Introduction

Davide Vione and Paola Calza

Department of Chemistry, University of Torino, Via P. Giuria 5 10125, Torino, Italy

Table of Contents

1.1. Introduction 3

1.1.1. Basic Principles of Photophysics and Environmental
Photochemistry 3
1.1.2. Photosensitisers and Photoinduced Transients in Surface
Waters 7
1.1.3. Seasonal and Long-term Variations of the Main
Photosensitisers in Lake Water 11
1.1.4. Photoinduced Transformation of Organic Micropollutants 12
References 12

1.1. Introduction

In the following sections of this introductory chapter, some basic pieces of information are provided that will be useful for the understanding of the chapters that follow.

1.1.1. Basic Principles of Photophysics and Environmental Photochemistry

In the natural environment, sunlight is the source of radiation (λ>280nm) and, for photochemistry to be operational, one needs molecules that absorb radiation above 280 nm. When a molecule absorbs a photon, an electron is excited from the ground state (often it is the ground state from both an electronic and a vibrational point of view) to a vibrationally excited state of an electronically excited state. In the case of organic molecules, the ground electronic state is usually a singlet one (S0), where the electrons are all paired in full orbitals with antiparallel spin. When the molecule absorbs radiation, it reaches an excited state that is also a singlet one. It could be the first (S1), the second (S2) or a higher excited singlet state depending on the energy of the photon and of the molecular states (see Figure 1.1). To begin with, assume that the electron reaches a vibrationally excited state of S1 In some cases there might be an excess of vibrational energy, which may cause an excessive strain on the weakest molecular bond that is involved in the vibrational motions triggered by radiation absorption. In this case, bond breaking would take place: indeed, the very word photolysis suggests the occurrence of photoinduced bond breaking. As an alternative, the molecule can undergo fast vibrational deactivation to the S1 ground state. From here, the molecule could take part in chemical reaction, thermal or photophysical deactivation to S0, or inter-system crossing (ISC). Chemical reaction is quite rare in the case of the short-lived excited singlet states, but it accounts, for instance, for the photoisomerisation of 2-chlorophenol to 5-ring species. Thermal deactivation (internal conversion) can be of some importance in the field of environmental photochemistry because it would be enhanced by inter-molecular interactions that could, for instance, account for the limited photophysical activity and photochemical reactivity of high-molecular weight chromophoric dissolved organic matter (HMW CDOM).

In some cases, the S1 [right arrow] S0 transition takes place by emission of a fluorescence photon. If this is the case, the electron reaches an excited vibrational state of S0 and then the ground one by vibrational relaxation. The double loss of vibrational energy, in both S1 and S0, explains why the wavelength of the emitted (fluorescence) photon is higher (and, therefore, the associated energy is lower) compared to the absorbed photon. Finally, the electron in the S1 state can undergo ISC to the first excited triplet state (T1). The ISC implies spin inversion, and a further inversion would be required for the transition from T1 to S0. Such a transition is formally forbidden by the selection rules of quantum mechanics, which in practice means that its probability is low and that the T1 state is longer-lived than S1. The quite long lifetime means that T1 can undergo chemical reactions, for instance with the solvent or with dissolved molecules, as well as intramolecular rearrangements. A common energy-transfer reaction takes place with dissolved O2, which is favoured by the fact that the O2 ground state is a triplet one. The reaction yields singlet oxygen (1O2), while the molecule usually reaches the S0 state. Internal conversion from T1 to S0 is also possible, in which case the energy is thermally lost through e.g. collisions with the solvent. In some rare cases, the transition from T1 to S0 can take place by emission of a phosphorescence photon. Similarly to the case of fluorescence emission, in the case of phosphorescence the molecule reaches a vibrationally excited state of S0 from which a further vibrational loss of energy takes place.

