Hydrogen Bonding-Controlled Photoinduced Electron and Energy Transfer
- 1k Downloads
Photoinduced electron and energy transfer processes play an important role in photosynthesis and optoelectronic conversion. The investigation of photoinduced electron and energy transfer, wherein donor and acceptor are assembled via noncovalent interactions, has attracted much interest. Among these noncovalent interactions, H-bonding interaction has emerged as a powerful tool to construct high array of supramolecular architectures owing to their tunable binding constant, high directionality and selectivity. This chapter provides an overview of the recent developments in H-bonding-based energy/electron transfer. Advances in this research field, including (1) the basic theories for energy transfer and electron transfer processes, (2) recent progresses of energy transfer and electron transfer based on H-bonding, (3) their applications in constructing optoelectronic devices, light-harvesting systems, wide-range color display, and storage materials, are presented.
KeywordsElectron Transfer Energy Transfer High Occupied Molecular Orbital Lower Unoccupied Molecular Orbital Electron Transfer Reaction
Photosynthesis is a natural energy conversion system that converts solar energy into chemical energy, the primary processes of which are a cascade of photoinduced excitation energy transfer to the reaction center (RC) and the subsequent rapid electron transfer to generate charge-separated state. Over the past decades, a series of significant results have been obtained for the study of photosynthesis: (1) The structure of the protein subunits in the photosynthetic RC has been ascertained. The RC consists of a pair of chloroplast–protein complex and various quinone derivatives with electron transfer chains, which are located in the grana membrane. Its periphery is the light-harvesting (antenna) complexes made up of a large number of protein-embedded pigments of chlorophyll and carotenoid . Deisenhofer, Huber, and Michel were awarded jointly by the Nobel Prize in Chemistry in 1988 “for the determination of the three-dimensional structure of a photosynthetic reaction centre.” (2) Research revealed that photosynthesis initiates with the absorption of light by pigments bound to the antenna, which is transferred over nanometer distances to RC through efficient long-distance energy transfer, where it is converted into chemical energy via charge separation produced by rapid electron transfer [2, 3, 4, 5, 6, 7]. (3) A series of theories on energy transfer and electron transfer have been established, such as Förster [8, 9] and Electron Exchange mechanism  of energy transfer, as well as Marcus theory [11, 12, 13, 14, 15] and Rehm-Weller equation [16, 17] of electron transfer. These theories will provide certain directions for photosynthesis studies.
The success and importance of photosynthesis have inspired researchers to develop artificial molecular systems for solar energy conversion based on efficient energy/electron transfer processes, which require the strictest control of the direction and efficiency of the transductions. Such a complex goal can be achieved by integrating energy/electron transfer donor–acceptor systems in a single molecular structure [18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34] or by self-assembling them through highly selective and directional supramolecular interactions [35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51]. However, the donor–acceptor system based on covalent bond connection has become more and more complex, which will make the synthesis much more complicated. In this context, the investigation of the photoinduced energy and electron transfer, wherein donor and acceptor are assembled via noncovalent interactions, has attracted much interest. Among various noncovalent interactions (π–π stacking, hydrogen (H)-bonding, ion–ion, ion–dipole, dipole–dipole, electrostatic interactions), H-bonding interaction has emerged as a powerful tool to construct high array of supramolecular architectures [52, 53, 54, 55, 56]. In particular, the energy and electron transfer in biological photosynthesis in nature are regulated through a network of H-bonds. H-bonding is directional and has a wide range of interaction energies (4–120 kJ mol−1) that are tunable by adjusting the number and type of H-bonds, their relative orientation, and their position in the overall structure. Donor–acceptor chromophores undergoing photoinduced energy and electron transfer are well organized through H-bonding. Studies on the energy/electron transfer process within H-bonding system will give insights into how various factors, such as driving force, H-bonding pathways, and interchromophore orientations can influence the transfer rate, and thus regulate its biological charge separation processes.
This chapter is intended to describe the major developments and breakthroughs in the H-bonding-based energy/electron transfer in the last two decades. We will start with the introduction of the basic theory for energy transfer and electron transfer processes, then review the most recent progress of energy transfer and electron transfer based on H-bonding and their applications.
1.2 Basic Theories for Photoinduced Electron and Energy Transfer
In this section, a brief discussion of electron transfer and energy transfer theories is presented. The purpose of this short exposition is to provide the basis for a discussion regarding the features of theory that are addressed by current H-bonding systems for photoinduced electron and energy transfer.
1.2.1 Photoinduced Electron Transfer (PET) Processes
1.2.2 Photoinduced Energy Transfer Processes
It has been generally accepted that two main mechanisms are involved by which a donor (D) in the excited state can pass its energy to a proximal acceptor (A) in the ground state, which results in the decrease of the emission intensity of the donor and transfers the energy to the fluorescent or nonfluorescent acceptor: Fluorescence resonance energy transfer (FRET, also named as Förster mechanism) [8, 9] and electron exchange (Dexter interactions) .
