Design of Electronic Devices Using Redox-Active Organic Molecules and Their Porous Coordination Networks (Springer Theses) 981163906X, 9789811639067

Table of contents :
Supervisor’s Foreword
Abstract
Parts of this paper have been published in the following journal articles.
Acknowledgements
Contents
1 General Introduction
1.1 Redox-Active Organic Molecule
1.1.1 Redox (Oxidation–Reduction) Reaction
1.1.2 Redox-Active Organic Molecules
1.1.3 Phenalenyl Chemistry
1.2 Resistive Random Access Memory (ReRAM)
1.2.1 Conventional Memory Devices
1.2.2 Resistive RAM (ReRAM or RRAM)
1.2.3 Structure and General Characteristic of ReRAMs
1.2.4 Materials and Switching Mechanisms in ReRAMs
1.3 Redox-Active Porous Coordination Networks (PCNs)
1.3.1 Role of Redox-Activity in Metal–Ligand Cooperated System
1.3.2 Redox-Active Ligands
1.3.3 Porous Coordination Networks
1.3.4 Ligand-Based Redox-Active PCNs
1.3.5 Conductive PCNs
1.3.6 Chemiresistive Sensor Based on PCNs
1.4 Post-synthetic Modification (PSM) of PCNs
References
2 Resistive Switching Memory Devices Based on a Redox-Active Organic Molecule
Abstract
2.1 Introduction
2.2 Results and Discussion
2.2.1 Selective Formation of Redox-Active TPDAP Films
2.2.2 Current–Voltage (I–V) Characteristics of TPDAPs
2.2.3 Ex Situ Study of Resistive Switching Memory Devices
2.2.4 Selective Formation of Redox-Inert TPHAP Films
2.2.5 Current–Voltage (I–V) Characteristics of TPHAPs
2.2.6 Switching Mechanism
2.3 Conclusion
2.4 Experimental Section
2.4.1 General Experimental Information
2.4.2 Syntheses
2.4.3 Device Fabrication
2.4.4 X-ray Structure Analysis
References
3 Chemiresistive Sensor Based on Redox-Active Porous Coordination Networks
Abstract
3.1 Introduction
3.2 Results and Discussion
3.2.1 Selective Formation of Redox-Active PCNs
3.2.2 Film Fabrication of Redox-Active PCNs
3.2.3 Electrical Properties of Redox-Active PCNs
3.2.4 Chemiresistive Humidity Sensing Ability of Redox-Active PCN
3.2.5 Mechanism Investigation
3.3 Conclusion
3.4 Experimental Section
3.4.1 General Experimental Information
3.4.2 Syntheses
3.4.3 X-Ray Structure Analysis
3.4.4 Electrical Measurement
References
4 Tunable Electrical Properties of Redox-Active Porous Coordination Networks via Post-synthetic Modification
Abstract
4.1 Introduction
4.2 Results and Discussion
4.2.1 Pore Opening of a Redox-Active PCN by PSMs
4.2.2 Efficient Oxidation of Open Structure After PSMs
4.2.3 Tunable Electrical Properties via PSMs
4.3 Conclusion
4.4 Experimental Section
4.4.1 General Experimental Information
4.4.2 Syntheses
4.4.3 X-Ray Structure Analysis
4.4.4 Electrical Measurement
References
5 Summary and Outlook
5.1 Summary
5.2 Limitations and Future Work
References

Citation preview

Springer Theses Recognizing Outstanding Ph.D. Research

Jaejun Kim

Design of Electronic Devices Using RedoxActive Organic Molecules and Their Porous Coordination Networks

Springer Theses Recognizing Outstanding Ph.D. Research

Aims and Scope The series “Springer Theses” brings together a selection of the very best Ph.D. theses from around the world and across the physical sciences. Nominated and endorsed by two recognized specialists, each published volume has been selected for its scientific excellence and the high impact of its contents for the pertinent field of research. For greater accessibility to non-specialists, the published versions include an extended introduction, as well as a foreword by the student’s supervisor explaining the special relevance of the work for the field. As a whole, the series will provide a valuable resource both for newcomers to the research fields described, and for other scientists seeking detailed background information on special questions. Finally, it provides an accredited documentation of the valuable contributions made by today’s younger generation of scientists.

Theses may be nominated for publication in this series by heads of department at internationally leading universities or institutes and should fulfill all of the following criteria • They must be written in good English. • The topic should fall within the confines of Chemistry, Physics, Earth Sciences, Engineering and related interdisciplinary fields such as Materials, Nanoscience, Chemical Engineering, Complex Systems and Biophysics. • The work reported in the thesis must represent a significant scientific advance. • If the thesis includes previously published material, permission to reproduce this must be gained from the respective copyright holder (a maximum 30% of the thesis should be a verbatim reproduction from the author's previous publications). • They must have been examined and passed during the 12 months prior to nomination. • Each thesis should include a foreword by the supervisor outlining the significance of its content. • The theses should have a clearly defined structure including an introduction accessible to new PhD students and scientists not expert in the relevant field. Indexed by zbMATH.

More information about this series at http://www.springer.com/series/8790

Jaejun Kim

Design of Electronic Devices Using Redox-Active Organic Molecules and Their Porous Coordination Networks Doctoral Thesis accepted by Tokyo Institute of Technology, Tokyo, Japan

123

Author Dr. Jaejun Kim Department of Chemistry Tokyo Institute of Technology Tokyo, Japan

Supervisor Prof. Masaki Kawano Department of Chemistry Tokyo Institute of Technology Tokyo, Japan

ISSN 2190-5053 ISSN 2190-5061 (electronic) Springer Theses ISBN 978-981-16-3906-7 ISBN 978-981-16-3907-4 (eBook) https://doi.org/10.1007/978-981-16-3907-4 © Springer Nature Singapore Pte Ltd. 2021 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Singapore Pte Ltd. The registered company address is: 152 Beach Road, #21-01/04 Gateway East, Singapore 189721, Singapore

Supervisor’s Foreword

Jaejun’s Ph.D. research focused on the design of electronic devices based on a redox-active organic molecule and its porous coordination networks. The suitability of the thesis as a doctoral dissertation was judged on the basis of the impact and rarity of the thesis on the research field. In addition, the appropriate experimental methods and theoretical discussion were considered as a criterion. Firstly, the thesis described the resistive switching memory device based on a redox-active organic molecule which showed the exceptional switching stability and high retention. In particular, the molecular orientation and its correlation with switching phenomena were well determined by structure analysis and I-V characteristics measurement. In addition, the mechanistic study was included with elaborate discussion based on theoretical calculation, control experiments, and ex-situ measurements. This research would be expected as a high impact for the industrial field as well as research area. In second part, the redox-active porous coordination networks were utilized for the fabrication of chemiresistive humidity sensor. The synthesis of materials and fabrication methods were described in detail and the performance as a humidity sensor were well characterized. Especially, the sensing mechanism was nicely investigated by X-ray crystallography and spectroscopic measurement using proper experimental conditions. The third part described the structural design for tunable electrical properties using post-synthetic modification. Along with the previous chapter, the limitation of synthesized porous coordination network was overcome by synthetic approaches. Although post-synthetic modification was well established as a synthetic method, the attempt to tune the electrical properties would be useful to design electronic devices based on redox-active porous coordination networks. With the three research chapters, the introduction and outlook of thesis were properly described to show the background and current situation, and the future works.

v

vi

Supervisor’s Foreword

Overall, this thesis is valuable to be a Ph.D. dissertation in terms of rarity of electronic devices based on redox-active organic molecule and nobility for understanding electrical properties in redox-active system.

Tokyo, Japan May 2021

Prof. Masaki Kawano

Abstract

Redox-active organic molecules can play a promising role as flexible electronic elements in addition to the lineup of rigid inorganic electronic elements because the electronic state may be readily controlled by the external energy. Although various kinds of redox-active organic molecules were synthesized, the attempt to apply them to electronic devices has been limited because of lack of proper structural design. Moreover, ligand-based redox-active porous coordination networks (PCNs) have not been well explored because of the limited availability of redox-active ligands. In addition to developing new redox-active organic molecules, to design the electronic devices based on redox-active organic molecules/PCNs it is essential to understand the relationship between the molecular arrangement, electrical properties, and its redox-activity. This thesis aims at the development of electronic devices using a redox-active organic molecule and its PCNs and highlights the importance of molecular arrangement. Redox-active organic molecule, 2,5,8-tri(4-pyridyl)1,3-diazaphenalene (TPDAP), having a large p-plane and multi-intermolecular interactivity was utilized for developing a resistive switching memory device. Furthermore, its PCNs were synthesized to fabricate chemiresistive sensors. The electrical properties were modulated by the post-synthetic modification. Each mechanism was systematically investigated by structural determination together with well-defined control experiments. The general guideline for designing the electronic devices using redox-active organic molecules was proposed. Chapter 1 describes the general introduction to the main contents of this thesis: redox-activity, memory device, porous coordination networks, and post synthetic modification. Chapter 2 presents the design of resistive switching memory device using redox-active organic molecule, TPDAP. The fabrication method of TPDAP films and their microstructure are described. The relationship between the molecular arrangement/redox-activity and electrical properties, especially resistive switching phenomena, was investigated by X-ray analysis, spectroscopic and I-V characteristic measurements. The switching mechanism was proposed based on the control experiments. vii

viii

Abstract

Chapter 3 describes the development of chemiresistive sensor using redox-active PCN composed of TPDAP and Cd2+. The synthesis of redox-active PCNs and their intrinsic electrical properties were investigated. The fabricated PCN films were applied for the humidity chemiresistive and its sensing mechanism was investigated by the single crystal X-ray analysis and IR spectroscopic measurement. In Chap. 4, the synthetic method to modify the redox-active PCN prepared in Chap. 3 was described. The electronic state of modified PCN was characterized by the electron paramagnetic resonance (EPR) spectroscopy. The control of electrical properties by post-synthetic modification was investigated by the X-ray analysis and I-V characteristic measurements.

Parts of this paper have been published in the following journal articles. J. Kim, H. Ohtsu, T. Den, K. Deekamwong, I. Muneta, M. Kawano, “Control of Anisotropy of a Redox-active Molecule-based Film Leads to Non-volatile Resistive Switching Memory”, Chem. Sci. 2019, 10, 10883–10893. J. Kim, J. Y. Koo, Y. H. Lee, T. Kojima, Y. Yakiyama, H. Ohtsu, J. H. Oh, M. Kawano, “Structural Investigation of Chemiresistive Sensing Mechanism in Redox-Active Porous Coordination Network”, Inorg. Chem. 2017, 56, 8735–8738. J. Y. Koo, Y. Yakiyama, J. Kim, Y. Morita, M. Kawano, “Redox Active Diazaphenalenyl-Based Molecule and Neutral Radical Formation”, Chem. Lett. 2015, 8, 1131–1133.

ix

Acknowledgements

I will begin by thanking my advisor, Prof. Masaki Kawano, for providing warm guidance during my Ph.D. course at Tokyo Institute of Technology. His great support has made my research and graduate studies successful and working alongside him for the last six years has been a pleasure owing to not only academic but also personal relationship. His guidance made me not to leave this long rough journey. I sincerely admire him as a scientist, teacher, and senior. Further I would like to thank Prof. Osamu Ishitani, Prof. Tetsuo Okada, Prof. Shinya Koshihara, Prof. Gaku Fukuhara for serving on my thesis committee and offering precious comments and discussion on my thesis. I thank my colleagues in our group who have shared pain, happiness, pleasure and sorrows of last years: Dr. Hiroyoshi Ohtsu, Dr. Yumi Yakiyama, Dr. Tatsuhiro Kojima, Dr. Pavel Usov, Dr. Jin Young Koo, Dr. Gil Ryeong Lee, Dr. Wanuk Choi, Dr. Jooyeon Ha, Mr. Krittanun Deekomwoong, Mr. Taizen Den, Ms. Miho Takakusagi, Mr. Nozomu Odagawa, Mr. Keisuke Nakanishi, Mr. Tatsuya Kanamaru, Mr. Yuki Wada, Mr. Kensuke Ona, Ms. Nana Furuno, Ms. Marie Okuyama. Thanks for my eternal supporter, my family. Their absolute support always protected me from a weak mind and keep my life right. I give my family all glory. I gratefully acknowledge Japanese Government (Monbukagakusho) for funding the my Ph.D. course as a MEXT scholarship student. February 2019

