordinated to the outer

sphere of the complex, which also lowers ΔH‡ for the dissociation step (Table ST3

of the SI (Figures S21−22 of the SI). The ΔH‡

values for the Cr‒OglyH dissociation

of the precursor complex with an explicit outer sphere water molecule is lower

(Figure S22) by 5 kJ mol‒1

compared to bond dissociation without the outer sphere

water molecule (Figure S21). Nevertheless, the activation energy for this pathway

remains substantially higher than for the interchange pathway discussed earlier. The

Cr…OglyH distances in the TS, without and with the outer-sphere water molecule

are quite similar (2.69 Å to 2.66 Å, respectively) (Figures SF21 to SF22 of the SI),

reflecting the relatively small contribution the additional H-bonding makes to this

process.

In our previous work30

, we found that the UV-Vis spectrum of fac-[Cr(gly)3] in

the solid state showed peaks at 510 nm (ν1) and 388 nm (ν2), whereas in DMSO the

two spin-allowed peaks were shifted to longer wavelengths (518 nm and 393 nm,

respectively). Bryan et al.19

assigned the spin allowed d-d transition at 503 nm to

4A2g →

4T2g (ν1) and 384 nm to

4A2g →

4T1g (ν2). As noted here, the aqueous

solution UV-Vis spectrum shows bands at 527 nm (44.2 M‒1

cm‒1

) and 395 nm (23.6

M‒1

cm‒1

) and the pH of the solution is 6.02. One explanation for this is that in acidic

aqueous solution, the tris complex undergoes Cr−N ring opening via an acid

catalysed aquation pathway with coordination of a water molecule, and that this

structure is in equilibrium with the closed structure. Protonation of the glycine

ligands may occur via intramolecular hydrogen transfer to form the conjugate base.

Chapter 6: *


Cr3+

O

NN

O

O

N

O

H

H

H

H

O

HH

H

H O

H

H

HH

fac-[Cr(gly)3]0

Cr

N

O

N

O

O

NO

H

HH

H

O

H

H

H

H

H

H

H

H

O

H

H

H

O

[Cr(H2O)(gly)2(O-glyH)]+

Scheme 6.3:

H3O+ H+

Cr

N

O

N

O

O

NO

H

HH

H

O

H

H

H

H

H

H

H

H

OH

H

O

[Cr(HO)(gly)2(O-glyH)]0

Cr

N

O

N

O

O

NO

H

HH

H

O

H

H

H

H

H

H

H

H

O

H

H

H

O

[Cr(H2O)(gly)2(O-glyH)]+

Scheme 6.4:

Cr

N

O

N

O

O

NO

H

HH

H

O

H

H

H

H

H

H

H

H

O

H

H

H

O

Cr

N

O

N

O

O

N

O

HH

H

H

O

H

H

H

H

H

H

H

H

O

H

H

H

O

TS I

Cr

N

O

N

O

O

N

O

HH

H

H

O

H

H

H

H

H

H

H

H

O

H

H

H

O

H2O

O

HH

P

Aquation of the [Cr(H2O)(gly)2(O−glyH) complex occurs via a two-step process.

Firstly, direct Cr…O-glyH bond breaking and formation of an intermediate as shown

in Scheme 6.3 (see Figures S23−S24). Secondly, this intermediate can easily add

H2O to form the corresponding product as shown in Scheme 6.4. The ∆H‡ value for

dissociation of the Cr‒gly bond for this pathway is 97 kJ mol1

(Scheme 6.4 and

Figure S23 of the SI), with the amines being cis and the COO‒ groups having trans

configuration. Furthermore, the overall activation enthalpy is 97 kJ mol1

, which is

Chapter 6: *


in good agreement with the experimental23

value (91.4 ± 2 kJ mol1

) (Table ST4 and

Figure S24).

The aquated complexes can also undergo reaction via replacement of the aqua

ligands. Therefore, we investigated Cr‒OH2 dissociation for cis and trans-

[Cr(H2O)4(gly)]2+

with up to three explicit outer sphere water molecules

([Cr(H2O)4(gly)]2+

…nH2O, n = 0 to 3) (Table 6.2 and Figures S25 to S33 of the SI).

Dissociation of the Cr−OH2 bond (63 kJ mol−1

) is a much lower energy process than

dissociation of the Cr-OglyH bond (138 kJ mol−1

). Overall, the addition of explicit

outer-sphere water molecules has a relatively small effect on the activation enthalpy

for this process. However, there is a noticeable difference in the activation

parameters for the two isomers. For example, the dissociation of the Cr‒OH2 bond

(TS) (Figure 10a−10b) for cis-[Cr(H2O)4(gly)]2+

…3H2O has ∆H‡ = 62 kJ mol

1,

whereas for the trans complex ∆H‡ = 52 kJ mol

1 (Table 6.2 and Figure 6.10). No

interchange pathway was identified for water-water exchange on this complex.


‒1), Gibbs energies (ΔG

‡, kJ mol

‒1), and

entropies (ΔS‡, J mol

‒1K

‒1) of activation for dissociation pathway of

[Cr(H2O)4(gly)]2+

with nH2O in the solution phase at 298.15 K.a

cis-[Cr(H2O)4(gly)])]2+

trans-[Cr(H2O)4(gly)])]2+

nH2O ΔH‡ ΔG

‡ ΔS

‡ ΔH

‡ ΔG

‡ ΔS

‡

0 63 67 ‒16 52 60 ‒26

1 61 63 4 56 61 ‒16

2 66 71 ‒15 59 60 ‒3

3 62 63 ‒3 52 55 ‒8

a Optimized structures defined in Figures S25−S33 of the SI.

The structures for the corresponding dissociation of [Cr(OH)(H2O)3(gly)] +

are

shown in Figures S34 to S39 of the SI. The ΔH‡

values for the Cr‒OH2 dissociation

in the [Cr(OH)(H2O)3(gly)] +

…nH2O (n = 0‒2) complexes are lower than for the

Chapter 6: *


parent species by 7 to 11 kJ mol‒1

(Table ST5 of the SI). In both systems, the values

for Cr…OH2 dissociation from the trans position are 9 to 12 kJ mol‒1

lower than the

corresponding values for the cis-systems (Table ST5 and Figures S34‒S39).


) for the dissociation pathway of (a) cis and

(b) trans position dissociation of [Cr(H2O)4(gly)]2+

…nH2O (n =0‒3) in solution

phase obtained at PBE0/cc-pVDZ.

Chapter 6: *


Figure 6.11: Structures of cis (A‒B) and trans (C) position dissociation of

[Cr(H2O)(gly)2]+

…H2O and maps of electrostatic potential (0.02 electrons Bohr−3

)

(red = electron-rich,blue = electron-deficient).

The bis-glycinatochromium(III) complex can also undergo reaction via release of

an inner-sphere water molecule. The transition structures for the Cr(III)…OH2+ bond

dissociation step of cis-[Cr(+NH3CH2COO

‒)2(H2O)2]

+ (A and B) (Figures S40‒S42

of the SI), and trans-[Cr(+NH3CH2COO

‒)2(H2O)2]

+ (C) (Figures S43‒S45) are

shown in Figure 6.11. The enthalpies of activation for these different conformations

are 57, 62 and 80 kJ mol1

, respectively (Table ST6 of the SI). The oxygen atom of

the departing water molecule is more negatively charged in the TS of the trans-

system (−0.962) than in either of the cis-TS structures investigated (−0.920 and

Chapter 6: *


−0.926), which suggests a trans-effect in these complexes in the order H2O < NH2R

< −O2CR. (Figure 6.11). Furthermore, NBO

33 charge analysis of the TSs shows an

increase in the positive charge on Cr(III) compared to R as shown in Figure S46 of

the SI. The ΔH‡ values for the corresponding conjugate base systems are 54, 52 and

41 kJ mol1

(Table ST7), respectively, due to changes in the electronic environment

and the trans-effect of OH− in system C (see Figures S47−S49 of the SI).

Discussion

In this study, we have carried out a detailed investigation of the associative

interchange (Ia) and dissociation (D) mechanisms for the aquation of mono- and tris-

(glycinato) chromium(III) complexes. The activation enthalpies for aquation of

[Cr(H2O)5(NH3+CH2COO

−)]

3+ and its conjugate base were found to be 88 and 76 kJ

mol1

respectively, via the Ia pathway with inclusion of explicit outer sphere

solvation (3H2O), which is in good agreement with experimental12

results (87 and 75

kJ mol1

). Outer-sphere solvation makes a significant contribution to ΔH‡, as shown

by the decreases in the activation enthalpies (26 and 21 kJ mol1

, respectively),

compared to the corresponding one water systems. Significantly, higher enthalpies of

activation were obtained via the dissociative pathway for the mono-

glycinatochromium(III) complex (138 kJ mol‒1

and 144 kJ mol‒1

, respectively).

However, for the sterically crowded [Cr(gly)3] complex, the dissociative mechanism

was identified as the preferred aquation pathway.

A key feature of the UV-Vis experimental investigation of the mono-, bis- and

tris-glycinatochromium(III) bidentate complexes is the effect of pH. In the highly

acidic environment of the stomach (pH = 3−4), orally ingested nutritional

supplements will likely undergo slow aquation, leading to a mixture of mono-, bis-

Chapter 6: *


and tris-glycinatochromium(III) complexes. However, when these complexes enter

the duodenum they will encounter an environment with a much milder pH the

monoglycinatochromium(III) complex is expected to undergo rapid deprotonation to

form the corresponding conjugate base species.

Conclusions

We have prepared mono-, bis- and tris-glycinatochroium(III) species and

characterized them by thermogravimetry, UV-Vis (in the range pH of ~3.0 to ~8.5),

EPR (both solid and frozen states), ATR-FTIR and Raman spectroscopy. Vibrational

frequencies calculated at PBE0/cc-pVDZ are in good agreement with experimental

values. The TG-DSC analysis indicates a progressive structural change along the

decomposition reaction, with major endothermic transitions found at 370, 390, and

424.0 ⁰C30 for mono-, bis- and tris-glycinatochromium(III) species, respectively.

Geometries (bond lengths and bond angles) at PBE0/cc-pVDZ are in good

agreement with the X-ray crystal data of fac-[Cr(gly)3]19

. Aquation of the mono-

glycinatochromium(III) complex is found to follow the associative interchange (Ia)

pathway for the ring-opened isomer with ΔH‡ values in good agreement with the

experimental data12

, for both parent and conjugate base forms of the complex. In

comparison, tris-glycinatochromium(III) undergoes aquation via the dissociative (D)

pathways. These results may provide further insight into whether or not these

complexes can either undergo binding to proteins or can form oligomers in

biological systems.

Chapter 6: *



Full experimental details and characterization data for mono-, bis- and tris-

glycinatochromium(III) complexes by UV-Vis, electron paramagnetic resonance

(EPR), ATR-FTIR, Raman spectroscopy and TG-DSC curves are available in the

Supporting Information and Appendix C. Also included are energies of all pertinent

structures are reported in the accompanying supporting information.

Acknowledgements

We thank to Professor Bogdan Dlugogorski for providing access to the EPR and

TGA-DSC facilities. In addition, we are grateful to the Australian National

Computational Infrastructure (NCI) facility and the Atlantic Computational

Excellence Network (ACENET) for computer time.

References

(1) Mertz, W., Chromium occurrence and function in biological systems.

Physiol. Rev.;(United States) 1969, 49 (2).