It can be seen that the rate and efficiency of the electron/energy transfer is directly related to the distance, their relative spatial orientation, and the conformation of the bridge between donor and acceptor, as well as the direct orbital overlap between the components. Integrating energy/electron transfer donor–acceptor systems by covalent bonds has allowed reasonably precise knowledge of the above information. Such studies have contributed a great deal to our understanding of the fundamental photophysical processes in photosynthetic center and the development of artificial photosynthesis. However, the donor–acceptor assemblies have evolved from binary, ternary to multiple systems. The preparation of large, sophisticated species by covalent bonds will make the synthesis much more complicated, sometimes beyond the scope of more conventional synthetic method. Moreover, nature relies on supramolecular methods of assembly. In natural light-harvesting assemblies, the crucial parameters, such as distance and orientations, are satisfied by organizing key pigment molecules in a fixed spatial relationship by the surrounding medium with the help of noncovalent interactions. In this context, self-assembling donor and acceptor through highly selective and directional supramolecular interactions has attracted much interest, in which H-bonding is undoubtedly the most compelling and well-studied supramolecular motif. In the following sections, we will focus on photoinduced electron and energy transfer processes based on various H-bonds, as well as their applications in constructing optoelectronic devices, light-harvesting system, wide-range color display, and storage materials.
1.3 Hydrogen Bonding-Controlled Photoinduced Electron Transfer
More and more attentions have been paid to H-bonding-based electron transfer in recent years. These systems have been extended to the donor–acceptor assemblies linked by two-point H-bonds, triple H-bonds, or multiple H-bonds. In this section, we will first present the important mechanistic progress in the last two decades and then discuss recent representative examples based on H-bonding.
1.3.1 Mechanistic Studies
It has been long recognized that H-bond is not only the bridge between electron donor and acceptor in biological matrix, but also involved directly in the electron transfer process and plays an important role in mediating electronic tunneling. However, the binding strength of a single H-bond is rather weak compared to those of the popularly studied covalent bond. It has been always a primary focus whether H-bonding is an appropriate electronic coupling tunnel. This question was solved until three supramolecular bischromophoric model systems 1–3 were synthesized by Therien to evaluate the relative magnitudes of electronic coupling provided by H, σ, and π bonds . (Porphinato)zinc donors and (porphinato)iron(III) chloride acceptors in 1–3 are separated by a virtually constant distance, thus establishing uniform driving force and reorganization energy for the ET reaction. Measurement of the photoinduced electron transfer rate constants enables a direct comparison of how well these three types of chemical interactions facilitate electron tunneling due to their exactly the same number of bonds and identical electron tunnelings. The electron transfer rates and driving forces are k ET = 4.3 × 109 s−1, ∆G 1 = –0.87 eV for σ bond interface in 1; k ET = 8.8 × 109 s−1, ΔG 2 = –0.87 eV for π bond interface in 2; and k ET = 8.1 × 109 s−1, ΔG3 = –0.70 eV for H-bond interface in 3 respectively. The surprising result is the electronic coupling modulated by the carboxylic acid/H-bond interface is superior to that provided by an interface composed of an analogous number of σ bonds, despite its relatively smaller driving force (ΔG 3 = –0.70 eV) than σ bonds (ΔG 1 = –0.87 eV). These results demonstrated the key role played by H-bonds in biological electron transfer processes and helped to analyze the through-protein electron transfer rate data as well as to strengthen the theory in predicting the path traversed by the tunneling electron in a biological matrix.
Similar H-bonding-mediated PET was also observed by Nocera et al. in the Zn′′PCO2H porphyrin and 3,4-dinitrobenzoic acid electron donor–acceptor system 4 . Fluorescence quenching and time-resolved absorption experiment demonstrated that photoinduced electron transfer occurred from photoexcitable Zn′′PCO2H porphyrin donor to the electron acceptor 3,4-dinitrobenzoic acid through a protonated (1-H) or deuterated (1-D) dicarboxylic acid interface. It is significant to observe the pronounced deuterium isotope effect of k H/k D = 1.7 and 1.6 for the charge separation and recombination rates, respectively. Thus, H-bonding network formed between dicarboxylic acid must have been involved in the electron transfer process. The results indicated that H-bonding interfaces are not only important in the supramolecular preorganization of acceptor/donor pairs for energy and electron transfer but also may directly mediate the electron transfer event.
