Recent progress in homogeneous multielectron transfer ... - JKU

Coordination Chemistry Reviews 304–305 (2015) 102–108

Contents lists available at ScienceDirect

Coordination Chemistry Reviews journal homepage: www.elsevier.com/locate/ccr

Review

Recent progress in homogeneous multielectron transfer photocatalysis and artificial photosynthetic solar energy conversion Günther Knör ∗ Institute of Inorganic Chemistry, Johannes Kepler University Linz (JKU), Altenbergerstr. 69, A-4040 Linz, Austria

Contents 1. 2.

3.

4.

5.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Modelling the key-steps of natural photosynthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1. Light-harvesting and energy transfer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2. Photoinduced electron transfer and charge separation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3. Multielectron catalysis and energy storage in chemical bonds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Replacing the function of photosynthetic redox enzyme cascades . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. Photochemical enzyme model compounds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2. The light-driven plastocyanin: ferredoxin oxidoreductase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3. Ferredoxin-NADP+ reductase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4. Subsequent dark-reactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . First examples of full photosynthetic reaction centre mimics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1. The design of a competitive substitute for photosystem I . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2. Quantum yields and energy-storage efficiencies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3. Beyond natural photosynthesis—An optimistic outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Concluding remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

a r t i c l e

i n f o

Article history: Received 31 July 2014 Received in revised form 10 September 2014 Accepted 19 September 2014 Available online 30 September 2014 Keywords: Solar energy conversion Artificial photosynthesis Multielectron transfer catalysis Long-wavelength photosensitization Sustainable fuel production

102 103 103 104 104 105 105 105 105 106 106 106 107 107 108 108 108

a b s t r a c t As a consequence of global population growth, our modern civilization is facing many upcoming challenges such as gradually depleting energy resources, anthropogenic climate change and other serious environmental issues. In this context, the development of artificial photosynthetic systems for solar energy harvesting and sustainable fuel production represents one of the most attractive long-term strategies to address these important topics. Despite the indisputable benefits, which such a man-made counterpart of the biological solar energy conversion and storage machinery could provide, progress in mimicking the essential functions of natural photosynthesis still remains quite difficult to achieve. While ongoing efforts in replacing the light-harvesting and charge separation function of natural photosystems have led to some remarkable success, many aspects of powering the endergonic chemical reactions required have to be much further elaborated. The current limitations of most artificial photosynthetic systems are related to an inefficient coupling of the catalytic steps necessary for chemical bond formation and for an accumulation of energy rich product molecules (solar fuels). In the present review, some recent breakthroughs in these directions are briefly discussed. © 2014 Elsevier B.V. All rights reserved.

1. Introduction

∗ Tel.: +43 (0)732 2468 5100. E-mail address: [email protected] http://dx.doi.org/10.1016/j.ccr.2014.09.013 0010-8545/© 2014 Elsevier B.V. All rights reserved.

Turning photons and abundant feedstocks into fuels by mimicking the light-dependent functions of green plants has been a central goal of photochemical research for more than a century [1]. Despite the enormous efforts in this direction, many of the

G. Knör / Coordination Chemistry Reviews 304–305 (2015) 102–108

103

Table 1 Artificial and natural photosynthetic transformations and their thermodynamicsa . Energy storing (endergonic) reaction

ne−

G◦ (kJ/mol)

G◦ (eV)b

2 H2 O → H2 + H2 O2 2 H2 O → 2 H2 + O2 2 H2 O + 2 NAD+ → 2 NADH/H+ + O2 CO2 + 2 H2 O → CH3 OH + 3/2 O2 CO2 + 2 H2 O → CH4 + 2 O2 6 CO2 + 6 H2 O → C6 H12 O6 + 6 O2

2 4 4 6 8 24

355 475 436 701 818 2880

1.84 1.23 1.13 1.21 1.06 1.24

a b

thr (nm)b 674 1008 1098 1025 1170 997

Adopted from Refs. [3,4] Calculated per electron transferred.

