Substantia. An International Journal of the History of Chemistry 3(2) Suppl. 6: 49-57, 2019 Firenze University Press www.fupress.com/substantia ISSN 2532-3997 (online) | DOI: 10.13128/Substantia-765 Citation: M. Bellardita, R. Ceccato, S. Dirè, V. Loddo, L. Palmisano, F. Par- rino (2019) Energy Transfer in Het- erogeneous Photocatalysis. Substantia 3(2) Suppl. 6: 49-57. doi: 10.13128/ Substantia-765 Copyright: © 2019 M. Bellardita, R. Ceccato, S. Dirè, V. Loddo, L. Palmisa- no, F. Parrino. This is an open access, peer-reviewed article published by Firenze University Press (http://www. fupress.com/substantia) and distributed under the terms of the Creative Com- mons Attribution License, which per- mits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability Statement: All rel- evant data are within the paper and its Supporting Information files. Competing Interests: The Author(s) declare(s) no conflict of interest. Energy Transfer in Heterogeneous Photocatalysis Marianna Bellardita1, Riccardo Ceccato2, Sandra Dirè2, Vittorio Loddo1, Leonardo Palmisano1,*, Francesco Parrino2,* 1 Dipartimento di Ingegneria, Università di Palermo, Viale delle Scienze ed. 6, 90128 Palermo, Italy 2 Dipartimento di Ingegneria Industriale, Università di Trento, Via Sommarive 9, 38123 Trento, Italy *Corresponding authors: L.P: E-mail: leonardo.palmisano@unipa.it; F.P: E-mail: franc- esco.parrino@unitn.it. Abstract. Electron transfer reactions constitute one of the main pillars of chemistry and numerous examples can be found in nature and in technological applications. Energy transfer induced reactions are more elusive but equally important in phenom- ena related with natural photosynthesis and in general when light-matter interactions are relevant. Heterogeneous photocatalysis is generally considered based on electron transfer reac- tions. In fact, absorption of photons of suitable energy induces formation of photogen- erated charges (electron and holes) which in turn initiate redox reactions through inter- facial electron transfer to (or from) surface species. However, rare examples of photo- catalytic reactions induced by prevailing energy transfer have been recently reported in literature. Investigation in this field may be still defined at a nascent level, and the mechanistic aspects of energy transfer, widely investigated in photochemistry of homo- geneous or colloidal systems should be clarified in heterogeneous photocatalysis. In the manuscript the basic principles of energy transfer will be presented along with some known examples. These concepts will be inferred in the field of heterogeneous photocatalysis, by considering the excited solid semiconductor as the energy donor. Some rare examples of energy transfer induced heterogeneous photocatalytic reactions will be presented along with some tentative mechanistic hypotheses. Keywords. Energy transfer; Heterogeneous photocatalysis; Green chemistry. 1. INTRODUCTION Processes based on energy transfer are nowadays of great actuality. Even if the theoretical description of these phenomena has been clarified last cen- tury,1,2 these basic processes find application in innovative technologies con- necting chemistry, biology and physics in interdisciplinary approaches. For instance, techniques based on energy transfer processes have been used to monitor DNA hybridization and sequencing, protein conformation, enzyme activity, and cellular dynamics.3-5 Other applications concern photonic logic 50 Marianna Bellardita et al. gates6,7 and energy harvesting.8,9 The reason for this widespread interest in energy transfer based technologies lies on the extremely high sensitivity to conformational changes in the distance and orientation between energy donors and acceptors. The distance changes observable in this way range from 0.5 to 10 nm so, for this reason, energy transfer based techniques are also referred to as “molecular rulers”.10 Moreover, energy transfer mecha- nisms have been found to be fundamental for the capture and transmission of light in natural photosynthesis.11 In fact, photosynthetic organisms use specialized complex- es which are able to transfer the captured light energy through an efficiently distributed hierarchy of proteins. This energetic cascade efficiently proceeding in precise time and distance scales, eventually reaches the reaction centres where solar energy is fixed into chemical bonds. The energy transfer process12 can be summarized through Eq. 1: D* + A → D + A* (1) in which the excited donor (D*) is quenched to its ground state (D) and the released energy is absorbed by an acceptor (A) which in turn is promoted to its excited state (A*). The simplest energy transfer mechanism is radiative and is known with the name of “trivial”. In this case, the emission spectrum of *D must overlap the absorption spectrum of A in order to make possible the absorption of the photon emitted