(2) Mertz, W. Nutr. Rev. 1975, 33, 129‒135.

(3) Rollinson, C. L. In Problems of chromium reactions, Andrews Radioactive

Pharmaceuticals US Atomic Energy Commission Conference, 1966; p 429.

(4) Rollinson, C.; Rosenbloom, E.; Lindsay, J. Proc. VII Intnl. Congr. Nutr.

1967, 5.

(5) Schroeder, H. A. Am. J. Clin. Nutr. 1968, 21, 230‒244.

(6) Schwarz, K.; Mertz, W. Arch. Biochem. Biophys. 1959, 85, 292‒295.

(7) Vincent, J., The nutritional biochemistry of chromium (III). Elsevier: 2011.

(8) Vincent, J. B. Proc. Nutr. Soc. 2004, 63, 41‒47.


17, 632‒640.

(10) Tang, H.-y.; Xiao, Q.-g.; Xu, H.-b.; Zhang, Y. J. Trace Elem. Med Biol. 2015,

29, 136‒144.

(11) Khan, I. A. J. Inorg. Nucl. Chem. 1981, 43, 1082‒1085.

Chapter 6: *


(12) Abdullah, M.; Barrett, J.; O'Brien, P. J. Chem. Soc., Dalton Trans. 1984, 8,

1647‒1650.

(13) Ramasami, T.; Taylor, R.; Sykes, A. Inorg. Chem. 1976, 15, 2318‒2320.

(14) Guindy, N.; Abou-Gamra, Z.; Abdel-Messih, M. J. Chim. Phys. Phys.- Chim.

Biol. 1999, 96, 851‒864.

(15) Guindy, N. M.; Abou-Gamra, Z. M.; Abdel-Messih, M. F. Monatsh. Chem.

2000, 131, 857‒866.

(16) Banerjea, D.; Chaudhuri, S. D. J. Inorg. Nucl. Chem. 1968, 30, 871‒880.

(17) Khan, I. A. Int. J. Chem. Kinet. 1985, 17, 1263‒1272.

(18) Ramasami, T.; Wharton, R. K.; Sykes, A. G. Inorg. Chem. 1975, 14, 359‒365.


1971, 10, 1468‒1473.

(20) Ley, H.; Ficken, K. Ber. Dtsch. Chem. Ges. 1912, 45, 377‒382.

(21) Earnshaw, A.; Lewis, J. J. Chem. Soc. 1961, 396‒404.

(22) Veal, J. T.; Hatfield, W. E.; Jeter, D. Y.; Hempel, J. C.; Hodgson, D. J. Inorg.

Chem. 1973, 12, 342‒346.

(23) Kita, E.; Marai, H.; Muzioł, T.; Lenart, K. Transit. Metal. Chem. 2011, 36,

35‒44.

(24) Kita, E.; Lisiak, R. Transit Metal Chem 2011, 36, 855‒860.

(25) Kiersikowska, E.; Marai, H.; Mątewska, M.; Wrzeszcz, G.; Kita, E. Transit.

Metal. Chem. 2014, 39, 361‒371.

(26) Kita, E.; Kiersikowska, E.; Marai, H.; Mierkiewicz, K. Transit. Metal. Chem.

2014, 39 , 63‒70.



88‒97.

(29) Uddin, K. M.; Henry, D. J. ChemistrySelect 2016, 1, 5236‒5249.

(30) Uddin, K. M.; Habib, A. M.; Henry, D. J. ChemistrySelect 2017, 2,

1950−1958.


2013.

(32) Miertuš, S.; Scrocco, E.; Tomasi, J. Chem. Phys. 1981, 55, 117‒129.

(33) Tomasi, J.; Mennucci, B.; Cammi, R. Chem. Rev. 2005, 105, 2999‒3094.

(34) Reed, A. E.; Curtiss, L. A.; Weinhold, F. Chem. Rev. 1988, 88, 899‒926.

(35) Emerson, K.; Graven, W. J. Inorg. Nucl. Chem. 1959, 11, 309‒313.

(36) Bjerrum, N. Z. Phys. Chem. 1907, 59, 336‒83.


Chapter 6: *


(38) Cooper, J. A.; Blackwell, L. F.; Buckley, P. D. Inorg. Chim. Acta 1984, 92,

23‒31.

(39) Shuttleworth, S.; Sykes, R. J. Am. Leather Chem. Assoc. 1959, 54, 259‒268.

(40) Freitas, A. R.; Silva, M.; Ramos, M. L.; Justino, L. L.; Fonseca, S. M.;

Barsan, M. M.; C. Brett, M.; Silva, M. R.; and Burrows, H. D. Dalton Trans.

2015, 44, 11491‒11503.

(41) Weckhuysen, B. M..; Schoonheydt, R. A., Mabbs, F. E.; Collison, D. J.

Chem. Soc., Faraday Trans. 1966, 92, 2431‒2436.

(42) Weckhuysen, B. M..; De Ridder, L. M.; Grobet,P. J.; Schoonheydt, R. A. J.

Phys. Chem. 1995, 99, 320‒326.

(43) Andriessen, W. T. M.; Groenewege, M. P. Inorg. Chem. 1976, 15, 621‒626.

(44) Sen, D. N.; Mizushima, S.; Curran, C.; Quagliano, J. V. J. Am. Chem. Soc.

1955, 77, 211.

(45) Mateescu, C.; Gabriel, C.; Raptopoulou, C. P.; Terzis, A.; Tangoulis, V.;

Salifoglou, A. Polyhedron 2013, 52, 598.

(46) Nakamoto, K. Infrared and raman spectra of inorganic and coordination

compounds, 5th

ed., J. Wiley, New York, 1997.

(47) Condrate, R. A.; Nakamoto, K. J. Chem. Phys. 1965, 42, 2590‒2598.

(48) Herlinger, A. W.; Long, T. V. J. Am. Chem. Soc. 1970, 92, 6481‒6486.

(49) Herlinger, A. W.; Wenhold, S. L.; Long, T. V. J. Am. Chem. Soc. 1970, 92,

6474‒6481.

Chapter 7: *

Uddin, K. M.; Henry, D. J. ACS Journal Inorg. Chem. (revised). 200

Chapter 7

Electron Paramagnetic Resonance Spectroscopy and New

Insights into the Substitution Reaction of Chromium(III)

Picolinate*

Abstract

Chromium(III) picolinate, [Cr(pic)3] was characterised by solid EPR at both room

and frozen temperature. Broad EPR signals for the mer-[Cr(pic)3] complex were

found for the solid with g values of 3.45 and 3.47 at room temperature and frozen at

100K, respectively. This result indicates that the mer-[Cr(pic)3] system has rhombic

character and a large zero field splitting D value. In addition, its purity was

confirmed by elementary analysis, atomic absorption spectroscopy (AAS), HPLC,

UV-Vis, ATR-FTIR, Raman spectroscopy and ESI-mass spectrometry. DFT

calculations reveal that the most likely pathway for the acid-catalyzed [Cr(pic)3]

aquation is via two-step mechanism, which is in excellent agreement with the

experimental result. Furthermore, the importance of outer sphere solvation was

investigated with the addition of up to three explicit water molecules in the second

coordination sphere. The ΔH‡

for water-water exchange on the D pathway of

Chapter 7: *


[Cr(pic)(H2O)4]2+

…nH2O complexes was found to decrease for n = 0‒3 (64, 58, 56,

and 53 kJ mol1

, respectively).

Introduction

The biological activity of tris(picolinate)chromium(III) [Cr(pic)3] (pic =

picolinato (−)) has been extensively studied for use in nutritional supplements.1−9

Furthermore, preclinical and clinical studies suggest that Cr(III) provides benefits to

both humans and animals including fat reduction, decreased serum glucose, lowering

blood lipids, reducing cholesterol and triglyceride levels in the blood and lowering

glycated haemoglobin (HbA1c) of subjects with type 2 diabetes.10−21

It has been

reported that, for a trial of 180 patients with type 2 diabetes, a daily treatment of 200

µg/tablet of [Cr(pic)3] had a significant effect on blood sugar control and 1000

µg/tablet for a period four months in humans showed no toxic reactions or abnormal

liver or renal function in any of the patients.16,18,22−30

Another study found that, for a

28-year-old female, 600 µg/day of Cr(III) (in the form of chromium picolinate)

treatment for type 1 diabetes reduced HbA1C from 11.3% to 7.9% over a three

month period.22

Nevertheless, there is significant controversy surrounding the

transport of Cr(III) in humans, through the stomach and intestinal fluids, blood

plasma, or intact cells and some reports suggest chromium(III) at higher dosages can

cause DNA fragmentation (clastogenic effect).21,23−36

Furthermore, there is

conflicting evidence, involving ligand substitution and the justification of the

proposed Cr(III) nutritional values, leading to questions about its reactivity,

absorptivity and toxicity in physiological environments.

For example, there is some evidence indicating genotoxic, mutagenic37

and

carcinogenic effects of [Cr(pic)3] at high concentrations. Biological oxidation of

Chapter 7: *


[Cr(pic)3] supplements would lead to the formation of genotoxic Cr(V/IV)−pic

intermediates,38−39

or carcinogenic Cr(VI).30

For example, it has been shown that the

oxidation reaction between [Cr(pic)3] and H2O2 in neutral aqueous media at 37 °C

results in the slow formation of Cr(VI).40

Trivalent [Cr(pic)3] complex has additional

reactive forms: [Cr(pic)2(H2O)2]+ and [Cr(pic)2OH]2, which may form in the high

acidity environment of the stomach or via enzyme-assisted metabolism in the

liver.25,41

mer-[Cr(pic)3]0 (I) fac-[Cr(pic)3]0 (II)

[Cr(pic)3]0 (III) [Cr(pic)3]0 (IV)

Cr3+

OO

OC

O

NC

ON

C

O

NCr3+

O

O

CO

N

CO

N

O CO

N

Cr3+O

O O

O

CO

N

NC ON

Cr3+

O

O

C

O

N

C

O

NO

ON

Cr3+

O

OC

O

N

O

ON

NC O

[Cr(pic)3]0 (V)

Figure 7.1: Structures for conformers of [Cr(pic)3].

The red meridional isomer of [Cr(pic)3] and the purple binuclear complex

[Cr(pic)2OH]2 have been identified from X-ray diffraction studies by Stearns et al.,3

along with the previously assigned facial isomer of the red [Cr(pic)3] complexes.42−44

In solution a range of [Cr(pic)3] isomers (I – V) could potentially exist as shown in

Figure 7.1 (see Figures SF1−SF2 of the Supporting Information (SI)). Picolinate(−)

Chapter 7: *


in the anionic form can coordinate to Cr(III) through the tertiary amine and the

carboxylate group to form meridional or facial isomers, Cr(III)-(pic)3 (mer-(I) or

fac-(II)). The carboxylate can also potentially act as a bidentate ligand to give

Cr(pic)3 (III), Cr(pic)3 (IV) or Cr(pic)3 (V)) (see Figure 7.1). Picolinic acid can also

exist in a number of different forms. The most stable form in acid solution is the

cation (picH+) (VI−VIII). Under neutral conditions picolinic acid (pic) will exist in

forms (IX−X) or the zwitterion form (XI) and at high pH the anion (pic‒) (XII) of the

picolinic acid will be important (see Figures 7.2 and SF3 of the SI):

NO

O HH

NO

O HH

(VII)

NO

O H

H

(VIII)

NO

OH

+H H

Picolinic acid (pic) (IX)

NO

OH

NO

O H

(X)

+H H

NO

O

Deprotonated picolinic acid (pic ) (XII)

Protonated picolinic acid (picH):

(VI)

(XI)

Figure 7.2: Structures for conformers of protonated, neutral, zwitterion and

deprotonation of picolinic acid.

