Ii ION-PAIR BEHAVIOR BETWEEN POLYOXOMETALATES ANION

Home , Ion association, Keggin structure

ION-PAIR BEHAVIOR BETWEEN POLYOXOMETALATES ANION AND

ALKALI METAL CATION

A Thesis

Presented to

The Graduate Faculty of The University of Akron

In Partial Fulfillment

of the Requirement for the Degree

Master of Science

Songtao Ye

May 2018

ION-PAIR BEHAVIOR BETWEEN POLYOXOMETALATES ANION AND

ALKALI METAL CATION

Songtao Ye

Thesis

Approved Accepted:

______Advisor Dean of the College Dr. Tianbo Liu Dr. Eric Amis

______Committee Member Dean of the Graduate School Dr. Toshikazu Miyoshi Dr. Chand Midha

______Department Chair Date Dr. Colleen Pugh

iii

ABSTRACT

Ion-pair behavior describes the partial association of oppositely charged ions in electrolyte solutions. Previous study mainly focused on the ion-pair behavior between simple ions, such as ion pairing in NaCl solution as well as ion-pair interactions in supramolecular complexes and biological associations. However, very few attentions have been placed on the solution system with particle sizes in between.

Recently, a group of well-defined, huge anionic cluster named polyoxometalates (POMs) have been synthesized and well characterized. The size of POMs is around nanometer scale, which is exactly between simple ions and large colloids. The solution behavior for

POMs is much different form simple electrolyte solutions or large colloids. As a result, it is interesting to study the ion-pair behavior for POMs in solution. Herein, ion-pairs between

Lacunary Keggin type POMs and alkali metal cations are investigated. The result showed that ion-pairs are formed between alkali cations and the “pocket” area on the surface of

Lacunary Keggin type POMs – K7PW11O39. Electrostatic interaction and the entropy gain during the solvation shell lost were considered major driving forces during the ion-pair formation. Smaller alkali cations (e.g., Li+ and Na+) tended to form contact ion- pair (CIP) which result in an elevated enthalpy change measured by Isothermal Titration Calorimetry

(ITC). Larger alkali cations (e.g., Rb+ and Cs+) favored a loose type of ion-pair – solvent separated ion-pair (2SIP) and solvent shared ion-pair (SIP). Size exclusion between the

“pocket” area on K7PW11O39 POM surface and alkali cation also played a significant role in determining the ion-pair structure. Results were further confirmed by Nuclear Magnetic

Resonance Spectroscopy (NMR).

ACKNOWLEDGEMENT

First, I would like to express my gratitude to my advisor Prof. Tianbo Liu for his patient

guidance and strong support for my study at the University of Akron. He brought me this

area and offered me great opportunities to study the complex solution behavior. Meanwhile,

I also want to thank my committee member Dr. Toshikazu Miyoshi for his advice to my research and constructive feedback to my dissertation.

Besides, I would like to appreciate Jiancheng Luo, a senior PhD student in our group. He

is so generous to share precious experience and useful methods with me. This dissertation

would never be finished without his help. And I also want to all the group members for

their assistance during my research.

Finally, I would like to extend my appreciation to my family and my friends, especially my parents. I appreciate all your support from both mentally and physically.

TABLE OF CONTENTS

LIST OF TABLES ...... vi

LIST OF FIGURES ...... vii

CHAPTER ...... 1

I. INTRODUCTION AND BACKGROUND ...... 1

1.1 Introduction of ion-pair behavior ...... 1

1.2 Experimental methods for studying ion-pair ...... 5

1.3 Solution behaviors of macroions ...... 7

1.4 Study motivation ...... 11

II. EXPERIMENT...... 12

2.1 Sample preparation ...... 12

2.1.1 Synthesis of Na3[PW12O40] ·13H2O ...... 12

2.1.2 Synthesis of K7[PW11O39] ·13H2O ...... 12

2.2 Isothermal Titration Calorimetry (ITC) ...... 13

2.3 Nuclear Magnetic Resonance Spectroscopy ...... 15

III. RESULT AND DISCUSSION ...... 16

3.1 Ion-pair formation monitored by Isothermal titration calorimetry (ITC) ...... 16

3.2 Ion-pair formation monitored by Nuclear Magnetic Resonance Spectroscopy

(NMR) ...... 22

IV. CONCLUSION...... 24

REFERENCE ...... 25

vii

LIST OF TABLES

Table Page

1. Types of POMs which can form Blackberry in aqueous solution.1 (Reprinted from

ref. 34, copyright ACS Publication.) ...... 8

2. Radius of bare alkali cations and alkali cations in water ...... 18

viii

LIST OF FIGURES

Figure 1 The Eigen–Tamm scheme for stepwise formation from free solvated cations Xx+ and solvated anions Yy− of 2SIP ion pairs, then SIP ion pairs, and finally CIP ion pairs, with elimination of solvent molecules from the solvation shells of the ions.5 (Reproduced from ref. 5, copyright IUPAC 2008) ...... 1

Figure 2 Dielectric loss (ε") spectrum for 0.37 M NiSO4(aq) at 25 °C.