by D* by the acceptor A. This pro- cess depends on the extent of overlapping of the spectra, on the emission efficiency of D* (i.e. its quantum yield of emission) and on the concentration of A in the path of photons emitted by D*. Energ y transfer can also occur non-radiatively through Förster or Dexter mechanisms. Förster reso- nance energy transfer (FRET) occurs from a fluores- cent donor to a lower energy acceptor via long-range dipole-dipole interactions. Therefore, FRET mechanism is favoured at a specific geometric orientation and it is highly sensitive to donor-acceptor distances. The efficiency (E) of FRET can be described by Eq. 2: E R R R 0 6 0 6 6 (2) where R is the distance between donor and acceptor, and R0, typical for each donor-acceptor couple, is the dis- tance at which the efficiency decreases of 50% (Forster distance). The FRET efficiency can be also expressed in terms of life time (τ) of the donor or of the fluorescence intensity (F) according to Eq. 3: E F F DA D DA D 1 1 (3) where the subscripts D and DA refer to as the donor and the donor in the presence of the acceptor, respectively. According to Eq. 3, the decrease of the lifetime (or of the fluorescence intensity) of the donor in the presence of an acceptor indicates the existence of FRET. Finally, the rate (kt) of FRET is described by Eq. 4: k R Rt D 1 0 6 6 (4) FR ET mechanism requires a large overlapping between donor emission and acceptor absorption spec- tra. Energy transfer can also proceed through a “colli- sion” mechanism which is often referred to as “Dexter” mechanism. In other words, the orbitals of the excited donor and of the acceptor in its ground state can overlap giving rise to a double electron transfer which does not modify the total charge of the system, but only its elec- tronic configuration. In this case, the initial steps of the interaction are the same giving rise to electron transfer, but the system evolves differently along with the reaction coordinate and no net charge separation is obtained. The rate (k) of Dexter energ y transfer can be described by Eq. 5: k KJ R L � � � � � � � �exp 2 (5) where K is a constant related to the specific donor-accep- tor couple, J is the normalized spectral overlap integral, R is the distance between donor and acceptor relative to their radii of van der Waals (L). By comparing Eqs. 4 and 5 it is evident that the rate of energy transfer decreases with increasing R but with different dependence laws. In particular, unlike FRET, Dexter energy transfer becomes negligible when the dis- tance between donor and acceptor increases of few Ang- stroms. This is quite obvious by considering that Dexter mechanism requires orbital (not only spectral) overlap. Any fluorescent species such as organic dyes, fluo- rescent proteins, or nanoparticles can be potentially an energy donor.13 In particular, semiconductor nanoparti- cles of few nanometres (quantum dots) have been deeply investigated due to their excellent photo-physical fea- tures, stability and versatility.14-16 However, the donor behaviour of larger semiconductor nanoparticles in 51Energy Transfer in Heterogeneous Photocatalysis energy transfer processes is poorly understood as dis- cussed throughout the text. Demonstrating and oppor- tunely exploiting the presence of energy transfer mecha- nisms in heterogeneous systems is relevant, for instance, in the fields of photocatalysis and photoelectrocataly- sis. These applications are generally based on electron transfer processes occurring on the surface of irradiated semiconductor nanoparticles suspended in a reacting medium or immobilized onto a support.17,18 Both tech- nologies can be performed in mild conditions of temper- ature and pressure, with water as the solvent, solar light as the driving force, and by using cheap, abundant and robust semiconductor nanoparticles. These technologies found traditional application in the field of environmen- tal remediation for both water and gaseous effluents.19 However, applications for the synthesis of high value added compounds are gaining increasing attention due to the appealing features of these processes in terms of conversions, selectivity and sustainability.20 The product distribution of this traditionally “electron transfer driv- en” processes can be significantly different when energy transfer processes become the prevailing mechanisms. For these reasons, understanding these mechanisms in heterogeneous photocatalytic systems and developing the technological tools to control and switch them from electron to energy transfer driven processes is highly desired and could open the route to unexplored and exciting novel organic syntheses as green alternative to the traditional ones. 2. ELECTRON AND ENERGY TRANSFER IN HETEROGENEOUS PHOTOCATALYSIS Charge recombination is generally seen as det- rimental in photocataly tic reactions as the absorbed energ y does not induce interfacia l electron trans- fer but is radiatively or not radiatively emitted. This energy could be in principle transferred to other spe- cies in the reacting mixture similarly to colloidal or homogeneous systems. However, even if the formal similarity with energy transfer processes in homoge- neous systems is plausible, in heterogeneous systems such as photocatalytic suspensions, it is often difficult to discriminate between electron and energy transfer, mainly due to the presence of solid particles. First of all, light scattering phenomena limit the use of spec- troscopic techniques and the consequent achievement of the parameters required for a rigorous characteriza- tion of the energy transfer process. For these reasons, up to now energy transfer mechanisms have been high- lighted indirectly by considering the product distri- bution of particular (and rare) reactions. The radiant field distribution in an irradiated slurry suspension is intrinsically not homogeneous and depends on features of the light absorbing species, reactor geometry and configuration, type of radiation and physico-chemical and optic characteristics of the photocatalyst.21 Indeed, unlike reactants which can be mechanically mixed within the system, photons do not possess mass and their “concentration” decreases with the distance from the radiation source.22 In these systems it is challenging to retrieve the local value of the rate of photon absorp- tion (RPA) which in turn determines the intrinsic reac- tion rate.23 In fact, only the average value of RPA is experimentally accessible while the local one, describ- ing the intrinsic kinetics, i.e. the events occurring at a molecular level not depending on mass and energy transport phenomena, can be retrieved rigorously only by the laborious and time demanding solution of the radiation transfer equation (RTE) by means of numeri- cal methods. Monte Carlo simulations24 and discrete ordinate methods (DOM) are usually used.25 This prob- lem has been recently approached by demonstrating that the average values satisfactorily approximate the local ones at sufficiently low optical thickness of the suspension.23 However, even if RPA could be approxi- matively retrieved, it is impossible to attribute it to the sole energ y transfer events because the macroscopic chemical transformation observed in a photocatalyt- ic reactor is the result of various processes differently interacting in a complex way. Only in few cases it has been possible to unequivocally attribute the formation of an intermediate product to energy transfer mecha- nisms. As a matter of fact, these cases are rare because the product distribution of electron and energy trans- fer reactions is often similar. Moreover, even if specific energy transfer derived products could be produced, they must be selectively obtained. This is not trivial, as the hydroxyl radicals photocatalytically produced oxi- dize almost any organic species with only few excep- tion, and the selective formation of a specific com- pound is the result of the complex interaction between light, reactants, products, and the irradiated surface of the photocatalyst. Even if the specific compound is selectively obtained, consecutive reactions and adsorp- tion phenomena could mask the identification of the prevailing mechanism. For these reasons, energy trans- fer mechanisms in heterogeneous photocatalysis have been often vaguely invoked but up to now never direct- ly evidenced. The extent and the nature of the energy transfer processes, deeply clarified in the photochemis- try of homogeneous systems, need to be investigated in heterogeneous photocatalysis. 52 Marianna Bellardita et al. In principle it is possible to hypothesize that the energy transfer mechanisms occurring in homogeneous systems also hold when an energy acceptor locally inter- act with the excited semiconductor acting as the energy donor. Only some mechanistic details for the forma- tion of singlet oxygen in heterogeneous photocatalysis have been tentatively proposed while less is known when more complex systems are under investigation.26 Upon absorption of a photon of suitable energy, an electron (e-CB) and a hole (h+VB) are localized respectively in the conduction and valence band of a semiconductor (SC) according to Eq. 6:27 SC + hν → SC*(e-CB, h+VB) (6) The photogenerated charges migrate to the surface of the semiconductor where they undergo interfacial electron transfer (Eq. 7). In fact, they can reduce or oxi- dise electron donors (D) or acceptors (A) to the corre- sponding radical cation (D+∙) and anion (A-∙): SC*(e-CB, h+VB) + A + D → SC + A-∙ + D+∙ (7) The interfacial electron transfer is thermodynami- cally feasible if the redox potentials of the couples A/A-∙ and D+∙/D lie within those of the photogenerated charg- es. However, kinetic limitations or the absence of suit- able A and D species