These differing conditions provide opportunities for picolinic acid to adopt different

possible chelating formations (I−V).

The biochemical pathways of [Cr(pic)3] supplements is of significant interest for

understanding the stability and reactivity of chromium(III)-picolinato complexes in

Chapter 7: *


aqueous solution at high acid concentration (pH ~1.0). In this work, our results

indicate that the formation of intermediates (XIII−XIV) occurs through ring opening

(i.e. Cr−N bond breaks) and subsequently takes the following ionized forms in acid

solution (see Scheme 7.1).

CrO O

O

CO

N

CO

N

C

O

N

0

H+

CrO OC

O

N

+

mer-[Cr(pic)3]0 [Cr(pic)2(OpicH)]+ (XIII)

H+

C

N

O

H

H+

H+

2+

[Cr(pic)2(OpicH2)]2+ (XIV)

N

O

C OCr

O OC

O

N

C

N

OH

N

OC O

H

Scheme 7.1:

There are a number of experimental kinetic investigations46−58

of the reaction

between chromium(III) and picolinic acid in acidic and basic aqueous solutions,46−47

but no prior theoretical study of their intimate mechanisms. Kita and Szablowicz

carried46

performed kinetic and equilibrium studies of aquation in the range

0.1−1.0M HClO4 at ionic strength I = 1.0 M (NaClO4) and temperature = ~20 to ~45

°C, observing the formation of [Cr(pic)2(H2O)2]+. They proposed that the reversible

acid-catalysed aquation proceeds by the chelate ring Cr−N bond breaking, followed

by Cr−O bond dissociation occurring by slow reaction, and reported enthalpies of

activation (ΔH‡) and entropies of activation (ΔS

‡) of 68.6 ± 2.2 kJ mol

1 and −76.4

± 7.3 J mol1

K‒1

, respectively.46

From the corresponding trends of base hydrolysis

of mer-[Cr(pic)3] complex, Marai et al. found that the ΔH‡ = 81 ± 4 kJ mol

1 and ΔS

‡

= −38 J mol1

K‒1

.47

However, it has not been demonstrated whether or not the

aquation of [Cr(pic)3] occurs via either an interchange (I), an associative (A) or a

dissociative (D) pathway. The mechanism for the substitution of ligands depends on

the primary coordination sphere of the precursor (R), the transition state (TS), and

Chapter 7: *


whether the electrostrictive forces surrounding the ligands are smaller or larger in the

TS than that for the precursor (i.e. positive or negative volume of activation (ΔV‡).

However, interpretation of the volume of activation (ΔV‡) in terms of a pathway is a

difficult and controversial subject. Kirk et al.58

determined that the ΔV‡ for the

photoaquation of [Cr(bpy)3]3+

complexes in basic media occurs through either an

associative or dissociative pathway, with an ΔV‡ of ‒1.6 or +2.9 ml mol

−1. This

finding is consistent with water entering from pockets between the ligands via either

an associative or dissociative process involving one or both bonds to a bipyridal

ligand. Consequently, there is interest in understanding the photoaquation reaction

and intimate mechanisms of ligand exchange processes involving [Cr(bpy)3]3+

in

acid solution (pH ˂ 6 media).56

In our previous studies59−61

of chromium(III) chloride

and other dihalides we investigated the mechanism for aquation under acidic

conditions, as might be found in the stomach after oral ingestion and under alkaline

conditions where formation of conjugate-base systems of chromium(III) is

favourable.61

Several experimental studies have investigated3,6,41,42,62−65

the synthesis and

characterization of [Cr(pic)3]. In particular, a number of studies5,6,66−67

have reported

high resolution electron impact mass spectra (Cr(pic)3: calculated m/z 418.0131,

found 418.0151). Chakov et al.63

have reported the electronic spectrum of [Cr(pic)3]

in different solvents (i.e. H2O, 2M HCl, 2M HNO3 and DMSO) and compared them

to previous work.3,42,43,66−70

Parajόn-Costa et al.65

observed that both [Cr(pic)3] and

[Cr(pic)2(OH)]2 complexes undergo a one-electron reversible reduction process,

finding that the half wave potential (E1/2) for the binuclear [Cr(pic)2(OH)]2 complex

is more negative (E1/2 = −1.49 V) than the mononuclear [Cr(pic)3] complex (E1/2 =

−1.23 V). Vincent et al.63

investigated 1H and

2H NMR spectral bands for [Cr(pic)3].

Chapter 7: *


They observed paramagnetically broadened and shifted resonance signals at +18 and

+24 ppm for the 1H. For the

2H they observed five downfield signals between +15

and +30 ppm and four upfield signals with three between −25 and −50 ppm and

another broad signal at −90 ppm).62

However, to date no electron paramagnetic

resonance (EPR) spectroscopy studies have been reported. Therefore, in this work

the EPR of [Cr(pic)3] is reported for the first time as well as elemental analysis,

atomic absorption (AAS), HPLC, TGA, UV-Vis, ATR-IR and Raman

spectroscopies.

The major aim of this study is to provide new insights into the intimate

mechanisms of the aquation of the chromium picolinate ([Cr(pic)3]) nutritional

supplement. This will lead to the development of a better understanding of ligand

exchange of [Cr(pic)3] systems, which may play a primary role in the biological

action of Cr(III) metal-based drugs. In addition, [Cr(pic)3] was investigated in the

solid state and frozen using electron paramagnetic resonance to determine the

environment of the metal center and give insight into its structural arrangement for

[Cr(pic)3] nutritional supplement.

Experimental

Reagents

Chromium(III) nitrate nonahydrate, Cr(NO3)3.9H2O (Sigma-Aldrich, ≥99.99%),

2-Picolinic acid (Sigma-Aldrich, ≥99%), Trifluoroacetic acid, TFA (Sigma-Aldrich,

HPLC, ≥99%) and acetonitrile (Sigma-Aldrich, HPLC Plus, ≥99.9%) were used

without further purification. Millipore filtered deionized water was used throughout

the experimental work.

Chapter 7: *


Analytical Methods

Elemental analysis (C, H, and N) was performed using the microanalytical unit, J

Science lab JM 11 analyzer. Determination of Cr was carried out using atomic

absorption spectroscopy (a Varian AA50 spectrometer with air/C2H2 flame

atomization) after digestion of the samples with 69% HNO3 (Merck). The liquid

samples were analysed with high-performance liquid chromatography (HPLC) on an

Agilent 1220 Infinity LC HPLC system equipped with a Apollo C18 (150 mm × 4.6

mm, 5 µm) column. The amounts of sample (30−50 mg) and mobile phase were

added into a 50 mL volumetric flask and filtered with 0.45 µm filter.

Chromatographic conditions were as follows: the ambient temperature was 30°C and

an isocratic mobile phase consisting of 0.05% trifluoroacetic acid in acetonitrile and

water (60:40, v/v) also filtered with 0.45 µm filter, flow rate of 0.5 mL/min for 8

min, injection volume = 20 µL and the UV detector wavelength = 264 nm. Thermal

analysis was conducted using a PerkinElmer STA 8000 TGA-DSC instrument.

Before the heating routine program was activated, the entire system was purged with

inert gas for 10 min at a rate of 20 mL/min, to ensure that the desired environment

was established. Thermogravimetric analysis (TGA) experiments were performed

with a heating rate of 20 °C min−1

from 30 to 1000 °C. High-purity argon gas was

used at a constant flow rate of 20 mL/min.

Mass spectra were acquired on a Waters Xevo G2 Q-TOF spectrometer equipped

with an electrospray ionization (ESI) probe, with positive reflector mode acquisition.

Solid-state ultraviolet-visible (UV-vis) spectra were recorded on a UV-3100PC,

SHIMADZU, UV-VIS-NIR scanning spectrophotometer over the range 200−1450

nm. Liquid-state Ultraviolet-visible (UV-Vis) spectra were recorded on a HP 8453

UV-Vis scanning spectrophotometer over the range 190−1100 nm. A Bruker EMX

Chapter 7: *


EPR spectrometer running the Xenon software with a Bruker ER 036TM NMR

teslameter was used to measure the X-band spectra of the solid Cr(III) complex with

EP parameters as follows: center field, 3200 G; sweep width, 6000 G; width TM,

200 G; FrequencyMon, 9.79 GHz; microwave power, 2.0 mW; microwave

attenuation, 20.0 dB; conversion time, 4.0 ms; gain, 30dB; modulation amplitude,

4.0 G; modulation frequency, 100 kHz; resolution, 15000 and sweep time, 60 s.

Infrared analysis was carried out using a Perkin Elmer FT-IR with a universal

ATR sampling accessory. ATR spectra were generated using 4 scans with a

resolution of 4 cm‒1

in a range of 4000 ‒ 350 cm‒1

. A constant pressure between the

ATR foot, sample and ATR crystal was achieved using an in-built pressure gauge

and software monitor. Raman analysis was conducted using a Nicolet 6700 FT-IR

with NXR FT-Raman module. Raman spectra were collected with: 1064 nm

excitation wavelength, 90o detection angle, CaF2 beam splitter, InGaAs detector,

gain of 1, optical velocity of 0.3165, aperture of 150, focus and side-to-side settings

optimised, laser power of 1.5 W and 256 scans with resolution of 8 cm‒1

in the range

4000 ‒ 200 cm‒1

.

Preparation of [Cr(pic)3]0

Chromium picolinate [Cr(pic)3] was prepared according to the method of Stearns

et al.3 A quantity of Cr(NO3)3.9H2O (6.40 g, 16 mM) was placed in a three-neck

flask and dissolved in 8 mL of 10.14 mM 70% HNO3. picolinic acid (6.77 g, 55

mM) was added slowly to the reaction mixture with continuous stirring and refluxed

at 60−65 °C for 30−40 min until the solution turned red-purple. The red solid

product was collected by vacuum filtration, recrystallized in CH3CN and dried in a

desiccator. The product was obtained in yield of the reaction was 45%. Anal. Calcd

Chapter 7: *


for C18H12N3O6 Cr: C, 51.67; H, 2.89; N, 10.05; Cr, 12.44. Found: C, 51.41; H, 2.53;

N, 9.98; Cr, 12.23. ESI-MS (CH3CN): MH+ m/z: calc. 419.0210; found: 419.0310.


Standard hybrid density functional theory calculations were carried out with

Gaussian 09.71

The geometries of all complexes in this study were fully optimized in

the gas phase and solvent (water) at the PBE0/cc-pVDZ level of theory, which we

have previously shown to give excellent agreement with experiment for the

activation enthalpies of related species.59−61

The effect of solvent (water) on the

structures and energetics of each complex was investigated using the polarisable

continuum model (PCM). Vibrational frequencies were obtained for all optimized

structures to check for the absence of imaginary frequencies for reactants,

intermediates and products and for the presence of a single imaginary frequency for

each transition state. For all the reaction pathways discussed in this study, the

transition states were also analyzed using the intrinsic reaction coordinate (IRC)

method. The final structures obtained from each IRC were further optimized in order

to positively identify the reactant and product complexes to which each transition

state is connected.