fundamental aspects of such a process still remain to be discovered. Oxygenic photosynthesis is a sunlight-driven enzymatic reaction sequence which directly converts carbon dioxide and water into food and biomass. Sugars such as glucose (C6 H12 O6 ) are the typical energy-rich storage chemicals produced by natural photosynthetic carbon fixation. For every glucose molecule formed, 24 electrons and protons (2 hydrogen equivalents or 4e− /4H+ per assimilated CO2 molecule) have to be released from H2 O acting as a reductant (“sacrificial donor”). The thermodynamic constraints for powering such a process [2,3] require to capture the energy of 24 photons with a maximum threshold wavelength thr of around 1000 nm (Table 1), which would then correspond to 100% efficiency of solar energy conversion into carbohydrate fuel. Natural oxygenic photosynthesis, however, where two photosystems (PS I and PS II) have been coupled in a row, requires capturing the energy of at least 48 red-light photons of about 680 nm (1.8 eV) in order to produce one molecule of glucose, which considerably reduces the theoretical maximum energy-storage efficiency to about 30% [4]. For polychromatic irradiation with sunlight, this value further drops to a theoretical limit of 12% [5], and plants can usually achieve no more than 5–6% solar power to carbon-based fuel conversion efficiency under optimum growth conditions [6]. Nevertheless, more than 80% of the current world energy supply are covered by burning the chemicals formed in the course of natural photosynthesis. In an artificial photosynthetic system, many useful energy storage molecules other than carbohydrates could be produced. Instead of the net 24 electron process required for glucose production, several more easily accessible solar fuels such as alkanes, methanol or hydrogen would require a much less demanding level of multielectron catalysis. Some alternative photosynthetic reactions are listed in Table 1. It should be pointed out, however, that for all of these examples the initial extraction of reduction equivalents (2e− /2H+ or H2 ) from H2 O is the energetically limiting step. Therefore, one of the major challenges in the field of artificial photosynthesis is to achieve an efficient water oxidation reaction driven by photons in the long-wavelength region of the solar spectrum.

Scheme 1. Basic functional requirements for the construction of photosynthetic model systems.

2.1. Light-harvesting and energy transfer The first step in solar energy conversion is light absorption. Light-harvesting complexes in photosynthetic antenna systems contain a set of specialized pigments including chlorophylls strongly absorbing in the long-wavelength part of the visible spectral region and beyond (Fig. 1). These pigments are acting as photosensitizers to transduce their excitation energy to the photochemically active reaction centres. The reaction centres of oxygenic photosynthetic organisms are operating near the rededge of the solar spectrum at around 680 nm (PS II) and 700 nm (PS I). Any excess energy of light absorbed at shorter wavelengths is converted to heat. Modelling the spectroscopic features of the natural light-harvesting chlorophyll pigments with robust synthetic compounds is an important task for the construction of artificial photosynthetic devices [2,8]. It is remarkable, that the long-wavelength limits of photochemical energy conversion in different photosynthetic organisms have only recently been found to extend significantly into the far-red region of the solar spectrum [9–11]. Such an extended threshold wavelength region closer to the absolute thermodynamic limits (Table 1) should also be considered as a benchmark value for synthetic light-harvesting chromophores. While many photosensitizers investigated in the field of artificial photosynthesis are still absorbing at much shorter wavelengths, several alternative dye molecules with a longer

2. Modelling the key-steps of natural photosynthesis From a chemical perspective, natural photosynthesis is an endergonic photocatalytic redox reaction, which has been optimized for the conversion of far-red light photons into chemical energy. The basic functional requirements of such a process can be rationally analysed [7] and these considerations may serve as a starting point for the development of fully operating artificial photosynthetic devices. Such an ambitious endeavour, however, requires a perfect coupling of all building blocks to orchestrate the necessary key-steps ranging from fundamental photophysical processes to light-mediated redox chemistry and biomimetic catalysis [2,7,8]. The state of the art of modelling these individual steps, which are summarized in Scheme 1, will be briefly discussed in the following sections.

Fig. 1. Absorption spectrum of the far-red light harvesting chlorophyll Chl f [9].

104


Fig. 2. Different strategies for charge separated state (D•+ + A•− ) formation. (a) Quinone acceptor photoreduction upon red-light excitation in a covalently linked pentade architecture [15]; and (b) direct quinone photoreduction with far-red light in a metal-to-ligand charge transfer system [8,16].

threshold wavelength are available. For example, the far-red light harvesting function of chlorophyll f (Fig. 1) could be readily replaced in a synthetic system by more robust photoreactive metallo-phthalocyanine derivatives with similar absorption characteristics [2,12]. In general, a broad range of deeply coloured dye molecules such as porphyrin derivatives, perylene diimides and several others, as well as inorganic chromophores such as semiconductor nanoparticles can be employed for replacing the red light absorption (low-bandgap) function of natural photosynthetic pigments. However, still in many cases only a moderate long-term stability of the selected chromophores against photobleaching or surface corrosion is observed. Unlike natural systems, the synthetic counterparts are lacking a suitable repair or replacement mechanism of the chromophores. Another even more serious drawback in manmade systems is the frequently very low efficiency of coupling constructive electron transfer steps to the excited state processes involved in artificial photosynthetic light-harvesting architectures.

as the primary electron acceptors [14]. Visible-light induced formation of charge-separated states in a large variety of synthetic chromophore-donor–acceptor systems has been studied during the last decades [15]. Typical approaches to induce charge separation at the molecular level are illustrated in Fig. 2. One possible strategy to avoid rapid charge recombination after photoexcitation is to increase the distance between the donor and the acceptor sites in a series of vectorial, energetically downhill electron transfer steps (Fig. 2a). Another approach [16,17] is to achieve instantaneous charge separation by a direct optical charge transfer (CT) transition between reducing (donor) and oxidizing (acceptor) subunits in supramolecular assemblies, ion pairs or coordination compounds (Fig. 2b). In the latter case, further charge stabilization is usually achieved by a subsequent quenching process of the primary CT excited state generated involving sacrificial donor or acceptor species. In this context it is interesting to note that nature also exploits the advantages of direct optical charge transfer (CT) excitation in the far-red spectral region to significantly improve the efficiency of charge separation in PS I reaction centres [11].