(Reproduced from ref. 1, copyright ACS Publications) ...... 6

Publishing Group.) ...... 9

n- Figure 4 Structure of ɑ-Keggin [XM12O40] anion. Red balls represent oxygen atoms,

blue balls represent M atoms and green ball represents X atom. The overall POM

anion carries n negative charges. (Reprinted from ref. 39, copyright ACS

Publications) ...... 10

n- Figure 5 Structure of ɑ-Lacunary Keggin [XM11O39] anion. One tetrahedral unit is

removed from ɑ-Keggin type structure ...... 10

Figure 6 Basic configuration of an isothermal titration calorimetry42 ...... 13

n- Figure 7 Structure of ɑ-Lacunary Keggin [XM11O39] anion...... 15

ix 7- Figure 8 Titrating 10 mM KCl solution into 0.5 mM [PW11O39] POM solution...... 16

7- Figure 9 Titrating 10 mM RbCl solution into 0.5 mM [PW11O39] POM solution...... 17

7- Figure 10 Titrating 10 mM RbCl solution into 0.5 mM [PW11O39] POM solution...... 17

7- Figure 11 Titrating 10 mM LiCl solution into 0.5 mM [PW11O39] POM solution...... 19

7- Figure 12 Titrating 10 mM NaCl solution into 0.5 mM [PW11O39] POM solution...... 20

Figure 13 Contribution form enthalpy and entropy to the Gibbs free energy during ion- pair formation...... 21

Figure 14 NMR spectrum for titrating Li+ into K7[PW11O39] POM solution. 22

+ Figure 15 NMR spectrum for titrating K into K7[PW11O39] POM solution ...... 23

CHAPTER I

INTRODUCTION AND BACKGROUND

1.1 Introduction of ion-pair behavior

Ion-pair behavior describes the partial association of oppositely charged ions in

2 electrolyte solutions to form distinct chemical species called ion-pairs.

Typically, an ion pair can be classified as 3 types, a double solvent-separated ion pair

(2SIP), if the primary solvation shells of the both ions remain intact, a solvent-shared ion

pair (SIP), if both ions share one solvent layer, or a contact ion pair, which two ions contact directly with each other.3 The development in understanding the relationships between the

4-5 various types of ion pairs was based on ultrasonic absorption data by Eigen et al.

Figure 1 The Eigen–Tamm scheme for stepwise formation from free solvated cations Xx+

and solvated anions Yy− of 2SIP ion pairs, then SIP ion pairs, and finally CIP ion pairs, with elimination of solvent molecules from the solvation shells of the ions.6 (Reproduced from

ref. 5, copyright IUPAC 2008)

1 This mechanism describes that the ion-pair formation occurs in an equilibrium format.

Initially, two oppositely charged ions come together driving by the electrostatic interaction and form a solvent-separated ion-pair (2SIP). During this process, the solvation shell for two ions are remain intact. In the next step, the solvent molecules between two ions are further expelled out of the solvation shell if the association constant for two ions are relatively higher. In this case, a solvent-separated ion-pair (SIP) would be formed.

Ultimately, a contact ion-pair (CIP) would be the last format of the association species and no solvent molecule will exist between two ion domains and the ions are considered in physical contact.

It is worth noticing that the quantity of a specific type of ion-pair in a given system is not isolated with the others, and the quantity is calculable if stepwise equilibrium constant Ki

(i=1, 2, and 3) and overall equilibrium constant KA is required. The equations are described

7 below.

[ ( )+( )] = [ +( 𝑥𝑥−)]𝑦𝑦[ ( )] 𝑋𝑋𝑋𝑋 𝑎𝑎𝑎𝑎 𝐴𝐴 𝑥𝑥 𝑦𝑦− 𝐾𝐾 𝑋𝑋 𝑎𝑎𝑎𝑎 𝑌𝑌 𝑎𝑎𝑎𝑎 [2 ] [ ] [ ] 1 = 2 = 3 = [ +( )][ ( )] [2 ] [ ] 𝑆𝑆𝑆𝑆𝑆𝑆 𝑆𝑆𝑆𝑆𝑆𝑆 𝐶𝐶𝐶𝐶𝐶𝐶 𝑥𝑥 𝑦𝑦− 𝐾𝐾 𝑋𝑋 𝑎𝑎𝑎𝑎 𝑌𝑌 𝑎𝑎𝑎𝑎 𝐾𝐾 𝑆𝑆𝑆𝑆𝑆𝑆 𝐾𝐾 𝑆𝑆𝑆𝑆𝑆𝑆 = 1 + 1 2 + 1 2 3

𝐴𝐴 𝐾𝐾 𝐾𝐾 𝐾𝐾 𝐾𝐾 𝐾𝐾 𝐾𝐾 𝐾𝐾 KA is calculable form stepwise equilibrium constant Ki, but the reverse cannot be derived.