Natural bonding orbital (NBO)72

analysis was performed and the charges were

calculated. The HOMO and LUMO energies were determined using minimized

singlet geometries to approximate the ground state. Enthalpies of activation and

entropies of activation were calculated at 298.15 K. In the figures, all distances are in

angstroms (Å), angles in degrees (º) and the energies in kJ mol‒1

. Unless otherwise

stated, all values given in the text were obtained at the PBE0/cc-pVDZ level in

solution (PCM). The optimized structures and the relative energies of reactants,

Chapter 7: *


intermediates, transition states, and products for all pathways are shown in Table S1

and Figures SF1 to SF17 of the Supporting Information (SI). The volume of

activation (∆V‡) and reaction volume (∆V) were correlated to the change in the Cr−L

bond lengths of the TS, reactant and product as reported previously.59−61

Figure 7.3: (a) Optimized structures of the conformers for [Cr(pic)3] at PBE0/cc-

pVDZ; (b) maps of electrostatic potential (0.02 electrons Bohr−3

) (red = electron-

rich, blue = electron-deficient).

The [Cr(pic)3] complex can exist as meridional and facial isomers in two

diastereomeric configurations. The meridional isomer of mer-[Cr(pic)3] complex

(Figures 7.1 and 7.3, (I)) is more energetically stable than the facial isomer (Figures

7.1 and 7.3, (II)) by 8.4 kJ mol1

at PBE0/cc-pVDZ. A range of different precursor

complex conformations (Figures 7.1 and SF1, (III−V)) including [Cr(pic)3] (III),

[Cr(pic)3] (IV), and [Cr(pic)3] (V) were investigated and found to also be less

Chapter 7: *


energetically stable than the structure with meridional conformation by 46.9, 47.6

and 51.5 kJ mol1

.

The acidic, neutral and basic forms of picolinic acid in aqueous solution were

further investigated (Figure SF2 of the SI). The most stable form in acidic solution is

VI, which was found to be 4.8 and 18.3 kJ mol1

more stable than VII and VIII,

respectively. The neutral structure may be in the forms shown in IX or X and as the

zwitterion (XI). However, form IX is more stable than X and XI by 19.4 and 34.7 kJ

mol1

, respectively.

Table 7.1: Selected experimental and computed bond distances (in Å)

for mer-[Cr(pic)3]0 and fac-[Cr(pic)3]

0 complexes.

a

Bond Distances (Å) mer‒ fac‒ Exptlb

Cr‒O1 1.945 1.946 1.957

Cr‒O2 1.948 1.945 1.949

Cr‒O3 1.956 1.946 1.950

Cr‒N1 2.073 2.085 2.047

Cr‒N2 2.059 2.085 2.053

Cr‒N3 2.075 2.086 2.058

a Optimized structures of meridional and facial

complexes at PBE1PBE/cc-pVDZ

using the PCM solvation model defined in Figure 7.3. b

Ref. 3.

The calculated average values for the Cr−O and Cr−N bond lengths are 1.950

(1.946) and 2.069 (2.085) Å for mer-(fac-)[Cr(pic)3]0, respectively, and therefore in

good agreement (Table 7.1) with the experimental X-ray structure.3

Several conformations of the open-type complex, [Cr(pic)2(picH)]+ were

investigated and (XV) is energetically more stable than that of XVI by 68.4 kJ mol1

(Figure 7.4). However, the lowest energy conformation of precursor structure

[Cr(pic)2(picH2)(H2O)]2+

is (XVII), which is energetically more stable than the

isomers, XVIII and XIX by 16.1 and 22.2 kJ mol1

(Figure SF3 in the SI).

Chapter 7: *


Figure 7.4: Optimized structures for isomers of [Cr(pic)2(picH)]+.

For the HOMO (t2g*) and LUMO (eg

*) surfaces for mer- and fac-[Cr(pic)3]

isomers, it was found that the HOMO (−7.62 and −7.71 eV) consists of a mixture of

picolinato(−) and Cr(III) metal orbitals while the LUMO (−1.99 and −2.02 eV) is

spread over all three picolinato(−) ligands as shown in Figure 7.5 (Figure SF4 of the

SI). The HOMO−LUMO gap in the fac isomer is higher than in the mer-[Cr(pic)3]

by 0.1 eV.

Figure 7.5: Frontier orbitals for mer-[Cr(pic)3]0 at PBE0/cc-pVDZ (0.02 electrons

Bohr−3

surface).

Chapter 7: *


Atomic charges were determined using natural bond orbital (NBOs) analysis (see

Figure 7.3a). The positive charge on Cr is similar in mer-[Cr(pic)3]0 (+1.029) to Cr in

fac-[Cr(pic)3]0 (+1.026) (Figure 7.3a), which is consistent with the charge in fac-

[Cr(gly)3]0 (+1.020).

73 However, the nitrogen and oxygen atoms of the complex

exhibit relatively different degrees of electronegativity, with oxygen being more

electronegative than nitrogen. Therefore, oxygen has shorter bond lengths due to

increased electron density, resulting in Cr(III) metal forming a stronger bond

between Cr(III)−O compared to Cr(III)−N (see Figure 7.3).


Identification of [Cr(pic)3]0

Chromium(III) tris(picolinate) (mer-[Cr(pic)3]) was synthesized using the

procedure of Stearns et al.3 Elemental analysis (% C, H, N, and Cr) and electrospray

mass spectrometry (ESI-MS) data confirm the formation of the mer-[Cr(pic)3]

complex. In particular, the ESI mass spectrum of mer-[Cr(pic)3] complex (Figure

7.6) shows a peak at m/z = 419.0310 (calc. m/z = 419.0210), corresponding to

[Cr(C18H12N3O6) + H+] for the Cr-52 isotope. The peak at 295.9970 corresponds to

[Cr(C12H8N2O4 + H+)]. Additional peaks at m/z = 186.2248 and m/z = 142.1631,

correspond to [C6H5NO2.H2O + CO2 +H+] and a picolinic acid with hydrate fragment

[C6H5NO2.H2O +H+], respectively (see Figure 7.6b−7.6c). The purity of [Cr(pic)3],

was further confirmed by a single peak at 2.8 min (Figure 7.6a) using high

performance liquid chromatography (HPLC).

Chapter 7: *


Figure 7.6: (a) Analysis of HPLC chromatograms and ESI mass spectrum of mer-

[Cr(pic)3] in CH3CN for (b) simulation and (c) experimental.

The stability of the mer-[Cr(pic)3] complex was also analysed using

thermogravimetry (TG) under argon. The differential scanning calorimetry (DSC)

curves (Figure SF4 of the SI) show five endothermic stages of mass loss in the

temperature ranges (a) 85 °C to 180 °C, (b) 350 °C to 400 °C, (c) 400 °C to 460 °C,

(d) 475 °C to 556 °C, and (e) 831 °C to 978 °C. The first stage shows a gradual

weight loss around 140 °C indicating the release of lattice water. The second to

fourth mass loss stages correspond to the decomposition of the tris-, bis-, and mono-

picolinate ligands, respectively, with the largest loss occurring around 516 °C

(Figure S4). The fifth stage, occurring around 890 °C, is due to the thermal

decomposition of the complex to chromium oxides. Recently we reported73

that for

tris-(glycinato)chromium(III), the second stage shows the greatest fraction of weight

loss, with a corresponding sharp endothermic peak at around 424 °C.

Chapter 7: *


The experimental IR (Raman) spectra of the solid picolinic acid and mer-

[Cr(pic)3]0, as well as mer-[Cr(pic)3]

0 and fac-[Cr(pic)3]

0, complexes in gas phase

showed are shown in Figures 7.7 and SF6‒SF9 of the SI. We have also calculated

Figure 7.7: ATR-IR (a−b) and Raman (c−d) spectra (1800‒1300 cm‒1

) for picolinic

acid and [Cr(pic)3].

the vibrational spectra for mer-[Cr(pic)3]0

and fac-[Cr(pic)3]0 complexes in gas

phase. These spectra show picolinic acid coordinates with chromium(III) via the

carboxylic and pyridine ring of the picolinic acid ligand. Both antisymmetric and

symmetric vibrations of the carboxylate group were observed for all of the

complexes coordinated with picolinic acid ligands. In the spectrum of [Cr(pic)3]0

the antisymmetric COO‒ vibrations for the C=O stretching band appeared in the

range 1678‒1667 (1698‒1664) cm‒1

, whereas the symmetric COO‒ vibration for the

C‒O stretching band appeared in the range 1352‒1321 (1356‒1329) cm‒1

(Figure

7.7). Previously, Dorsett et al.74

found the νas(COO−) mode located at 1675 cm

‒1.

Chapter 7: *


We note the difference between the symmetric and asymmetric stretches was greater

than 300 cm‒1

, indicating that resonance occurs in the carboxylate group of the

picolinic ligand coordinated to the [Cr(pic)3]0

complex. Therefore, the oxygen-metal

bond in this system involves electrostatic character. Previously, Parajón-Costa et

al.65

reported that the IR (Raman) spectra of solid Cr(pic)3 showed a shift of the

carboxyl group (∆ν = νas − νs) of ~300 cm‒1

, as well as observing a band shift of 219

cm‒1

in solid Cr(me)3.75

There is also good agreement between the experimental and

theoretical data (Figures SF8‒SF9 of the SI).

Figure 7.8: UV-Vis spectrum of mer-[Cr(pic)3]0 in CH3CN (3.3 mM); in DMF (10.7

mM) and (b) DMSO (9.2 mM) solutions.

The UV/Vis spectrum of the mer-[Cr(pic)3] complex obtained in aqueous

solutions (H2O, 2M HNO3, 2M HCl, and ratio of 2M HNO3:H2O) and different

polar aprotic solvents (CH3CN, DMF, and DMSO) are presented in Table S1 and

Figures 7.8 and SF10, and compared to those previously measured.3,42,63,75

Two main

peaks are assigned to the spin allowed d-d transitions. Firstly, the 500−600 nm

transition is assigned to 4A2g →

4T2g (ν1) and the second transition observed between

300−500 nm is assigned to 4A2g →

4T1g (ν2) (see Table S1 of the SI). The ligand field

Chapter 7: *


splitting 10 Dq (or ∆o = ν1)73,77−80

obtained from the spectrum in CH3CN (519 nm),

DMF (518 nm), and DMSO (519 nm) are 19.27 × 103 cm

-1, 19.31 × 10

3 cm

−1 and

19.27 × 103 cm

−1, respectively, corresponding to an energy of ~2.39 eV, assigned to

the d (t2g) orbitals, centered in the Cr(III) metal ion, to π orbitals in the picolinato

ligand. The ligand field splitting value in polar aprotic solvents are consistent with

an (CO2−)3(pyridine-N)3 environment around Cr(III). Yuen et al.

42 reported, that in

DMF, the main band shift occurs at 515 nm (∆N = 21.7 × 103 cm

‒1). The spin

forbidden transitions were not observed in our experiment.