2.2. Photoinduced electron transfer and charge separation 2.3. Multielectron catalysis and energy storage in chemical bonds One of the basic functions common to all photosynthetic reaction centres is to trigger photoinduced electron transfer (PET) between electron donor (D) and electron acceptor sites (A). This fundamental process leads to the formation of a charge-separated state with a lower energy than that of the excited state species involved (Eq. 1): D + A + h → D•+ + A•−

(1)

The primary donor of native photosystems usually consists of a special pair of chlorophyll pigments [13] able to release an electron upon red-light irradiation. Tightly-bound oxidants such as quinones present in reaction centres of the PS II type are acting

While modelling important functions of photosynthetic reaction centres such as light-absorption across the visible spectrum, excitation energy transfer and subsequent generation of charge separated states with high quantum yields has already reached a considerable level within the last decades [18], the development of complete artificial photosynthetic model systems capable of converting light energy into solar fuels still remains a major challenge [18,19]. One of the most serious problems encountered is to couple photoinduced charge separated states to functional building blocks acting as efficient catalysts for multielectron transfer (MET) and proton coupled electron transfer (PCET) steps, which are required


to achieve permanent energy storage in sufficiently stable chemical bonds (Scheme 1). One possible step towards linking charge separation and multielectron reactivity in molecular systems is given by the approach of charge photo-accumulation [20], where after the first excitation and reductive quenching process a second photon has to be absorbed by the system in its one-electron reduced form. This requires the presence of a suitable electron acceptor site, which can accumulate and store several electrons with only a modest difference in potentials between the successive reduction steps [12]. Another, quite different but very powerful strategy to achieve the required MET reactivity is to create light-sensitive multielectron transfer reagents, which are able to form a net two-electron photoredox product upon absorption of only one photon [7]. These systems, which have been introduced as so-called multielectron transfer photosensitizers [21], are based on special building blocks, which are reaching an energetically unfavourable and thus chemically reactive intermediate oxidation state upon one-electron transformation. The reactive intermediate formed photochemically can then be stabilized by a thermally induced second electron transfer step, resulting in the two-electron reaction product [2,7,8,21]. Successful design of such multielectron transfer photosensitizers requires components with stable oxidation states differing by more than one unit, and is usually assisted by building blocks that can partially stabilize the unfavourable one-electron radical intermediates involved in the process. In the pioneering studies of such MET-sensitizers [7], the combination of redox-active main group elements acting as earthabundant non-precious metal catalytic sites [22] combined with radical-stabilizing non-innocent organic ligands [23,24] has been successfully introduced as a new concept for modelling multielectron transfer photoreactions. The accumulation of permanent energy-rich photoproducts following this strategy has also been demonstrated [2]. More recent applications of such systems will be described in the next sections. 3. Replacing the function of photosynthetic redox enzyme cascades

105

Fig. 3. Schematic representation of the electron transport pathways in a type I reaction centre such as PS I: Light-induced oxidation of an electron donor (D) leads to the formation of a more reactive one-electron reduced acceptor (A), which can be accumulated as a soluble electron carrier for further catalytic processes (protein structure PDB-code 4L6 V, adapted from Ref. [34]).

that the biomimetic performance of several light-driven enzyme model systems based on low-molecular weight components is comparable or even better than that of the native enzymes to be substituted under physiological conditions (aqueous aerobic or anaerobic solution, ambient temperature, atmospheric pressure). For example, a sunlight-exposed artificial photoenzyme replacing the two-electron transfer redox function of the copper enzyme galactose oxidase shows an almost 200-fold specific activity compared to the natural counterpart under identical conditions [28]. Photochemical control also allows to trigger and regulate the artificial endonuclease reactivity of otherwise biocompatible and non-cytotoxic reagents for efficient catalytic DNA double strand cleavage driven by visible light [32]. Moreover, an artificial multi-enzyme cascade catalysis with photochemical oxidoreductase model compounds could be achieved, which enables a full reaction control over two-electron vs. four-electron substrate conversion [33]. In the following sections, the application of photochemical enzyme models based on multielectron transfer sensitizers to mimic the redox enzyme functions observed in natural photosynthesis will be described. 3.2. The light-driven plastocyanin: ferredoxin oxidoreductase