in the reacting medium can favour quenching of SC* and, possibly, consequent energy transfer to a generic species B (Eq. 8): SC*(e-CB, h+VB) + B → SC + B* (8) It is worth to note that the difference between the redox potentials of the SC bands and those of D and A species is the driving force of electron transfer while the downhill character of the process in terms of energy (E(SC*) > E(B*)) mainly determines the rate of energy transfer. Moreover, while the electron transfer pro- cess creates a large charge re-distribution by generating charged species (A-∙ and D+∙), energy transfer generates a neutral excited state (B*). As a consequence, for instance the polarity of the solvent affects more the electron than the energy transfer processes. The quenching of the excited semiconductor can occur radiatively by emis- sion of a photon. In this case the energy transfer to the species B occurs by absorption of the emitted photon similarly to the trivial mechanism expressed in homoge- neous systems. This process is summarized in Eqs. 9-10 and in Figure 1: SC*(e-CB, h+VB) → SC + hν (9) B + hν → B* (10) This mechanism does not require electronic interac- tion or even contact between the irradiated semiconduc- tor and the energy acceptor. The efficiency of this type of energy transfer does not depend on the distance between B and the surface of the semiconductor, but mainly on its concentration in the suspension. In fact, at higher B concentrations it increases the number of B molecules in the path of the emitted photons. Other factors influencing the efficiency of the trivial mechanism are the charge recombination rate and the probability to emit photons rather than heat. Moreover, it is required that the emission spectrum of the semi- conductor and the absorption spectrum of B overlap in order to avoid mismatch between the excited semicon- ductor and the acceptor. Unlike the trivial mechanism, Förster and Dexter energy transfer are radiationless processes and depend on the distance of the acceptor from the semiconductor (even if with different dependence laws, see Eqs. 4-5). No electron exchange between acceptor and donor occurs when Förster mechanism takes place (Figure 2). Indeed, the oscillating electric field locally produced by the separated charges behaves as a virtual photon which excites the acceptor through dipole-dipole interactions. Notably, the oscillating field can be generated both by charges localized in the valence and conduction bands upon band to band transitions, or within the conduction band by vibrational states transitions. On the other hand, Dexter energy transfer mecha- nism occurs when simultaneously two electrons move in opposite directions (from donor to acceptor and vicever- sa) without net charge exchange (Figure 3). It is generally accepted that in heterogeneous pho- tocatalysis electronic interaction occurs through surface adsorption of a substrate, which generally gives rise to surface metal coordination compounds. Adsorption per- turbates the electronic structure of both semiconduc- tor and adsorbate at different extents.28 This situation is hν B* TiO2* B e- h+ Figure 1. Trivial energy transfer mechanism. 53Energy Transfer in Heterogeneous Photocatalysis similar to what occurs in Dexter processes at the initial stage of interaction but, as below detailed, generally it results in interfacial electron transfer. Strong electronic coupling between adsorbate and semiconductors generates new energy levels.29 Combina- tion of the HOMO level of the adsorbate with conduc- tion band surface states creates a surface hybrid HOMO- LUMO system between the adsorbate and the semicon- ductor which can extend deeply into the semiconductor due to its band structure. The resulting electronic sys- tem is generally characterized by novel ligand to metal electronic transitions, usually in the visible light region, which can be classified as optical electron transfer (OET). In this case, neither the semiconductor nor the adsorbate alone are able to absorb visible light, while the resulting charge transfer complex does. Various organic and some inorganic compounds present this behaviour when adsorbed onto a semiconductor such as aromatic 1,2 diols, some lignin components and SO2. The extent of this electronic interaction is evident when consid- ering the effects of the adsorption of these compounds onto colloidal semiconductor nanoparticles (ca. 3 nm sized). In fact, in this cases the electronic alteration of the band structure is extended to the whole particle and a significant red shift of the band gap is obtained rather than novel absorption bands. Notably, similar behaviour is reported also for other strongly interacting electronic systems such as solid solutions of semicon- ductors.30 Weaker electronic coupling generally favours electron transfer from the adsorbate to the semiconduc- tor or, in rare cases, viceversa. This