Jaejun Kim

xi

Contents

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

1 1 1 2 4 6 6 7 8 9 10

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

10 10 11 12 13 14 15 16

2 Resistive Switching Memory Devices Based on a Redox-Active Organic Molecule . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.1 Selective Formation of Redox-Active TPDAP Films . . 2.2.2 Current–Voltage (I–V) Characteristics of TPDAPs . . . . 2.2.3 Ex Situ Study of Resistive Switching Memory Devices 2.2.4 Selective Formation of Redox-Inert TPHAP Films . . . . 2.2.5 Current–Voltage (I–V) Characteristics of TPHAPs . . . . 2.2.6 Switching Mechanism . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

21 21 23 23 26 27 31 33 33

1 General Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.1 Redox-Active Organic Molecule . . . . . . . . . . . . . . . . . . . . . 1.1.1 Redox (Oxidation–Reduction) Reaction . . . . . . . . . . . 1.1.2 Redox-Active Organic Molecules . . . . . . . . . . . . . . . 1.1.3 Phenalenyl Chemistry . . . . . . . . . . . . . . . . . . . . . . . 1.2 Resistive Random Access Memory (ReRAM) . . . . . . . . . . . 1.2.1 Conventional Memory Devices . . . . . . . . . . . . . . . . . 1.2.2 Resistive RAM (ReRAM or RRAM) . . . . . . . . . . . . 1.2.3 Structure and General Characteristic of ReRAMs . . . . 1.2.4 Materials and Switching Mechanisms in ReRAMs . . . 1.3 Redox-Active Porous Coordination Networks (PCNs) . . . . . . 1.3.1 Role of Redox-Activity in Metal–Ligand Cooperated System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.3.2 Redox-Active Ligands . . . . . . . . . . . . . . . . . . . . . . . 1.3.3 Porous Coordination Networks . . . . . . . . . . . . . . . . . 1.3.4 Ligand-Based Redox-Active PCNs . . . . . . . . . . . . . . 1.3.5 Conductive PCNs . . . . . . . . . . . . . . . . . . . . . . . . . . 1.3.6 Chemiresistive Sensor Based on PCNs . . . . . . . . . . . 1.4 Post-synthetic Modification (PSM) of PCNs . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

xiii

xiv

Contents

2.3 Conclusion . . . . . . . . . . . . . . . . . . . . . . . 2.4 Experimental Section . . . . . . . . . . . . . . . 2.4.1 General Experimental Information 2.4.2 Syntheses . . . . . . . . . . . . . . . . . . 2.4.3 Device Fabrication . . . . . . . . . . . . 2.4.4 X-ray Structure Analysis . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

37 37 37 37 38 39 41

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

43 43 44 44 45 47

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

48 51 54 55 55 55 56 58 59

4 Tunable Electrical Properties of Redox-Active Porous Coordination Networks via Post-synthetic Modification . . . . 4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2.1 Pore Opening of a Redox-Active PCN by PSMs . . 4.2.2 Efficient Oxidation of Open Structure After PSMs . 4.2.3 Tunable Electrical Properties via PSMs . . . . . . . . . 4.3 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.4 Experimental Section . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.4.1 General Experimental Information . . . . . . . . . . . . 4.4.2 Syntheses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.4.3 X-Ray Structure Analysis . . . . . . . . . . . . . . . . . . . 4.4.4 Electrical Measurement . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

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

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

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

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

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

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

61 61 62 62 64 65 68 68 68 69 69 70 71

5 Summary and Outlook . . . . . . . . 5.1 Summary . . . . . . . . . . . . . . . 5.2 Limitations and Future Work References . . . . . . . . . . . . . . . . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

73 73 75 76

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

3 Chemiresistive Sensor Based on Redox-Active Porous Coordination Networks . . . . . . . . . . . . . . . . . . . . . . . . 3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2 Results and Discussion . . . . . . . . . . . . . . . . . . . . . 3.2.1 Selective Formation of Redox-Active PCNs 3.2.2 Film Fabrication of Redox-Active PCNs . . . 3.2.3 Electrical Properties of Redox-Active PCNs 3.2.4 Chemiresistive Humidity Sensing Ability of Redox-Active PCN . . . . . . . . . . . . . . . . 3.2.5 Mechanism Investigation . . . . . . . . . . . . . . 3.3 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4 Experimental Section . . . . . . . . . . . . . . . . . . . . . . 3.4.1 General Experimental Information . . . . . . . 3.4.2 Syntheses . . . . . . . . . . . . . . . . . . . . . . . . . 3.4.3 X-Ray Structure Analysis . . . . . . . . . . . . . . 3.4.4 Electrical Measurement . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

Chapter 1

General Introduction

1.1 1.1.1

Redox-Active Organic Molecule Redox (Oxidation–Reduction) Reaction

A redox (oxidation–reduction) reaction is a type of chemical reaction that involves a transfer of electrons between two species in which the oxidation number of a molecule, atom, or ion changes by gaining or losing an electron. Oxidation is the loss of electrons or an increase in oxidation state by a molecule, atom, or ion, whereas reduction is the gain of electrons or a decrease in oxidation state. Electrons can move spontaneously from higher energy levels to lower energy levels within an atom. A similar movement can take place between two different chemical reactants. If there are electrons in one reactant that are at higher energy than unfilled orbitals of the other reactant, the higher energy electrons can transfer to the unfilled orbitals at lower energy. There are three common types of redox reactions as follows. Combustion Reaction A combustion reaction is a redox reaction between a compound and molecular oxygen (O2) to form oxygen-containing products. When one of the reactants is a hydrocarbon, the products include carbon dioxide and water. The following reaction is the combustion of octane, a hydrocarbon. Octane is a component of gasoline, and this combustion reaction occurs in the engine of most cars: 2C8 H18 þ 25O2 ! 16CO2 ðgÞ þ 18H2 O Disproportionation Reactions A disproportionation reaction (or auto-oxidation reaction) is a reaction in which a single reactant is both oxidized and reduced. The following reaction is for the disproportionation of hypochlorite, ClO−: © Springer Nature Singapore Pte Ltd. 2021 J. Kim, Design of Electronic Devices Using Redox-Active Organic Molecules and Their Porous Coordination Networks, Springer Theses, https://doi.org/10.1007/978-981-16-3907-4_1

1

2

1 General Introduction  3ClO ðaqÞ ! ClO 3 ðaqÞ þ 2Cl ðaqÞ

The reactant ClO− is oxidized to ClO 3 where the oxidation number of chlorine increases from +1 to +5, at the same time, the chlorine of ClO− is reduced to Cl where the oxidation number decreases from +1 to −1. Replacement Reactions Displacement reactions, also known as replacement reactions, involve compounds and the “replacing” of elements. They occur as single and double replacement reactions. A single replacement reaction (or single displacement reaction) involves two elements trading places within a compound. For example, many metals react with dilute acids to form salts and hydrogen gas. The following reaction shows zinc replacing hydrogen in the single replacement reaction between zinc metal and aqueous hydrochloric acid: ZnðsÞ þ 2HClðaqÞ ! ZnCl2 ðaqÞ þ H2 ðgÞ A double replacement reaction is similar to a single replacement reaction, but involves “replacing” two elements in the reactants with two in the products: Fe2 O3 ðsÞ þ 6HClðaqÞ ! 2FeCl3 ðaqÞ þ 3H2 O Redox reaction is common and essential to some of the basic functions of life, including, photosynthesis, respiration, combustion, corrosion, and metabolism. In fact, our technology, from fire to redox-flow batteries, is largely based on redox reactions, which takes place through either a simple process, such as the burning of carbon in oxygen to yield carbon dioxide (CO2), or a more complex process such as the oxidation of glucose (C6H12O6) in the human body through a series of electron transfer processes and the charge/discharge process equipped with the membrane in the electrochemical cell. When a chemical reaction is caused by an externally supplied current, as in electrolysis, or if an electric current is produced by a spontaneous chemical reaction as in battery, it is called an electrochemical reaction. In general, electrochemistry describes the overall reactions when individual redox reactions are separate but connected by an external electric circuit and an intervening electrolyte.

1.1.2

Redox-Active Organic Molecules

Organic molecules are typically composed of earth-abundant elements, and thus, their cost and availability are less constrained by the production and reserves of rarer elements. Transition metals must be mined and refined in ore; however, organic molecules can conceivably be synthesized in a sustainable way using green chemistry routes. More importantly, while the known periodic table limits the

1.1 Redox-Active Organic Molecule

3

number of inorganic redox couples, a wider array of redox-active organic molecules is available, allowing for the realization of new redox couples. It is also considered as a “designable” structure by modification of redox moieties, substitute, neighboring molecules, and molecular orientation, enable to tune the redox potential and physical properties [1]. Due to those advantages, numerous redox-active organic molecules have been synthesized in various applications such as catalysts, redox-flow cells, capacitors, and redox medicine. In general, redox activity was determined by the functional group in the molecules. Scheme 1.1 describes the representative redox-active organic molecules classified by the functional groups. Quinones are oxidized derivatives of aromatic compounds and are often readily made from reactive aromatic compounds with electron-donating substituents such as phenols and catechols, which increase the nucleophilicity of the ring and contributes to the large redox potential needed to break aromaticity. The N–H center for imides derived from ammonia is acidic and can participate in hydrogen bonding. The thiol group contains R–SH functional group. Thiols are structurally similar to the alcohol groups, but these functional groups have very different chemical properties. Thiols are more nucleophilic, more acidic, and more readily oxidized. Carbazole is readily oxidized but may form various products depending on the reaction conditions [2]. Triphenyl amine can be easily oxidized to form stable radical cations as long as the para-position of the phenyl rings is protected, and the oxidation process is always associated with a strong color change.

Scheme 1.1 Representative redox-active organic molecules

4

1.1.3

1 General Introduction

Phenalenyl Chemistry

Phenalenyl, which arises from a triangular fusion of three benzene rings, is one of the most fundamental delocalized neutral radicals, discovered in 1956 [3]. Phenalenyl chemistry has long played an important role in organic radical chemistry because of its high stability with highly symmetric (D3h) structure and unique electronic structure. An unpaired electron in the resonance structure of phenalenyl radical is delocalized on six equivalent carbon atoms (1, 3, 4, 6, 7, and 9-positions). This non-bonding character of the singly occupied molecular orbital (SOMO) provides the same p-electron delocalization energy in all the redox states (cation, neutral radical, and anion), so that it is expected to behave as a good amphoteric redox system [4]. Phenalenyl radical can be generated by the air oxidation of phenalene and later by the air oxidation of phenalenyl anion. The long-lived radicals were stable in the absence of air and can be dimerized in a condensed phase or in the solution state at low temperature. Many researchers have focused on this amphoteric character of phenalenyl radical, and Haddon reported its derivatives would be good candidates for single-component organic metals and superconductors [5]. Since phenalenyl chemistry had been initialized, synthetic efforts have been made to extend this by preparating phenalenyl derivatives such as alkoxy, hydroxyl, N–S–N groups, and spiro-conjugated to isolate the radical itself in the crystalline state. Especially, Morita et al. synthesized 1,3-diazaphenalenyl (DAP) and its derivatives which are typical examples of the isoelectronic mode of heteroatomic modification for phenalenyl [6]. The incorporation of nitrogen atoms into a phenalenyl radical may affect the stabilization of radical species owing to the higher electronegativity of nitrogen than that of carbon [7]. Su et al. reported another form of phenalenyl, hexaazaphenalenyl (HAP) anion, introducing the six nitrogen atoms into phenalenyl skeleton [8, 9]. HAP is a highly symmetric heterocycle with full nitrogen substitution in all of the a sites of phenalenyl. Because of the directionality of lone-pair electrons, a radially extended mode is expected for metal-coordination and hydrogen bonding interactions. In addition, a HAP anion showed exceptionally higher stability among the phenalenyl anions because of its low HOMO energy level compared to DAP. Recently, HAP and DAP derivatives by substitution with three pyridines were synthesized as next-generation molecules. In 2012, Yakiyama et al. synthesized 2,5,8-tri(4′-pyridyl)-1,3,4,6,7,9-hexaazaphenalene (Scheme 1.2, TPHAP) and showed its multi-interactivity could generate various kinds of metastable coordination networks owing to its multi-interactivity [10, 11]. Since TPHAP still has a low HOMO level similar with HAP, it was difficult to apply in relation to the redox property. This limitation had encouraged to prepare redox-active molecule, 2,5,8-tri (4-pyridyl)1,3-diazaphenalene (Scheme 1.2, TPDAP), in which three pyridyl groups are introduced into the DAP skeleton [12]. Cyclic voltammetry of TPDAP in a solution state (Fig. 1.1a) revealed two irreversible oxidation peaks at +0.30 V and +0.68 V (vs. Fc0/Fc+ in DMF). A TPDAP radical (7) was generated

1.1 Redox-Active Organic Molecule

5

Scheme 1.2 Molecular structures of 2,5,8-tri(4′-pyridyl)-1,3,4,6,7,9-hexaazaphenalene (TPHAP) and 2,5,8-tri(4-pyridyl)1,3-diazaphenalene (TPDAP)

at +0.30 V and the cation (7+) at + 0.68 V. Both of the peaks showed anodic shifts from those of DAP (E1OX = −0.04 V, E2OX = +0.27 V in DMF) attributable to the substitution effect with three electron withdrawing pyridine rings as predicted by the energy diagram. The first oxidation potential is similar to that of 1,4-hydroquinone (E1OX = +0.26 V), indicating the donating ability of 7−. However, because of the successive reaction of the oxidized species, no reduction peak was observed. On the other hand, the solid-state CV using crystals showed two reversible redox peaks (E1red = − 0.35 V, E1OX = − 0.04 V; E2red = +0.28 V, E2OX = +0.44 V) (Fig. 1.1b). In addition, its porous coordination networks show the anisotropic conductivity which only appears in the p–p stacking direction, and radical-induced oxidation which initiated the formation of dimerization and further generates radicals [13]. The further research using redox-active TPDAP is topic in this thesis (Chaps. 2, 3, and 4).