Mimicking photosynthesis requires not only a successful modelling of all the necessary key-steps described above, but also a productive coupling of these steps to a functional energy conversion system. The catalytic multielectron transfer and proton transfer processes of oxygenic photosynthesis are mediated by the two photosystems PSI and PS II, which can be functionally characterized as light-driven oxidoreductase enzymes [25,26]. It therefore seems appropriate to briefly discuss the current state of the art in the development of light-powered artificial enzymes. 3.1. Photochemical enzyme model compounds Artificial bio-inspired systems constructed to mimic the catalytic performance of native enzymes upon exposure to visible light or diffuse daylight (artificial photoenzymes) have been introduced about a decade ago [27,28]. The concept of this very versatile method is to create photoresponsive low-molecular weight compounds with an excited state electronic structure and reactivity replacing the function of short-lived active site intermediates involved in biocatalytic substrate conversion cycles. The design principles of such homogeneous photocatalytic systems have been reviewed in more detail elsewhere [2,8]. Although still a limited number of such artificial photoenzymes has been fully characterized to date [8,29], with a main focus on light-driven oxidoreductases [28,30] and nuclease enzymes [31,32], the proof of concept provided is very encouraging. It could already be demonstrated

In natural photosynthesis, the terminal light-reactions are mediated by highly organized pigment protein complexes of the photosystem I (PS I) type [26,34], where solar energy is collected to photocatalyze the generation of strongly reducing chemicals by oxidizing weaker reductants (Fig. 3). Plastocyanine (PC), a soluble blue copper protein with a reversible Cu(II)/Cu(I) redox couple and ferredoxin (Fd), a soluble iron-sulfur protein with a reversible Fe(III)/Fe(II) redox couple are acting as the one-electron donor (D) and one-electron acceptor (A) pairs in PS I of green plants [35]. Thus light-harvesting, photoinduced electron transfer and terminal quenching of the charge separated state mediated by PS I can be described as the catalytic function of a natural photoenzyme called the light-driven plastocyanine: ferredoxin oxidoreductase (EC 1.97.1.12) [26]. These fundamental catalytic functions can also be addressed with artificial enzyme model compounds driven by light [2,7]. 3.3. Ferredoxin-NADP+ reductase The last step in the photosynthetic electron transfer chain producing the two-electron reduced form of nicotinamide adenine dinucleotide phosphate (NADPH) as a solar fuel is catalyzed by a second oxidoreductase enzyme called ferredoxin-NADP+ reductase (EC 1.18.1.2) [35,36]. This enzyme takes up the one-electron reduction equivalents produced in PS I (Fig. 3), and efficiently mediates

106


Fig. 5. Stepwise two-electron redox cascade for a NADH-dependent abiotic CO2 fixation process generating long-term storable carbon-based fuels [30,42].

Fig. 4. Schematic representation of electron and proton transfer steps catalyzed by the flavoprotein ferredoxin-NADP+ -reductase (protein structure PDB-code 2BSA, structural data from Ref. [37]).

the required process of charge-accumulation and proton-coupled multielectron transfer catalysis (Fig. 4). Thus, nature has developed a visible light-driven enzymecascade reaction directly converting solar energy into storable chemical energy. This is also the final goal of all artificial photosynthetic systems (Table 1). As will be shown in Section 4, such a coupled process, replacing the functions of both the light-driven plastocyanine: ferredoxin oxidoreductase (PS I) and the twoelectron redox catalyst ferredoxin-NADP+ -reductase, has recently been achieved in a synthetic system based on molecular components [30]. 3.4. Subsequent dark-reactions The light-reactions of natural photosynthesis, partially described above, result in the formation of the energy-rich chemicals ATP and NADPH [25,26]. These compounds are only stable for a limited time, and therefore have to be immediately consumed or further converted into long-term storable chemicals. Nature has chosen energy-rich carbon-based molecules such as sugars for this purpose, which are formed in enzymatic dark-reactions where carbon dioxide fixation is coupled to the consumption of the primary solar fuels produced in the light-reactions. Catalytic formation of glucose by CO2 assimilation formally consumes 24 electrons and protons or 12 equivalents of photogenerated NADPH. Since each of the coupled enzymatic 2e− /2H+ steps required can only operate with less than 100% efficiency, a considerable share of the initially captured solar energy is lost in this kind of long-term storage processes. This explains the rather low maximum efficiency with which photosynthesis can convert solar energy into biomass [38]. In artificial photosynthesis, other carbon-based storage molecules may be chosen to fix the reduction equivalents generated with sunlight (Table 1). An attractive liquid fuel, requiring a six electron reduction of carbon dioxide, is methanol [39]. Biocatalytic synthesis of methanol from CO2 involving a set of different dehydrogenase enzymes as catalysts has already been demonstrated and optimized [40–42]. Such an exergonic catalytic dark-reaction, however, requires continuous supply with the twoelectron reduced nucleotide cofactors NADPH or NADH. Utilizing an efficient artificial photosynthetic [30] or natural photosynthetic [42] recycling process of nucleotide cofactors such as NADH can therefore lead to a sustainable production of methanol as an easily storable and transportable liquid solar fuel (Fig. 5). 4. First examples of full photosynthetic reaction centre mimics As described above, the solar energy conversion process catalyzed by natural photosynthetic reaction centres and redox

proteins leads to an accumulation of the two-electron reductant NAD(P)H powered by red light. The corresponding reaction sequence can be summarized as a photocatalytic step (Eq. (2)) mediated by the light-driven enzyme PS I (EC 1.97.1.12), and a directly connected second catalytic step (Eq. (3)) accelerated by a ferredoxin (Fdred ) dependent reductase (EC1.18.1.2): D + A(Fdox ) + h(700 nm) → D•+ + A•− (Fdred )