mechanism, also known as photoinduced electron transfer (PET), usually takes place when chromophore species such as dyes are adsorbed on a semiconductor.31 Visible light radiation absorption promotes the dye to its excited state which in turn injects an electron in the conduction band of the semiconductor. This weak electronic coupling deter- mines high electron transfer efficiency and low charge recombination and, therefore, these systems are often used in dye sensitized solar cells. Notably, these consid- erations justify the necessity to avoid the use of dyes as model compounds when one needs to estimate the vis- ible light photocatalytic efficiency of novel semiconduc- tors. In fact, the dye degradation rate expresses the effi- ciency of electron injection from the excited dye to the conduction band of the semiconductor rather than the activity of the semiconductor. It is generally accepted that the above reported elec- tronic interactions favour optical or photoinduced elec- tron transfer. On the other hand, as far as the energy transfer in photocatalysis is concerned, indirect obser- vations suggest that energy transfer is prevailing when the surface of the semiconductor is grafted with species which block or substitute the surface hydroxyl groups, as in the examples reported in the next Section. Therefore, it seems that the direct contact between semiconductor and substrate favours electron transfer, while a mediated contact or a certain distance between them could favour energy transfer. For this reason, it seems plausible that some energy transfer processes observed by blocking the surface of the semiconductor could be of trivial or Först- er nature. Even if, up to now, the few examples of energy transfer driven photocatalytic reactions reported do not allow to further discriminate between them, it is pos- sible to conclude that surface grafting can be proposed as a tool to switch between energy and electron transfer processes. Hydroxyl groups mainly origin from dissociative adsorption of water at the surface of the semiconductor.32 They determine the acidity and basicity of the surface, are responsible for the hydrophilicity (and superhydro- philicity) of the surface and influence the water dynamics and the competition between different substrates, greatly affecting the selectivity of photocatalytic reactions. In some cases, the hydroxylation density favours the pho- tocatalytic activity even if not all of the surface hydroxyl groups are able to generate hydroxyl radicals. However, for the purpose of this paper, it is worth to stress the role of hydroxyl radicals in determining the electronic fea- tures of the semiconductor. First of all, protonation or deprotonation of hydroxyl groups changes the charge of the surface. Therefore, pH changes can shift the poten- tials of the valence and conduction band and influence the redox ability of the irradiated semiconductor. More- TiO2* B TiO2 B* e- e- h+ e- e- h+ Figure 2. Förster energy transfer mechanism. Figure 3. Dexter energy transfer mechanism. 54 Marianna Bellardita et al. over, hydroxyl groups can act as traps for the photogen- erated charges which thus prolong their life time and facilitate interfacial electron transfer. Moreover, as men- tioned above, they act as anchoring sites for organic spe- cies and mediate the electronic modifications induced by the interaction. As a consequence, blocking these sites may favour charge recombination thus reducing the probability of efficient charge transfer.26 3. EXAMPLES OF ENERGY TRANSFER DRIVEN PHOTOCATALYTIC REACTIONS Some significant examples of energy transfer driven processes have been obtained in plasmonic photocata- lytic composites used as sensors or as oxygen getters, but only rarely energy transfer driven processes have been invoked in classical heterogeneous photocatalysis for degradation or synthetic applications. One of the first examples of energy transfer process in heterogeneous photocatalysis has been reported by Wang et al.33 Authors hypothesized, without providing conclusive demonstration, that an array of vicinal TiO2 nanoparticles could enable energy transfer through an antenna mechanism similar to the natural photosynthet- ic process of the same name (Figure 4). An interesting case of photocatalytic mechanism which can be possibly interpreted as an energy trans- fer process is the hole induced oxidation of alcohols.34 This reaction generally occurs through abstraction of a hydrogen atom from the α position, giving rise to α-hydroxyalkyl radicals (Eq. 11): R−CH2 –OH + h+ →R−ĊH−OH + H+ (11) These radicals are generally powerful reducing spe- cies which can then inject an electron into the conduc- tion band of the semiconductor according to a photo- electrochemical process often referred to as “current doubling effect” which eventually produce the carbonyl compound.35 This example is paradigmatic