In 2000, Isied et al. reported a series of transition metal complexes with H-bonding molecular recognition sites capable of forming noncovalent donor acceptor complexes (k a = (2−5) × 105 M−1 in CH2Cl2) . They determined for the first time the dependence of the rate of intramolecular electron transfer on the reaction driving force across a H-bonding interface. Fluorescence lifetime measurement indicated the electron transfer reactions from the excited states of the *MIIG to H-′MIII in MIIG–H–′MIII complexes 5. The emissions of all complexes decay biexponentially with short lifetimes corresponding to the intramolecular electron transfer (τ = 4–7 ns) and longer lifetimes corresponding to the emission of the uncomplexed *MIIG (M = Ru, τ′ = 770 ns; M = Os, τ′ = 90 ns). The forward and reverse electron transfer reaction rates of the MIIG–H–′MIII complexes 5 were plotted against their respective free energies. Despite that it is incapable of distinguishing the similar values of λel and Had obtained from the classical Marcus equation, this work demonstrated how the driving force dependence of the rates of intramolecular electron transfer reactions in these H-bonded complexes can be used to determine the electronic coupling matrix element for electron transfer. Comparison of the H-bonded systems with similar covalently bound complexes 6 at similar distances shows that an H-bonding interface can be as effective a bridge for electron transfer as covalently bound bridges.
The above examples show that H-bonding is effective electron conduit for donor–acceptor interaction. This discovery provides strong support for the assumption that H-bonding plays an important role in the long-range electron transfer in biological system, which is a milestone for the development of electron transfer.
1.3.2 Electron Transfer Based on Double and Triple Hydrogen Bonding
In the past decades, porphyrins and metalloporphyrins have been probed as integrative components in photosynthetic RC models due to their rich photo- and redox chemistry. They function as light harvesting dyes through most of the visible part of the solar spectrum and as donors to electron transfer and transport. On the other hand, fullerene was frequently employed as electron acceptor due to its small reorganization energy in electron transfer reactions, as well as its formation of unprecedented long-lived radical ion pair states through ultrafast charge separation together with very slow charge recombination feature. Thus, porphyrins/metalloporphyrins and fullerenes are molecular architectures ideally suited for electron donor–acceptor systems to transmit and process photoenergy. Photoexcitation of the porphyrin/metalloporphyrin by visible light is readily followed by an electron transfer to fullerenes, guaranteeing the formation of a radical ion pair state.
Guldi et al. reported a set of two-point amidinium–carboxylate H-bonded C60–porphyrin ensembles (8) . 1H NMR, absorption and fluorescence spectroscopy confirmed the formation of 1:1 complex between porphyrin and C60. The binding constant deduced from nonlinear least-square analyses of the fluorescence intensity versus concentration of C60 is 2.1 × 107 M−1 in toluene. Transient absorption studies reveals that the singlet excited states of zinc porphyrin in 8b decay with rates of about 1010 s−1, from which an electronic coupling of 36 cm−1 between zinc porphyrin and C60 was calculated. A fast, efficient, and longer-lived formation of radical ion pair states (10 μs in THF) was facilitated by such strong electronic couplings. It can be seen that H-bonding is advantageous in facilitating electron transfer processes.
Bis(zinc porphyrin)-fullerene supramolecular triad 9 could also be constructed using a diacetylamidopyridine/uracil complementary H-bonding motif . The formation constant for the supramolecular triad, calculated by the Benesi-Hildebrand plot of fluorescence quenching, was 6.2 × 103 M−1 in acetonitrile/o-dichlorobenzene (6:4) with 1:1 complexation between the (ZnP)2 and C60 entities. Both zinc porphyrin entities of the dimer seem to be involved in the electron transfer process. The measured k CS and k CR values were found to depend on the positioning of the porphyrin entity with respect to the fullerene entity (near or far), thus delineating the structural importance of the studied supramolecular triad in controlling the electron transfer rates. The calculated damping factor was suggestive of through-space ET in the triad.
Considerable efforts in Sessler group have been devoted to constructing various electron donor–acceptor systems based on cytosine–guanine base-pairing motif to model the basic electron transfer process. Ensemble 10 is an H-bonded assembly formed between a zinc porphyrin appended to guanine and quinine appended to cytosine . PET was observed from the excited zinc porphyrin to quinone with k ET = 4.2 ± 0.7 × 108 s−l. However, the low association constant (3,100 ± 470 M−1) and the large degree of flexibility in the system would complicate the analyses. It is possible that electron transfer in 10 occurs by diffusional encounter between partners within a H-bonded complex. A more rigid system 11, in which the flexible alkane chain was removed, was then synthesized to avoid the problem . The corresponding association constant in the ensemble was calculated to be 8,990 ± 600 M−1, which is significantly higher than that observed for the more flexible aggregate 10. Due to the rigidity, electron transfer most probably occurred via a through-bond process involving the H-bonded network with k ET = 8 × 108 s−l.