Fig. 1.1 Solution- and solid-state CVs of H+1− at 298 K. a Solution-state CV measured in 0.1 M −1 Et4 N þ PF 6 anhydrous DMF solution using a glassy carbon (GC) electrode. Scan rate: 100 mVs . b Solid-state CV measured in 0.1 M n-Bu4 N þ PF CH CN solution using a GC electrode. Scan 3 6 rate: 500 mVs−1. The x signs and arrows indicate the starting points and the direction of the forward scan, respectively. Reproduced from Ref. [12] with permission

6

1.2 1.2.1

1 General Introduction

Resistive Random Access Memory (ReRAM) Conventional Memory Devices

An electronic memory is a component, device, or recording medium used for data storage. As the information generation has been growing, memory devices are getting closer to our life. The technological advancement in almost all devices and products has become possible with the development of memory devices. Over the past two decades, memory technology has been rapidly developed according to demand for small cell size, cheap price, and high retention. Figure 1.2 shows the classification of memory devices including conventional and emerging memory devices [14]. Depending on the recyclability, memory devices were classified by random access memory (RAM) and read-only memory (ROM). RAM can be written to or read from any cells without read/write limitations, whereas ROM cannot be re-written. RAMs can be further classified into volatile RAM and non-volatile RAM. In general, volatile RAMs, all ROMs, and flash are considered as conventional memories. In particular, DRAM and flash memory are widely produced and the largest market in current device market. DRAM stores each bit of data into a separate capacitor which can be discharged or charged, and these two states represent the two values of a bit (0 and 1). Because the capacitor leaks charge, the information cannot be kept unless the capacitor charge is refreshed periodically. The advantage of DRAM resides in its simplicity: Only one capacitor and one transistor (metal-oxide semiconductor field effect transistor, MOSFET) are required per bit. This enables DRAM to reach very high

Fig. 1.2 Classification of memory devices

1.2 Resistive Random Access Memory (ReRAM)

7

densities. DRAM was traditionally used for long time in the semiconductor industry. However, flash memory has caught up with DRAMs and now are scaling ahead in terms of storage density as well as minimum feature size. Flash memory was invented by Dr. Fujio Masuoka in 1980 at Toshiba [15]. A basic flash memory cell consists of a storage transistor with a control gate and a floating gate, which is insulated from the rest of the transistor by a thin dielectric material or oxide layer. The floating gate stores the electrical charge and controls the flow of the electrical current. Electrons are added to or removed from the floating gate to change the storage transistor’s threshold voltage. Changing the voltage affects whether a cell is programmed as a 0 or a 1. Since it can keep the charge in the floating layer, it is non-volatile. Although DRAM and flash memories were widely developed in the current industry, they have some important disadvantages. Flash memory has long program time (write and erase times of 1 ms and 0.1 ms, respectively) and high cost (0.5 $/ MB). DRAM has even faster with program time in the range of ns and cheaper (1.5 $/MB); however, it has inexcusable problem, volatility. More importantly, both devices with complicated architecture are getting closer to the limitation of integrated density, which is leading to finding new memory device.

1.2.2

Resistive RAM (ReRAM or RRAM)

ReRAM is composed of an electrode/insulator/electrode sandwich structure. By applying the voltage, the high resistance state of the active layer is changed to the low resistance state. With a simple architecture, additionally, it has good complementary metal-oxide semiconductor (CMOS) compatibility, which is essential for its practical applications and mass productions. In 1962, for the first time, Hickmott [16] observed large negative differential resistance in five thin anodic oxide films including silicon oxide (SiOx), aluminum oxide (Al2O3), tantalum pentoxide (Ta2O5), zirconium dioxide (ZrO2), and titanium dioxide (TiO2). Subsequently, more materials were demonstrated to show resistive switching, and the switching mechanisms started to be explored as well. However, high operation voltage and current, poor endurance, and manufacturing sensitivity have made it less favorable than silicon-based memory devices. Thanks to the remarkable progress in material science, however, ReRAM has started to attract the academia again. In 2000, Liu et al. [17] and Beck et al. [18] found an electric-field-induced resistance change effect in a perovskite-type oxide. This finding accelerated the research on ReRAM using various kinds of materials, such as dielectric, paraelectric, ferroelectric, ferromagnetic, and semiconducting materials. Zhuang et al. reported the first possibility for practical application of ReRAM, which showed the 640 bit ReRAM array using 0.5 mm CMOS process [19]. In that study, ReRAM first showed the “resistive” results using perovskite oxides such as Pr0.7Ca0.3MnO3 enabled memory switching through modulation of voltages with small voltage and current. In 2004, transition metal oxides replaced perovskite oxides because of the critical

8

1 General Introduction

manufacturing limitations, such as a difficulty of controlling stoichiometry and contamination [18]. In 2010s, as many researchers have intensively studied, ReRAM has reached to exhibit excellent miniaturization potential down to 10 nm [20], sub-nm operation speed [21], very low energy consumption ( 2r(I), 280 parameters, 2.598º < h < 25.054º, R1 = 0.0659, wR2 = 0.1752, GOF = 0.932.

References

41

References 1. Meijer GI (2008) Science 319:1625–1626 2. Burr GW, Kurdi BN, Scott JC, Larn CH, Gopalakrishnan K, Shenoy RS (2008) IBM J Res Dev 52:449–464 3. Akinaga H, Shima H (2010) Proc IEEE 98:2237–2251 4. Waser R, Dittmann R, Staikov G, Szot K (2009) Adv Mat 21:2632–2663 5. Waser R, Aono M (2007) Nat Mat 6:833–840 6. Chang T-C, Chang K-C, Tsai T-M, Chu T-J, Sze SM (2016) Mater Today 19:254–264 7. Lelmini D, Philip Wong H-S (2018) Nat Electron 1:333–343 8. Mittal S (2018) Mach Learn Knowl Extr 1:75–114 9. Cho B, Song S, Ji Y, Kim T-W, Lee T (2011) Adv Funct Mater 21:2806–2829 10. Zhu L, Zhou J, Guo Z, Sun Z (2015) J. Materiomics 1:285–295 11. Wang C, Gu P, Hu B, Zhang Q (2015) J Mater Chem C 3:10055–10065 12. Pan F, Gao S, Chen C, Song C, Zeng F (2014) Mater Sci Eng R 83:1–59 13. Gao S, Yi X, Shang J, Liu G, Li R-W (2019) Chem Soc Rev, Advance Article 14. Zhang M, Long S, Wang G, Li Y, Xu X, Liu H, Liu R, Wang M, Li C, Sun P, Sun H, Liu Q, Lü H, Liu M (2014) Chin Sci Bull 59:5324–5337 15. Ding Y, Chang C, Zhang L, Zhou Y, Yu G (2018) Chem Soc Rev 57:69–103 16. Zhu H, Li Q (2016) Appl Sci 6:1–15 17. Araujo RB, Banerjee A, Panigrahi P, Yang L, Strømme M, Sjödin M, Araujo CM, Ahuja R (2017) J Mater Chem A 5:4430–4454 18. Koo JY, Yakiyama Y, Lee GR, Lee J, Choi HC, Morita Y, Kawano M (2016) J Am Chem Soc 136:1776–1779 19. Wei X, Xu W, Huang J, Zhang L, Walter E, Lawrence C, Vijayakumar M, Henderson WA, Liu T, Cosimbescu L, Li B, Sprenkle V, Wang W (2015) Angew Chem Int Ed 54:8684–8687 20. Armstrong CG, Toghill KE (2018) Electrochem Commun 91:19–24 21. Koo JY, Yakiyama Y, Kim J, Morita Y, Kawano M (2015) Chem Lett 44:1131–1133 22. Kim J, Ohtsu H, Den T, Deekamwong K, Muneta I, Kawano M (2019) Chem Sci 10:10883– 10893 23. Yakiyama Y, Ueda A, Morita Y, Kawano M (2012) Chem Comm 48:10651–10653 24. Wang ZS, Zeng F, Yang J, Chen C, Yang YC, Pan F (2010) Appl Phys Lett 97:253301 25. Yakiyama Y, Lee GR, Kim SY, Matsushita Y, Morita Y, Park MJ, Kawano M (2015) Chem Commun 51:6828–6831 26. Wright GT (1958) Nature 182:1296–1297

Chapter 3

Chemiresistive Sensor Based on Redox-Active Porous Coordination Networks

Abstract By changing the rate of evaporation, two kinds of crystalline films composed of redox-active porous coordination networks (PCN 1 and 2) were selectively prepared on a gold-patterned substrate using a DMF solution of 2,5,8-tri (4-pyridyl)1,3-diazaphenalene and Cd(NO3)2. The film 1 showed highly sensitive humidity sensing ability. Single-crystal structures and infrared spectroscopic analyses before/after hydration of single crystal of PCN 1 revealed the sensing mechanism: Exchange of nitrate ions with water on Cd atoms occurred in hydrated conditions to generate conductive cationic network.

3.1

Introduction

Porous coordination networks (PCNs) [1–8] are self-assembled materials which are composed of metal ions and organic linkers, of which various combinations enable to construct designable structure. While PCNs have widely been developed because of their structural/functional interests, chemical sensor application has been considered as one of the fascinating targets because the large surface area and chemical tunability enable it to change chemical and physical properties which can be influenced by host–guest interaction in the presence of analyte molecules [9–14]. However, PCN-based chemiresistive sensors are rare because of their ordinary insulating property which comes from the poor orbital overlap between the p orbitals of organic ligands and the d orbitals of the metal ions. Although several conductive PCNs have been reported using redox-active molecules which provide unpaired electrons or facile charge transfer between nodes [15–20], no reported chemiresistive sensors have attested their sensing mechanisms with consideration of structural change [21–23]. Another barrier to apply PCNs for various electronic devices is a difficulty of integration with substrates. While PCNs are synthesized as bulk material, practical application in functional nanomaterials will require the use of PCNs film to increase the availability and efficient use of the pore in the PCNs. Although several growth methods have been developed in the last few years [18, 24–28], such as direct © Springer Nature Singapore Pte Ltd. 2021 J. Kim, Design of Electronic Devices Using Redox-Active Organic Molecules and Their Porous Coordination Networks, Springer Theses, https://doi.org/10.1007/978-981-16-3907-4_3

43

3 Chemiresistive Sensor Based …

44

growth/deposition from solvothermal synthesis [29–39], layer-by-layer (LBL) approach [40, 41] and electrochemical method [42, 43], they were generally a time-consuming process and sometimes cannot apply for specific linkers and metals. In this chapter, the author reports a redox-active PCN-based chemiresistive humidity sensor and its sensing mechanism investigated by single-crystal X-ray analysis and infrared (IR) spectroscopy [44]. The PCN film fabricated from a redox-active ligand, 2,5,8-tri(4-pyridyl)1,3-diazaphenalene (TPDAP) [45], and Cd (NO3)2 by the simple solvent evaporation showed resistivity change depending on a trace amount of water. The mechanism study using X-ray analysis/IR spectroscopy revealed that the dissociation of nitrate ions from Cd2+ under hydrated conditions plays a crucial role in achieving the electrical conductivity. Understanding of the correlation between conductivity and structure is essential to designing chemiresistive sensors using PCNs.