for at least two reasons. Firstly, by considering that the spatial charge separation of the photogenerated electrons and holes is only some Angstroms across the surface of the excited semiconductor, it is highly probable that the cur- rent doubling effect occurs in a concerted way rather than as a two-steps process. In this case it could be ten- tatively seen as a Dexter energy transfer. Secondly, the sole presence of the carbonyl product is not an evidence of an energy transfer mechanism, as the same product can be obtained also through other mechanisms. In fact, evidences of current doubling effects have been obtained by photocurrent measurement rather than by simple photocatalytic experiments. This example highlights the difficulties faced in trying to evidence energy transfer mechanisms in photocatalytic reactions. Direct or indirect detection of singlet oxygen in aqueous photocatalytic suspensions is an indirect dem- onstration of the existence of energy transfer triggered process in heterogeneous photocatalysis. Molecular oxy- gen exists in its ground state as a triplet. Two excited states of molecular oxygen can be obtained upon excita- tion. The energy of these two singlet states, denoted as 1Σg+ and 1Δg lie 158 and 95 kJ∙mol-1 above the energy of the ground state, respectively. The 1Δg state, common- ly denoted as 1O2, is the more stable of the two excited states and is enough long living to induce chemical transformation under mild conditions. Formation of singlet oxygen in the presence of irra- diated TiO2 as the semiconductor has been explained by Nosaka et al.36 in terms of double electron transfer in opposite direction. In fact, molecular oxygen can be first reduced by a photogenerated electron at the surface of TiO2 to superoxide anion radical which in turn can be oxidized by a photogenerated hole giving rise to singlet oxygen. It is evident that the result of this mechanism is a neutral species (singlet oxygen) and no net charge exchange occurs. To the best of our knowledge, there are no clear evidences that this double electron trans- fer occurs in a consecutive or in a concerted way. In the second case, similarly to the current doubling effect mentioned above, the formation of singlet oxygen could resemble a Dexter type energy transfer mechanism. Other authors hypothesized that trivial mechanism could be responsible for the formation of singlet oxygen, i.e. that the radiative emission occurring upon charge recombination in the semiconductor could afford tri- plet to singlet excitation of molecular oxygen.37 How- ever, Daimon et al. pointed out the energetic mismatch between the band gap of the considered semiconductor (TiO2) and the energy difference between triplet and sin- glet oxygen states.38 By taking into account this obser- vation, recently Macyk et al.39 proposed the formation of singlet oxygen in the presence of surface modified A A-∙ D D+∙ e- e- e- e- e- h+ h+ h+ h+ h+ TiO2 TiO2 TiO2 TiO2 TiO2 Figure 4. Photocatalytic antenna mechanism. 55Energy Transfer in Heterogeneous Photocatalysis TiO2, through energy transfer from excited Ti3+ species to molecular oxygen. In this case the energetic differ- ences between the states in the donor (semiconductor) and acceptor (molecular O2) are compatible. The semi- conductor quenching responsible for the transfer, in this case, is an intra-band transition (within the conduction band) rather than band to band recombination (Figure 5). Unfortunately, the nature of this energy transfer has not been highlighted. Also in this case surface modifications performed by substituting hydroxyl with fluoride groups or by anchor- ing organosilanes, organic molecules or platinum com- plexes are reported to enhance the production of singlet oxygen. For instance, Janczyk et al.37 observed efficient photocatalytic degradation of cyanuric acid only in the presence of surface modified TiO2. In fact, cyanuric acid is one of the few compounds which cannot be photocat- alytically degraded by bare TiO2, due to its stability even in the presence of hydroxyl radicals, superoxide anions and peroxides, but it can be oxidized in the presence of singlet oxygen. These reports elegantly demonstrate that (i) singlet oxygen can be produced in aqueous irradiated suspen- sions of TiO2, and that (ii) modification of the surface of TiO2 promotes the production of singlet oxygen. A similar indirect demonstration of significant sin- glet oxygen formation in the presence of surface modi- fied TiO2 has been recently reported by Ciriminna et al. [40,41] In this case a synthetic approach has been used. The considered reaction is the epoxidation of limonene (see Figure 6). This natural terpene is gaining increasing attention because it can be used as a raw material for the pro- duction of a promising biopolymer, i.e. polylimonene carbonate (PLC). PLC is biodegradable, presents high thermal resistance (transition temperature up to 180°C), high transparency, and exceptional gas permeation abil- ity. These features make PLC a green alternative to the petrochemical derived polycarbonates with applications in the field of breathing glasses and food safe plastics. Moreover, the presence of an exocyclic C=C double bond, easy to functionalize, makes the properties of PLC extremely tunable for innovative applications such as sea water soluble polymers and antibacterial polymers. The bottle neck of the PLC production is its starting materi- al, i.e. limonene epoxide (LE). LE is obtained industrially from limonene in low yields and harsh operative condi- tions (Prileshayew reaction). Recently, limonene epoxide has been obtained with high conversion and selectivity (up to 90%) in the presence of surface modified com- mercial TiO2 (Evonik) under simulated solar light and in acetonitrile as the solvent. While OH radicals formed in the presence of bare TiO2 mainly induced overoxidation of limonene, alkyl silane modification of TiO2 favoured formation of singlet oxygen which selectively oxidizes limonene to 1,2 limonene epoxide. Another example of energy transfer driven process is the photocatalytic isomerization of caffeic acid in aqueous TiO2 suspensions and under nitrogen atmos- phere42 (Figure 7). UV irradiation of aqueous solutions of trans-caf- feic acid in the absence of the semiconductor induces photochemical formation of the cis isomer until a pho- TiO2* e- h+ 3O2 O2-∙ 1O2 TiO2* e- h+ TiO2* h+ Ti 3+ Ti3+* 3O2 1O2 IR UV light D+∙ D e- A B Figure 5. Mechanisms of singlet oxygen formation. A: Nosaka mechanism; B: Macyk mechanism. Notably, the TiO2 surface has been functionalized in the experiments carried out by Macyk et al. 39 Figure 6. Limonene epoxidation. Trans-caffeic acid Cis-caffeic acid Figure 7. Isomerization of caffeic acid. 56 Marianna Bellardita et al. tostationary cis/trans ratio is achieved. In the pres- ence of TiO2 a higher cis/trans ratio has been observed. This finding suggests that photocatalysis contributes to isomerization along with the photochemical process. Various tests have been performed in order to highlight this result. Photocatalytic degradation of caffeic acid is suppressed being the reaction carried out under nitrogen atmosphere. This factor supports the hypothesis of ener- gy transfer due to the high recombination probability in deoxygenated suspensions. Moreover, in the presence of 2-propanol as the hole scavenger, isomerization was totally inhibited and again surface modification of TiO2 further enhanced the cis/trans ratio. 4. CONCLUSION AND PERSPECTIVES Heterogeneous photocatalysis has been generally considered as the result of electron transfer reactions mainly occurring at the surface of the irradiated semi- conductor. The possibility of addressing the product distribution of photocatalytic reactions and of obtain- ing selectively some reaction intermediates as high value added compounds, recently moved the attention of the scientific community towards photocatalytic syntheses rather than photocatalytic degradation. As a matter of fact, after ca. 50 years of scientific investigations on pho- tocatalytic water purification, the gap between laboratory solutions and the needs of the water purification indus- try is evident. A change of direction is required in this field. It makes sense to use photocatalysis, possibly cou- pled with other advanced oxidation processes, to get rid of compounds harmful at low concentrations but only as a final treatment, after the application of technolo- gies capable to efficiently treat high volumes of effluents. Robust and reusable photocatalysts should be used rath- er than elegant and complex composites and applicative and engineering issues must be faced rather than basic research. The situation is different when considering pho- tocatalysis as a green alternative to traditional organic synthesis methods. In this case basic research is still needed to develop tools to control and address the selec- tivity of the process and to propose novel and sustainable synthetic solutions. From the few examples summarized hereby it is possible to conclude that when the substrate electronically interacts with the surface of the semicon- ductor, electron transfer is generally the prevailing pro- cess, even if Dexter-like double electron transfer pro- cesses cannot be excluded. On the other hand, trivial or Forster energy transfer processes likely occur at the sur- face of modified semiconductors where orbital overlap- ping is less probable. These preliminary results, however, require further efforts to understand the energy transfer processes occurring in irradiated photocatalytic systems. In our opinion this knowledge will induce the develop- ment of tools allowing to efficiently switch from electron to energy transfer reactions and will open novel possibili- ties especially in the field of the green photocatalytic syn- thesis of high value compounds. REFERENCES 1. T. Förster, Ann. Phys. 1948, 437, 55. 