The complementary imide–diaminotriazine triple H-bonding was also widely utilized in the construction of electron donor–acceptor system. Schenning et al. reported the collective and hierarchical self-assembly of oligo(phenylenevinylene) (OPV) and perylenediimide (PDI) into chiral fibers through diaminotriazine-diimide triple H-bonding and π–π interaction . Supramolecular OPV-PDI-OPV entity 12 was initially formed via H-bonding and subsequently self-assembled into chiral stacks by π–π interaction. UV/vis, fluorescence, and CD titration experiments confirmed the formation of the 2:1 complex and its further stacking into J-aggregates with a helical screw sense. Upon photoillumination of these fibers, electron transfer from the OPV-donor to the perylene-acceptor chromophore took place, leading to charge separation within the aggregated dyes. This process was ascertained by transient absorption spectroscopy. Later, their further studies of ensemble 13 indicated that there is a distinct dependence of aggregate stability on OPV conjugation length [67, 68]. Different CD effects and electron recombination rates were observed due to different packing ways and their modes of connection (H-bonded or covalent) of the donor and acceptor dyes. Such well-defined co-aggregated dyes may serve as valuable nanoscopic functional building blocks for solid-state devices. These studies carried out with the electron donor OPV and electron acceptor PDI couple shed light on the importance of the chromophore organization in thin films for application in efficient organic photovoltaic devices.
In 2013, Rodríguez-Morgade et al. reported the assembly of phthalocyanine and bifunctional PDI through malamine-imide triple H-bonding, to afford electron donor–acceptor array 14 . In this study, 14 was probed in the ground and excited state by steady-state and time-resolved techniques. It was demonstrated that electronic interaction between phthalocyanine and PDI in 14 gave rise to intramolecular electron transfer upon selective photoexcitation of PDI, affording a several-ns-lived PDI•−/H2Pc•+ species. It is also interesting to observe the presence of the corresponding “SYN” and “ANTI” isomers in the arrays by 1H NMR.
A charge-separated (CS) state will be produced in a photoinduced electron transfer event to drive further chemical reaction in natural photosynthesis. Obtaining long-lived CS state following electron transfer is a key for improving the efficiency of solar energy conversion, which is in essence to accelerate the forward ET and to slow down the charge recombination (CR). Over the past decades, researches have been focused on designing versatile electron donor–acceptor arrays to give rise to long-lived CS states with high quantum yield upon photoexcitation.
Sessler et al. synthesized a rigid ensemble 16, in which dimethylaniline and anthracene donor–acceptor couple was constructed via noncovalent guanosine–cytidine base-pairing interaction [72, 73]. Steady-state fluorescence quenching measurements, time-resolved fluorescence quenching and transient absorption measurements indicated that photoinduced electron transfer occurred from the dimethylaniline donor to the singlet excited state of the anthracene acceptor upon excitation at 420 nm. The rate constants for photoinduced intraensemble electron transfer and subsequent charge recombination are k CS = (3.5 ± 0.03) × 1010 s−1 and k CR = (1.42 ± 0.03) × 109 s−1, respectively, which led to a short lifetime for charge-separated state as 705 ps . Later, the same group developed another porphyrin–fullerene donor–acceptor ensemble 17 based on Watson–Crick hydrogen bonding paradigm to further improve the CS state . Fullerene was employed here because small reorganization energy in ET reactions was required which should accelerate forward ET and to slow down back ET, resulting in the formation of long-lived CS states. Steady-state fluorescence quenching spectroscopic analyses indicated an efficient electron transfer occurs from the excited zinc porphyrin to fullerene with binding constant as (5.1 ± 0.5) × 104 M−1 in CH2Cl2. Time-resolved fluorescence measurement indicated a bi-exponential decay (τ 1 = 2.1 ns, τ 2 = 0.6 ns) of porphyrin fluorescence in the presence of fullerene, which implies the PET process within the ensemble. From the derived lifetimes, a rate constant (ca. k cs = 1.2 × 109 s−1) for the forward electron transfer was estimated. The oxidized zinc porphyrin radical cation (ZnP•+) and the fullerene radical anion (C60 •−) were observed from transient absorption spectroscopic measurements, with weak bands at 600–800 nm range and 1,000 nm, respectively. Kinetic analysis of the transient absorption data resulted in a long-lived CS state of 2.02 μs. This value is higher than those reported for related covalently linked C60-ZnP dyads as a result of the beneficial effect of the hydrogen bonds.
Mendoza et al. reported a series of C60–TTF ensembles 18–21, in which a photoexcited-state acceptor (C60) and an electroactive donor (TTF) were held together through complementary guanidinium-carboxylate H-bonding . Transient absorption spectroscopy supported the formation of C60 •−·TTF•+ by the characteristic fullerene radical anion (1,000 nm) and TTF•+ (450 nm) transitions. The lifetimes of the radical pairs are in the range of hundreds of nanoseconds to microseconds, several orders of magnitude higher than those reported for covalently linked C60–TTF dyads [75, 76, 77]. Charge recombination rates in the C60·TTF ensembles are typically around 106 s−1 depending on electronic coupling between donor and acceptor. Photophyscial investigation of C60·TTF ensembles with different chemical spacers (i.e., phenyl vs. biphenyl) and two functional groups (ester and amide) supports through-space electron transfer.