3.2 3.2.1

Results and Discussion Selective Formation of Redox-Active PCNs

Two kinds of PCNs composed of TPDAP, nitrate, and Cd2+ were selectively prepared by controlling the rate of solvent evaporation. Considering the solubility of TPDAP and proper evaporation conditions, DMF was selected for the crystallization. Because DMF is readily decomposed in the presence of water/metal even at temperatures lower than the boiling point, the evaporation of decomposed products resulted in network formation. PCN 1, [Cd2.5(NO3)5(TPDAP)3(DMF)3(H2O)7.4], (orthorhombic Cmma) was synthesized by heating a DMF solution of TPDAP and Cd(NO3)2
4H2O in sealed conditions at 90 °C. The reddish block crystals of PCN 1 were densely crystallized on the bottom of a container. The single-crystal X-ray analysis revealed that PCN 1 has a p–p stacking columnar structure and 8 Å-sized diagonal channels along the [1 1 0] direction (Fig. 3.1). In contrast to the sealed conditions, the fast evaporation under open conditions at 90 °C produced reddish thin plate crystals of a totally different structure, [Cd1.5(NO3)2.77(HTPDAP)0.77(TPDAP)0.23(DMF)2(H2O)] (PCN 2; monoclinic C2/c). The single-crystal X-ray analysis revealed that PCN 2 has a 2D slipped p–p stacking structure possessing 3D-tangled pores with the window size of 7 Å (Fig. 3.2).

3.2 Results and Discussion

45

Fig. 3.1 Crystal structure of PCN 1: a asymmetric unit b packing c simulated void space using contact surface calculated on Mercury 3.7 (23.4% (8462.80 Å3), cavity dimension: 18.38 Å  8.25 Å). Hydrogen atoms and solvents in a pore are omitted for clarity. Blue, N; Yellow, Cd; Red, O. Reprinted with permission from Ref. [44]. Copyright (2017) American Chemical Society

Fig. 3.2 Crystal structure of PCN 2: a asymmetric unit b packing c simulated void space using contact surface calculated on Mercury 3.7 (12.6% (1024.61 Å3), cavity dimension: 7.01 Å  7.04 Å). Hydrogen atoms and solvents in a pore are omitted for clarity. Blue, N; Yellow, Cd; Red, O. Reprinted with permission from Ref. [44]. Copyright (2017) American Chemical Society

3.2.2

Film Fabrication of Redox-Active PCNs

The simple solvent evaporation method was applied for the device fabrication based on the redox-active PCNs of 1 and 2 on an Au-patterned SiO2/Si wafer. By changing the rate of solvent evaporation, dense and uniform crystalline films of PCN 1 and 2 were selectively prepared on the substrate by replacing a glass plate with a wafer under the same condition as their single-crystal growth conditions (Fig. 3.3a). The solvent evaporation method is not conventional for film fabrication of ordinary PCNs because PCNs are apt to crystallize as a free-standing crystal, which means the specific substrates are mostly needed to control its growth. However, a large p-planar ligand, TPDAP, and crystallization mechanism of networks enable to deposit crystalline films on the non-functionalized substrate uniformly and densely. The powder diffraction patterns of both films are well matched with those of both single crystals (Fig. 3.3b). However, the 2D layer structure of

46

3 Chemiresistive Sensor Based …

Fig. 3.3 a Optical microscopic images of films. b Powder X-ray diffractions of single crystals and films measured under air. It should be noted that powder X-ray diffractions of film 2 and single crystals 2 measured under air do not indicate the structure of initial PCN 2. Reprinted with permission from Ref. [44]. Copyright (2017) American Chemical Society

PCN 2 drastically changed after exposure to air. It was observed that severe cracks and color change gradually appeared in a single crystal of PCN 2 by exposure to air from the mother solution. The powder X-ray diffraction of single crystals of PCN 2 measured under air was significantly different from the pattern simulated from single-crystal X-ray analysis measured at 100 K. This result can be understood that faster reaction generated a kinetic network in multi-interactive system [46, 47]. In addition, film 1 showed preferred orientation along the p–p stacking direction corresponding to 2h = 25.8°. Therefore, the pore of film 1 was efficiently open for guest encapsulation, and it encouraged to apply the film 1 for the chemical sensing application.

3.2 Results and Discussion

3.2.3

47

Electrical Properties of Redox-Active PCNs

In order to investigate the intrinsic electrical properties of redox-active PCNs, the current–voltage curves of both single crystals were measured with silver paste as an electrode. As expected from the redox activity of TPDAP, single crystals of PCN 1 and 2 showed electron conductivity of 1.80  10−8 S cm−1 (23 ± 1 °C, 72% RH, in air) and 1.05  10−7 S cm−1 (23 ± 1 °C, 72% RH, in air), respectively, along the p–p stacking direction (Fig. 3.4). Koo et al. reported that radical TPDAP species in Cd-1mono generated by air oxidation considerably increased the conductivity along the p–p stacking direction [48]. Both PCN 1 and 2 were synthesized under air, and it also contained the radical species which contributed the electrical conductivity. The difference of the conductivity value between PCN 1 and 2 can be rationalized by the p–p distance: The p–p distance (3.3463(73) Å or shorter) in the 2D structure of PCN 2 is shorter than that of PCN 1 (3.4664(280) Å). The simple architecture of the deposited films on Au-patterned wafers enabled to measure current–voltage characteristics by applying voltage via the gold electrode. Although film 1 and 2 showed conductivity (Fig. 3.5), the accurate conductivity value could not be obtained because the determination of channel area was difficult. However, the stability of films could be confirmed by monitoring the change of conductance in one channel under air. While the current of film 1 slightly decreased during the measurement, that of film 2 significantly decreased (Fig. 3.6). This result was expectedly in line with the structural stability. In the other words, film 1 composed of 3D network showed better electrical stability than film 2.

Fig. 3.4 Current–voltage characteristics of a single crystal of PCN 1 and b single-crystal PCN 2. The current measurements of both crystals were performed along the p–p stacking direction. Crystal dimensions of PCN 1: A = 100 lm  80 lm, L = 105 lm; PCN 2: A = 20 lm  20 lm, L = 105 lm. Reprinted with permission from Ref. [44]. Copyright (2017) American Chemical Society

3 Chemiresistive Sensor Based …

48

Fig. 3.5 Current–voltage characteristics of a film 1 and b film 2. The nonlinearity might originate from the relatively poor contact of the active layer with the electrode. The exact conductivity value could not be estimated because of the difficulty in determining the precise contact area. Reprinted with permission from Ref. [44]. Copyright (2017) American Chemical Society

Fig. 3.6 Conductance change of film 1 and film 2 under air. The initial value was measured in N2. Reprinted with permission from Ref. [44]. Copyright (2017) American Chemical Society

3.2.4

Chemiresistive Humidity Sensing Ability of Redox-Active PCN

Film 1 composed of a very dense crystalline layer with open channels showed a significant change in its resistivity for humidity with high sensitivity and selectivity (vide infra). Firstly, air/vacuum-dependent conductivity of film 1 was investigated. Conductive film 1 in air (on-state) was changed to an electrically quenched state (off-state) under vacuum (10−6 torr). The conductivity recovered when film 1 was exposed to air again, even though the value slightly decreased due to the decay in crystallinity (Fig. 3.7). To study the electric switching mechanism, current change under humid/dry N2 conditions was measured. When the humid N2 gas was introduced to the devices by passing the N2 stream through a water bath, the current significantly increased by

3.2 Results and Discussion

49

Fig. 3.7 Current–voltage characteristics of film 1 under air (black), vacuum (red), and re-air (blue) (initial condition: 15 ± 1 °C, 68.1% RH). Reprinted with permission from Ref. [44]. Copyright (2017) American Chemical Society

Fig. 3.8 a Current change of film 1 under humid N2 (red)/dry N2 (green) when the applied voltage was 10 V (initial condition: 12 ± 1 °C, 50% RH) b current change of film 1 under vaporized EtOH/acetone when the applied voltage was 10 V (initial condition: 12 ± 1 °C, 52% RH). Yellow area indicates the duration of gas flowing. Reprinted with permission from Ref. [44]. Copyright (2017) American Chemical Society

two orders of magnitude. In contrast, when dry N2 gas was introduced to the device, the current decreased by two orders of magnitude compared with the initial state (Fig. 3.8a). Notably, similarly to the results upon exposure to dry N2, other vapors, such as ethanol and acetone, also reduced the conductivity, leading to the electrical off-state (Fig. 3.8b). It is considered that they decreased the relative humidity compared to air condition, indicating that only water can promote high conductive state among vapors that we investigated. On the other hand, noteworthy is that, as a control experiment, we confirmed that a non-porous film composed of only TPDAP showed no humidity sensing ability

50

3 Chemiresistive Sensor Based …

(Fig. 3.9). Because the TPDAP film has highly packing structure which does not allow any pore, the water cannot influence the electronic state of molecules. Even though it is extremely difficult to measure the exact concentration of water in a N2 gas, it is expected that the device could sensitively detect water. Instead of the quantitative measurement, the conductance of film 1 was measured depending on the vacuum level to investigate the sensitivity (Fig. 3.10). Considering that the change in humidity as a function of the vacuum level is relatively small, the remarkable difference in the conductance clearly indicates that film 1 is highly sensitive to water.

Fig. 3.9 Current change of TPDAP film under humid N2 when the applied voltage was 10 V. Yellow area indicates the duration of gas flowing (initial condition: 13 ± 1 °C, 48% RH). Reprinted with permission from Ref. [44]. Copyright (2017) American Chemical Society

Fig. 3.10 Current–voltage characteristic of film 1 depending on the vacuum level. Left and right graphs were plotted on linear and logarithmic scales, respectively (initial condition: 12 ± 1 °C, 68% RH). Reprinted with permission from Ref. [44]. Copyright (2017) American Chemical Society

3.2 Results and Discussion

51

Fig. 3.11 Powder X-ray diffraction of film 1 before and after 30 cycles of humid N2 sensing. Reprinted with permission from Ref. [44]. Copyright (2017) American Chemical Society

In addition, the structural stability of film PCN 1 as a sensing material was confirmed by powder X-ray diffraction which showed that crystallinity was retained after 30 cycles of humid N2 sensing (Fig. 3.11).

3.2.5

Mechanism Investigation

Single-crystal X-ray analysis before/after hydration of a single crystal of PCN 1 revealed that the key mechanism of humidity sensing is the dissociation of NO3− from Cd ions under hydration. The unit cell volume of a hydrated crystal of PCN 1 (1-hyd) under fully hydrated conditions in a capillary became a quarter compared with that of 1 before hydration. One of the two mono-dentate ligands being part of the columnar layer in an asymmetric unit cell was disordered in PCN 1, while every mono-dentate ligand in 1-hyd was severely disordered (Fig. 3.12). It should be noted that the previously reported Cd-1mono synthesized under dry conditions has only ordered mono-dentate TPDAP layers, indicating that hydration makes mono-dentate TPDAPs disordered. In this sense, PCN 1 is a kind of intermediate state between Cd-1mono (ordered mono-dentate TPDAP layer) and PCN 1-hyd (disordered one). It is reasonable that the fully hydrated network (1-hyd) showed that every mono-dentate layer was disordered. In addition, importantly NO3− on Cd2, Cd3, Cd5, and Cd6 facing to a pore was partially replaced by water (Fig. 3.13). Both disordering of mono-dentate ligand and dissociation of NO3− were also confirmed by powder X-ray diffraction, clearly indicating that the diffraction intensities corresponding to each plane for columnar layer and coordinated NO3− drastically changed after hydration (Fig. 3.14). The distinguishable peaks of 2h = 18.223° corresponding to (12 0 0) in PCN 1 were shifted to 2h = 17.959° after hydration, indicating disordering of mono-dentate TPDAP layers and dissociation of NO3− on Cd3 and Cd6 on that plane. In addition, the relative intensity of 2h = 4.186° corresponding to (0 4 0) in PCN 1 became smaller, indicating the parts of nitrates on Cd2 and Cd5 on the plane were replaced by water. The partial

52

3 Chemiresistive Sensor Based …

Fig. 3.12 Crystal structures of a 1 and b 1-hyd. Schematic drawing of c 1 and d 1-hyd. The direction of arrow indicates coordination with Cd. Three kinds of coordination arrangement in columnar layers were shown in (c). One of the two mono-dentate ligands changed to disorder after hydration, which was probably induced by the dissociation of bridging NO3−. Reprinted with permission from Ref. [44]. Copyright (2017) American Chemical Society

hydration in sensing measurement induced this event in a local part. The locally dissociated NO3− made the framework positively charged, leading to an electronically deficient state in the network framework, resulting in increasing the electronic interaction between TPDAPs in the pillar. In other words, the water acted as a dopant to decrease the resistivity in PCN 1.