The [Cr(pic)3] is paramagnetic and the solid at room temperature has a magnetic

moment of 3.88 µB by Stearns and Armstrong3 and 3.84 µB by Dorsett and Walton,

74

as expected for a d3 (or magnetically dilute t2g

3 (S = 3/2)) system. In order to obtain

some insight into the coordination geometry of mer-[Cr(pic)3] in different

environments we have performed EPR on the solid state powder sample at room

temperature and frozen in liquid nitrogen at 100K. Chromium(III) complexes are

paramagnetic and can be identified by two different X-band EPR signals (g = 2 β‒

signal is a broad isotropic signal and g = 3.5−5.5 δ‒signal is at higher fields as a

positive lobe in the region) from an electronic spin S = 3/2, with moderately large

zero field splitting D and small rhombic splitting E (D > hν, E/D ≈ 0).81−83

Previously, Andriessen and Groenewege reported84

the EPR spectra for a series of

cis complexes of the type cis-[Cr(bpy)2XY]Z, cis- [Cr(phen)2XY]Z, and cis-

[Cr(ox)2XY]Z, where XY is H2O−H2O, H2O−HO−, Cl

−−Cl

−, bpy, phen, ox, en, or

acac. These results indicated very low zero-field D values, with the three series

exhibiting correlations with the ligand field strengths of XY ligands: phen > bpy (en)

> acac > ox > Cl− > H2O > OH

−, and they observed rhombic distortion differences of

0.05 to 0.33 D.84

Similarly, Jacquamet et al.86

recently reported the EPR spectrum of

Chapter 7: *


a mixture of closely related complexes Cr(acac)3 (2.7 mM ), which showed

significant broadening of signals with g = 3.8. This is consistent with a S = 3/2 center

that exhibits splitting correlations with the ligand field strengths.

Figure 7.9: Analysis of X-band solid state powder simple for EPR spectra of mer-

[Cr(pic)3] complex in solid state (a) room temperature and (b) frozen state (100 K).

In this study, the broad solid state powder EPR signal of the mer-[Cr(pic)3]

complex has a g value of 3.45 at room temperature at 298K and 3.47 in frozen state

at 100K, respectively (Figure 7.9a−7.9b). This result indicates that the mer-[Cr(pic)3]

system has rhombic character and a large D value. This signal corresponds to the

(CO2)3(pyridine −N), coordination geometry identified in the molecular structure

determined from single-crystal XRD by Stearns et al.3 These results also suggest that

there is no change in coordination environment between solid and solution. The EPR

spectrum of 1 mM Cr(acac)3 at 77 K shows85

a signal of g ≈ 4.3, which arises from

Chapter 7: *


isolated Cr(III) species in a distorted geometry. Similarly, Jacquamet et al.86

recently

reported the EPR spectrum of a mixture of closely related complexes Cr(acac)3 (2.7

mM ), which showed significant broadening of signals with g = 3.8.

Mechanistic study

Cr(III) nutritional supplements for mammals and human life are generally in the

form of stable tris-chelate [Cr(pic)3] complex.3,87

When in acidic media such as the

stomach or by enzyme-assisted metabolism in the liver,41

[Cr(pic)3] can easily get

into the blood stream under physiological environment conditions. Nevertheless, it is

likely that the physiological activity is linked to acid-catalyzed aquation involving

exchange of one or more picolinic acid. Therefore we have investigated the

mechanisms for aquation of [Cr(pic)3] complexes. The results provide a detailed

understanding of the mechanism for the hydrolysis of Cr(III) supplements, which

could be useful in understanding the speciation, prior to binding to biomolecules in

physiological environments.

Scheme 7.2: Aquation of [Cr(pic)2(OpicH)]+

with acidic aqueous solution

(H3O+).

[Cr(pic)2(2H2O)...picH2]2+ (P)

[Cr(pic)2(OpicH)]+...H3O+ (R)

H3O

2+

Cr

O

O

C

O

N

C

N

OH

N

OC O

H

O H

H

2+

Cr

O

O

C

O

N

C

N

OH

N

OC O

H

OH

H

[Cr(pic)2(H2O)(OpicH2)]2+ (I1)TS1

Cr

OO

C

O

N

CNO

N

O

C O

H

OH

H

Cr

O

O

C

O

N

C

N

OH

N

OC O

H

O H

HCrO

O

C

O

N

C

N

OH

N

O

C O

O

H

H

H

TS2

H2O

CrO

O

C

O

N

C

N

OH

N

OC O

O

H

H

HO

HH

[Cr(pic)2(H2O)...(picH2)]2+ (I2)

2+2+2+

H

2+

Chapter 7: *


For the acid catalyzed reaction path, initially [Cr(pic)3] proceeds via a

dissociative mechasnism with Cr−N picolinato(−) chelate-ring opening and

protonation of the ligand to give a precursor reactant as shown in Scheme 7.1

(structure (XIII)). Aquation proceeds via a stepwise reaction mechanism as shown

in Scheme 7.2 (Figure SF11 of the SI). The first step involves nucleophilic attack by

a water molecule on the pentacoordinate Cr(III), with the acid catalysed aquation

consisting of a rapid protonation equilibrium with the transition state (TS1), possibly

occurring via the tautomer’s path.

The stability of the associated reactant complex (R) involving outer sphere

coordination of the attacking water molecule and acid catalysed hydronium water

hydrogen bond with OpicH ligand (see Scheme 7.2) ([Cr( pic)2(OpicH)]

+…H3O

+)

were explored. The environment of the acidic proton is consistent with a Zundel

cation within the inner sphere (H+(H2O) (see Scheme 7.2 and Figure SF11 of the

SI). The activation enthalpy (∆H‡) for the interaction of [Cr(

pic)2(OpicH)]

+ with

acidic aqueous solution (H3O+) is 34 kJ mol

1 with PCM (see Figure 7.10). The

Cr−OpicH2 distance increases from 2.06 Å for the intermediate (I1) system through

to 3.23 Å for the TS2 system (Figure SF11). As noted earlier, the second rate-

determining step involves dissociation of the [Cr(pic)2(H2O)+…(picH2)]

+ and the

∆H‡ value (68 kJ mol

1) for this process is in excellent agreement with the

experimental value (68.6 ± 2.2 kJ mol1

).46

Therefore, we propose that the aquation

pathway of the [Cr(pic)2(OpicH)]+…H3O

+ complex undergoes an acid equilibrium

followed dissociative release of the monodentate picH2. Finally, this process can

easily add H2O to form the corresponding intermediate product (P) as shown in

Scheme 7.2 and Figure SF11 of the SI.

Chapter 7: *



) for aquation via stepwise (two-step)

pathways in solution phase obtained at PBE0/cc-pVDZ.

The changes in geometry throughout the process are accompanied by changes in

the charges on several key atoms. To explore the importance of atomic charges on

the aquation reaction, natural bonding orbital (NBO) analysis was performed and the

calculated charges are shown in (Figure SF12 of the SI). The positive charge on

Cr(III) and the magnitude of the negative charge on oxygen increase from

intermediate (I1) through to transition state TS2. However, the charges on

Cr(+1.078) for intermediate (I1) increase to Cr(+1.318) for TS2, whereas O(‒0.608)

for I1 shows a small increase (‒0.648) for TS2 in magnitude in these systems due to

the steric picolinato(−) ligand effect. A similar process is predicted for the second

picH2 dissociation and H2O addition to form [Cr(pic)(H2O)4]2+

complex.

To further explore the properties of this system, water exchange at the mono-

picolinatochromium(III) complex was investigated. Notably, for the closed-type

bidentate system, no interchange pathway was identified. Therefore, Cr‒OH2 bond

Chapter 7: *


dissociation with explicit outer sphere water molecules ([Cr(pic)(H2O)4]2+

…nH2O, n

= 0 to 3) (Figures S13 to S16 of the SI) was examined.


‒1), Gibbs Energies (ΔG

‡, kJ mol

‒1), and

Entropies (ΔS‡, J mol

‒1 K

‒1) of Activation for Cr-OH2 Dissociation Pathway of

trans-[Cr(pic)(H2O)4]2+

with nH2O in the Solution Phase at 298.15 K.a

trans-[Cr(pic)(H2O)4])]2+

nH2O ΔH‡ ΔG

‡ ΔS

‡

0 64 70 ‒31

1 58 66 ‒19

2 56 68 ‒33

3 53 66 ‒29

a Optimized structures defined in Figures SF13 to SF16 of the SI.

Outer sphere solvation is an important factor in the mechanism of this process.

This is demonstrated clearly by explicit solvation for the dissociation pathway of

trans-[Cr(pic)(H2O)4]2+

…nH2O complexes (n = 1‒3) (Table 7.2). Hydrogen bonding

between the departing H2O and the neighbouring water molecules plays a role,

assisting in the ligand transfer in these systems (Figures 7.11 and SF13 to SF16 of

the SI)). This is clearly demonstrated by the changes in overall activation enthalpies

for the D pathway of trans-[Cr(pic)(H2O)4]2+

…nH2O complexes (64, 58, 56, and 53

kJ mol1

, for n =0‒3, respectively). These results indicate that three water molecules

are required to stabilise the transition states for this process with additional explicit

solvation having minimal effect. The oxygen atom of the leaving H2O molecule

exhibits relatively little change in charge across the one, two, and three water

systems as shown in the Figures SF15−SF16 of the SI for the D mechanism.

Furthermore, the charges on chromium change slightly in magnitude (+1.321 to

+1.302) across one to three water systems (Figures SF15−SF16). However, the

solvation with three explicit water molecules appears to be sufficient to stabilize the

Chapter 7: *


transition state. Nevertheless, the position of these three water molecules is also

important to facilitate H-bonding between outer sphere water molecules and the

inner sphere through bridging bifurcated H-bonds. We note that when more than

three water molecules are included the additional water molecules have the weakest

interaction with the inner sphere and therefore there is a small difference (3 kJ mol1

)

in ΔH‡ between the two to three waters. Overall the lowest energy n = 3 for the most

stable TS conformer for the monodentate system has ΔH‡ = 53 kJ mol

1 (Table 7.2

and Figure 7.11).

Figure 7.11: Optimized transition structures of trans-[Cr(pic)(H2O)4]+…nH2O

(n =0‒3) via dissociation pathway in solution phase obtained at PBE0/cc-pVDZ.

Chapter 7: *


Discussion

In this work the solid [Cr(pic)3] nutritional supplement was investigated by EPR

(at both room temperature and frozen at 100 K), elementary analysis, AAS, ES-MS,

HPLC, ATR-FTIR, Raman, UV-Vis and ESI-mass spectrometry. Structural

parameters were calculated by DFT for neutral mer- and fac-[Cr(pic)3] isomers and

compared to the X-ray data for mer-[Cr(pic)3].3 The agreement between calculated

and experimental structural parameters is good. The EPR spectra obtained on solid at

both room temperature and frozen at 100 K showed no appreciable difference (g =

~3.46). This suggests that there is no change in coordination environment, and is

consistent with other six-coordinate octahedral Cr(III) complexes.

Mechanistic investigations were carried out, showing tris-(picolinato)chromium

(III) reacts slowly with acidic aqueous solution to form a bis-(picolinato)

chromium(III), and further aquation leads to mono-(picolinato)chromium(III)

species. A multi-step mechanism was identified beginning with Cr−N ring opening

and H2O addition. The second rate-determining step involves dissociation of the

[Cr(pic)2(H2O)…(picH2)]2+

with ∆H‡ = 68 kJ mol

1, in excellent agreement with the

experimental result (68.6 ± 2.2 kJ mol1

).46

The UV-Vis spectrum revealed two spin-

allowed d‒d transitions with a shift in the band at 514 nm to 526 nm, followed by a

shift to 533 nm.46

It has been reported that the Cr−N bond breaking is evident from

the peak at 266 nm,46

whereas we observed that free picolinic acid in aqueous

solution has a peak at ~266 nm (XI) by UV-Vis and HPLC (see Figure SF17 of the

SI). Therefore, acid-catalyzed aquation proceeds from the open [Cr( pic)2(OpicH)]

2+

complex, with water entering in acidic media (H3O+) to form a six coordinate

inetermediate (I1), followed by dissociation [Cr( pic)2(H2O)…(picH2)]

2+, which is in

Chapter 7: *


good agreement with volume of activation of −1.6 and +2.9 ml mol−1

for the

photoaquation of [Cr(bpy)3]3+

.58

The ∆H‡ value for dissociation of the Cr‒OH2 bond

(TS1) for trans-[Cr(pic)(H2O)4]2+…3H2O complex with three waters is 53 kJ mol

1.