Kim et al. reported the formation of supramolecular complex 22 of the benzo-annulated TTF calix  pyrrole (TTF-C4P) as an electron donor with porphyrin as an electron acceptor in benzonitrile . The TTF-C4P binds to the carboxylate moiety of the porphyrin through H-bonding with a 1:1 stoichiometry and a binding constant of 6.3 × 104 M−1 in this solvent at 298 K. Photoexcitation of complex 22 formed between these two components in PhCN at 298 K afforded the CS state, which was characterized by forward and backward intramolecular ET rate constants of 2.1 × 104 and 3.6 × 102 s−1, respectively. The triplet CS state produced upon photoirradiation of porphyrin was found to be 2.8 ms, one of the longest known lifetimes for a CS generated via a photoinduced electron transfer process within a noncovalently bound complex. It is interesting to see that free metal porphyrin acts as electron acceptor instead of donor in this case.
1.3.3 Photoinduced Electron Transfer Based on Quadruply and Multiple Hydrogen Bonding
In 1996, Meijer et al. built the first donor–acceptor–donor–acceptor (DADA) quadruple H-bonding array by acylated triazines and pyrimidine derivatives, whose highest association constant reaches 2 × 105 M−1 [79, 80]. In 1998, they succeeded in building a self-complementary acceptor–acceptor–donor–donor (AADD) quadruple H-bonding array based on 2-ureido-4[1H]-pyrimidinone (UPy) unit, which can exist as homodimers in nonpolar solvents with the association constant higher than 107 M−1 . The UPy module represents a fascinating self-complementary AADD quadruple H-bonding module. The association ability of UPy is much higher than that of the DADA array. Owing to their great binding strength and directionality, quadruple H-bonding arrays have shown extensive applications in assembling and disassembling supramolecular systems, as well as in creating H-bonded donor–acceptor dyads with efficient energy/electron transfer process.
Multiple H-bonding motifs, such as Barburic acid–Hamiliton receptor pairing, whose K a is in the range of 103–1012 M−1 in apolar solvents, have also been employed in constructing electron donor–acceptor assemblies. In 2007, Guldi and Hirsch et al. developed porphyrin–fullerene electron donor–acceptor systems 25 held together by a Hamilton-receptor-based H-bonding motif . In this system, fullerene derivative is a monomalonate, where one branch carries a cyanuric acid moiety. The corresponding linkers are either propylene or hexylene chains. The second malonate branch bears dendritic termini to improve solubility. The Hamilton-receptor counterpart is coupled to a library of porphyrin derivatives involving either tin or zinc as central metals leading to new porphyrin building blocks. The association constants of the corresponding 1:1 complexes 25 connected by six H-bonds were determined to be in the range between 3.7 × 103 and 7.9 × 105 M−1 in CHCl3 by 1H NMR and fluorescence titration experiments. In response to visible light irradiation, 25 gave rise to a fast charge separation (k CS = 4.3 × 109 s−1) evolving from the photoexcited ZnP chromophores.
Within this context, Guldi and co-workers further studied the electron transfer along π-conjugated spacers (p-phenyleneethynylene, p-phenylene-vinylene, p-ethynylene, and fluorine) in a series of supramolecular porphyrin/fullerene hybrids (26) through Hamilton receptor/cyanuric acid binding motif . Selective photoexcitation of the porphryins (2.0 eV) triggers electron transfer in most hybrids to yield one-electron-reduced fullerenes and one-electron-oxidized porphyrins (1.4 eV). Electronic communications in the systems are controlled by the conjugated spacers (length and nature). In terms of simple distance dependence, transient absorption measurements confirm that the charge separation rate constants are 3.1 × 109 and 1.1 × 1010 s−1 for 26a and 26c, respectively. For 26b, no electron transfer activity is found on the time scale of up 3,000 ps, which suggests that incorporation of an extra p-phenylene-ethynylene shuts down electron transfer in 26b. On the other hand, the rate constants in 26a, 26d, and 26f increase with decreasing attenuation factor of the spacer. Charge recombination rates also follow the resembling trend. In p-phenylene-ethynylene-based systems, the values of 1.4 × 107 (26a) and 3.3 × 107 s−1 (26c) reflect the change in spacer length, 12.2 versus 7.9 Å. On the other hand, varying the spacer from p-phenylene-ethynylene, fluorene, to p-phenylene-vinylene (i.e., 26a, 26f, and 26d) results in steady increase in charge recombination rate constant from 1.4 × 107 to 1.9 × 107 and to 2.5 × 10−7 s−1. This study demonstrates the charge separation recombination kinetics is dependent of either the length or the β-factor of the employed spacer.
de Cola and co-workers studied strong ground-state electron donor–acceptor (EDA) interactions in 1:1 H-bonded assemblies 27, constructed by a barbituric acid-substituted fullerene derivative and corresponding Hamiliton receptor bearing thienylenevinylene (TV) units . The observed intense EDA absorption feature between the two TV and C60 moieties comes from the close proximity of the redox-active moieties within the assembly. The binding constant for the assemblies is 5,500 M−1. The photoinduced electron transfer from electron-rich thienylenevinylene subunits to the fullerene is very fast (k et = 5.5 × 1012 s−1), as determined by fs-time-resolved transient absorption spectroscopy.
