Introduction

Fabrication of molecular wire and the electron transfer property through the molecular wire are important in chemical and biological processes and future application for molecular electronics. We have reported redox active bis(terpyridine)metal complex wires prepared by the stepwise coordination method with a surface-attaching ligand containing thiol and terpyridine moieties and an azobenzene-bridged bisterpyridine ligand and iron or cobalt ion as metal center of the terpyridine complex. In this thesis, general utility of the stepwise coordination method and electrochemical properties of the complex wires prepared by this method are described.

Procedure of the stepwise coordination method

The preparation of bis(terpyridine)iron, Fe(tpy)2, complex wires was carried out as follows (Figure 1). Flame-annealed Au(111)/mica electrode was immersed in a chloroform solution of a surface-attaching ligand ((A)2, (AH)2 or(AF)2). The electrode was subsequently immersed in Fe(BF4)2 aq, followed by immersion in a chloroform solution of bridging ligand (L1, L2 or LF) to make a film with a length of a single complexation cycle. These two processes were repeated stepwisely to grow films of multi-complexation cycles (described in Chapter 2). Preparation of Fe(tpy)2 complex wires with redox active terminal was accomplished by further immersion process. The gold electrode modified with desired number of complex cycles was soaked in Fe(BF4)2 aq and then in a solution of terminal ligand (TFc, THRu or TFRu) to fix a redox active terminal of the molecular wire (described in Chapters 3 and 4). Co(tpy)2 complex wires with Fe(tpy)2 as the terminal redox active moiety with Co(NH3)6Cl3 and L1 were also prepared in the manner similar to that of ferrocene-terminated Fe(tpy)2 complex wires noted above (described in Chapter 3).

Synthesis of linear and dendritic bis(terpyridine)iron complex wires by the surface bottom-up method and the electron transfer mechanism (Chapter 2)

Establishment of the fabrication techniques of desired molecular architectures is one of the important topics for the molecular device. Accomplishment of the structure control of the wire is very significant to realize elaborate molecular interconnection such as a multiarmed wire which is needed for future application of complex wires.

Stepwise preparation of linear (with L1) and dendritic (with L2) metal complex wires on gold electrode has been succeeded by the bottom-up method as shown in Figure 1, according to the contact angle measurement, STM measurement, and estimation of surface coverage from the redox peak area of cyclic voltammogram (Figure 2). Owing to the highly organized structures, we observed unique i - t characteristics by potential step chronoamperometry (PSCA). Such structures have not been reported previously for redox polymers with a low ordering of molecular wires. These i - t characteristics of both linear and branched redox-oligomer wires could be interpreted by a mechanism for through-bond electron transport by using two kinetic factors: k1 (s-1), for the electron transfer between the nearest redox site and the electrode, and k2 (cm2mol-1s-1), for the electron transfer between neighboring redox sites in a molecular wire (Figure 3).

Long-range electron transfer through a π-conjugated bis(terpyridine)metal complex wire (Chapter 3)

Recently, many studies have been carried out regarding the electron transport ability of molecular wires such as alkyl chain, β-conjugated polymers, DNA, and polypeptide. The long-range electron transfer is important for many future applications in solar energy harvesting and molecular electronic devices. Electron portability of a molecular wire can be evaluated with the attenuation factor,β, which is expressed in eq. 1,k = k0exp{-β(d-d0)} (1) where k denotes the electron transfer rate constant when the distance between donor and acceptor units is d (A), and k0 (s-1) is the rate constant when the donor and acceptor units are in the closest distance d0. The value of β can be obtained for many kinds of molecular chains, but little has been known about redox metal complex polymer chains.

Figure 4 shows the cyclic voltammograms of [A-1FeTFc] and [A-2FeL1-1FeTFc]. The current peak around 0.1 V (vs. ferrocenium/ferrocene (Fc+/Fc)) is assigned to the terminal ferrocene moiety of the wires. The other peak around 0.7 V (vs. Fc+/Fc) originates from redox reaction of FeIII(tpy)2/FeII(tpy)2. The ratio of the surface coverage of Fc to Fe(tpy)2 should be 1:1 for a film with n = 1 and 1:3 for a film with n = 3. The surface coverageT was estimated from the anodic current area of cyclic voltammogram; TFc andTFe(tpy)2 were 1.5 × 10-10 and 1.7 × 10-10 molcm-2, respectively, for n = 1, and 1.4 × 10-10 and 3.9 × 10-10 molcm-2, respectively, for n = 3. Surface coverages for the wires with a ruthenium complex terminal were similar to that of ferrocene-terminated wires. Results of IRRAS spectra for [A-1FeTFc], [A-2FeL1-1FeTFc], and [A-4FeL1-1FeTFc] also indicated that the molecular wires were successfully prepared with the desired conformation on gold surface.