(2)

2 Fdred + NAD(P)+ + H+ → 2 Fdox + NAD(P)H

(3)

The same overall function, coupling multielectron transfer photochemistry to permanent energy storage in the chemical bonds of NADH, has recently been demonstrated for an artificial photosynthetic system [30]. This functional counterpart, which is based on red-light responsive coordination compounds, will now briefly be discussed. 4.1. The design of a competitive substitute for photosystem I Photoinduced charge separation, reductive quenching by an electron donor and intermediate negative charge stabilization at the one-electron level, as described in Eq. (2) for natural photosynthesis, can also be successfully addressed with small molecular building blocks based on earth-abundant components. We have employed the very robust water-soluble tin chlorin derivative dihydroxo-(2,3-dihydro-5,10,15,20-tetrakis(Nmethylpyridinium-4-yl)-porphyrinato)tin(IV)-tetrachloride (SnC) with attractive chlorophyll-like absorption characteristics for this purpose [30]. This homogeneous photocatalyst is able to fully replace important functional properties of PS I such as red-light powered accumulation of a soluble one-electron carrier substituting the role of reduced ferredoxin (Fdred ) in nature: D + SnC + h(600–700 nm) → D•+ + SnC•−

(4)

2SnC•− + H+ → SnC + SnCH−

(5)

Moreover, carrying the functional subunits of a typical multielectron transfer photosensitizer (Section 2.3), the redox-active tin complex is able to participate in subsequent electron and proton transfer steps (Eq. (5)), efficiently competing with unproductive back-electron transfer to the starting compounds [7]. Thus, after the primary charge separation (Eq. (4)), the metastable radical anion intermediate (SnC•− ) is further stabilized in a reaction sequence replacing the crucial two-electron accumulation function of the enzyme-bound flavin cofactor [36] present in native ferredoxin: NADP+ reductase (Fig. 4, Eq. (3)). The red-light driven formation of the two-electron reduced and mono-protonated form (SnCH− ) of the tin chlorin photocatalyst (Fig. 6) is considered to be the key-step in this unprecedented artificial photosynthetic system, leading to an efficient production of the reduced cofactor NADH [30]. Generation of the biologically active 1,4-dihydro-nicotineamide form of the nucleotide cofactor NAD(P)H, replacing the energystoring processes occurring in the natural photosynthetic lightreactions, requires an additional regioselective hydride transfer step involving the accumulated photoproduct SnCH− (Eq. (6)). In our model system this is mediated by a soluble rhodium-based cocatalyst [30]. This last step of the reaction sequence regenerates


107

Table 2 Electron donor properties and calculated energy-storage efficiencies for photosynthetic NAD(P)H formationa . Donor red

PC TEOA EDTA

E◦ (V vs. NHE)b

E◦ (V)c

(nm)

hc/ (eV)

(%)

+0.38 +1.07 +1.17

0.70 1.39 1.49

700d 660e 660e

1.77 1.88 1.88

39.6 73.9 79.3

a

Assumptions and values from Refs. [45–47]. Redox potentials in water at pH 7. Calc. with E◦ = −0.32 V for NAD(P)+ [47]. d Excitation threshold for natural PS I. e Excitation threshold for the artificial photosynthetic system reported in Ref. [30]. b c

Fig. 6. Absorption spectrum of the water-soluble tin chlorin photocatalyst (SnC) providing both the light-harvesting function of chlorophylls and the proton-coupled multielectron transfer reactivity occurring in natural photosynthesis. Inset: twoelectron ring-reduced and meso-protonated form (SnCH− ) of the tin chlorin complex acting as the primary hydride transfer source for catalytic NADH production [30].

the multielectron transfer sensitizer SnC for further photocatalytic turnover: SnCH− + NAD(P)+ → SnC + NAD(P)H

(6)