This mechanism has been established for decades for charged ions with considerably strong association constant. However, not all types of ion-pairs are observable for different approaches. Marcus et al.2 has reported that using conventional spectroscopic techniques

to study of multistep ion-pair reactions can sometimes be misleading. The result obtained

2 from other work also concluded that conventional spectroscopies detect only CIP species

in most cases7.

Theoretical treatment of ion-pairs was first introduced by Bjerrum in early 20 century.8

In his approach, he followed the Debye-Hückel theory9, and assumed the ions hard spheres to simplify the system. He described spices as ion-pairs if two oppositely charged ions stay together at a distance, r, which is shorter than a critical distance, q, and longer than a closest approach, ɑ. Distance shorter than ɑ cannot be reached due to the strong repulsive forces of the electron shells of the ions. Ions further apart than distance q, energy required to form ion pairs is not sufficient to overcome the thermal movement, so they are considered free.

The distance q, also called Bjerrum length, depends on the charge of central ion and the charge of the surrounding ion, the temperature, and the permittivity of the solvent. The equation is listed below.

q = 2 𝑧𝑧𝑖𝑖 𝑧𝑧𝑗𝑗𝑒𝑒 𝐵𝐵 𝜀𝜀𝜀𝜀 𝑇𝑇 Here is the Boltzmann constant and is the Kelvin temperature. The effect of the

𝐵𝐵 solvent 𝑘𝑘is expressed through its bulk permittivity,𝑇𝑇 , and the solvent is considered as a

dielectric continuum. It is interesting to notice that, 𝜀𝜀the ratio between electrostatic energy

and thermal energy is a factor of 2 when the distance between the counterions is . In other words, once ion pair is formed, the electrostatic energy is at least two times larger𝑞𝑞 than the energy required for thermal fluctuation. The definition is, however, not precise because counterions having an energy between and 2 can still form relatively long-lived

𝐵𝐵 𝐵𝐵 associations which they are not included𝑘𝑘 under𝑇𝑇 this𝑘𝑘 criterion.𝑇𝑇

3 Based on the Bjerrum approach, simple ionic solution (such as dilute alkali halide

aqueous solution) have been well-studied. It is widely accepted that for those alkali halides in water, such as dilute KCl solution, no ion-pair would occur. This is because in those system, only very short Bjerrum length q is derived because of their insufficient charge as shown in equation above. Sometimes the q values in those system is close or even shorter than the closest approach a. For example, the Bjerrum length for KCl is 0.357 nm, and the closest approach between potassium cation and chloride anion is = + + = − 𝐾𝐾 𝐶𝐶𝐶𝐶 0.319 . Considering the tremendous hard sphere repulsion, 0.319𝑎𝑎 nm𝑟𝑟 can𝑟𝑟 never be approached.𝑛𝑛𝑛𝑛 In this case, q and a values have only tiny differences, which means ion-pairs

are very hard to form. In another case, for larger alkali halides, such as CsI, Bjerrum length is even shorter than the closest approach. So that all ions can be considered free in the

10 solution. This result is further corroborated by conductivity data.

The competition between ion-pairs and solvation of the ions is always exist11.

Electrostatic interaction and ion-dipole interaction are considered the major driving forces for each behavior. In fact, the nature behind above two interactions can be attributed to the simple attractions between positive and negative charge. In some certain cases, an oppositely charge ion can replace some of the solvent in the solvation shell of a given ion.

During this process, free solvent molecules would be expelled out of the ion solvation shell.

As a result, an entropy gain of total system can be observed.

4 1.2 Experimental methods for studying ion-pair

Historically, several different approaches12-16 have been applied to determine ion-pairs

formation as well as distinguishing the association constant value for each types of ion-

pair. Among them, conductivity measurement is the first experimental evidence to

determine ion-pairs.17 For ion-pairs in the solution, regardless of which types of ion-pair it

formed, will decrease the solution conductivity.18 However, the drawbacks for this

measurement are also straightforward. People cannot separate different types of ion-pair through conductivity measurement since all types of them contribute equally to the conductivity. Subsequently, other approaches are designed to achieve this goal. In this article, spectroscopy methods19 and relaxation methods will be introduced.

Spectroscopy have been widely used to study the ion-pair behavior because they are

applicable to various systems without considering the theoretical support. The nature of

spectroscopy methods relies on the observation of a new well-separated peak, or in other

cases, the modification of their original peaks. However, those popular techniques also

have some significant limitations. Specifically, the result could be misleading20 when the

system contains considerably amount of 2SIP and SIP types of ion-pair. This is because

conventional spectroscopies are generally sensitive to only CIP species.

In contrast, dielectric relaxation spectroscopy (DRS) have shown to have unusual

capabilities for the study of ion-pair behavior.21-22 DRS measures the electric permittivity of a sample corresponding to an applied electromagnetic field. The frequency is often in the microwave region (0.01GHz~100GHz). DRS measures frequency-dependent complex dielectric response, ε*(υ), of a solution which related to the motions of all dipolar species

5 in the solution including the solvent molecules. In addition, DRS favors those dipole

species with larger dipole moment (μ).