3.2 Results and Discussion

53

Fig. 3.13 X-ray structures of coordination parts in a PCN 1 and b 1-hyd. After hydration, NO3− was partially replaced by water. For clarity, cadmium ions, nitrate ions, and coordinated waters were plotted except for the pyridine groups of TPDAP. Blue, N; Yellow, Cd; Red, O. Reprinted with permission from Ref. [44]. Copyright (2017) American Chemical Society

Fig. 3.14 Synchrotron powder X-ray diffraction of single crystals of PCN 1 and 1-hyd and patterns simulated from single-crystal data of PCN 1 and 1-hyd. The distinguishable peaks of 2h = 18.223° corresponding to (12 0 0) in PCN 1 were shifted to 2h = 17.959° after hydration, indicating disordering of mono-dentate TPDAP layers and dissociation of NO3− on Cd3 and Cd6 on that plane. In addition, the relative intensity of 2h = 4.186° corresponding to (0 4 0) in PCN 1 became smaller, indicating the parts of nitrates on Cd2 and Cd5 on the plane were replaced by water. Both peak changes well match with the patterns simulated from single-crystal X-ray data. Reprinted with permission from Ref. [44]. Copyright (2017) American Chemical Society

IR spectroscopic measurements of 1 depending on the vacuum level revealed that all bands related to NO3− vibration modes ma(NO2) at 1420 cm−1, ms(NO2) at 1280 cm−1, m(NO) at 1030 cm−1, and out-of-plane mode at 820 cm−1 were shifted, indicating the coordination of dissociated NO3− to Cd ions in low humid conditions (Fig. 3.15). Both X-ray and IR analysis results indicate that higher humidity

3 Chemiresistive Sensor Based …

54

Fig. 3.15 FT-IR spectra of PCN 1 depending on the vacuum level. Measurements were performed in the ATR mode. All bands related to NO3− vibration modes corresponding to ma(NO2) at 1420 cm−1, ms(NO2) at 1280 cm−1, m(NO) at 1030 cm−1, and out-of-plane mode at 820 cm−1 were shifted, indicating the coordination of dissociated NO3− in air to Cd ions in relatively low humid conditions. Reprinted with permission from Ref. [44]. Copyright (2017) American Chemical Society

conditions induced the dissociation of NO3− from Cd ions by coordination of water. The coordination/dissociation of water/NO3− plays a crucial role in conductivity change.

3.3

Conclusion

In conclusion, stable and dense crystalline film 1 prepared by the simple solvent evaporation method showed highly sensitive and selective humidity sensing ability as a chemiresistive sensor. Single-crystal X-ray analysis and IR spectroscopy revealed that the key mechanism of sensing is the dissociation of NO3− on Cd ions under humid conditions. The dissociation of anions from the metals makes the network positively charged like a doping state. This knowledge will help to realize the electrical conduction in the redox-active PCNs and open the way for the practical use of redox-active PCNs as a chemiresistive sensor.

3.4 Experimental Section

3.4 3.4.1

55

Experimental Section General Experimental Information

All reagents were purchased from commercial sources and used without further purification unless otherwise noted. Gold electrodes (Cr/Au = 4 nm/40 nm) were deposited on the SiO2/Si wafer (300-nm-thick SiO2) with shadow mask (W = 200 lm, L = 10 lm) by thermal evaporator. Elemental analyses were performed on an Elementar vario MICRO cube at the Technical Support Center of the Pohang University of Science and Technology (POSTECH). The optical microscope images were captured by Olympus BX53F. Fourier transform infrared (FTIR) spectroscopic measurements depending on the vacuum level were carried out on a BOMEM DA8 FTIR spectrometer equipped with a golden gate diamond ATR (Specac) and a liquid nitrogen-cooled mercury–cadmium–telluride (MCT) detector at Pohang Accelerator Laboratory. 2,5,8-tri(4-pyridyl)1,3-diazaphenalene (TPDAP) was synthesized by five steps as reported in the procedure.

3.4.2

Syntheses

Synthesis of 1 ([Cd2.5(NO3)5(TPDAP)3(DMF)3(H2O)7.4]) Single crystals of 1 were crystallized at 90 °C on a hot plate under air from the solution of TPDAP (7.16 mg, 0.018 mmol), Cd(NO3)2
4H2O (13.8 mg, 0.045 mmol), and DMF (1 ml) in a sealed 8 ml vial for 2 h. Reddish block crystals of 1 were collected by filtration and dried with N2 gas. Anal. Calcd for C87H86.8Cd2.5N23O25.4 (=Cd2.5(NO3)5(C26H17N5)3((CH3)2NCHO)3(H2O)7.4): C, 48.78; H, 4.08; N, 15.04. Found: C, 49.06; H, 4.09; N, 14.74. The chemical formula was confirmed by elemental analysis and X-ray structure analysis. Synthesis of 2 ([Cd1.5(NO3)2.77(TPDAP)0.77(TPDAP−)0.23(DMF)2(H2O)]) Single crystals of 2 were crystallized at 90 °C on a hot plate under air from the solution of TPDAP (7.16 mg, 0.018 mmol), Cd(NO3)2
4H2O (13.8 mg, 0.045 mmol), and DMF (1 ml) in an open 8 ml vial for 30 min. Reddish thin plate crystals of 2 were collected by filtration dried with N2 gas. Anal. Calcd for C32H32.77Cd1.5N9.77O11.31 (=Cd1.5(NO3)2.77(C26H17N5)0.77(C26H16N5)0.23((CH3)2NCHO)2(H2O)): 42.53; H, 3.65; N, 15.14. Found: C, 42.43; H, 3.72; N, 15.01. The chemical formula was confirmed by elemental analysis, NMR, and X-ray structure analysis. Synthesis of Film 1 A mother solution was prepared by the procedure described in the synthesis of 1. The 0.2 ml of solution was transferred to a 4 ml vial containing a gold-patterned

3 Chemiresistive Sensor Based …

56

SiO2 wafer. After heating at 90 °C for 2 h with sealed condition, the wafer was taken from the vial, then washed with acetone, and dried under N2 flow for 1 h. Synthesis of Film 2 A mother solution was prepared by the procedure described in the synthesis of 2. The 0.2 ml of solution was transferred to a 4 ml vial containing a gold-patterned SiO2 wafer. After heating at 90 °C for 30 min with open condition, the wafer was taken from the vial, then washed with acetone, and dried under N2 flow for 1 h. TPDAP Ligand Film TPDAP was deposited on a gold-patterned SiO2 wafer by thermal evaporation. The following condition was used: sample temperature, 230 °C; substrate temperature, 100 °C; vacuum level, 3  10−6 torr; rate, 0.1 Å/s.

3.4.3

X-Ray Structure Analysis

Single-Crystal X-Ray Analysis of PCN 1 The diffraction data for PCN 1 was recorded on a ADSC Quantum 210 CCD diffractometer using synchrotron radiation (k = 0.7000 Å) at 2D SMC beamline of Pohang Accelerator Laboratory (PAL). The diffraction images were processed using HKL3000. The structures were solved by direct methods (SHELXS-97) and refined by full-matrix least squares calculations on F2 (SHELXL-2014) using the SHELX-TL program package. C83.31H48Cd2.5N20.56O16.94, Mr = 1889.04, crystal dimensions 0.01  0.03  0.04 mm3, orthorhombic, Cmma, a = 40.522(1) Å, b = 58.798(1) Å, c = 15.1871(4) Å, V = 36,184.6(2) Å3, T = −173 °C, Z = 16, qcalcd = 1.387 g cm−3, l = 6.24 cm−1, 23,493 unique reflections out of 7292 with I > 2r(I), 1409 parameters, 2.016º < h < 29.532º, R1 = 0.1071, wR2 = 0.3927, GOF = 1.028, and CCDC deposit number 1541112. Single-Crystal X-Ray Analysis of PCN 2 The diffraction data for PCN 2 was recorded on a RIGAKU/MSC Mercury CCD X-ray diffractometer using synchrotron radiation (k = 0.6889 Å) at PF-AR (NW2A beamline) of the High Energy Accelerator Research Organization (KEK). The diffraction images were processed using HKL2000. The structures were solved by direct methods (SHELXS-97) and refined by full-matrix least squares calculations on F2 (SHELXL-97) using the SHELX-TL program package. C30.50H23Cd1.5N9.12O11.62, Mr = 871.93, crystal dimensions 0.10  0.04  0.02 mm3, orthorhombic, C2/c, a = 18.246(6) Å, b = 32.036(4) Å, c = 15.089(2) Å, V = 8144(3) Å3, T = −173 °C, Z = 8, qcalcd = 1.424 g cm−3, l = 7.07 cm−1, 8680 unique reflections out of 2068 with I > 2r(I), 1409 parameters, 1.878º < h < 26.880°, R1 = 0.1360, wR2 = 0.4569, GOF = 0.999, and CCDC deposit number 1541115.

3.4 Experimental Section

57

Single-Crystal X-Ray Analysis of PCN 1-hyd A single crystal of PCN 1 was inserted into a 0.5-mm-diameter capillary, and then wet cotton was placed at a distance of 4 cm from the single crystal. The capillary containing a single crystal and wet cotton was sealed by flame to keep the hydrate condition inside of the capillary. After kept at room temperature for 30 min, it was set on the goniometer with clay. The diffraction data for PCN 1-hyd was recorded on a ADSC Quantum 210 CCD diffractometer using synchrotron radiation (k = 0.8500 Å) at 2D SMC beamline of Pohang Accelerator Laboratory (PAL). The diffraction images were processed using HKL3000. The structures were solved by direct methods (SHELXS-97) and refined by full-matrix least squares calculations on F2 (SHELXL-2014) using the SHELX-TL program package. C78H32Cd2.25N16.50O15, Mr = 1693.10, crystal dimensions 0.055  0.040  0.040 mm3, orthorhombic, Cmmm, a = 20.5544(5) Å, b = 29.4116(6) Å, c = 15.2146(4) Å, V = 9197.8(4) Å3, T = −173 °C, Z = 4, qcalcd = 1.228 g cm−3, l = 8.62 cm−1, 4137 unique reflections out of 2510 with I > 2r(I), 494 parameters, 2.752° < h < 29.587°, R1 = 0.1480, wR2 = 0.4093, GOF = 1.618, and CCDC deposit number 1541116 (Fig. 3.16). X-Ray Powder Diffraction of Single Crystals of PCN 1, Single Crystals of PCN 2, Film 1, and Film 2 The X-ray powder diffractions of single crystals of PCN 1, single crystals of PCN 2, film 1, and film 2 were measured on the Bruker D8 Advance system equipped with a Cu-sealed tube (k = 1.549 Å).

Fig. 3.16 Microscopic images of sample for single-crystal X-ray analysis of PCN 1-hyd

3 Chemiresistive Sensor Based …

58

X-Ray Powder Diffraction Measurement of 30 Cycles of Humid Sensing The X-ray powder diffraction of film 1 before/after 30 cycles of humid sensing was measured on the Rigaku SmartLab (Cu Ka, k = 1.5418 Å) with transmittance mode after removing the substrate. X-Ray Powder Diffraction Measurement of PCN 1 and 1-hyd in Capillary The single crystals of PCN 1 were inserted into two 0.3–mm-diameter capillaries, and then in the case of PCN 1-hyd, one drop of water was added into the capillary. The diffraction data for PCN 1 and 1-hyd was measured using synchrotron radiation (k = 1.08 Å) at BL44B2 beamline of Super Photon ring-8 GeV (SPring-8).