Upon radiation absorption, other excited states different from S1 may be reached. If the energy is high enough (which is more common in the case of the environmentally non-relevant UVC radiation, λ<280nm), the electron may be abstracted from the molecule to produce the photoionisation. Interestingly, the irradiation of chromophoric dissolved organic matter (CDOM) is well known to yield aquated electrons. If the photon energy is not high enough to cause photoionisation, higher excited states than S1 may be reached (S2, S3 and so on). In the case of most organic molecules, the electron reaches a vibrationally excited state of the singlet state Sn (n>1), from which it undergoes vibrational deactivation to the ground vibrational level of S1. This is quite important for photochemistry and photophysics because, independently of the excitation wavelength, the molecule always reaches the ground state of S1, from which its further evolution will be the same independently of the details of the excitation process. It is the so-called Kasha's rule, according to which, for instance, the fluorescence emission wavelength is always the same (following the S1 [right arrow] S0 transition) independently of the excitation wavelength (which may cause transitions of the kind S0 [right arrow] S1, S0 [right arrow] S2 and so on, depending on the photon energy).

An interesting application of Kasha's rule can be seen in the case of the irradiation of anthraquinone-2-sulfonate (AQ2S). This molecule is not fluorescent (its ISC is much more efficient than the energy deactivation by fluorescence emission), but upon irradiation AQ2S forms fluorescent hydroxyderivatives. The fluorescence excitation-emission matrix spectrum of irradiated AQ2S is reported in Figure 1.2, showing a single emission band (in the range of 550–600 nm) that, however, corresponds to several excitation bands at different wavelengths, ranging from 200 to 500 nm and suggesting various S0 [right arrow] Sn transitions.

In some cases, Kasha's rule is apparently not followed. This is the case, for instance, of the fluorescence spectrum of 4-phenoxyphenol, which has four paired emission bands (see Figure 1.3). One pair has emission at 300–350 nm, which corresponds to two excitation bands at 200–250 and 250–300 nm. The second pair has emission at 350–425 nm, corresponding to two excitation bands around 250 and 300 nm. According to Kasha's rule, one would expect a single emission wavelength (S1 [right arrow] S0 transition), linked to the occurrence of one pair only or to four bands that should be vertically placed, as in the case of Figure 1.2. However, 4-phenoxyphenol exists in aqueous solution in three different rotational conformations that undergo very slow interconversion (compared to the typical fluorescence lifetimes). It has been shown that a conformer accounts for the bands with emission at 300–350 nm, the other two for the bands with emission at 350–425 nm.

1.1.2. Photosensitisers and Photoinduced Transients in Surface Waters

Several photochemical processes can take place in sunlit surface waters. Some of them involve the absorption of sunlight by the molecule(s) that are transformed, following reaction pathways that have been partly described in section 1.1.1 and that will be more extensively dealt with in Chapter 4. However, additional reactions are triggered by the absorption of sunlight photons by naturally occurring photoreactive species called photosensitisers. Among these naturally occurring compounds there are CDOM, nitrate, nitrite, Fe species and H2O2. The (photo)chemistry of Fe, which also partially involves that of H2O2, is quite complex and will be the subject of a separate chapter. The present introduction will provide a general overview of the photochemistry of CDOM, nitrate and nitrite, to be further elucidated later on in this book.

CDOM is, almost beyond any doubt, the single most important photosensitiser (or better, class of sensitisers) that occurs in natural waters. Its complex and not yet completely elucidated molecular structure has been the subject of intense debate. Today, many scientists accept that it could be a supramolecular (rather than a macromolecular) combination of smaller compounds, which form aggregates that vary for the apparent molecular size. The most important photoactive moieties of CDOM are its humic and fulvic parts, which are responsible for an important fraction of sunlight absorption by CDOM itself.

The absorption of sunlight by CDOM causes the excitation of its photoactive moieties to produce the excited singlet states (1CDOM*). In the case of aromatic aldehydes, benzophenones, quinones and other chromophores, the singlet excited states tend to undergo an efficient ISC to produce the longer-lived triplets (3CDOM*), which are often involved in chemical reactions that cause the transformation of other dissolved molecules including dissolved pollutants.