It has already been demonstrated that the light energy stored in the form of NAD(P)H equivalents can be utilized for powering coupled enzymatic dark reactions in situ [30], which opens the stage for generating different types of solar fuels through artificial photosynthesis (Section 3.4). As will be discussed below, the functional counterpart of the natural energy-trapping and chemical bond formation processes described here is characterized by a very promising performance. Furthermore, this artificial photosynthetic reaction centre tolerates a broad spectrum of primary electron donors. 4.2. Quantum yields and energy-storage efficiencies Absolute quantum efficiencies (ϕ) for the photocatalytic accumulation of reduced chemical species in natural photosynthesis are typically reaching maximum values between 1% and 6% depending on the irradiation conditions [43]. Reasonably high quantum yield values between 1% and 2%, which are already in the efficiency range of the native counterparts, have also been reported for the artificial photosynthetic generation of NADH under monochromatic red-light exposure [30]. Moreover, when the generation of the primary photoproducts was directly coupled to an enzymatic NADH-consuming dark-reaction, the light-powered formation of reduced chemical species was found to be almost ten times more efficient. This significant increase of the product formation quantum yield up to around 10% (ϕ = 0.1) can be rationalized by the fact that an additional dark-reaction successfully competes with otherwise unproductive reverse processes of cyclic electron flow. The primary photoproduct (NADH), which is a much better electron donor than the initially employed reductive quencher (D), is always partially re-oxidized under steady-state turnover conditions, since in contrast to the much more complex natural reaction centres no separation of the accumulated energy-storage products across a membrane barrier is possible in homogeneous solution. Still the quantum yields found for this rather simple model system are quite impressive. Even more promising than the high quantum yields achieved is the theoretical maximum energy-storage efficiency of the light-reactions observed with the artificial photosynthetic model

system described above (Section 4.1). The efficiency parameter calculated for the generation of NAD(P)H as a terminal product of the photosynthetic light-reactions characterizes the energy change per photon associated with moving one electron through the potential difference separating the sacrificial electron donor D applied and NAD(P)+ [44]. Note, however, that much lower long-term energy-storage efficiencies are finally obtained for the production of organic biomass from NAD(P)H [4,38]. In Table 2, the maximum energy-storage efficiencies of natural photosystem I with reduced plastocyanine (PCred ) as an electron donor, and a limiting threshold excitation wavelength of 700 nm for the corresponding reaction centre, are directly compared to our recently characterized artificial photosynthetic system [30], which is limited to approximately 660 nm light absorption (Fig. 6). As can be seen from the data provided in Table 2, compared to the native photosystem (PS I) an approximately twice as high maximum energy-storage efficiency for the light-reactions has been achieved with the artificial photosynthetic model system. While the same energy-rich reduction product is formed in both cases, the oxidation power of the artificial photosystem employed obviously is much higher than that of PS I. This feature can significantly improve the overall performance, since much weaker donors than plastocyanine are tolerated, which leads to a higher potential difference E◦ to NAD(P)H. Moreover, the efficiency loss through dissipation of excess energy from the photoexcited reaction centre is at the same time minimized, since the excited state involved is less reducing than PS I* . 4.3. Beyond natural photosynthesis—An optimistic outlook Modified forms of photosynthetic reactions will certainly play an important role for providing a sustainable energy supply in the near future. It has been analysed recently that numerous points of inefficiency in the natural systems are amenable to further modifications [48]. Synthetic biology could be used to enhance the performance of natural photosynthesis for improved solar energy conversion. It might even become possible to beat nature at her own game and to generate artificial photosynthetic systems with an efficiency much closer to the thermodynamic limits. Indeed, some first encouraging steps into this direction have been reported in the present review. Moreover, some of the data shown in Table 2 are indicating that it should even be possible to leave behind the natural limits of a two-photosystem (Z-scheme) architecture for driving oxygenic photosynthesis [48]. Since photosystem II, which is the most powerful oxidant known in biology, displays an electrochemical midpoint potential of approximately + 1.2 V vs. NHE and water as a sacrificial electron donor under natural photocatalytic conditions is oxidized at + 0.82 V vs. NHE [47], there might be a reasonable chance to set up an artificial photosynthetic system based on only one red-light powered multielectron transfer sensitizer involving H2 O as an electron source and NAD+ as an electron acceptor. Such an ambitious approach will require a perfect adjustment of the ground and excited state redox potentials of the photocatalyst