1.3.4 Applications of Hydrogen Bonding-Based Photoinduced Electron Transfer
Much effort has been devoted to applying electron transfer in H-bonding system for the construction of optoelectronic devices and organic solar cells. For example, fullerene derivative and perylene bisimide have been assembled to form a H-bonded supramolecular system 28 through triple amino-carboxylic acid interaction . Under 63.2 mW/cm2 white light irradiation, the film made from the assembly on indium tin oxide (ITO) electrodes generated a steady and rapid photocurrent. The response of on/off cycling was prompt and reproducible.
The group of Meijer and Schenning has constructed ambipolar field-effect transistors from imides-diaminotriazines H-bonded p-n dyad complexes 29 based on OPV4T in combination with PBI-2 . The transistors show two independent pathways for charge transport. In contrast, processing of OPV and PBI that are not connected by H-bonds formed charge transfer donor–acceptor complexes. They showed no mobility in field-effect transistors, presumably due to an unfavorable supramolecular organization.
Xue et al. reported complex 30 of π gelator and fullerene derivative with photoinduced electron transfer for photocurrent generation . In this study, π-gelator as an electron donor formed a complex with a fullerene derivative (electron acceptor). In the hybrid gel phase, the complex self-assembled into nanofibers in which C60CO2H and the gelator were packed into 1D superstructures being interdigitated each other. Such an ordered microstructure ensured efficient charge-carrier transport so that large photocurrents were achieved. The method provides a convenient way to fabricate active layers with high performance in a photovoltaic system by utilizing the gelation of D-A-type π-conjugated gelators.
Qin and co-workers reported the preparation of polythiophene block copolymers (BCPs) which were selectively functionalized with diaminopyrimidine moieties (electron donor) and thymine-tethered fullerene derivative (electron acceptor) . It was shown that the stability of the polymer solar cell (PSC) devices employing these copolymers (ensemble 31) was significantly improved through the “three-point” complementary H-bonding between diaminopyrimidine and thymine moieties. More interestingly, bulk heterojunction (BHJ) morphologies could be systematically adjusted by varying the blend ratio of BCPs to fullerene derivatives. However, the overall device efficiency was still quite low compared with benchmark poly(3-hexylthiophene) (P3HT)/phenyl-C61-butyric acid methyl ester (PCBM) BHJ devices. To improve power conversion efficiencies (PCEs), the authors further synthesized new diblock polythiophene copolymer, P4, having a relatively shorter functionalized block carrying isoorotic acid moiety and a diaminopyridine tethered fullerene (PCBP) . Self-assembly between P4 and PCBP (ensemble 32) through “three-point” complementary hydrogen bonding interactions is utilized to control and stabilize blend morphologies. Solar cells employing these materials show not only comparable PCEs with standard P3HT/PCBM devices but also much enhanced stability and tunable active layer morphologies by simply varying polymer/fullerene weight ratios.
1.4 Hydrogen Bonding-Controlled Photoinduced Energy Transfer
1.4.1 Photoinduced Energy Transfer Based on Triple Hydrogen Bonding
Watson–Crick nucleobase pairing is one of the most popular and well-studied triple H-bonding motifs. Sessler and co-workers assembled triple H-bonding ensembles 33–35 based on Watson–Crick nucleobase pairing between guanine (G) and cytosine (C) and studied their energy transfer processes [95, 96, 97]. Zinc porphyrins and free base porphyrins were employed as energy donor and acceptor. The association constant for G and C association is around 2 × 104 M−1 in CH2Cl2. In the first generation ensembles 33, photoinduced TTET was observed within the H-bonded complex following excitation of the porphyrin . However, ensembles 33 proved to be very flexible so that no SSET was observed. The intraensemble diffusional encounter between the donor and acceptor in such flexible system could not be excluded for the observation of TTET. They further developed rigid ensembles 34 and 35, in which donor and acceptor are connected by a phenyl group to G and C recognition units respectively [96, 97]. Both singlet and triplet energy transfer were observed in ensembles 34 and 35. The singlet energy transfer dynamics was consistent with Förster mechanism, with transfer rate as k ET = 9 × 108 M−1 and transfer efficiency as Φ = 0.6. Due to their rigid structures and large center-to-center distance between donor and acceptor (ca. 22.5 Å), intraensemble triplet energy transfer with electron exchange mechanism occurred through H-bonding interface rather than through space. The results indicate that H-bonding is not involved in mediating FRET (Förster mechanism) process but play an important role in mediating triplet energy transfer process through electron exchange mechanism.