𝜇𝜇 𝑧𝑧𝑧𝑧𝑧𝑧 In this equation, ze is the charge of the ions and d is their separation distance. For different types of ion-pairs, the distance between the charge centers can be described in the following order 2 > > . As a result, DRS has a sensitivity toward the various ion-pair

𝑆𝑆𝑆𝑆𝑆𝑆 𝑆𝑆𝑆𝑆𝑆𝑆 𝐶𝐶𝐶𝐶𝐶𝐶 23 types 𝑑𝑑in the following𝑑𝑑 𝑑𝑑 order: 2SIP > SIP > CIP.

Figure 2 Dielectric loss (ε") spectrum for 0.37 M NiSO4(aq) at 25 °C. (Reproduced from

ref. 1, copyright ACS Publications)

6 1.3 Solution behaviors of macroions

Solution state is one of the most important state which more than 90% of chemical

reactions and all the biological process are undertaken in solution state. However, understanding the solution behavior has always been a challenging task for physical chemist because of the complexity of the solution. This situation could be even worse if the solutes are charged. Historically, two different theories have been developed to describe complex solution behavior. On one hand, Debye-Hückel theory9 is quite well-known and

successful to describe simple ionic solution (such as dilute NaCl aqueous solution), and its

extended form can be applied to higher ionic concentrations. On the other hand, when the

charged particles are reaching colloidal region, Derjaguin-Landau-Verwey-Overbeek theory24-25 (or DLVO theory) can be applied to these meta-stable colloids system. However, only few experiments had been done with intermediate particles because of unsuitable models. Recently, development in synthetic inorganic chemistry has produced a large series of giant molecules named polyoxometalates (POMs) which serve as perfect samples for us to study complex solution behavior.

POMs are a large group of rigid, well-defined, huge anionic cluster.26 Their geometrically

perfect structures are built through the corner-, edge- or face-sharing of early transition metal (Mo, W, Nb, Ta, V, Pd…etc.) oxygen polyhedra into larger architectures with high symmetry. The size and shape of POMs span a broad range, from the smallest

2- hexatungstate [W6O19] anion to the giant nanocapsules that contain 368 molybdenum

atoms introduced by Müller et al.27 POMs have found application in a variety of disciplines, including molecular magnetism28, protein sequencing29, solar cells30, stimuli responsive

materials31, catalysis32, medicine33, and storage devices34.

7 Table 1. Types of POMs which can form Blackberry in aqueous solution.1 (Reprinted from

ref. 34, copyright ACS Publication.)

POMs can be easily dissolved in water and other polar solvent because of their charged

hydrophilic nature. While dissolving in water forming dilute solution, all the counterions

are considered released. As a result, POMs anions are fully charged and tend to deprotonate upon solvation, thus their total charge can be easily controlled by adjusting the pH of the solution. It has been recently report by Liu et al35 that some of the hydrophilic POMs do not stay as single clusters or colloids in water or mixed polar solvents. While instead, POMs tend to self-assemble into large, single layer, hollow and spherical like structure35-36 shown

in figure 3. More interesting, POMs possess other unique solution behavior such as self-

37-38 recognition.

More importantly, POMs are optimal templates for studying the ion-pair behavior with

nanometer scale systems. Most of POMs are soluble in water and hydrophilic solvent with tunable negative charges.

Figure 3 “Giant Wheel” POMs self-assemble into Blackberry like structure in aqueous

Group.)

Keggin-type POMs are used to study the POM-alkali metal ion pairs due to the relatively detailed simulation study for its solution behavior.39-40 Keggin POMs are a family of

n- heteropolyanions with general formula of [XM12O40] where M is normally Mo or W and

X is a heteroatom that is most commonly known to be P5+ and Si4+.41The Keggin structure

was also proved to be sensitive to the pH. Upon increasing the pH, the Keggin POM will

transform into mono-, di- or tri-lacunary Keggin POMs by the removal of M-O unit.

In our experiment, a type of mono-lacunary Keggin POM with the formula K7PW11O39

was used. A “pocket ” area could be found on the surface of the lacunary Keggin POM as

shown in Figure 5.

n- Figure 4 Structure of ɑ-Keggin [XM12O40] anion. Red balls represent oxygen atoms, blue balls represent M atoms and green ball represents X atom. The overall POM anion carries n negative charges. (Reprinted from ref. 39, copyright ACS Publications)

n- Figure 5 Structure of ɑ-Lacunary Keggin [XM11O39] anion. One tetrahedral unit is removed from ɑ-Keggin type structure.

1.4 Study motivation

The purpose of this project is to study the ion-pair behavior between macroions and simple ions in solution, thus filling up the blank page between small ionic solution and large supramolecular complexes. Different approaches including thermodynamic and spectroscopy methods have been applied to identify the ion-pair formation and the types of ion-pair.