3.4.4

Electrical Measurement

I–V Characteristics of Single Crystal of PCN 1 and 2 The current–voltage characteristics of single crystals were measured in air using Keithley 4200-SCS semiconductor parameter analyzer. Single crystals were placed on a SiO2 (300 nm) wafer by attaching conductive silver paste at the two edges of crystals. By applying voltage through the conductive silver paste, the current– voltage characteristics were recorded as shown in Scheme 3.1. The electrical resistance which is defined as the ratio of voltage across a single crystal (V) to current through it (I) was obtained from the current–voltage curve. The conductivity of single crystals was calculated by the following equation: r¼

1 l ¼ q RA

R: electrical resistance of a uniform specimen of a crystal l: length of crystal A: cross-sectional area of a crystal.

Scheme 3.1 Schematic of current–voltage measurement experimental apparatus for single crystal

3.4 Experimental Section

59

Scheme 3.2 Schematic of current–voltage measurement experimental apparatus for single crystal

I–V Characteristics of Film 1 and 2 The current–voltage characteristics of films were measured in air using Keithley 4200-SCS semiconductor parameter analyzer. Since the crystalline films 1 and 2 were directly prepared on the gold-patterned electrode, the voltage could be applied though the god electrode as shown in Scheme 3.2. Sensing Measurements The electrical performance and sensing tests were measured under ambient conditions and vacuum conditions in dark using a Keithley 4200 semiconductor parametric analyzer. For measurements of current changes under gas flow, dry/ humid/ethanol/acetone vapors were produced by passing dry nitrogen gas through a flask filled with molecular sieve/water/liquid ethanol/liquid acetone, respectively. The films were exposed to each vapor stream coming from a 3-mm-diameter tube. The end of the tube was installed at a distance of 5 cm above the surface of the device. While the flow of vapors was on, the gas flow was maintained at 2 sccm.

References 1. Eddaoudi M, Moler D, Li H, Chen B, Reineke TM, O’Keeffe M, Yaghi OM (2001) Acc Chem Res 34:319–330 2. Kitagawa S, Kitaura R, Noro S-I (2004) Angew Chem Int Ed 43:2334–2375 3. Férey G (2008) Chem Soc Rev 37:191–214 4. Robson R (2008) Dalton Trans 38:5113–5131 5. Zhou H-CJ, Kitagawa S (2014) Chem Soc Rev 43:5415–5418 6. Farha OK, Hupp JT (2010) Acc Chem Res 43:1166–1175 7. Cook TR, Zheng Y-R, Stang PJ (2013) Chem Rev 113:734–777 8. Furukawa H, Cordova KE, O’Keeffe M, Yaghi OM (2013) Science 341:123044 9. Allendorf MD, Bauer CA, Bhakta RK, Houk RJT (2009) Chem Soc Rev 38:1330–1352 10. Chen B, Xiang S, Quan G (2010) Acc Chem Res 43:1115–1124 11. Cui T, Yue Y, Qian G, Chen B (2012) Chem Rev 112:1126–1162 12. Kreno LE, Leong K, Farha OK, Allendorf M, Van Duyne RP, Hupp JT (2012) Chem Rev 112:1105–1125 13. Hu Z, Deibert BJ, Li J (2014) Chem Soc Rev 43:5815–5840 14. Yu F-Y, Chen D, Wu M-K, Han L, Jiang H-L (2016) ChemPlusChem 81:675–690 15. Takaishi S, Hosoda M, Kajiwara T, Miyasaka H, Yamashita M, Nakanishi Y, Kitagawa Y, Yamaguchi K, Kobayashi A, Kitagawa H (2009) Inorg Chem 48:9048–9050 16. Kobayashi Y, Jacobs B, Allendorf MD, Long JR (2010) Chem Mater 22:4120–4122 17. D’Alessandro DM, Kanga JRR, Caddy JS (2011) Aust J Chem 64:718–722 18. Stavila V, Talin AA, Allendorf MD (2014) Chem Soc Rev 43:5994–6010

60

3 Chemiresistive Sensor Based …

19. Talin AA, Centrone A, Ford AC, Foster ME, Stavila V, Haney P, Kinney RA, Szalai V, Gabaly FEI, Yoon HP, Léonard F, Allendorf MD (2014) Science 343:66–69 20. Sun L, Campbell MG, Dincă M (2016) Angew Chem Int Ed 55:3566–3579 21. Achmann S, Hagen G, Kita J, Malkowsky IM, Kiener C, Moos R (2009) Sensors 9:1574– 1589 22. Campbell MG, Sheberla D, Liu SF, Swager TM, Dincă M (2015) Angew Chem Int Ed 54:4349–4352 23. Campbell MG, Liu SF, Swager TM, Dincă M (2015) J Am Chem Soc 137:13780–13783 24. Betard A, Fischer RA (2012) Chem Rev 112:1055 25. Gliemann H, Woll C (2012) Mater Today 15:110 26. Zacher D, Shekhah O, Woll C, Fischer RA (2009) Chem Soc Rev 38:1418 27. Shekhah O, Liu J, Fischer RA, Wöll C (2011) Chem Soc Rev 40:1081–1106 28. Falcaro P, Ricco R, Doherty CM, Liang K, Hill AJ, Styles MJ (2014) Chem Soc Rev 43:5513–5560 29. Hermes S, Schröder F, Chelmowski R, Wöll C, Fischer RA (2005) J Am Chem Soc 127:13744–13745 30. Zacher D, Baunemann A, Hermes S, Fischer RA (2007) J Mater Chem 17:2785–2792 31. Arnold M, Kortunov P, Jones DJ, Nedellec Y, Karger J, Caro J (2007) Eur J Inorg Chem 60– 64 32. Hausdorf S, Baitalow F, Seidel J, Mertens F (2007) J Phys Chem A 111:4259–4266 33. Hermes S, Witte T, Hikov T, Zacher D, Bahnmuller S, Langstein G, Huber K, Fischer RA (2007) J Am Chem Soc 129:5324–5325 34. Gascon J, Aguado S, Kapteijn F (2008) Micropor Mesopor Mat 113:132–138 35. Liu Y, Ng Z, Khan EA, Jeong H-K, Ching C, Lai Z (2009) Micropor Mesopor Mat 118:296– 301 36. Li YS, Bux H, Feldhoff A, Li GL, Yang WS, Caro J (2010) Adv Mater 22:3322–3326 37. Kozachuk O, Yusenko K, Noei H, Wang Y, Walleck S, Hlaser T, Fisher RA (2011) Chem Commun 47:8509–8511 38. Wade CR, Li MY, Dincă M (2013) Angew Chem Int Ed 52:13377–13381 39. Kung CW, Chang TH, Chou LY, Hupp JT, Farha OK, Ho KC (2015) Chem Commun 51:2414–2417 40. Shekhah O, Wang H, Kowarik S, Schreiber F, Paulus M, Tolan M, Sternemann C, Evers F, Zacher D, Fischer RA, Wöll C (2007) J Am Chem Soc 129:15118–15119 41. Shekhah O, Wang H, Strunskus T, Cyganik P, Zacher D, Fischer R, Wöll C (2007) Langmuir 23:7440–7442 42. Buchan I, Ryder MR, Tan JC (2015) Cryst Growth Des 15:1911–1999 43. Stassen I, Styles M, Assche TV, Campagnol N, Fransaer J, Denayer J, Tan JC, Falcaro P, Vos DD, Ameloot R (2015) Chem Mater 27:1801–1807 44. Kim J, Koo JY, Lee YH, Kojima T, Yakiyama Y, Ohtsu H, Oh JH, Kawano M (2017) Inorg Chem 56:8735–8738 45. Koo JY, Yakiyama Y, Kim J, Morita Y, Kawano M (2015) Chem Lett 44:1131–1133 46. Yakiyama Y, Ueda A, Morita Y, Kawano M (2012) Chem Comm 48:10651–10653 47. Kitagawa H, Ohtsu H, Kawano M (2013) Angew Chem Int Ed 52:12395–12399 48. Koo JY, Yakiyama Y, Lee GR, Lee J, Choi HC, Morita Y, Kawano M (2016) J Am Chem Soc 138:1776–1779

Chapter 4

Tunable Electrical Properties of Redox-Active Porous Coordination Networks via Post-synthetic Modification

Abstract Redox-active PCN 1 with small 8 Å-sized diagonal pore due to high p–p stacking in the columnar structure was converted into PCN 1open with 1D channels by immersing the crystals in ethanol-containing water. By removing mono-dentate TPDAP ligands in PCN 1, 13 Å
4 Å-sized channels surrounded by tri-dentate TPDAPs were formed in PCN 1open that was more efficiently oxidized by air than PCN 1 because of the larger surface area, resulting in increase of the spin density of neutral radical states. However, because PCN 1open became insulator, the p–p stacking columnar structure was proved to be a conductive pathway in PCN 1. In fact, PCN 1open containing neutral radicals was changed into the high conductivity of 3.18
10−4 S cm−1 from that of 1.23
10−8 S cm−1 in PCN 1 by intercalation of TPDAPs into the open channels.

4.1

Introduction

In principle, a porous coordination network (PCN) [1–8] which has been in the spotlight as a porous material for various applications is considered to enable to produce designable structures by controlling metal nodes and bridge ligands. Although many combinations of ligands and metals produced enormous numbers of PCNs, in many case it is impossible to control and predict the final structure because of their self-assembly nature dependent on the molecular intrinsic properties such as solubility, intermolecular interaction, and stability, which gives rise to various geometries. Especially, large p-plane molecules are apt to form non-porous structures or very small-sized pores. It is not trivial to use them as functional ligands with redox activity. In order to realize the rational chemistry, several synthetic approaches using PCNs have been introduced, categorized in post-synthetic modification (PSM) that is a process by which a preformed PCN is converted into a new PCN via crystal-to-crystal transformation [9–12]. Although there are specific methods depending on materials and purposes, PSM commonly aims to modify an undesirable structure for target properties. PSM can be a way to ultimately implement a designable structure and properties by providing the possibility of © Springer Nature Singapore Pte Ltd. 2021 J. Kim, Design of Electronic Devices Using Redox-Active Organic Molecules and Their Porous Coordination Networks, Springer Theses, https://doi.org/10.1007/978-981-16-3907-4_4

61

4 Tunable Electrical Properties of Redox-Active …

62

employing various organic transformations and the ability of reagents to access the interior of solids. In this chapter, the author describes the tunable electrical conductivity of redox-active PCNs via post-synthetic approach. PCN 1 with small pores due to the high p–p stacking of the columnar structure was converted into PCN 1open with 1D channels by immersing the crystals in ethanol-containing water. By removing mono-dentate TPDAP ligands in PCN 1, 13 Å
4 Å-sized channels surrounded by tri-dentate TPDAPs were formed in PCN 1open. Owing to a large open pore, PCN 1open could be more readily oxidized by air as compared to PCN 1. The insulating PCN 1open because of the absence of p–p stacking was converted to the conductive PCN 1spin by re-encapsulation of TPDAPs. The conductivity of PCN 1spin was four orders of magnitude higher than that of PCN 1. These results will provide a comprehensive understanding for designing conductive PCNs using redox-active organic molecules.

4.2 4.2.1

Results and Discussion Pore Opening of a Redox-Active PCN by PSMs

A high packing structure of a redox-active PCN was converted to an open structure possessing large 1D channels by simple immersion of wet ethanol. PCN 1, [Cd2.5(NO3)5(TPDAP)3(DMF)3(H2O)7.4], (orthorhombic Cmma) prepared by the solvent evaporation method described in Chap. 3 has a p–p stacking columnar structure and 8 Å-sized diagonal channels along the [1 1 0] direction as shown in Fig. 4.1. A large p-plane of TPDAP induced a high packing structure with a p–p distance of 3.3463(73) Å, which limited efficient applications as a porous material and functionality of redox activity of ligands. As described in Chap. 3, the electrical properties of PCN 1 were modulated by the dissociation of nitrates by water. In addition, the single-crystal X-ray analysis before/after hydration revealed the water could also dissociate the mono-dentate ligands, making every mono-dentate ligand layers disordered (Fig. 4.2). It implies that the TPDAP can be dissociated but cannot be extracted from the network because of its hydrophobicity. These results led to the idea of removing the mono-dentate ligands with a water-containing organic solvent to prepare a large open channel. Considering the solubility of TPDAP and stability of PCN 1, ethanol with a proper amount of water was used. As a result, the mono-dentate layers in PCN 1 were removed by the wet ethanol, producing a 1D channel of 13 Å
4 Å (Fig. 4.3). In order to investigate the influence of water, powder X-ray diffraction patterns of PCN 1 after immersion in wet/dry ethanol were measured (Fig. 4.4). While the diffraction pattern after immersion in wet ethanol was noticeably changed that after immersion in dry ethanol was retained. Especially, a peak of 2h = 25.8° in PCN 1 corresponding to the periodicity of p–p stacking disappeared in the diffraction pattern after

4.2 Results and Discussion

63

Fig. 4.1 a Crystal structure of PCN 1. b Simulated void spaces by Mercury 3.10.: 23.4% (8462.80 Å3), cavity dimension: 8.38 Å
8.25 Å. c Projection of each independent layer in PCN 1. Gray, C; Blue, N; Yellow, Cd, Red, O. Hydrogen atoms and solvents are omitted for clarity

Fig. 4.2 a Coordination changes after hydration. b Projection of each independent layer in PCN 1. Notably, every mono-dentate layer was changed to disordered layers by hydration, indicating the water could dissociate the mono-dentate ligands. Gray, C; Blue, N; Yellow, Cd, Red, O. Hydrogen atoms and solvents are omitted for clarity

immersion in wet ethanol, indicating formation of PCN 1open. This result indicated that water played an important role in dissociating the mono-dentate ligands and was consistent with the result of Chap. 3.