108


to supply sufficient driving force (200–300 mV) for both the water cleavage and the NADH-formation process, which requires a minimum energy corresponding to approximately 1.13 eV (Table 1). Further investigations in this direction are currently underway in our group. 5. Concluding remarks It has been shown here that artificial photosynthetic solar energy conversion and storage with yields comparable to the natural counterparts can be readily achieved in homogeneous solution with a small set of low-molecular weight components. The first successful synthetic model for a biomimetic accumulation of NADH with red-light photons has been described [30], and the conceptual background leading to this development has been reviewed [2,7]. With such a man-made photocatalytic system, the overall function of native photosystem I in green plant chloroplasts (PS I and ferredoxin: NADP+ reductase reactions) can be completely replaced to drive subsequent biochemical dark-rection cascades with enzymes requiring the photochemically produced NAD(P)H as a cofactor. These findings will open many possibilities for future investigations reaching from hybrid photo-biochemical CO2 reduction processes generating carbon-based solar fuels to the development of novel light-driven artificial enzyme reactions for asymmetric catalysis and green chemistry [8]. While comparably high quantum yields in the range of the naturally evolved photosynthetic energy conversion systems have already been achieved [30], the artificial photosynthetic reaction sequence described here can operate with a more versatile set of electron delivering donors than PS I. Since donors with much less reduction power than that of the natural substrate plastocyanine can be coupled to the artificial photosynthetic system, the light-reactions catalyzed are able to reach even better maximum energy-storage efficiencies and fractional energy yields than those observed in nature, while providing the identical chemical storage product NAD(P)H as a solar fuel. Although obviously some significant progress in homogeneous multielectron transfer photocatalysis and artificial photosynthetic solar energy conversion has been made in the last decades, there are still many unresolved problems to be addressed. The next important stage of development in this promising direction towards “powering the planet” with artificial photosynthetic devices will require a successful coupling of an efficient bio-inspired water oxidation process to the solar fuel generating reactions described here. Some of the most difficult catalytic key-steps, such as visible-light induced O O bond formation involving water as the terminal electron donor, have already been addressed with light-driven enzyme models based on strongly oxidizing, high-valent metal-oxo bond containing multielectron transfer photosensitizers [2,21,49,50]. A fully operating synthetic counterpart of natural photosystem II (PS II), the light-driven water: plastoquinone oxidoreductase, however, still remains to be discovered more than hundred years after the first attempts to mimic green plants [1]. Acknowledgements Financial support by the Austrian Science Foundation (FWF project P21045 “Bioinspired Multielectron Transfer Photosensitizers”), the Austrian Climate and Energy Funds (FFG project 841186 “Artificial Photosynthesis”), and the European Commission COST Action CM1202 “Supramolecular Photocatalytic Water Splitting” (PERSPECT-H2 O) is gratefully acknowledged. References [1] G. Ciamician, Science 36 (1912) 385–394. [2] G. Knör, Chem. Eur. J. 15 (2009) 568–578.