CHAPTER II

EXPERIMENT

2.1 Sample preparation

2.1.1 Synthesis of Na3[PW12O40] ·13H2O The procedure for this POM is based on a published literature.41 100.0 g (0.3 mol)

Na2WO4·2H2O powder was dissolved in 150 mL water. 50.0 g (0.28 mol) solid

Na2HPO4·2H2O powder was then added into beaker. The solution was heated in a water bath at 80 °C for 4 h. To the solution, 160 mL of 7.3 M aqueous HCl solution was slowly added, followed by stirring for 30 min in a water bath at 80 °C. The colorless solution was changed to pale yellow, and crystalline white powder was formed. The solution was concentrated to ca. 200 mL. After cooling to room temperature, crystalline powder was formed at the bottom of the beaker. The white crystalline powder was collected and re- dissolved in ca. 100 mL water with an oil bath at over 90 °C. The colorless clear solution was cooled to room temperature and allowed to stand overnight in a refrigerator at 4 °C.

Crystalline white powder formed was collected and dried in a small vial.

31 P NMR (25 °C, D2O): δ -15.04

2.1.2 Synthesis of K7[PW11O39] ·13H2O

20.0 g (6.3 mmol) Na3[PW12O40] ·13H2O crystalline powder was dissolved in 300 mL of water (adjusted to pH = 4.9 by adding a solid of Na2CO3 with a small portion) where into a solid KCl (50.0 g (0.7 mol)) was added, followed by heating with an oil bath at 90 °C.

12 The resulting colorless clear solution was cooled to room temperature and allowed to stand overnight in a refrigerator at 4 °C. White powder formed at the bottom of the beaker was

31 collected on a glass vial by filtering and dried for 2 h. P NMR (25 °C, D2O): δ -10.77

All the salts for synthesizing above two POMs and other salts (LiCl, NaCl, KCl, RbCl

and CsCl) used for studying the ion-pair behavior were bought from Sigma-Aldrich

Corporation.

2.2 Isothermal Titration Calorimetry (ITC)

Isothermal titration calorimetry is an ideal technique for measuring ion-pair interactions.

The technique relies only on the detection of a heat effect upon binding, it can be used to measure the binding constant, K, the enthalpy of binding, ∆H° and the stoichiometry, or number of binding sites, n. A scheme describing the structure of Isothermal titration calorimetry is listed in figure 6 below.

Figure 6 Basic configuration of an isothermal titration calorimetry42.

13 Isothermal titration calorimetry is a quantitative technique which is used to determine

the enthalpy change (ΔH), the binding affinity (Ka), and the binding stoichiometry (n).

Once these initial thermodynamic parameters have been measured, entropy change (ΔS)

and Gibbs free energy (ΔG) can be calculated using the relationship:

𝑎𝑎 𝛥𝛥𝛥𝛥 = −𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅 𝛥𝛥𝛥𝛥 − 𝑇𝑇𝛥𝛥𝛥𝛥 Where R is the gas constant and T is the absolute temperature.

A commercial Isothermal Titration Calorimetry from TA Instrument, equipped with a

hastelloy cell was used for ITC experiment. 1.0 mL 0.5 mg/mL K7[PW11O39] POM solution was added into the calorimetric cell, 10.0 mM, 250.0 μL LiCl solution in a 250 μL calorimetric syringe was titrated into POM solution. The titration temperature was set at

25.0 °C. The mode of ITC was incremental mode, and the injections were made of 25 times

with 10.0 μL each time, the interval between two injections was set as 500 s, and the cell

+ was at 250 rpm stirring. After titrating Li into K7[PW11O39] solution, 10 mM, 250.0 μL

NaCl, KCl, RbCl and CsCl solutions were titrated into Na3[PW12O40] solution in the same condition respectively. Another group of 1.0 mL H2O was titrated with same concentration alkali halide salt solution as the background.

After the titration, the data were analyzed by Nano Analyze Software from TA

Instrument.

14 2.3 Nuclear Magnetic Resonance Spectroscopy

31P-NMR spectrum was performed to monitor ion-pair behavior. From the figure 7 below, the pink ball in the center of this K7[PW11O39] POM is a phosphorus atom. A close binding

would change the chemical environment of the center phosphorus atom, which will lead to

a chemical shift on 31P-NMR spectrum.

n- Figure 7 Structure of ɑ-Lacunary Keggin [XM11O39] anion.

30 mg/mL K7[PW11O39] POM D2O solution was prepared as the background solution.

Another set of solution with the molar ratio of alkali cation versus POM solution equals to

one was prepared to show the chemical shift of the phosphorus signal. A commercial 1H

resonance 500 MHz Varian, Inc. NMR spectroscopy was used to perform above

experiment.

CHAPTER III

RESULT AND DISCUSSION

3.1 Ion-pair formation monitored by Isothermal titration calorimetry (ITC)

7- For titrating different monovalent alkali chloride salt solution into [PW11O39] type POM

anions, result follows two different trends.

+ + + (1) Heat release rate trend: K < Rb < Cs

The heat release rate represents an increasing ion pair formation process as the increment of atom number form K+ to Cs+. The result is shown in figure 8, 9 and 10.

7- Figure 8 Titrating 10 mM KCl solution into 0.5 mM [PW11O39] POM solution.