The attenuation factor β was estimated by investigating the rate of electron transfer between gold electrode to the redox active terminal moiety through the complex wires with varying the number of bis(terpyridine)metal complex units within a molecular wire. The electron transfer rate constant was directly measured from the slope of ln i - t plots of PSCA. The electron transfer rate constant values of the terminal ferrocene moiety through Fe(tpy)2 complex wires [A-(n-1)FeL1-1FeTFc] are shown in Figure 5(A). From the results, the attenuation factor β was calculated at 0.015 ± 0.007 (Figure 5(B)). This value is considerably smaller compared with theβ・ values of other molecular wire systems previously reported. Consequently, the above results strongly indicate that the complex wire has a distinguished long range electron portability. In addition, the dependences ofβ on the measurement condition, surface-attaching ligand, terminal redox active moiety, bridging moiety of ligand, potential of complex within the wire, and metal center of the metal complex wires were investigated. These results revealed that the value ofβ was an intrinsic parameter of the composition of the metal complex wire and controllable by bridging moiety and metal center of the complex wire. Theβ value of Fe(tpy)2-terminated Co(tpy)2 complex wire was estimated at 0.004 ± 0.002 which is smaller than that of Fe(tpy)2 wire. Therefore, the Co(tpy)2 complex wire has a higher long range electron portability compared with the Fe(tpy)2 complex wire.

Trials of highly functional metal complex wires (Chapter 4)

In Chapter 4, I tried to build highly functional systems by exploiting hetero-metals, hetero-ligands, and a three-way ligand to bring out novel functions in the molecular assembly.

Electron transfer behaviors of ferrocene-terminated hetero-metal complex wires, [A-1FeL1-1CoL1-TFc] and [A-1CoL1-1FeL1-TFc] showed that the interesting insight that the long-range electron transfer ability of the complex wires can be controlled with the sequence of the metal ions within the complex wire. The fact indicates that hetero-metal alignment within metal complex wires has a potential to fabricate complex wires with a wide variety of properties by using many kinds of metal ions.

Potential staircase of complexes within the wire was introduced by hetero-ligand fabrication of the complex wires with L1 and fluorinated ligand LF by taking an advantage of electron-withdrawing effect of fluorine. Although the wires with potential staircase are anticipated to exhibit rectification behavior, they did not have any rectification property. However, the wires with the terminal ruthenium moiety have superior long-range electron portability compared with the homo-ligand wires.

The dependence of long-range electron transportability on the structure of the complex wire was investigated. Dendritic complex wires with the terminal ruthenium complex, [AH-(n-1)FeL2-1FeTHRu] can be prepared on surface by the stepwise coordination method using a branched ligand, L2. The electrochemical behavior of the ruthenium complex through the dendritic complex wire was investigated. The results revealed that the dendritic complex wire had a rectification property that can flow electron from the electrode to the complex wire faster compared with the flow in the inverse direction (Figure 6).

Conclusion

The results of the thesis show the excellent properties of π-conjugated metal complex wire, conductivity, intra-system electron transfer, easily controllable length and structure. In addition, the insights into the correlation between electrochemical property and hetero-metal, potential structure within the wire, and structure of wire would serve as a guidance of developing highly functionalized molecular wires.

Figure 1 Procedure of stepwise coordination method and chemical structures of ligands used in this thesis.

Figure 2 Plots of the surface coverage of redox-active sites, T, versus n, for [A-nFeL1] A), [A-nFeL2] B). The lines denote the relationships T= C×n for A, T= C×(2n-1) for B.

Figure 3 Schematic illustration of the electron transfer mechanism and the analyses of chronoamperograms.

Figure 4 Cyclic voltammograms of [A-1FeTFc] (A) and [A-2FeL1-1FeTFc] (B) in 0.1 M Bu4N-ClO4-CH2Cl2.

Figure 5 Plots of k versus n (A) and ln k versus distance (B) for [A-(n -1)FeL1-1FeTFc].

Figure 6 Schematic illustration of rectification property through the dendritic complex.