These results indicate that three water molecules are required to stabilise the

transition states for this process, with additional explicit solvation having a minimal

effect on the Cr(III) metal complex. However, in the highly acidic environment of

the stomach, [Cr(pic)3] is only slightly soluble in water,6 so the complex will be

mostly passed through to the duodenum which has a milder pH. The complex may

eventually enter the blood stream, which results in the increased insulin enhancing

treatment in type 2 diabetics. These results could be useful to the development of

these treatments in humans and animals, considering that chromium(III) can form

biologically active and stable species of [Cr(pic)3] under physiological pH.

Conclusions

The mer-[Cr(pic)3] complex was detected with EPR, resulting in octahedral

Cr(III) resonance signals being proposed and discussed in this work. This complex

was confirmed by thermogravimetry, elemental analysis, AAS, HPLC, UV-Vis,

ATR-FTIR, Raman and ESI-mass spectrometry spectroscopy. In addition, the TG-

DSC analysis indicated the occurrence of progressive structural changes during the

decomposition reaction. Peak temperatures (384, 431, 516, and 890 ⁰C) of the

endothermic process in the non-isothermal DSC curves were obtained using a

heating rate of 20 ⁰C/min. Geometries (bond lengths) at PBE0/cc-pVDZ are in good

agreement, with minor differences in the ligand configurations (i.e. mer-isomer and

fac-isomer), compared with the X-ray crystal data of mer-[Cr(pic)3].3

The computed

ΔH‡ values found for acid-catalyzed aquation of the open type tris system in PCM

Chapter 7: *


are in good agreement with the experimental data.46

Water exchange on

[Cr(pic)(H2O)4]2+

was investigated with explicit outer sphere H2O molecules. The

dissociative (D) mechanism with three water molecule has the lowest activation

enthalpy. Therefore, the mer-[Cr(pic)3] is likely to be the biologically active form of

Cr(III). However, [Cr(pic)2(H2O)]+ and [Cr(pic)(H2O)4]

2+ species should also be

investigated experimentally for treatment of both types 1 and 2 diabetes.


Full experimental details and characterization data for mer-[Cr(pic)3] complex by

UV-Vis, ATR-FTIR, Raman spectroscopy and TG-DSC curves are available in the

Supporting Information and Appendix D. Also included are energies of all pertinent

structures are reported in the accompanying supporting information.

Acknowledgements

I would like to thank Professor Bogdan Dlugogorski for providing access to the

EPR and TGA-DSC facilities. In addition, I am grateful to the Australian National

Computational Infrastructure (NCI) facility and the Atlantic Computational Excellence

Network (ACENET) for computer time.

References

(1) Wasser, W. G.; Feldman, N. S.; D'Agati, V. D. Ann. Intern. Med. 1997, 126,

410.

(2) Anderson, R. A. Regul. Toxicol. Pharmacol. 1997, 26, S35−S41.

(3) Stearns, D. M.; Armstrong, W. H. Inorg. Chem. 1992, 31, 5178−5184.

(4) Mertz, W. Science 1981, 213, 1332−1338.

(5) Press, R. I.; Geller, J.; Evans, G. W. West. J. Med. 1990, 152, 41.

(6) Evans, G.; Pouchnik, D. J. Inorg. Biochem., 1993, 49, 177−187.

(7) Vincent, J. The nutritional biochemistry of chromium (III), Elsevier, 2011.

Chapter 7: *



Toxicol. 1999, 12, 483−487.

(9) Hepburn, D. D.; Vincent, J. B. Chem. Res. Toxicol. 2002, 15, 93−100.

(10) Page, T.; Ward, T.; Southern, L. J. Anim. Sci. 1991, 69, 356.

(11) Page, T.; Southern, L.; Ward, T.; Thompson, D. J. Anim. Sci. 1993, 71,

656−662.

(12) Evans, G. W.; Bowman, T. D. J. Inorg. Biochem. 1992, 46, 243−250.

(13) Evans, G. W. West. J. Med. 1991, 155, 549.

(14) Saiyed, Z. M.; Lugo, J. P. Food Nutr. Res. 2016, 60.

(15) Tsave, O.; Yavropoulou, M.; Kafantari, M.; Gabriel, C.; Yovos, J.;

Salifoglou, A. J. Inorg. Biochem. 2016, 163, 323−331.

(16) Cefalu, W. T.; Hu, F. B. Diabetes Care, 2004, 27, 2741−2751.

(17) Anderson, R. A. Biol Trace Elem Res. 1992, 32, 19−24.


Chi, J.; Feng, J. Diabetes, 1997, 46, 1786−1791.

(19) Martin, J.; Wang, Z. Q.; Zhang, X. H.; Wachtel, D.; Volaufova, J.; Matthews,

D. E.; Cefalu, W. T. Diabetes Care, 2006, 29, 1826−1832.

(20) Patal, P. C.; Cardino, M. T.; Jimeno, C. A. Consultant, 2010, 48.

(21) Levina, A.; Lay, P. A. Dalton Trans. 2011, 40, 11675−11686.

(22) Fox, G. N.; Sabovic, Z. J. Fam. Pract. 1998, 46, 83−86.

(23) Coryell, V. H.; Stearns, D. M. Mutat. Res. 2006, 610, 114−123.

(24) Andersson, M. A.; Grawé, K. V. P.; Karlsson, O. M.; Abramsson-Zetterberg,

L. A.; Hellman, B. E. Food. Chem. Toxicol. 2007, 45, 1097−1106.

(25) Weeks, C. L.; Levina, A.; Dillon, C. T.; Turner, P.; Fenton, R. R.; Lay, P. A.

Inorg. Chem. 2004, 43, 7844−7856.

(26) Nguyen, A.; Mulyani, I.; Levina, A.; Lay, P. A. Inorg. Chem. 2008, 47,

4299−4309.

(27) Stout, M.; Nyska, A.; Collins, B.; Witt, K.; Kissling, G.; Malarkey, D.;

Hooth, M. Food. Chem. Toxicol. 2009, 47, 729−733.

(28) Cupo, D. Y.; Wetterhahn, K. E. Cancer Res. 1985, 45, 1146−1151.

(29) Kato, I.; Vogelman, J. H.; Dilman, V.; Karkoszka, J.; Frenkel, K.; Durr, N.

P.; Orentreich, N.; Toniolo, P. Eur. J. Epidemiol. 1998, 14, 621−626.

(30) Levina, A.; Codd, R.; Dillon, C. T.; Lay, P. A. Prog. Inorg. Chem. 2003, 51,

145−250.

(31) Wu, L. E.; Levina, A.; Harris, H. H.; Cai, Z.; Lai, B.; Vogt, S.; James, D. E.;

Lay, P. A. Angew. Chem. 2016, 128, 1774−1777.

(32) Stearns, D. M. Biofactors 2000, 11, 149−162.

(33) Vincent, J. B. J. Trace. Elem. Med. Biol. 2014, 28, 397−405.

(34) Vincent, J. B. Dalton Trans. 2010, 39, 3787−3794.

(35) Levina, A.; Lay, P. A. Chem. Res. Toxicol. 2008, 21, 563−571.

Chapter 7: *


(36) Bailey, C. H. Biol. Trace Elem. Res. 2014, 157, 1−8.

(37) Stearns, D. M.; Silveira, S. M.; Wolf, K. K.; Luke, A. M. Mutat. Res. 2002,

513, 135-142.

(38) Cawich, C. M.; Ibrahim, A.; Link, K. L.; Bumgartner, A.; Patro, M. D.;

Mahapatro, S. N.; Lay, P. A.; Levina, A.; Eaton, S. S.; Eaton, G. R. Inorg.

Chem. 2003, 42, 6458−6468.

(39) Codd, R.; Lay, P. A.; Levina, A. Inorg. Chem. 1997, 36, 5440−5448.

(40) Mulyani, I.; Levina, A.; Lay, P. A. Angew. Chem. Int. Ed., 2004, 43,

4504−4507.

(41) Kareus, S. A.; Kelley, C.; Walton, H. S.; Sinclair, P. R. J. Hazard. Mater.

2001, 84, 163−174.

(42) Yuen, G.; Heaster, H.; Hoggard, P. E. Inorg. Chim. Acta 1983, 73, 231−234.

(43) Dorsett, T. E.; Walton, R. A. J. Chem. Soc., Dalton Trans. 1976, 347−350.

(44) Oki, H.; Yoneda, H. Inorg. Chem. 1981, 20, 3875−3879.

(45) Stephen, W.; Basolo F; Pearson, R.G. Mechanisms of Inorganic Reactions: A

Study of Metal Complexes in Solution, 2nd Edn.: J. Wiley and Sons, Inc.,

New York, 1967, xi+ 701 pp., price 144 s. Elsevier: 1969.

(46) Kita, E.; Szabłowicz, M. Transition Met. Chem. 2003, 28, 698−706.

(47) Marai, H.; Kita, E.; Wiśniewska, J. Transition Met. Chem. 2012, 37, 55−62.

(48) Pazderska-Szabłowicz, M.; Karpiński, A.; Kita, E. Transition Met. Chem.

2006, 31, 1075−1080.

(49) Kita, E.; Lisiak, R. Transition Met. Chem. 2010, 35, 441−450.

(50) Abdel-Messih, M.; Abou-Gamra, Z. Monat. Chem. 2012, 143, 211−216.

(51) Szabłowicz, M.; Kita, E. Transition Met. Chem. 2005, 30, 623−629.

(52) Pazderska-Szabłowicz, M.; Białkowska, A.; Kita, E. Transition Met. Chem.

2006, 31, 413−420.

(53) Kiersikowska, E.; Kita, E.; Kita, P.; Wrzeszcz, G. Transition Met. Chem.

2016, 41, 435−445.

(54) Marai, H.; Kita, E.; Kiersikowska, E.; Kuchta, S.; Bajek, A.; Drewa T.

Transition Met. Chem. 2012, 37, 337−344.

(55) Maestri, M.; Bolletta, F.; Serpone, N.; Moggi, L.; Balzani, V. Inorg. Chem.

1976, 15, 2048−2051.

(56) Maestri, M.; Bolletta, F.; Moggi, L.; Balzani, V.; Henry, M.; Hoffman, M. J.

Am. Chem. Soc. 1978, 100, 2694−2701.

(57) Jamieson, M. A.; Serpone, N.; Maestri, M. Inorg. Chem. 1978, 17,

2432−2436.

(58) Kirk, A.; Porter, G. B.; Scandola, M. R. Inorg. Chim. Acta 1984, 90,

161−164.

(59) Uddin, K. M.; Ralph, D.; Henry, D. J. Comp. Theor. Chem. 2015, 1070,

152−161.

Chapter 7: *



88−97.

(61) Uddin, K. M.; Henry, D. J. ChemistrySelect 2016, 1, 5236−5249.