7- Figure 9 Titrating 10 mM RbCl solution into 0.5 mM [PW11O39] POM solution.

7- Figure 10 Titrating 10 mM RbCl solution into 0.5 mM [PW11O39] POM solution.

17 For K+ and Rb+, there are basically no interaction can be found by the titration curve.

The heat release rate for titrating salt in to POM solution is similar with background level

heat (which titrate same concentration of salt solution in water). For Cs+, a relatively low

heat release was found. The difference between above three alkali cations can be illustrate

by the charge density theory.

Table 2. Radius of bare alkali cations and alkali cations in water.

+ + + + + Li Na K Rb Cs

Bare Ion Radius / pm 76 102 138 152 167

Hydrated Ion Radius / pm 340 276 232 228 226

It is widely accepted that, for alkali elements, the effective charge would increase with

the increment of the atomic numbers. The reason is because for smaller element with higher local charge density, strong ion-dipole interaction will attract solvent molecules around it.

As a result, a thicker solvation shell will screen a part of the net charge, leaving a weakened effective charge. Thus, the effective charge for K+, Rb+, and Cs+ coincide with the heat

release trend. It is convincing that the nature for above mentioned exothermic heat release peaks can be contribute to the electrostatic interactions. In this case, a solvent separated or solvent shared ion-pair is formed.

18 + + + (2) Heat release rate trend: K << Na ~ Li

For trend one, it is expected and easily explained by the charge nature of alkali cations.

However, when it comes to the trend two, the result is surprising. The unexpected heat

release rate cannot be explained the previous theory. ITC result are listed on figure 11 and

12 below.

7- Figure 11 Titrating 10 mM LiCl solution into 0.5 mM [PW11O39] POM solution.

+ + 7- By titrating 10 mM Li and Na alkali solution into [PW11O39] POM solution, an

unexpected heat release was obtained, which is against our previous result. As ion pairs

were formed, heat release for the first several injections could reach the level of 300 µJ for

Li+ and 200µJ for Na+.

7- Figure 12 Titrating 10 mM NaCl solution into 0.5 mM [PW11O39] POM solution.

After several injections, the curve for both cases experience a dramatic decrease all the

way down to background level. In both two cases, a critical molar ratio around 1 arise our

attention. The reasonable explanation is that the appearance of this unusual phenomenon

is related to the ion insertion (in other words, contact ion-pair) of small alkali cations

7- towards lacunary Keggin structure of [PW11O39] POM anions. The specified ratio 1 is

because of the only “pocket” on the POM surface. Once the “pocket” is occupied, other alkali cations cannot continue to form this specific type of ion-pair. Several simulation results have been published introducing the charge density and solvation shell around normal Keggin anion. Based on one of the published paper, for normal Keggin anions, the

7- terminal oxygens are the least negatively charged. The lacuna for [PW11O39] POM anion,

20 as shown in Figure 7, is surround by 4 terminal oxygen. As a result, the highest charge

density can be obtained at the lacuna position. When the molar ratio for alkali cation over

POM anion reaching the level of 1, POM anion are treated saturate based on this modal.

Entropy gain during the ion-pair formation is considered another major driving force.

Solvent molecules in the ion solvation shell are well-ordered. Upon ion-pair formation,

those ordered solvent molecules (in this case, H2O) are expelled out of the overlapped

region. As a result, ordered water molecules are released and become free water molecules.

During this process, the entropy of this system will increase. By calculating and fitting ITC data, the entropy contributions to the Gibbs free energy for titrating different alkali salt solution into K7PW11O39 POM solution are plotted below. The data showed that entropy

gain contribute significantly when CIP is the formed.

Thermodynamic Fitting Data for Different Alkali Cations

kJ/mol 30

25 20 15

5 0 Li+ Na+ Cs+

- ΔH TΔS - ΔG

Figure 13 Contribution form enthalpy and entropy to the Gibbs free energy during ion-pair formation.

21 3.2 Ion-pair formation monitored by Nuclear Magnetic Resonance Spectroscopy (NMR)

To further confirm previous ion-pair type conclusion, a set of NMR study with variable

7- salt concentration was conducted. In this experiment, 30 mg/ml [PW11O39] POM solution with different molar of alkali salt solution was used to study the chemical shift of center phosphorus atom.

Figure 14 NMR spectrum for titrating Li+ into K7[PW11O39] POM solution.

+ Figure 15 NMR spectrum for titrating K into K7[PW11O39] POM solution.

From above two figures, we can conclude that a contact ion-pair was formed when titrate

+ + Li into K7[PW11O39] POM solution. The inserted Li cation is much closer than the free

cations outside. As a result, the chemical shift of center phosphorus atom showed a upfield change. However, for K+ cation, its large size restrict itself to be accommodated into the

“pocket” area. Thus, no chemical shift change was observed.