[3] A.L. Lehninger, Bioenergetics: The Molecular Basis of Biological Energy Transformations, second ed., W.A. Benjamin, Menlo Park, 1971. [4] J. Barber, Chem. Soc. Rev. 38 (2009) 185–196. [5] D.A. Walker, J. Appl. Phycol. 21 (2009) 509–517. [6] X.-G. Zhu, S.P. Long, D.R. Ort, Annu. Rev. Plant Biol. 61 (2010) 235–261. [7] G. Knör, Coord. Chem. Rev. 171 (1998) 61–70. [8] G. Knör, U. Monkowius, Adv. Inorg. Chem. 63 (2011) 235–289. [9] M. Chen, H. Scheer, J. Porphyrins Phthalocyanines 17 (2013) 1–15. [10] D.M. Niedzwiedzki, G.S. Orf, M. Tank, K. Vogl, D.A. Bryant, R.E. Blankenship, J. Phys. Chem. B 118 (2014) 2295–2305. [11] E. Schlodder, F. Lendzian, J. Meyer, M. Çetin, M. Brecht, T. Renger, N.V. Karapetyan, J. Am. Chem. Soc. 136 (2014) 3904–3918. [12] C. Uslan, K.T. Oppelt, L.M. Reith, B.S. Sesalan, G. Knör, Chem. Commun. 49 (2013) 8108–8110. [13] G. Stochel, M. Brindell, W. Macyk, Z. Stasicka, K. Szaciłowski, Bioinorganic Photochemistry, Wiley, Chichester, 2009. [14] M. Flores, A. Savitsky, M.L. Paddock, E.C. Abresch, A.A. Dubinski, M.Y. Okamura, W. Lubitz, K. Möbius, J. Phys. Chem. B 114 (2010) 16894–16901. [15] D. Gust, T.A. Moore, A.L. Moore, in: V. Balzani (Ed.), Electron Transfer in Chemistry, vol. 3, Wiley-VCH, Weinheim, 2001, pp. 272–336. [16] M. Leirer, G. Knör, A. Vogler, Z. Naturforsch. 54b (1999) 1039–1041. [17] E. Wöß, U. Monkowius, G. Knör, Chem. Eur. J. 19 (2013) 1489–1495. [18] B.D. Sherman, M.D. Vaughn, J.J. Bergkamp, D. Gust, A.L. Moore, T.A. Moore, Photosynth. Res. 120 (2014) 59–70. [19] M.D. Kärkäs, E.V. Johnston, O. Verho, Björn Åkermark, Acc. Chem. Res. 47 (2014) 100–111. [20] Y. Pellegrin, F. Odobel, Coord. Chem. Rev. 255 (2011) 2578–2593. [21] G. Knör, A. Vogler, S. Roffia, F. Paolucci, V. Balzani, Chem. Commun. (1996) 1643–1644. [22] P.P. Power, Nature 463 (2010) 171–177. [23] P.J. Chirik, K. Wieghardt, Science 327 (2010) 794–795. [24] V.K.K. Praneeth, M.R. Ringenberg, T.R. Ward, Angew. Chem. 124 (2012) 10374–10380. [25] T.J. Wydrzynski, K. Satoh (Eds.), Photosystem II: The Light-Driven Water:Plastoquinone Oxidoreductase, Advances in Photosynthesis and Respiration, vol. 22, Springer, Dordrecht, 2005. [26] J.H. Golbeck (Ed.), Photosystem I: The Light-driven Plastocyanine: Ferredoxin Oxidoreductase, Advances in Photosynthesis and Respiration, vol. 24, Springer, Dordrecht, 2006. [27] G. Knör, Habilitation Thesis, University of Regensburg, 2001. [28] G. Knör, ChemBioChem 2 (2001) 593–596. [29] G. Stochel, M. Brindell, W. Macyk, Z. Stasicka, K. Szacilowski, Bioinorganic Photochemistry, Wiley, Chichester, 2009, pp. 197–200. [30] K.T. Oppelt, E. Wöß, M. Stiftinger, W. Schöfberger, W. Buchberger, G. Knör, Inorg. Chem. 52 (2013) 11910–11922. [31] G. Knör, Inorg. Chem. Commun. 4 (2001) 160–163. [32] M. Häubl, L.M. Reith, B. Gruber, U. Karner, N. Müller, W. Schöfberger, G. Knör, J. Biol. Inorg. Chem. 14 (2009) 1037–1052. [33] M. Hajimohammadi, C. Schwarzinger, G. Knör, RSC Adv. 2 (2012) 3257–3260. [34] Y. Mazor, D. Nataf, H. Toporik, N. Nelson, eLife 3 (2014) e01496, http://dx.doi. org/10.7554/eLife.01496.001. [35] B. Ke (Ed.), Photosynthesis: Photochemistry and Photobiophysics, Advances in Photosynthesis and Respiration, vol. 10, Kluwer, Dordrecht, 2001, pp. 605–634 (Chapter 34). [36] N. Carrillo, E.A. Ceccarelli, Eur. J. Biochem. 270 (2003) 1900–1915. [37] J. Teijero, I. Pérez-Dorado, C. Maya, M. Martínez-Júlvez, J. Sanz-Aparicio, C. Gómez-Moreno, J.A. Hermoso, M. Medina, Biochemistry 44 (2005) 13477–13490. [38] X.-G. Zhu, S.P. Long, D.R. Ort, Curr. Opin. Biotechnol. 19 (2008) 153–159. [39] G. Olah, A. Goeppert, G.K.S. Prakash, Beyond Oil and Gas: The Methanol Economy, Wiley-VCH, Weinheim, 2009. [40] R. Obert, B.C. Dave, J. Am. Chem. Soc. 121 (1999) 12192–12193. [41] A. Dibenedetto, P. Stufano, W. Macyk, T. Baran, C. Fragale, M. Costa, M. Aresta, ChemSusChem 5 (2012) 373–378. [42] R. Cazelles, J. Drone, F. Fajula, O. Ersen, S. Moldovan, A. Galarneau, New. J. Chem. 37 (2013) 3721–3730. [43] I. Terashima, T. Fujita, T. Inoue, W.S. Chow, R. Oguchi, Plant Cell Physol. 60 (2009) 684–697. [44] This value is closely related but not identical to the fractional energy yield (FEY) depending on the free energy loss G between the excited state level of the photosystem involved and the redox product formed; see also: H. Dau, I. Zaharieva, Acc. Chem. Res. 42 (2009) 1861–1870. [45] S.N. Datta, J. Sudhamsu, A. Pandey, J. Phys. Chem. B 108 (2004) 8007–8015. [46] J. Barber, Personal Communication. [47] S.P. Mielke, N.Y. Kiang, R.E. Blankenship, M.R. Gunner, D. Mauzerall, Biochim. Biophys. Acta 1807 (2011) 1231–1236. [48] R.E. Blankenship, D.M. Tiede, J. Barber, G.W. Brudvig, G. Fleming, M. Ghirardi, M.R. Gunner, W. Junge, D.M. Kramer, A. Melis, T.A. Moore, C.C. Moser, D.G. Nocera, A.J. Nozik, D.R. Ort, W.W. Parson, R.C. Prince, R.T. Sayre, Science 332 (2011) 805–809. [49] W. Schöfberger, F. Lengwin, L.M. Reith, M. List, G. Knör, Inorg. Chem. Commun. 13 (2010) 1187–1190. [50] L.M. Reith, M. Himmelsbach, W. Schöfberger, G. Knör, J. Photochem. Photobiol., A: Chem. 218 (2011) 247–253.