(62) Broadhurst, C. L.; Schmidt, W. F.; Reeves, J. B.; Polansky, M. M.; Gautschi,

K.; Anderson, R. A. J. Inorg. Biochem. 1997, 66, 119−130.

(63) Chakov, N. E.; Collins, R. A.; Vincent, J. B. Polyhedron 1999, 18,

2891−2897.

(64) Hakimi, M. J. Korean Chem. Soc. 2013, 57, 721−725.

(65) Parajón‐Costa, B. S.; Wagner, C. C.; Baran, E. J. Z. Anorg. Allg. Chem. 2003,

629, 1085−1090.

(66) Kim, S.-I.; Woo, D.-J.; Lee, M.-H.; Woo, G.-J.; Kang, D.-K.; Cha, K.-W.

Bull. Korean Chem. Soc. 2003, 24, 1517−1520.

(67) Koll, M.; Hoenen, H.; Aboul‐Enein, H. Y. Biomed. Chromatogr. 2005, 19,

119−122.

(68) Vinayakumar, C.; Dey, G.; Kishore, K.; Moorthy, P. Radiat. Phys. Chem.

1996, 48, 737−742.

(69) Kingry, K. F.; Royer, A. C.; Vincent, J. B. J. Inorg. Biochem. 1998, 72,

79−88.

(70) Gould, E. S.; Taube, H. J. Am. Chem. Soc. 1964, 86, 1318−1328.


2013.

(72) Reed, A. E.; Curtiss, L. A.; Weinhold, F. Chem. Rev. 1988, 88, 899−926.

(73) Uddin, K. M.; Habib, A. M.; Henry, D. J. ChemistrySelect 2017, 2,

1950−1958.

(74) Dorsett, T. E.; Walton, R. A. J. Organomet. Chem. 1976, 114, 127−134.


17, 632−640.

(76) Nakahara, M. Bull. Chem. Soc. Jpn. 1962, 35, 782−785.

(77) Pham, T. N.; Aitken, J. B.; Levina, A.; Lay, P. A. Inorg. Chem., 2014, 53,

10685−10694.


Keramidas, A. D.; Kabanos, T. A. Inorg. Chem. 2014, 53, 11404−11414.

(79) Lever, A. J. Chem. Educ. 1968, 45, 711.

(80) Dou, Y.-S. J. Chem. Educ. 1990, 67, 134.

(81) Weckhuysen, B. M.; Schoonheydt, R. A.; Mabbs, F. E.; Collison, D. J.

Chem. Soc., Faraday Trans. 1996, 92, 2431−2436.

(82) Bryliakov, K. P.; Lobanova, M. V.; Talsi, E. P. J. Chem. Soc., Dalton Trans.

2002, 2263-2265.

(83) F.; Mabbs, Collison, D.; Gatteschi, D. Magn. Reson. Chem. 1994, 32, 67−67.

(84) Andriessen, W.; Groenewege, M. Inorg. Chem. 1976, 15, 621−626.

Chapter 7: *


(85) Brückner, A.; Jabor, J. K.; McConnell, A. E.; Webb, P. B. Organometallics

2008, 27, 3849−3856.

(86) Weckhuysen, B. M.; De Ridder, L. M.; Grobet, P. J.; Schoonheydt, R. A. J.

Phys. Chem. 1995, 99, 320−326.

(87) Jacquamet, L.; Sun, Y.; Hatfield, J.; Gu, S. P.; Cramer, M. W.; Crowder, M.

W.; Lorigan, G. A.; Vincent, J. B.; Latour, J.-M. J. Am. Chem. Soc. 2003,

125, 774−780.

Chapter 8: Summary and Future Directions. 231

Chapter 8

Summary and Future Directions

8.1 Summary

A number of early scientific studies suggested that the chromium(III) ion is an

essential trace element in human nutrition. Subsequently a range of chromium(III)

containing nutritional supplements became commercially available. However, recent

studies have shown that Cr(III) is not an essential trace element in humans.

Chromium(III) chloride nutritional supplements have potential as an anti-depressant,

can reduce lipid deposits and are reported to lower blood glucose levels in type 2

diabetics. Therefore, understanding the stability and reactivity of dichlorotetraaqua

chromium(III) and chromium(III) with amino acid ligands can provide a basis for

exploring the biochemical pathways in physiological environments. Some recent

studies have suggested that the anti-diabetic effects of Cr(III) supplements might be

explained in terms of a stimulation by subtoxic doses of a toxic chemical. However,

the active species and mechanisms are not clearly identified. Therefore, the focus of

this work is on the speciation pathways of selected chromium(III) nutritional

supplements.


In Chapter 2, a comprehensive investigation was conducted to firstly establish

the reliability of selected methods (i.e. M06, B3LYP, CAM-B3LYP and PBE0) for

investigating reactions of Cr(III), by comparing calculated results with experimental

values, where possible. In addition, the role of solvent was investigated by exploring

both implicit solvation (PCM) and explicit solvation. However, the overall objective

of this study was to provide a detailed investigation of aquation of halopentaaqau-

Cr(III) species and to provide new insight into the observed trends that have

contributed to the debate of the intimate mechanism of this process. Features were

identified that contribute to the observed trends in each of the different pathways. As

a result of this analysis it was proposed that aquation of the highly basic fluoride

system proceeds via the Ia pathway, whereas systems with less basic leaving groups

(chloride, bromide and iodide) undergo aquation via the dissociation (D) pathway.

The activation enthalpies calculated at PBE0/cc-pVDZ are in excellent agreement

with experiment (MAD 1.0 kJ mol1

).

Chapter 3 presented an investigation of aquation of chromium(III) chloride

([CrCl2(H2O)4]Cl) and other dihalide complexes. The identification of the D

mechanism has some interesting implications for the speciation of chromium(III) in

physiological environments, since a range of biological molecules could potential ly

act as nucleophile and replace chloride after dissociation. Nevertheless, the

activation enthalpy for the dichloride system is 72 kJ mol‒1

, so this will be a

relatively slow process under physiological conditions. Therefore, it seems likely

that the active complex in chromium(III) chloride treatments is either [Cr(H2O)4Cl2]+

or [Cr(H2O)5Cl]2+

. Furthermore, the results in this study suggest that there may be

some value to explore the efficacy of other dihalotetraaquachromium(III) complexes


for treatment of type II diabetes and other physiological disorders due to their

different aquation/substitution rates.

Chapter 4 presented an extension of the chromium(III) chloride study to explore

the effects of pH and the formation of conjugate base species such as

[Cr(H2O)5(OH)]2+

, which in turn has been shown to catalyse conversion of glucose

to other biological molecules. Chromium(III) species that survive the highly acidic

environment (i.e. pH range of 1.0 to 3.7) of the stomach will pass on to the

duodenum, which has a milder pH ~4.6 Green chromium(III) chloride solution

reacts slowly with water to form a monochloro hydroxide, [Cr(H2O)4(OH)Cl]H2O+,

which undergoes slow aquation to form the penta-aqua hydroxide,

[Cr(H2O)5(OH)]2+

at higher pH, which can either undergo binding to bio-molecules

or can form hydrolytic polymers. Aquation studies of cis- and trans-

[Cr(H2O)4(OH)X]+ and other systems reveal formation of conjugate base species

with Lewis-acidic/Brønsted-basic forms that may eventually enter the blood stream,

which has a physiological pH of ~7.4. Therefore, the [Cr(H2O)5(OH)]2+

is key

derivative of chromium chloride nutritional supplement at mild and physiological

pHs.

Chapter 5 presented synthesis and fully characterization of facial tris-glycinated

Cr(III) complex by elemental analysis, atomic absorption spectroscopy (AAS), 1H

NMR, UV−Vis, EPR in both solid and frozen state, and ESI−MS spectroscopy.

Furthermore, the thermal decomposition of glycine ligands with Cr(III) complex was

explored by differential scanning calorimetry (DSC) under nonisothermal conditions,

at various heating rates from 5 to 20 °C/min over the temperature range from

30−1000 °C. Secondly, the electrochemical properties determined by cyclic

voltammetry indicate a reversible response of the [Cr(gly)3]/[Cr(gly)3]−1

redox


couple at the solid‒solvent (CH3CN)‒electrode interface. The redox behavior may

play a role in the biological activity of the complex.

Chapter 6 presented the synthesis of mono-, bis- and tris-glycinatochromium(III)

species and characterization by UV-Vis, TG-DSC, EPR in both solid and frozen

state. ATR-FTIR and Raman spectra were determined both experimentally and

theoretically. A key feature of the UV-Vis experimental investigation of the mono-,

bis- and tris-glycinatochromium(III) bidentate complexes is the effect of pH. In a

highly acidic environment like the stomach (pH = 3−4), the tris-glycinato complex

will undergo slow aquation to give bis- and mono-glycinato species. However, when

these complexes enter a much milder pH as found in the duodenum, the mono-

glycinatochromium(III) complex is expected to undergo rapid deprotonation to form

the corresponding conjugate base species and it may eventually enter the blood

stream at physiological pHs. These may be the active species for the treatment of

physiological disorders such as type 2 diabetes and depression. The theoretical study

reveals that activation via outer-sphere coordination is expected in all aqueous

environments including the stomach and gut. Secondly, once activated, the rate

determining step for the tris-complex is dissociation and is therefore independent of

the nature or concentration of the incoming nucleophile. However, aquation of the

mono-glycinatochromium(III) complex is found to follow the associative

interchange (Ia) pathway for the ring-opened isomer with ΔH‡ values in good

agreement with the experimental data, for both parent and conjugate base forms of

the complex. These results may provide further insight into whether or not these

complexes can either undergo binding to proteins or can form oligomers in

biological systems.


Chapter 7 presented new insights into the intimate mechanisms of the aquation of

the chromium picolinate ([Cr(pic)3]) nutritional supplement. This process initially

proceeds via a dissociative ring-opening step to form the [Cr(pic)2(H2O)(picH2)]+

intermediate in acidic media. Subsequently two possible pathways involving either

stepwise (two-step) or concerted (one-step) mechanism can occur. The most likely

pathway for the acid-catalyzed [Cr(pic)3] aquation via open type [Cr(

pic)2(OpicH)]2+

, is in excellent agreement with the experimental result. Electron

paramagnetic resonance spectroscopy found broad EPR signals for the mer-

[Cr(pic)3] complex with g values of 3.45 and 3.47 at room temperature and frozen at

100K, respectively. This result indicates that the mer-[Cr(pic)3] system has rhombic

character and a large zero field splitting D value. In addition, its purity was

confirmed by elementary analysis, atomic absorption spectroscopy (AAS), HPLC,

UV-Vis, ATR-FTIR.

The research in this thesis provides a detailed understanding of the aquation

mechanisms for several Cr(III) supplement systems. This information is useful in

understanding the speciation of Cr(III), prior to binding to biomolecules in

physiological environments, within the context of treatments for type 2 diabetes and

several other disorders. This work explores different reaction pathways (i.e., Ia, Id, A,

and D) using computational chemistry techniques to predict the overall activation

enthalpies of several Cr(III) nutritional supplements (e.g. chromium(III) chloride,

mono-, bis-, tris-glycinatochromium(III), their conjugate bases, and Cr(pic)3

species). The role of the solvent was also explored using both implicit (PCM) and

explicit solvation models. In order to determine the performance of the

computational methods and models used, comparisons with available experimental

results were conducted. Very good agreement between computational and


experimental observations was achieved when a combination of implicit (PCM) and

explicit (at least three water molecules) was used in the model systems.