64

4 Tunable Electrical Properties of Redox-Active …

Fig. 4.3 a Crystal structure of PCN 1open. b Simulated void spaces by Mercury 3.10.: 53.2% (4769.76 Å3), cavity dimension: 14.64 Å
4.64 Å. c Projection of each independent layers in PCN 1open. Gray, C; Blue, N; Yellow, Cd, Red, O. Hydrogen atoms and solvents are omitted for clarity

Fig. 4.4 Powder X-ray diffractions of PCN 1 after immersion into dry/wet ethanol

4.2.2

Efficient Oxidation of Open Structure After PSMs

The open structure, PCN 1open, was efficiently oxidized by air compared to PCN 1. The electron paramagnetic resonance (EPR) spectroscopic measurement indicated that the spin density of PCN 1open was considerably higher than that of PCN 1 (Fig. 4.5): While the spin density of PCN 1 was 0.029% after exposure to air for 3 days, that of PCN 1open was 2.369% (Table 4.1). Although redox-active TPDAP could be oxidized by air [13], the high packing structure of PCN 1 suppressed the oxidation of TPDAPs inside of a crystal. However, the open channel in PCN 1open which could accept a number of guest molecules, such as oxygen in air, efficiently generated paramagnetic species. The comparison of crystal size distributions

4.2 Results and Discussion

65

Fig. 4.5 EPR spectra of PCN 1, 1open, and 1spin. All samples were measured after exposure to air for 1 day

Table 4.1 Spin density calculation of PCN 1, 1open, and 1spin Weight (mg)

Number of TPDAPs (mmol)

TEMPO (reference) 3.0 0.0193 PCN 1 1.4 0.0019 1.3 0.0019 PCN 1open 1.4 0.0019 PCN 1spin The spin amounts were calculated using TEMPO as

Amplitude

Double intensity (DI)

Spin density (%)

3 30 30 30 a reference

27,299.2 783.59 64,906.33 29,071.31

100 0.029 2.369 1.061

between PCN 1 and PCN 1open from the microscopy images revealed that it is almost similar, implying that the significant increase of spin density in PCN 1open is responsible for the oxidation of TPDAPs inside of a crystal at the bulk level (Fig. 4.6). The spin density of PCN 1spin in Fig. 4.5 and Table 4.1 will be described in Sect. 4.2.3.

4.2.3

Tunable Electrical Properties via PSMs

The electrical conductivity significantly increased after the encapsulation of TPDAP into PCN 1open. The current–voltage characteristics of single crystals of PCN 1 and PCN 1open revealed that the semiconducting electrical conductivity of 1.23
10−8 S cm−1 (23 ± 1 °C, 72% RH, in air) in PCN 1 significantly decreased

66

4 Tunable Electrical Properties of Redox-Active …

Fig. 4.6 Microscopic images of PCN 1 and PCN 1open

by the removal of mono-dentate TPDAPs, resulting in insulator of PCN 1open. Although the PCN 1open had 90 times more radical species than PCN 1, the electrical conductivity of was extremely low (4.5
10−12 S cm−1 (20 ± 1 °C, 40% RH, in air)) because it did not have continuous p–p stacking which was a possible conductive pathway in PCN 1. In order to make a proper electron transfer pathway into PCN 1open, TPDAPs were intercalated (PCN 1spin, Fig. 4.7). Single-crystal X-ray diffraction analysis of PCN 1spin revealed that every intercalated TPDAP was disordered on the position same as the mono-dentate TPDAP layers. Notably, while PCN 1 possessed both ordered mono-dentate TPDAP layers, PCN 1spin possessed only disordered mono-dentate TPDAP layers. This result can be understood by the role of water during dissociation of the mono-dentate ligands in the previous process. Because water could dissociate bridging nitrates, the Cd3 which was coordination sites of intercalated TPDAPs also could be disordered. Figure 4.8 shows that each powder X-ray diffraction pattern of PCN 1, 1open, and 1spin well matches with those of simulated patterns. Especially, a peak of 2h = 25.8° corresponding to the periodicity of p–p stacking was recovered after encapsulation of TPDAPs. It should be noted that PCN 1spin could not be obtained directly using the same condition as the encapsulation of solvent and temperature, indicating single-crystal-to-single-crystal transformation. The spin density of PCN 1spin was comparable to that of PCN 1open, which means that the radical TPDAPs were retained during encapsulation of TPDAPs (see Fig. 4.5 and Table 4.1). Notably, the electrical conductivity of a single crystal of PCN 1spin (3.18
10−4 S cm−1 (26 ± 1 °C, 30% RH, in air) significantly increased by four orders of magnitude as compared with both PCN 1 and PCN 1open owing to the continuous p–p stacking layers and radical species which could facilitate the electron transfer (Fig. 4.9).

4.2 Results and Discussion

67

Fig. 4.7 a Crystal structure of PCN 1open. b Projection of each independent layer in PCN 1open. Gray, C; Blue, N; Yellow, Cd, Red, O. Hydrogen atoms and solvents are omitted for clarity

Fig. 4.8 Powder X-ray diffraction patterns of PCN 1, PCN 1open, and PCN 1spin with the simulated patterns from single-crystal structures. The distinguishable peak of 2h = 25.8° corresponding to (6 0 0) was absent in PCN 1open, implying the absence of mono-dentate ligand layers. It should be noted that while the powder X-ray diffraction patterns were measured under ambient conditions, the simulated patterns were generated from the single-crystal X-ray diffraction data measured at 100 K using Mercury 3.10

4 Tunable Electrical Properties of Redox-Active …

68

Fig. 4.9 Current–voltage (I–V) characteristic of PCN 1, 1open, and 1spin. Left and right graphs were plotted on linear and logarithmic scales, respectively. Crystal dimensions of PCN 1: A = 60 lm
50 lm, L = 110 lm; 1open: A = 80 lm
40 lm, L = 120 lm; 1spin: A = 100 lm
40 lm, L = 100 lm

4.3

Conclusion

In conclusion, the electrical conductivity of redox-active PCNs was modulated by post-synthetic modification which enabled efficient oxidation of PCNs during the process. Redox-active PCN 1 with small pores due to the high p–p stacking in the columnar structure was converted into PCN 1open with 1D channels by immersing the crystals in ethanol. By removing mono-dentate TPDAP ligands in PCN 1, the open channels surrounded by tri-dentate TPDAPs were formed in PCN 1open, and it facilitated the oxidation of TPDAPs. However, PCN 1open was electrically insulating because of the absence of the p–p stacking which was a conductive pathway in PCN 1. On the other hand, the insulating PCN 1open could be modulated into conductive PCN 1spin by intercalation of TPDAPs owing to high spin density of radical species and continuous p–p stacking. This study will help to understand the conduction mechanism in redox-active materials and will provide general guidelines for designing conductive PCNs using redox-active organic molecules.

4.4 4.4.1

Experimental Section General Experimental Information

All reagents were purchased from commercial sources and used without further purification unless otherwise noted. The optical microscope images were captured by Olympus BX53F. EPR was measured using JEOL JES-X310. 2,5,8-tri (4-pyridyl)1,3-diazaphenalene (TPDAP) was synthesized by five steps as reported in the procedure.

4.4 Experimental Section

4.4.2

69

Syntheses

Synthesis of PCN 1 ([Cd2.5(NO3)5(TPDAP)3(DMF)3(H2O)7.4]) The synthetic procedure of PCN 1 was the same as described in Sect. 3.4.2. Single crystals of PCN 1 were crystallized at 90 °C for 2 h on a hot plate under air from a solution of TPDAP (7.16 mg, 0.018 mmol), Cd(NO3)24H2O (13.8 mg, 0.045 mmol), and DMF (1 ml) in a sealed 8 ml vial. Reddish block crystals of 1 were collected by filtration and dried with N2 gas. Synthesis of PCN 1open PCN 1 prepared as described above was desolvated by heating at 90 °C under vacuum to remove guest molecules in the pores. Desolvated crystals were immersed into ethanol (95%), which was replaced until no color change was observed. Pale brown block crystals of PCN 1open were collected by filtration and dried with N2 gas. Synthesis of PCN 1spin PCN 1 prepared as described above was immersed into a saturated TPDAP solution of chloroform/DMF mixture (volume ratio: 20:1) under N2. After 1 week, reddish block crystals of PCN 1spin were collected by filtration and dried with N2.

4.4.3

X-Ray Structure Analysis

Single-Crystal X-Ray Analysis of PCN 1 The information was the same as described in Sect. 3.4.3. The diffraction data for PCN 1 was recorded on a ADSC Quantum 210 CCD diffractometer using synchrotron radiation (k = 0.7000 Å) at 2D SMC beamline of Pohang Accelerator Laboratory (PAL). The diffraction images were processed using HKL3000. The structures were solved by direct methods (SHELXS-97) and refined by full-matrix least squares calculations on F2 (SHELXL-2014) using the SHELX-TL program package. C83.31H48Cd2.5N20.56O16.94, Mr = 1889.04, crystal dimensions 0.01
0.03
0.04 mm3, orthorhombic, Cmma, a = 40.522(1) Å, b = 58.798(1) Å, c = 15.1871(4) Å, V = 36,184.6(2) Å3, T = −173 °C, Z = 16, qcalcd = 1.387 g cm−3, l = 6.24 cm−1, 23,493 unique reflections out of 7292 with I > 2r(I), 1409 parameters, 2.016º < h < 29.532º, R1 = 0.1071, wR2 = 0.3927, and GOF = 1.028. Single-Crystal X-Ray Analysis of PCN 1open The diffraction data for PCN 1open was recorded on a ADSC Quantum 210 CCD diffractometer using synchrotron radiation (k = 0.7000 Å) at 2D SMC beamline of Pohang Accelerator Laboratory (PAL). The diffraction images were processed

4 Tunable Electrical Properties of Redox-Active …

70

using HKL3000. The structures were solved by direct methods (SHELXS-97) and refined by full-matrix least squares calculations on F2 (SHELXL-2014) using the SHELX-TL program package. C46.22H28.44Cd2.25N11.28O16.58, Mr = 1262.74, crystal dimensions 0.04
0.04
0.05 mm3, orthorhombic, Cmmm, a = 20.0751 (10) Å, b = 20.4375(10) Å, c = 15.1578(4) Å, V = 8957.7(7) Å3, T = −173 °C, Z = 4, qcalcd = 1.255 g cm−3, l = 5.98 cm−1, 7027 unique reflections out of 2926 with I > 2r(I), 488 parameters, 1.793º < h < 29.538º, R1 = 0.1653, wR2 = 0.3927, and GOF = 1.101. Single-Crystal X-Ray Analysis of PCN 1spin The diffraction data for PCN 1open was recorded on a ADSC Quantum 210 CCD diffractometer using synchrotron radiation (k = 0.8000 Å) at 2D SMC beamline of Pohang Accelerator Laboratory (PAL). The diffraction images were processed using HKL3000. The structures were solved by direct methods (SHELXS-97) and refined by full-matrix least squares calculations on F2 (SHELXL-2014) using the SHELX-TL program package. C82.95H32Cd2.25N19.36O18.47, Mr = 1850.46, crystal dimensions 0.04
0.03
0.05 mm3, orthorhombic, Cmmm, a = 20.167(6) Å, b = 29.419(6) Å, c = 15.171(4) Å, V = 9000.85(40) Å3, T = −173 °C, Z = 4, qcalcd = 1.249 g cm−3, l = 7.43 cm−1, 4952 unique reflections out of 1360 with I > 2r(I), 484 parameters, 1.378º < h < 29.901º, R1 = 0.1209, wR2 = 0.3113, and GOF = 0.969. Powder X-Ray Diffractions The X-ray powder diffraction data of PCN 1, 1open, and 1spin was measured on the Rigaku SmartLab (Cu Ka, k = 1.5418 Å) with transmittance mode under ambient condition.