CHAPTER IV

CONCLUSION

In summary, ion-pairs between Lacunary Keggin type POMs and alkali metal cations are

investigated. Ion-pairs are formed between alkali cations and the “pocket” area on the surface of

Lacunary Keggin type POMs – K7PW11O39. Electrostatic interaction and the entropy gain during

the solvation shell lost are considered major driving forces during the ion-pair formation. Smaller

alkali cations (e.g., Li+ and Na+) will form contact ion-pair (CIP) which result in an elevated enthalpy change measured by Isothermal Titration Calorimetry (ITC). Larger alkali cations (e.g.,

Rb+ and Cs+) favor loose type of ion-pairs – solvent separated ion-pair (2SIP) and solvent shared ion-pair (SIP). Size exclusion between the “pocket” area on K7PW11O39 POM surface and alkali cation also played a significant role in determining the ion-pair structure.

REFERENCE

1. Yin, P.; Li, D.; Liu, T., Solution behaviors and self-assembly of polyoxometalates

as models of macroions and amphiphilic polyoxometalate–organic hybrids as

novel surfactants. Chemical Society Reviews 2012, 41 (22), 7368-7383.

2. Marcus, Y.; Hefter, G., Ion pairing. Chemical reviews 2006, 106 (11), 4585-4621.

3. Marcus, Y., Ion solvation. Wiley: 1985.

4. Eigen, M.; Tamm, K., Schallabsorption in elektrolytlösungen als folge chemischer

relaxation I. Relaxationstheorie der mehrstufigen dissoziation. Berichte der

Bunsengesellschaft für physikalische Chemie 1962, 66 (2), 93-107.

5. Eigen, M., Schallabsorption in Elektrolytlösungen als Folge chemischer Relaxation

II. Meßergebnisse und Relaxationsmechanismen für 2—2‐wertige Elektrolyte.

Berichte der Bunsengesellschaft für physikalische Chemie 1962, 66 (2), 107-121.

6. Buchner, R., What can be learnt from dielectric relaxation spectroscopy about ion

solvation and association? Pure and Applied Chemistry 2008, 80 (6), 1239-1252.

7. Hefter, G., When spectroscopy fails: the measurement of ion pairing. Pure and

applied chemistry 2006, 78 (8), 1571-1586.

8. Bjerrum, N., K. Dan. Vidensk. Selsk. 1926, 7, 9.

9. Debye, P.; Hückel, E., De la theorie des electrolytes. I. abaissement du point de

congelation et phenomenes associes. Physikalische Zeitschrift 1923, 24 (9), 185-

206.

10. Hanna, E.; Pethybridge, A.; Prue, J., Ion association and the analysis of precise

conductimetric data. Electrochimica Acta 1971, 16 (6), 677-686.

11. Gilkerson, W., Application of free volume theory to ion pair dissociation constants.

The Journal of Chemical Physics 1956, 25 (6), 1199-1202.

12. Strehlow, H., CB Monk. Electrolytic Dissociation. Academic Press, London und

New York 1961. 320 Seiten. Preis: 60 s. Berichte der Bunsengesellschaft für

physikalische Chemie 1962, 66 (1), 75-76.

13. Davies, C.; Association, I., Butterworths. London: 1962.

14. Nancollas, G. H., Interactions in electrolyte solutions: Metal complex and ion-pair

formation in solution. Elsevier Pub. Co.: 1966; Vol. 8.

15. Alcock, N. W., Bonding and structure: structural principles in inorganic and

organic chemistry. Ellis Horwood Limited: 1990.

16. Wachter, R.; Riederer, K., Properties of dilute electrolyte solutions from

calorimetric measurements. Pure and Applied Chemistry 1981, 53 (7), 1301-1312.

17. Arrhenius, S., Über die Dissociation der in Wasser gelösten Stoffe. Zeitschrift für

physikalische Chemie 1887, 1 (1), 631-648.

18. White, R. E.; Bockris, J.; Conway, B. E.; Yeager, E., Comprehensive treatise of

electrochemistry. Vol. 8: experimental methods in electrochemistry. 1984.

19. Rudolph, W.; Irmer, G.; Hefter, G., Raman spectroscopic investigation of

speciation in MgSO 4 (aq). Physical Chemistry Chemical Physics 2003, 5 (23),

5253-5261.

20. Buchner, R.; Chen, T.; Hefter, G., Complexity in “simple” electrolyte solutions:

ion pairing in MgSO4 (aq). The Journal of Physical Chemistry B 2004, 108 (7),

2365-2375.

21. Barthel, J.; Neueder, R.; Poepke, H.; Wittmann, H., Osmotic and activity

coefficients of nonaqueous electrolyte solutions. 1. Lithium perchlorate in the

protic solvents methanol, ethanol, and 2-propanol. Journal of solution chemistry

1998, 27 (12), 1055-1066.

22. Buchner, R.; Sipos, P.; Hefter, G.; May, P. M., Dielectric relaxation of concentrated

alkaline aluminate solutions. The Journal of Physical Chemistry A 2002, 106 (28),

6527- 6532.

23. Chen, T.; Hefter, G.; Buchner, R., Dielectric spectroscopy of aqueous solutions of

KCl and CsCl. The Journal of Physical Chemistry A 2003, 107 (20), 4025-4031.