Another classic triple H-bonding motif is the combination of either barbituric or cyanuric acid and melamine, which was originally described by Whitesides . In 2006, Schenning and Meijer et al. reported helical co-assemblies 36 yielded by π–π stacking of H-bonded porphyrin and oligo(p-phenylene vinylene)s (OPVs) based on this motif . In this study, a porphyrin derivative bearing enantiomerically pure 3,7-dimethyloctyloxy side chains was equipped with cyanuric acid (CN-Por), resulting in two sites available for H-bonding to OPVnT. The formation of 1:2 complexation was confirmed by fluorescence titration. This trimer was then organized into a helical structure by π–π stacking and H-bonding. Fluorescence exchange experiment in methylcyclohexane demonstrated the energy transfer from OPV4T to porphyrin.
Guldi and co-workers assembled another supramolecular phthalocyanine (Pc)–PDI trimer 37 by using the melamine/perylenediimide triple H-bonding motif . The association constant between PDI and ZnPc was calculated to be 2 × 105 M−1 in THF. Photoexcitation of the PDI component afforded transduction of singlet excited-state energy to the energetically lower lying phthalocyanine. The energy transduction process was investigated by fluorescence, time-resolved fluorescence spectroscopy, and transient absorption spectroscopy.
1.4.2 Energy Transfer Based on Quadruply Hydrogen Bonding and Its Applications
Energy transfer was not only observed in the supramolecular assembly with different lengths of OPV, but also among OPV and various other quadruply H-bonded functionalized chromophores. For example, Janssen and co-workers reported the synthesis of a perylenediimide with the UPy H-bonding motif (PERY-UPy) 38 and studied its photophysical properties in hetero-assemblies with UPy functionalized OPY (OPV-UPy) using fluorescence spectroscopy and femtosecond pump-probe spectroscopy . The association constant of 38 is as high as 108 M−1 in toluene. Photoluminescence studies revealed that a singlet energy transfer reaction occurred after excitation of the OPV chromophore. A time constant of 5.1 ps was obtained for this reaction, which is in fair agreement with Förster theory. Although exergonic, electron transfer did not occur after photoexcitation as a result of a too weak electronic coupling between OPV and PERY chromophores in the excited state.
Wu et al. also synthesized and assembled heterodimer 39, in which the chromophores were connected to the AADD module via only a methylene group . Due to the rigidity, directionality, and specificity of the linker, naphthalene and anthracene were arranged side-by-side with donor-to-acceptor edge-to-edge distance being 13 Å. Excitation of the naphthalene chromophore in 39 resulted in efficient inner-assembly singlet energy transfer from naphthalene to anthracene. The energy transfer efficiency and rate constant were calculated to be 89 % and 9.8 × 108 s−1, respectively.
To examine whether the UPy H-bonding module could mediate the TTET, Wu et al. designed another UPy-bridged assembly 40 . They reported that photoinduced intra-assembly TTET took place between the UPy-bridged benzophenone and naphthalene. The naphthalene group quenched the phosphorescence of the benzophenone efficiently upon selective excitation of the benzophenone at 77 K. A flash photolysis experiment at room temperature indicated that the intra-assembly TTET occurred with a rate constant of 3.0 × 106 s−1 and an efficiency of 95 % in CH2Cl2. It is known that TTET occurs via the Dexter electron exchange mechanism, and its rate constant decreases exponentially with donor–acceptor distance. When the donor–acceptor distance increases beyond the sum of their van der Waals radii, this kind of energy transfer process is generally negligible. In this case, the benzophenone and naphthalene units are arranged side-by-side with a much larger separation. Since the rigid structure prevents the donor and acceptor from any collisions through the solvent or space mechanism, the highly efficient intra-assembly TTET in 40 should occur through the H-bonds. As a result, UPy H-bonds here not only act as rigid scaffolds to fix the well-defined relative separations and orientations of the chromophores but also play a crucial role in mediating the photoinduced intra-assembly triplet–triplet energy transfer process.
Later, Wu et al. designed another UPy-bridged assembly 41, which have ferrocene and fullerene donor and acceptor, respectively . Intra-assembly electron transfer from the singlet or triplet excited fullerene chromophore to the ferrocene group was too efficient to be observed, despite that ferrocene and fullerene are typically an electron transfer pair for efficient electron transfer reactions. Instead, intra-assembly triplet–triplet energy transfer occurred through the UPy quadruply H-bonding module at room temperature with a rate constant of 9.2 × 105 s−1 and an efficiency of 73 % in CH2Cl2.