4.4.4

Electrical Measurement

I–V Characteristics of Single Crystals of PCN 1, 1open and 1spin The current–voltage characteristics of single crystals were measured in air using a Keithley 4200-SCS semiconductor parameter analyzer. Single crystals were placed on a SiO2 (300 nm) wafer with conductive silver paste attached on two edges of crystals. By applying voltage through the conductive silver paste, the current– voltage characteristics were recorded. The procedure was the same as described in Sect. 3.4.4 (See Scheme 3.1). The electrical resistance which is defined as the ratio of voltage across a single crystal (V) to current through it (I) was obtained from the current–voltage curve. The conductivity of a single crystal was calculated by the following equation:

4.4 Experimental Section

71

r¼

1 l ¼ q R
A

R: electrical resistance of a uniform specimen of a crystal l: length of a crystal A: cross-sectional area of a crystal.

References 1. Eddaoudi M, Moler D, Li H, Chen B, Reineke TM, O’Keeffe M, Yaghi OM (2001) Acc Chem Res 34:319–330 2. Kitagawa S, Kitaura R, Noro S-I (2004) Angew Chem Int Ed 43:2334–2375 3. Férey G (2008) Chem Soc Rev 37:191–214 4. Robson R (2008) Dalton Trans 38:5113–5131 5. Zhou H-CJ, Kitagawa S (2014) Chem Soc Rev 43:5415–5418 6. Farha OK, Hupp JT (2010) Acc Chem Res 43:1166–1175 7. Cook TR, Zheng Y-R, Stang PJ (2013) Chem Rev 113:734–777 8. Furukawa H, Cordova KE, O’Keeffe M, Yaghi OM (2013) Science 341:123044 9. Burrows AD (2011) CrystEngComm 13:3623 10. Wang Z, Cohen SM (2009) Chem Soc Rev 38:1135 11. Tanabe KK, Cohen SM (2011) Chem Soc Rev 40:498 12. Cohen SM (2010) Chem Sci 1:32 13. Koo JY, Yakiyama Y, Kim J, Morita Y, Kawano M (2015) Chem Lett 44:1131–1133

Chapter 5

Summary and Outlook

5.1

Summary

This study is set out to investigate how the electrical properties of redox-active materials are introduced/controlled using redox activity and molecular arrangement with the overall goal of gaining deeper insight into the electron transfer mechanism in the redox-active materials. This thesis includes synthesis, fabrication, characterization, and mechanistic study of electronic devices based on redox-active organic molecule, 2,5,8-tri(4-pyridyl)1,3-diazaphenalene (TPDAP) [1], having a large p-plane and multi-intermolecular interaction sites. This thesis fundamentally discussed three approaches to control the electrical properties of redox-active materials, considering the molecular arrangement: (1) by the external voltage (2) by the chemical environment (3) through the structural modification. Furthermore, each method was applied to fabricate the appropriate electronic devices such as a resistive switching memory device and a chemiresistive sensor. In addition, challenges to integrate materials into the devices such as a synthetic or fabrication method were approached. Another notable point of this thesis is an introduction of the advanced techniques to investigate the mechanisms such as single-crystal X-ray analysis/infrared (IR) spectroscopic measurement under certain conditions and grazing incidence X-ray wide angle scattering (GIWAXS) before/after applying voltages and current–voltage (I–V) measurement of single crystals. The electrical properties of redox-active organic molecule, TPDAP, were modulated by external voltages, which could be applied to the fabrication of resistive switching memory devices (ReRAMs). The anisotropic TPDAP film (aniso-TPDAP) and isotropic TPDAP film (iso TPDAP) were selectively prepared by thermal evaporation at different substrate temperatures. When the substrate temperature was 25 °C, the large p-plane and multi-intermolecular interactivity of TPDAP enabled to produce an anisotropic and extremely uniform film (aniso-TDPAP) stacked by p–p interaction on the out-of-plane and stabilized by hydrogen bonding on the in-plane. On the other hand, the substrate temperature of © Springer Nature Singapore Pte Ltd. 2021 J. Kim, Design of Electronic Devices Using Redox-Active Organic Molecules and Their Porous Coordination Networks, Springer Theses, https://doi.org/10.1007/978-981-16-3907-4_5

73

74

5 Summary and Outlook

80 °C induced randomly oriented crystal growth resulting in a rough surface. While the Au/aniso-TPDAP/Au/SiO2 showed non-volatile resistive switching memory behavior with high on–off ratio, retention, and endurance, the Au/iso-TPDAP/Au/ SiO2 did not show any switching phenomena. GIWAXS and UV–vis and IR spectroscopic study of on- and off-states indicated that the resistance modulation did not result from changes in structure/electronic state/chemical bond at the bulk level. Furthermore, control experiments using an isotropic TPDAP film and a redox-inert 2,5,8-tri(4′-pyridyl)-1,3,4,6,7,9-hexaazaphenalene (TPHAP) [2] having the same molecular shape as TPDAP verified that the anisotropic-oriented layers of TPDAP via p–p stacking and its oxidation/reduction process promoted two reversible resistance states. In other words, the external voltage reversibly generated conductive oxidized layers in the local region from the non-conductive neutral layers. This study can provide general guidelines for designing resistive switching memory devices using redox-active organic molecules, highlighting the importance of molecular arrangement. The electrical properties of redox-active porous coordination networks (PCNs) composed of TPDAP and Cd2+ could be also varied by chemical environment, which was applied to fabrication of a chemiresisive sensor. Two kinds of redox-active PCNs were selectively synthesized by the solvent evaporation method depending on the rate of evaporation. While slow evaporation produced [Cd2.5(NO3)5(TPDAP)3(DMF)3(H2O)7.4] (PCN 1) that has a p–p stacking columnar structure with a 6 Å-sized diagonal channel along the [1 1 0] direction, fast evaporation produced [Cd1.5(NO3)2.77(TPDAP)0.77(TPDAP−)0.23(DMF)2(H2O)] (PCN 2) having a 2D slipped p–p stacking structure possessing 3D-tangled pores with a window size of 7 Å. The crystalline films of PCN 1 and PCN 2 were fabricated under the same condition as their single-crystal growth on the Au-patterned SiO2/Si wafer, respectively. The intrinsic electrical properties of single crystals and films were confirmed by measuring the I–V characteristics. While the electrical current of film 2 decreased with exposure to air, that of film 1 was relatively retained because of structural stability of 3D networks. Film 1 composed of a very densely packed layer with open channels showed a significant change in resistivity for humidity with high sensitivity and selectivity. When wet N2 gas was introduced to the device by passing N2 stream through a water bath, the current significantly increased by two orders of magnitude. In contrast, when dry N2 gas was introduced to the device, the current decreased by two orders of magnitude compared with the initial state. Single-crystal X-ray analysis and IR spectroscopic measurement before/after hydration of a single crystals of PCN 1 revealed that the key mechanism of humidity sensing is the dissociation of NO3− from Cd ions under hydration. The locally dissociated NO3− made the framework positively charged, leading to an electronically deficient state in the network framework, resulting in increasing the electronic interaction between TPDAPs in the pillar. This research helps to understand the electron transfer mechanism in redox-active PCNs and provides the important message about the structural design for the application of chemiresistive sensing.

5.1 Summary

75

The electrical properties of PCN 1 were modulated by the post-synthetic modification (PSM). Redox-active PCN 1 with small pores due to high p–p stacking in the columnar structure was converted into PCN 1open with 1D channels by immersing the crystals in ethanol containing a proper amount of water. Water could dissociate NO3− and mono-dentate TPDAPs, and ethanol could extract the dissociated TPDAPs simultaneously. By removing mono-dentate TPDAP ligands in PCN 1, 13 Å  4 Å-sized channels surrounded by tri-dentate TPDAPs were formed in PCN 1open. Although PCN 1open had radical species by oxidation under air, it was insulating because of the absence of the p–p stacking which was a conductive pathway in PCN 1. On the other hand, the insulating PCN 1open could be modulated into conductive state (PCN 1spin) by intercalation of TPDAPs again owing to the formation of p–p stacking and radical species. The electrical conductivity of 3.18  10−4 S cm−1 in PCN 1spin was by four orders of magnitude higher than that of PCN 1, 1.23  10−8 S cm−1. This result supported the importance of conductive pathways and electroactive species for making conductive PCNs. The research presented in this thesis has investigated the control of electrical properties using redox activity and molecular arrangement in redox-active materials on their various applications. This study provides insight into the underlying mechanisms of electron transfer in redox-active materials, taking into account the molecular arrangement comprehensively. Eventually, this thesis provides an understanding of the relationship between the redox activity/molecular arrangements and electrical properties that will aid in the design of various electronic devices using redox-active organic molecules.

5.2

Limitations and Future Work

This thesis aims to provide general guidelines for designing electronic devices using redox-active organic molecules. There are several points to give feedback to the future works from the results of this thesis. The research in this thesis used the redox-active organic molecule, TPDAP; however, a better performance can be achieved by new organic molecules from the guidelines in this thesis. Therefore, a potential next step for this research can be a molecular design for target application considering proper redox activity and intermolecular interaction sites. For example, some molecules having higher highest occupied molecular orbital (HOMO) level than that of TPDAP can vary the available electrodes in the resistive switching memory devices although its singly occupied molecular orbital (SOMO) and lowest occupied molecular orbital (LUMO) should be considered together to make a stable switching and high retention. Intermolecular interactions are very important for forming a uniform film with electron pathways and should be considered in the molecular design process. Another point is placed on the selection of metal ions to design the PCNs for the detection of various analytes. Because cadmium ions have a strong affinity to water,

76

5 Summary and Outlook

nitrates could be easily replaced by water in PCN 1 and the redox activity of organic molecules can generate the electrical signal transduction. According to this guideline, the proper combination of target analytes and metal ions enables to fabricate various chemiresistive sensors using redox-active organic molecules. In common with the above, the molecular arrangement also should be considered to generate a proper electron pathway. Finally, technical problems such as low yield of materials, limitation of fabrication methods, and measurement techniques should be improved. Although organic molecules have big advantages in terms of designable structure, the yield of organic synthesis greatly affects commercialization as well as the research. In this sense, the synthesis method of TPDAP as a good candidate for various electronic devices should be improved for the intensive study and future commercialization. In addition, the method of integrating PCNs on a substrate also should be extensively studied to make nano-sized films as electronic devices. There are many reports about the fabrication of thin-film PCNs; however, they are still limited for functionalized ligands and substrates [3–8]. TPDAP enabled to make a microsized film, the practical application requires nanotechnology because of low-power consumption, high density, and compatibility with other electronics. Finally, the measurement techniques or methods which are not influenced by electrodes to investigate the mechanism should be developed and improved. The electrodes in ReRAMs restricted various technical measurements, so that it is difficult to investigate the switching mechanism using direct evidence. Various in situ measurements using specific architectures or methods to avoid electrode effects will help to understand the switching mechanism of ReRAM. Since organic molecules featuring a wide range of physical and chemical properties are designable, there are plenty of the opportunities for the rational design and fine-tuning of the desired materials for target applications. This thesis provides the guidelines for such designs, and it is expected that various electronic devices using redox-active organic molecule will be developed based on that.

References 1. 2. 3. 4. 5. 6. 7. 8.

Koo JY, Yakiyama Y, Kim J, Morita Y, Kawano M (2015) Chem Lett 44:1131–1133 Yakiyama Y, Ueda A, Morita Y, Kawano M (2012) Chem Comm 48:10651–10653 Betard A, Fischer RA (2012) Chem Rev 112:1055 Gliemann H, Woll C (2012) Mater Today 15:110 Zacher D, Shekhah O, Woll C, Fischer RA (2009) Chem Soc Rev 38:1418 Shekhah O, Liu J, Fischer RA, Wöll C (2011) Chem Soc Rev 40:1081–1106 Stavila V, Talin AA, Allendorf MD (2014) Chem Soc Rev 43:5994–6010 Falcaro P, Ricco R, Doherty CM, Liang K, Hill AJ, Styles MJ (2014) Chem Soc Rev 43:5513– 5560