2. D. L. Dexter, J. Chem. Phys. 1953, 21, 836. 3. Y. Zhang, T. H. Wang, Theranostics 2012, 2, 631. 4. I. L. Medintz, H. Mattoussi, Phys. Chem. Chem. Phys. 2009, 11, 17. 5. I. L. Medintz, H. T. Uyeda, E. R. Goldman, H. Mat- toussi, Nat. Mater. 2005, 4, 435. 6. J. C. Claussen, W. R. Algar, N. Hildebrandt, K. Susumu, M. G. Ancona, I. L. Medintz, Nanoscale 2013, 5, 12156. 7. J. C. Claussen, N. Hildebrandt, K. Susumu, M. G. Ancona, I. L. Medintz, ACS Appl. Mater. Interfaces 2014, 6, 3771. 8. S. Choi, H. Jin, J. Bang, S. Kim, J. Phys. Chem. Lett. 2012, 3, 3442. 9. P. V. Kamat, J. Phys. Chem. C 2008, 112, 18737. 10. K.F. Chou, A.M. Dennis, Sensors 2015, 15, 13288. 11. E. Collini, C. Curutchet, T. Mirkovic, G. D. Scholes, in Energy Transfer Dynamics in Biomaterial Systems Springer Series in Chemical Physics, vol. 93 (Eds.: I. Burghardt et al.) Springer-Verlag, Berlin, Heidelberg, 2009. 12. V. May, O. Kühn, Charge and Energy Transfer Dynamics in Molecular Systems, Wiley‐VCH Verlag GmbH, Weinheim, 2011. 13. K.E. Sapsford, L. Berti, I. L. Medintz, Angew. Chem. Int. Ed. 2006, 45, 4562. 14. K. E. Sapsford, T. Pons, I. L. Medintz, H. Mattoussi, Sensors 2006, 6, 925. 15. G. W. Chen, F. L. Song, X. Q. Xiong, X. J. Peng, Ind. Eng. Chem. Res. 2013, 52, 11228. 16. N.T. Chen, S. H. Cheng, C. P. Liu, J. S. Souris, C. T. Chen, C. Y Mou, L. W. Lo, Int. J. Mol. Sci. 2012, 13, 16598. 17. F. Parrino, V. Loddo, V. Augugliaro, G. Camera Roda, G. Palmisano, L. Palmisano, S. Yurdakal Catal. Rev. 2019, 61, 163. 18. L. Zhang, L. Liardet, J. Luo, D. Ren, M. Grätzel, X. Hu, Nat. Catal. 2019, DOI: 10.1038/s41929-019-0231-9. 19. V. Augugliaro, V. Loddo, M. Pagliaro, G. Palmisano, L. Palmisano, Clean by Light Irradiation: Practical 57Energy Transfer in Heterogeneous Photocatalysis Applications of Supported TiO2, RSC pubblishing, 2015. 20. F. Parrino, M. Bellardita, E.I. García-López, G. Marcì, V. Loddo, L. Palmisano, ACS Catal. 2018, 812, 11191. 21. G. Camera-Roda, V. Augugliaro, A.G. Cardillo, V. Loddo, L. Palmisano, F. Parrino, F. Santarelli, Catal. Today. 2016, 259, 87. 22. V. Balzani, G. Bergamini, P. Ceroni, Angew. Chem. Int. Ed. 2015, 54, 11320. 23. G. Camera-Roda, V. Loddo, L. Palmisano, F. Parrino, Catal. Today 2017, 281, 221. 24. C. A. Arancibia-Bulnes, J. C. Ruiz-Suárez, Appl. Opt. 1999, 38, 1877. 25. G. Sgalari, G. Camera Roda, F. Santarelli, Int. Com- mun. Heat Mass. 1998, 25, 651. 26. F. Parrino, C. De Pasquale, L. Palmisano, Chemsu- schem 2019, 12, 589. 27. H. Kisch, Semiconductor photocatalysis: principles and applications, Wiley-VCH Verlag GmbH, Weinheim, 2015. 28. W. Macyk, K. Szaciłowski, G. Stochel, M. Buchalska, J. Kuncewicz, P. Łabuz, Coord. Chem. Rev. 2010, 254, 2687. 29. P. Ji, M. Takeuchi, T. M. Cuong, J. Zhang, M. Mat- suoka, M. Anpo, Res. Chem. Intermed. 2010, 36, 327. 30. H. Lachheb, F. Ajala, A. Hamrouni, A. Houas, F. Par- rino, L. Palmisano, Catal. Sci. Technol. 2017, 7, 4041. 31. T. Sakata, K. Hashimoto, M. Hiramoto, J. Phys. Chem. 1990, 94, 3040. 32. A. D. Weisz, A. E. Regazzoni, M. A. Blesa, Solid State Ionics 2001, 143, 125. 33. C. Wang, R. Pagel, J. K. Dohrmann, D. W. Bahne- mann, C. R. Chimie 2006, 9, 761. 34. H. Schwarz, R. Dodson, J. Phys. Chem. 1989, 93, 409. 35. E. Kalamaras, P. Lianos, J. Electroanal. Chem. 2015, 751, 37. 36. Y. Nosaka, T. Daimon, A. Y. Nosaka, Y. Murakami, Phys. Chem. Chem. Phys. 2004, 6, 2917. 37. A. Janczyk, E. Krakowska, G. Stochel, W. Macyk, J. Am. Chem. Soc. 2006, 128, 15574. 38. T. Daimon, Y. Nosaka, J. Phys. Chem. C 2007, 111, 4420. 39. M. Buchalska, P. Łabuz, Ł. Bujak, G. Szewczyk, T. Sarna, S. Maćkowski, W. Macyk, Dalton Trans. 2013, 42, 9468. 40. R. Ciriminna, F. Parrino, C. De Pasquale, L. Palmisa- no, M. Pagliaro, Chem. Commun. 2018, 54, 1008. 41. F. Parrino, A. Fidalgo, L. Palmisano, L. M. Ilharco, M. Pagliaro, R. Ciriminna, ACS Omega 2018, 3, 4884. 42. F. Parrino, A. Di Paola, V. Loddo, I. Pibiri, M. Bellar- dita, L. Palmisano, Appl. Catal. B: Environ. 2016, 182, 347. Substantia An International Journal of the History of Chemistry Vol. 3, n. 2 Suppl. 6 - 2019 Firenze University Press Where does chemistry go? From mendeelev table of elements to the big data era Luigi Campanella1, Laura Teodori2,* Visualizing Solubilization by a Realistic Particle Model in Chemistry Education Antonella Di Vincenzo, Michele A. Floriano* Chemistry as building block for a new knowledge and participation Stefano Cinti Tissue Engineering Between Click Chemistry and Green Chemistry Alessandra Costaa#, Bogdan Walkowiakb, Luigi Campanellac, Bhuvanesh Guptad, Maria Cristina Albertinie* and Laura Teodori a, f* Chemistry Beyond the Book: Open Learning and Activities in Non-Formal Environments to Inspire Passion and Curiosity. Sara Tortorella,1,2,* Alberto Zanelli,2,3 Valentina Domenici2,4