Wong et al. reported the construction of highly luminescent hollow nanospheres from three aggregated π-conjugated oligomers (42–44) . The most remarkable characteristic of the system is that, by mixing the oligomers in a suitable ratio through quadruple H-bonding interactions, any luminescence color, including white light emission, could be formed from the aggregated objects. In detail, emission of 42 (10−4 M solution in THF) is efficiently quenched by the addition of micromolar amounts of 43 or 44, which is assigned to singlet energy transfer on the basis of the favorable overlap between the emission spectrum of 42 and the absorption shell of 43 or 44. Along similar lines, the emission spectrum of the individual vesicle-like aggregates could be tuned by adjusting the composition of the solution from which they were dropcast. To demonstrate this, the color coordinates of a series of samples in which the proportion of 42, 43, and 44 were varied were measured, and the vesicle-like aggregates thus formed gave the expected progression of colors from blue to yellow. Thanks to the possibility of combining three different colors, the gamut obtainable is very large. They cover more than 75 % of the gamut of a standardized red-green-blue (RGB) liquid crystalline color display.
1.5 Summary and Outlook
In summary, the principles, examples, and applications of H-bonding-controlled photoinduced electron and energy transfer have been reviewed. H-bonding clearly offers an attractive way to assemble donor and acceptor, to facilitate photoinduced electron/energy transfer processes, as well as to increase the charge separation lifetime. These systems are of great importance to mimic and understand the photosynthetic system in nature. H-bonding is not only the bridge between donor and acceptor in biological matrix, but also involved directly in the electron transfer/exchange process and plays an important role in mediating electronic tunneling. Electron/energy transfer based on H-bonding systems have found application in promising fields such as in constructing optoelectronic devices and organic solar cells, light-harvesting system, wide-range color display, and storage material. This is a thriving topic that will furnish future avenues as an outstanding assuming motif in nanoscience and nanotechnology. More applications might also be found in aqueous media and in biological matrix such as cells.
We are grateful for the financial support from the 973 Program (2013CB933800, 2013CB834505), the National Natural Science Foundation of China (91027041 21222210).
- 3.Grondelle R v, Dekker JP, Gillbro T, Sundstrom V (1994) Biochim Biophys Acta Bioenerg 1187:1–65Google Scholar
- 7.Scheuring S, Seguin J, Marco S, Levy D, Robert B, Rigaud JL (2003) Proc Natl Acad Sci USA 100:1690Google Scholar
- 9.Turro NJ (1978) Modern molecular photochemistry. Benjamin-Cummings, Menlo Park Chapter 9Google Scholar
- 13.Marcus RA (1957) J Chem Phys 26:86Google Scholar
- 16.Rehm D, Weller A (1969) Ber Bunsenges Phys Chem 73:834Google Scholar
- 32.Rajkumar GA, Sandanayaka ASD, Ikeshita K-i, Araki Y, Furusho Y, Takata T, Ito O (2006) J Phys Chem B 110:6516Google Scholar
- 38.Ballardini R, Credi A, Teresa Gandolfi M, Marchioni F, Silvi S, Venturi M (2007) Photochem Photobiol Sci 6:345Google Scholar
- 66.Schenning APHJ, Herrikhuyzen J v, Jonkheijm P, Chen Z, Würthner F, Meijer EW (2002) J Am Chem Soc 124:10252Google Scholar
- 67.Würthner F, Chen Z, Hoeben FJM, Osswald P, You C-C, Jonkheijm P, Herrikhuyzen J v, Schenning APHJ, Schoot PPAM v d, Meijer EW, Beckers EHA, Meskers SCJ, Janssen RAJ (2004) J Am Chem Soc 126:10611Google Scholar
- 69.Jimínez ÁJ, Calderón RMK, Rodríguez-Morgade MS, Guldi DM, Torres T (2013) Chem Sci 4:1064Google Scholar
- 73.Sessler JL, Jayawickramarajah J, Gouloumis A, Torres T, Guldi DM, Maldonadoa S, Stevenson KJ (2005) Chem Commun 1892 Google Scholar
- 74.Segura M, Sánchez L, Mendoza J d, Martín N, Guldi D M (2003) J Am Chem Soc 125:15093Google Scholar
- 85.McClenaghan ND, Grote Z, Darriet K, Zimine M, Williams RM, Cola LD, Bassani (2005) Org Lett 7:807Google Scholar
- 89.Jonkheijm P, Stutzmann N, Chen Z, Leeuw DM d, Meijer EW, Schenning APHJ, Würthner F (2006) J Am Chem Soc 128:9535Google Scholar
- 93.Li F, Yang J, QinY (2013) J Polym Sci Part A Polym Chem 51:3339Google Scholar
- 95.Harriman A, Magda D, Sessler JL (1991) J Chem Soc Chem Commun 345Google Scholar