Despite the importance of triplet-sensitised transformation for the removal of many pollutants from surface water environments, the main 3CDOM* sink is actually the reaction with O2 to produce singlet oxygen (1O2). The latter is another important transient that might be involved in the indirect photochemistry of dissolved compounds (e.g. chlorophenolates and aromatic amino acids such as tryptophan and tyrosine) or that, as an alternative, may undergo deactivation upon collision with water. On top of all this, irradiated CDOM can also induce the photochemical production of •OH, of which it is actually the main source in most surface waters. Although several details are still missing, it is now generally acknowledged that part of the •OH photoproduction by irradiated CDOM involves the photochemical generation of H2O2. The latter probably takes place upon dismutation of HO2•/O2-•, produced upon scavenging by oxygen of the reduced radical transients that are formed by reaction between 3CDOM* and dissolved compounds. Once photochemically generated, H2O2 can yield •OH through direct photolysis or (more probably) the Fenton reaction with H2O2. The Fenton process is actually much less "clean" than may be suggested by its traditional stoichiometric notation because superoxidised Fe species (e.g. ferryl) may be formed in competition with •OH. The actual •OH yield reaches ~60% under the most favourable conditions (pH 2–3) and it decreases with increasing pH. The reactions described so far are reported below, where S–H is a generic dissolved molecule that undergoes oxidation by reaction with 3CDOM*.

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (1.1)

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (1.2)

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (1.3)

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (1.4)

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (1.5)

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (1.6)

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (1.7)

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (1.8)

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (1.9)

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (1.10)

However, the photochemical production of •OH by irradiated CDOM is also known to follow an additional, •OH-independent pathway. It is possible that this pathway does not only yield genuine •OH because so-called low-level hydroxylating species are certainly formed in the process. However, a certain amount of •OH is also produced. To get an idea of what the additional hydroxylating species may be, one can consider the photochemistry of AQ2S, the triplet state of which (3AQ2S*) is certainly unable to produce •OH upon water oxidation. However, 3AQ2S* quickly reacts with water to produce an adduct that may either evolve into AQ2S hydroxyderivatives or transfer the OH group to other molecules. The reaction products are not different from those expected upon reaction with •OH, but the AQ2S–H2O adduct is a much less powerful oxidant compared to •OH.

The H2O2-independent pathway to •OH in the presence of irradiated CDOM might involve water oxidation by 3CDOM* (which is still controversial, as some triplet states can oxidise H2O but others cannot), or a preliminary CDOM oxygenation followed by e.g.•OH production upon photochemical excitation of some oxygen-containing groups. Anyway, many details of the process are missing at the moment.

An interesting issue is the detection of very high concentration levels of 1O2 in the inner hydrophobic cores of HMW CDOM. This concentrated seems to have few chances of escaping into the solution bulk because hydrophobic 1O2 probes (undergoing preferential partitioning into the waterless core environment) are needed to highlight it. However, it could play an important role in the degradation of hydrophobic pollutants that are partitioned inside CDOM cores. The occurrence of elevated 1O2 in HMW CDOM cores might be due to elevated photochemical formation or to longer lifetimes than in solution. Presently, little to no evidence is available to support the former hypothesis, while there is evidence for the latter.

Nitrate yields •OH upon absorption of UVB and (to a lesser extent) UVA radiation. A major process is generally agreed to be the following:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (1.11)

The quantum yield of this reaction is independent of the wavelength and it is around 0.01. The involvement of H+ in the •OH production might suggest that the process could depend on pH, which is not exactly the case [at least as far as reaction (1.11) is concerned]. Indeed, •OH has pKa ~ 12 and the acid-base equilibrium [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] can, therefore, only be of importance around pH 12. At the typical pH values of surface waters, the equilibrium is totally shifted toward •OH.

Still, the photogeneration of •OH by irradiated nitrate depends on pH in the ~ neutral range, which warrants a different explanation. A possibility is the contemporary photoisomerisation of nitrate to peroxynitrite, ONOO-, taking place along with reaction (1.11). The ONOO- is the conjugated base of peroxynitrous acid, HOONO, a weak acid with pKaa ~ 7. HOONO and ONOO- strongly differ for the inactivation pathways because, in addition to the common back-isomerisation to nitrate, HOONO decomposes into •OH + *NO2 while ONOO- reacts with dissolved CO2. Therefore, the prevalence of HOONO at pH < 7 and of ONOO- at pH > 7 could explain hy the photogeneration of •OH by nitrate decreases with pH in the ~ neutral range.