24. Derjaguin, B., Theory of the stability of strongly charged lyophobic sols and the

adhesion of strongly charged particles in solutions of electrolytes. Acta

Physicochim. USSR 1941, 14, 633-662.

25. Overbeek, J. T. G.; Verwey, E., Theory of the Stability of Lyophobic Colloids: The

interaction of Sol Particles Having an Electric Double Layer. 1948.

26. Müller, A.; Reuter, H.; Dillinger, S., Supramolecular inorganic chemistry: small

guests in small and large hosts. Angewandte Chemie International Edition 1995,

34 (21), 2328-2361.

27. Müller, A.; Beckmann, E.; Bögge, H.; Schmidtmann, M.; Dress, A., Inorganic

Chemistry Goes Protein Size: A Mo368 Nano‐Hedgehog Initiating Nanochemistry

by Symmetry Breaking. Angewandte Chemie International Edition 2002, 41 (7),

1162-1167.

28. Clemente-Juan, J. M.; Coronado, E.; Gaita-Ariño, A., Magnetic polyoxometalates:

from molecular magnetism to molecular spintronics and quantum computing.

Chemical Society Reviews 2012, 41 (22), 7464-7478.

29. Crans, D. C.; Mahroof-Tahir, M.; Anderson, O. P.; Miller, M. M., X-ray Structure

of (NH4) 6 (Gly-Gly) 2V10O28. cntdot. 4H2O: Model Studies for

Polyoxometalate- Protein Interactions. Inorganic Chemistry 1994, 33 (24), 5586-

5590.

30. Wang, S.-M.; Liu, L.; Chen, W.-L.; Wang, E.-B.; Su, Z.-M., Polyoxometalate–

anatase TiO 2 composites are introduced into the photoanode of dye-sensitized

solar cells to retard the recombination and increase the electron lifetime. Dalton

Transactions 2013, 42 (8), 2691-2695.

31. Jin, L.; Fang, Y.; Hu, P.; Zhai, Y.; Wang, E.; Dong, S., Polyoxometalate-based

inorganic–organic hybrid film structure with reversible electroswitchable

fluorescence property. Chemical Communications 2012, 48 (15), 2101-2103.

32. Kozhevnikov, I., Catalysts for fine chemical synthesis, catalysis by

polyoxometalates. Wiley: 2002; Vol. 2.

28 33. Rhule, J. T.; Hill, C. L.; Judd, D. A.; Schinazi, R. F., Polyoxometalates in medicine.

Chemical Reviews 1998, 98 (1), 327-358.

34. Gómez-Romero, P.; Cuentas-Gallegos, K.; Lira-Cantú, M.; Casan-Pastor, N.,

Hybrid nanocomposite materials for energy storage and conversion applications.

Journal of materials science 2005, 40 (6), 1423-1428.

35. Liu, T.; Diemann, E.; Li, H.; Dress, A. W.; Müller, A., Self-assembly in aqueous

solution of wheel-shaped Mo 154 oxide clusters into vesicles. Nature 2003, 426

(6962), 59.

36. Kistler, M. L.; Bhatt, A.; Liu, G.; Casa, D.; Liu, T., A Complete

Macroion−“Blackberry” Assembly− Macroion Transition with Continuously

Adjustable Assembly Sizes in {Mo132} Water/Acetone Systems. Journal of the

American Chemical Society 2007, 129 (20), 6453-6460.

37. Liu, T.; Langston, M. L.; Li, D.; Pigga, J. M.; Pichon, C.; Todea, A. M.; Müller, A.,

Self-recognition among different polyprotic macroions during assembly processes

in dilute solution. Science 2011, 331 (6024), 1590-1592.

38. Yin, P.; Zhang, Z.-M.; Lv, H.; Li, T.; Haso, F.; Hu, L.; Zhang, B.; Bacsa, J.; Wei,

Y.; Gao, Y., Chiral recognition and selection during the self-assembly process of

protein- mimic macroanions. Nature communications 2015, 6, 6475.

39. Leroy, F.; Miró, P.; Poblet, J. M.; Bo, C.; Bonet Á valos, J., Keggin polyoxoanions

in aqueous solution: Ion pairing and its effect on dynamic properties by molecular

dynamics simulations. The Journal of Physical Chemistry B 2008, 112 (29), 8591-

8599.

40. Lopez, X.; Nieto-Draghi, C.; Bo, C.; Avalos, J. B.; Poblet, J. M., Polyoxometalates

in solution: molecular dynamics simulations on the α-PW12O403-Keggin anion in

aqueous media. The Journal of Physical Chemistry A 2005, 109 (6), 1216-1222.

41. Keggin, J., The structure and formula of 12-phosphotungstic acid. Proc. R. Soc.

Lond. A 1934, 144 (851), 75-100.

42. Núñez, S.; Venhorst, J.; Kruse, C. G., Target–drug interactions: first principles and

their application to drug discovery. Drug discovery today 2012, 17 (1-2), 10-22.