8.2 Future directions

The findings of this research suggests that Cr(III) nutritional supplements can

undergo significant speciation via several pathways prior to binding to biomolecules

in physiological environments. In future, we plan to investigate model systems for

the interaction of these species with biomolecules:

Synthesis and characterization of Cr(III) with amino acids (e.g. Cr(me)3,

Cr(cys)3, Cr(glu)3 and Cr(D-phe)3) and mix ratios of amino acids and

other ligands.

Investigation of reaction pathways of Cr(III) with small organic and

inorganic molecules (e.g. methanol, ammine, amino acids, nicotinamide

and nicotinic acid).

Appendices 237

Appendix A

Figure A1: Optimized structures of (a) Zundel cation (H5O2+); (b) Eigen cation,

(H9O4+); (c) Zundel anion (H3O2

‒) and (d) Eigen anion (H7O4

‒) in solution phase

obtained at PBE0/cc-pVDZ. H-Bonds distances are in Å.

Appendices 238

Figure A2: Optimized TS structures of cis-[Cr(H2O)4(OH)Cl]+ with 2H2O, 3H2O,

4H2O and 5H2O in solution phase obtained at PBE0/cc-pVDZ.

H-Bonds distances are in Å.

Appendices 239

Appendix B

Experimental Section

Reagents

Chromium(III) chloride hexahydrate, CrCl3.6H2O (Sigma-Aldrich, ≥98.0%), glycine

(Sigma-Aldrich, ≥98.5%), ethanol (LabServ, 99.8%) and Sodium hydroxide (Sigma-

Aldrich, 98.5%) were used without further purification. Millipore filtered deionized

water was used throughout the experimental work.

Analytical Methods

Elemental analysis (C, H, and N) was performed using the microanalytical unit, J

Science lab JM 11 analyzer. Determination of Cr was carried out using atomic

absorption spectroscopy (a Varian AA50 spectrometer with air/C2H2 flame

atomization) after digestion of the samples with 69% HNO3 (Merck). Thermal

analysis was conducted using a PerkinElmer STA 8000 TGA-DSC instrument.

Before the heating routine was activated, the entire system was purged with argon

for 10 min at a rate of 20 mL/min, to ensure that the desired environment was

established. The differential scanning calorimetry (DSC) and thermogravimetric

analysis (TGA) experiments were performed with a heating rate of 5, 10, 15, and 20

Appendices 240

°C min−1

from 30 to 1000 °C. High-purity argon gas was used at a constant flow rate

of 20 mL/min. During the experiments, the sample mass loss and heat flow were

recorded continuously under dynamic conditions.

Mass spectra were acquired on a Waters Xevo G2 Q-TOF spectrometer

equipped with electrospray ionization (ESI) probe, with positive reflector mode

acquisition. Solid-state ultraviolet-visible (UV-vis) spectra were recorded on a UV-

3100PC, SHIMADZU, UV-VIS-NIR scanning spectrophotometer over the range

200−1450 nm. Liquid-state ultraviolet-visible (UV-Vis) spectra were recorded on a

HP 8453 UV-Vis spectrophotometer scanning over the range 190−1100 nm. A

Bruker EMX EPR spectrometer running the Xenon software with a Bruker ER

036TM NMR teslameter was used to measure the X-band spectra of the solid Cr(III)

complex with EP parameters for solid state as follows: center field, 3200 G; sweep

width, 6000 G; width TM, 200 G; FrequencyMon, 9.79 GHz; microwave power, 2.0

mW; microwave attenuation, 20.0 dB; conversion time, 4.0 ms; gain, 30dB;

modulation amplitude, 4.0 G; modulation frequency, 100 kHz; resolution, 15000 and

sweep time, 60 s. For frozen state and solution samples, the EPR tube was flushed

with with nitrogen gas for 2 min before analysis and the samples frozen by slow

immersion in a liquid nitrogen bath. EPR parameters for frozen state were gain, 1.0 ×

104; modulation frequency, 100 kHz; modulation amplitude, 1 G; conversion time,

0.41 ms; time constant, 81.92 ms; sweep time, 41.98 s; field center, 3350 G; sweep

width, 6400 G; frequency, 9.531415 GHz; and power, 2.0 mW. 1H NMR spectra

were recorded on a Varian 400 MHz spectrometer and referenced to internal

tetramethylsilane (TMS).

Cyclic voltammetry was performed using an ALS/HCH Model 620D

Electrochemical Analyzer with a glassy carbon working electrode, saturated

Appendices 241

Ag/AgCl reference electrode, and a Pt-wire auxiliary electrode. The supporting

electrolyte was 0.1 M of tetrabutylammonium hexafluorophosphate, TBAH in

CH3CN (or DMSO). Oxygen was removed by purging the solution with pure

nitrogen and continuous gas stream passed over the sample. The electrode was

calibrated against the ferrocene (Fe(C5H5)2), Fe+/Fe redox couple with a formal

reduction potential Eº = +0.4 V vs NHE.[46]

1‒3 mg of fac-Cr(gly)3 were placed on a

coarse grade filter paper and the glassy electrode was pressed on to the substance.[57‒

58] The E1/2 values were determined, E1/2 = (Epc + Epa)/2 and the electrochemical

reversibility of a redox couple, ΔEp = Epa – Epc, where Epa and Epc are anodic and

cathodic peak potentials, respectively.

Preparation of fac-[Cr(gly)3]

Tris-glycinatochromium(III) was prepared according to the method of Bryan et

al.[17]

A quantity of CrCl3.6H2O (2.66 g, 10 mM) was placed in a three-neck flask

and dissolved in 50 mL of H2O. Glycine (2.25 g, 30 mM) was added slowly to the

reaction mixture with continuous stirring and refluxed at 80−85 °C for 12 h until the

solution tuned red. Then, NaOH (1.2 g, 30 mM) solution was added dropwise with

stirring. The solution was then placed in a water bath (37 °C) and the red solid

product was collected by vacuum filtration, washed with ethanol and dried in a

desiccator. The yield of the reaction was 1.25 g (46%). Anal. Calcd for C6H12N3O6

Cr: C, 26.28; H, 4.41; N, 15.33; Cr, 18.98. Found: C, 26.21; H, 4.65; N, 14.97; Cr,

18.42. ESI-MS (H2O): MH+ m/z: calc. 275.0210; found: 275.2205.

Appendices 242

Figure B1: 1H NMR spectrum (400 MHz, D2O) of glycine for D2O with DMSO and

TMS.

Figure B2: 1H NMR spectrum (400 MHz, DMSO) of fac-[Cr(gly)3] with TMS and

added 1 µL H2O.

D2O

‒CH2‒

DMSO TMS

H2O

DMSO ‒CH2‒

TMS

Appendices 243

Figure B3: TG-DSC curve of the fac-[Cr(gly)3] complex at heating rates for (a) 5

°C/min; (b) 10 °C/min; (c) 15 °C/min, and (d) 20 °C/min.

Figure B4: Plot of ln(β/Tp

2) versus reciprocal peak temperature (1/Tp) and lnβ versus

1/Tp for (a) Kissinger’s method and (b) Ozawa’s method, respectively.

Appendices 244

Appendix C

Figure C1: UV-visible (UV-Vis) spectrum of [Cr(gly)(H2O)4]2+

, [Cr(gly)2(H2O)2]+

and fac-[Cr(gly)3]0 at different pH in the range 350‒800 nm, in water.The

concentration of the species are c = 21, 14, and 9 mM.

Appendices 245

Figure C2: ATR-IR (a‒d) spectra (4000‒350 cm‒1

) for glycine, fac-[Cr(gly)3],

[Cr(gly)2(H2O)]+ and [Cr(gly)(H2O)4]

2+ complexes.

Appendices 246

Figure C3: Raman (a‒d) spectra (4000‒200 cm‒1

) for glycine, fac-[Cr(gly)3],

[Cr(gly)2(H2O)]+

and [Cr(gly)(H2O)4]2+

complexes.

Appendices 247

Appendix D

Input File for Reactant of [Cr(pic)(H2O)4]2+

Complex %mem=2500MB

%nprocshared=4

# PBE1PBE/cc-pVDZ OPT FREQ

# SCRF=(PCM,solvent=water)

Title = “[Cr(pic)(H2O)4]2+ reactant state”

Complex Charge = 2 Multiplicity = 4

Symbolic Z-matrix or Cartesian (as usual)

Input File for TS of [Cr(pic)(H2O)4]2+

...H2O‡

%mem=2500MB

%nprocshared=4

# PBE1PBE/cc-pVDZ OPT=(TS,CalcFc,NoEigenTest) FREQ

# SCRF=(PCM,solvent=water)

Title = “[Cr(pic)(H2O)3]2+ with H2O Dissociation transition state”

Complex Charge = 2 Multiplicity = 4

Symbolic Z-matrix or Cartesian (as usual)

Appendices 248

Output File for Reactant of [Cr(pic)(H2O)4]2+

! Basis set = CC-pVDZ

Zero-point correction= 0.196350 (Hartree/Particle)

Sum of electronic and thermal Energies= -1785.168294

Sum of electronic and thermal Enthalpies= -1785.167350

Sum of electronic and thermal Free Energies= -1785.233044

Entropy = 138.265 Cal/Mol-Kelvin

Nuclear repulsion energy = 1360.8442809868 Hartrees.

FORMULA = C6H12CrNO6(2+,4)

Output File for TS of [Cr(pic)(H2O)4]2+

...H2O‡

! Basis set = CC-pVDZ

Zero-point correction= 0.197397 (Hartree/Particle)

Sum of electronic and thermal Energies= -1785.146296

Sum of electronic and thermal Enthalpies= -1785.145352

Sum of electronic and thermal Free Energies= -1785.207561

Entropy = 130.932 Cal/Mol-Kelvin

Nuclear repulsion energy = 1333.9359868094 Hartrees.

Imaginary frequency = 143.64i (cm**-1)

FORMULA = C6H12CrNO6(2+,4)

Appendices 249

Figure D1: UV‒vis spectral changes in solvents: (a) 2N HCl; (b) 2N HNO3; (c)

DMSO; (d) DMF; (e) 70% HClO4 and (f) 2N HNO3 : H2O (1:1) for chromium(III)

nitrate nanohydrate and CrPic complexes.

Appendices 250

Table D1: Electronic Spectral changes (in nm) in different solvents for

Chromium(III) Nitrate Nanohydrate (Cr(NO3)3.9H2O) and

Chromium Picolinate (CrPic) complexes.a

Cr(NO3)3.9H2O CrPic

Solvent λmax λmax ref.

H2O

576, 409

516

this work

513

b

2M HNO3

576, 406

527, 393

this work

521

b

2M HNO3 :

578, 405

527, 390

this work

H2O (1:1)

2M HCl 576, 406 532, 394 this work

527, 391 b

CH3CN

−

519

this work

DMSO 581, 412 519 this work

520

b

DMF

580, 412

518

this work

515

c

70% HClO4 358, 576 539, 388 this work

60% HClO4 543, 400 d a

UV-Vis spectra of mer-[Cr(pic)3] defined in Figures D1 and SF10 of the SI.

b

Ref. 63. c Ref. 42.

d Ref. 81.

Investigation of the Substitution Reaction Pathways of ...

Documents

Transcript of Investigation of the Substitution Reaction Pathways of ...