本論文は5章からなり、第1章では研究の背景および目的、第2章では表面上逐次錯形成反応の樹状型ワイヤーへの拡張性の検討および錯体ワイヤーからなる膜中での電子移動機構の解明、第3章では金属錯体連結ワイヤーの長距離電子輸送能の検討、第4章では単一ワイヤー中に異種金属や架橋配位子を組み込んだワイヤーおよび樹状構造を持つワイヤーの電子輸送特性、第5章では全体の総括が述べられている。以下に各章の概要を記す。

第1章では、研究の背景について述べている。現在に至るまで数十年にわたり半導体ベースのデバイスは微細化することにより高機能化を果たしてきた。しかしながら現在デバイス作製に用いられている技術は大きな材料から小さなデバイスを作製するトップダウン的な考え方に基づいており、その手法による微細加工技術は近い将来に理論的、実質的な限界が訪れると考えられている。そこで、これに代わる方法論として原子や分子を組み上げることによりデバイスを組み上げる、ボトムアップ法が注目を集めている。高度な機能を有するデバイスを作製するには、定量的かつ規則的に分子を配列させて機能の発現を実現することが求められる。そこで本研究では、金属錯体を規則的に連結させたワイヤーの逐次的かつ定量的錯形成による作製と、その錯体連結ワイヤーの電子伝導特性の研究を行った。

第2章においては、直線状および枝分かれを持ったテルピリジン架橋配位子を用いた金表面上逐次錯形成により錯体を直線状および樹状に連結した金属錯体連結ワイヤーの作製を行った。電気化学測定、STM測定、接触角測定の結果より表面上に1次元状ワイヤーおよび樹状型ワイヤーの構築が確認された。またそれらの金属錯体ワイヤーからなる膜中での電子移動機構の解析をクロノアンペロメトリーにより行い、その結果をシミュレーションによって再現することによって、これらの錯体ワイヤー膜中での電子移動が分子鎖内で起こることを明らかとした。

第3章では、π共役錯体連結ワイヤーを介した長距離電子輸送に関して述べている。具体的には、長さの異なる錯体ワイヤーの末端にレドックス活性基を固定し、そのレドックス活性基と電極間の電子移動速度の距離依存性から長距離電子輸送能の評価を行った。錯体ワイヤーの長距離電子輸送能を支配する要素を明らかとするため、測定条件、電極固定用配位子、末端レドックス活性基、架橋配位子の架橋部位、錯体の酸化還元電位、錯体の中心金属の種類による特性の違いの検討を行った。その結果、長距離電子輸送能は架橋配位子の架橋部位および錯体の中心金属により制御が可能であることが示された。更にコバルト錯体連結ワイヤーは距離に対して電子移動速度の減衰がほとんど起きない極めて高い長距離電子輸送能を持つことが明らかとなった。

第4章においては、異種金属を組み込んだワイヤー、ワイヤー内にポテンシャル勾配を持つワイヤー、樹状型ワイヤーを介した長距離電子輸送能について述べている。異種金属を組み込んだワイヤーではワイヤー中の金属イオンの順序によって電子移動特性が異なることが明らかとなった。ポテンシャル勾配を持つワイヤーは、ポテンシャル勾配を持たないワイヤーと比較して電子移動速度が減衰しないという結果が得られ、ポテンシャル勾配を持つワイヤーは高い長距離電子輸送能を持つことが示された。樹状型ワイヤーは、電極から錯体に電子を流しやすく、逆方向には電子を流しにくいという構造に由来する整流作用を持つことが明らかとなった。

第5章では、以上の結果を総括し、今後の研究展望を述べている。

以上、本論文では、表面上逐次的錯形成反応の汎用性を示すとともに作製される錯体連結ワイヤーが優れた伝導特性を持つこと、またワイヤー中の金属イオンの順序や架橋配位子の組み合わせ、ワイヤーの構造により伝導性を変化させることが可能であることを記述している。本博士論文において示されたπ共役錯体連結ワイヤーの高い長距離電子輸送能およびその構造との相関は高度に機能化された分子電子素子の実現に道筋をつけるものであり、分子科学の新領域開拓に大きく貢献するものと期待される。なお、本論文第2章は金井塚勝彦、村田昌樹、西原 寛との共同研究、第3章は、金井塚勝彦、栗田知周、長津聡明、瀬川 祐、利光史行、邨次 智、宇津野充弥、久米晶子、村田昌樹、西原 寛とのとの共同研究であり、一部は既に学術雑誌として出版されたものであるが、論文提出者が主体となって実験および解析を行ったもので、論文提出者の寄与が十分であると判断する。

したがって、博士(理学)の学位を授与できると認める。