Photodynamic Therapy: An overview

The use of light in medical treatment is nothing new. In the ancient world, roots of the plant Dorstenia were applied topically on to irritated skin which would clear following a few hours of sunshine. The active ingredient, psoralen, is now the basis of PUVA (psoralen + UVA) therapy used to treat the effects of psoriasis and other skin ailments. Psoralen acts by intercalating between base pairs of DNA and upon UV irradiation, the two double bonds form [2+2] cyclo-adducts with thymine, kinking and destroying the DNA of the replicating cells, which was causing the skin irritations in the first place.

In the early twentieth century, following significant progess in synthetic chemistry of coloured dyes throughout the nineteenth century, two German scientists completed work on the toxic effect of eoisin, a methylene based dye. The effect was only noted with the presence of light, and at a later date the presence of oxygen. Thus the modern day science of photodynamic therapy (PDT) was discovered, consisting of three components: photosensitising dye, light, and oxygen. Despite positive results from trials, the work went relatively unnoticed, and it wasn’t until the 1970s that it really picked up again.

Fundamentals

The basic principles of PDT are relatively easy to consider. A light absorbing dye is applied or injected into the patient and after a time appropriate for maximum uptake into tumour, the affected area is irradiated with light. The dye absorbs the incident light and an electronically excited state is formed. This subsequently generates a reactive oxygen species (ROS) which destroys the tumour. The concept is beautifully simple, and as the dye is non-toxic in the absence of light, does not carry the negative effect of traditional chemotherapies which are much less discriminate in their action. In PDT, only irradiated areas of body tissue will generate activity leading to cell destruction. As mentioned, there are three components to PDT: light, photosensitiser and oxygen. the latter two are considered in turn, below.

Oxygen

The net result of dye irradiation is the generation of reactive oxygen species, and it is generally considered that singlet oxygen 1O2 is the ROS responsible for cell destruction in PDT. 1O2 is formed when oxygen, which in the ground state exists in as a triplet (3O2 ) absorbs energy. According to the MO diagram of the ground state 3O2 shown, the singlet is formed with the pairing of electrons in the LUMO. This is energetically unstable relative to the ground state, as there is a vacant orbital of the same energy available to the paired electron. Hence, 1O2 is energetically higher (thus more reactive) and will return to its triplet ground state. The lifetime of the singlet state in a cell is of the order of 100s of ns, and it has been estimated that it can diffuse less than 50 nm in this time. Therefore, in cells of the order of microns, the action is limited to cellular dimensions.

1O2 is formed by energy transfer from the photoexcited dye. But an alternative is possible. If the photoexcited dye transfers an electron to molecular oxygen, a superoxide anion is formed. Therefore, a crucial aspect of drug development in PDT is the nature of the ROS formed. Electron transfer forming superoxide anion is called a Type I reaction. Energy transfer forming 1O2 is called a Type II reaction. We can distinguish between these by understanding the chemistry behind them. Importantly for us here. Type II is detected because as singlet oxygen returns to its triplet ground state, it emits a small amount of infrared phosphorescence, which can be detected (see figure – the emission maximum is approximately 1270 nm). Type I on the other hand, can be detected by monitoring the redox chemistry of Fe3+ and subsequent formation of hydroxyl radicals (the photo-Fenton mechanism).

Singlet oxygen emission

Photosensitiser

The photosensitiser has several functions. It must locate in the tumour. This involves considering both hydrophilic and lipophilic components in the molecular design that are not covered here. It should also absorb in the far visible region (600 – 800 and preferably 700 – 800 nm). Haemoglobin is a significant component of body tissue, and absorbs strongly in the mid visible region (580 nm). This is obvious when we shine a light through our hand – only red light passes through. Therefore the ideal will be one which absorbs light where the body does not, allowing them to be used deep within body tissue (e.g. liver, pancreas).

As it stands, clinically approved PDT drugs are not yet optimised for longer wavelength light absorption, and hence PDT treatment is currently limited to areas that are easy to expose to a light source: skin, lung, oesophagus, etc. This issue of “penetration depth” was the subject of a recent court case, whereby a doctor justifiably claimed that PDT treatment would not be suitable for treatments such as liver cancer, as the liver was just too big for light to pass through. (The doctor was subsequently acquitted of all charges). The bar chart (Ref 1) shows how photosensitiser absorption capacity affects penetration depth, and this is the focus of current research (below). Penetration doubles once light at longer wavelength than the absorption of haemoglobin is achieved (630 nm) and doubles again at 700 nm. An ideal photosensitiser will therefore absorb between 700 – 800 nm.

Other factors for an ideal photosensitiser include: low toxicity in the absence of light and little post-treatment side affects. One of the most significant side effects is post-treatment light sensitivity, whereby patients have to avoid light for fear that healthy body tissue which have residual amount of photosensitiser present will generate unwanted activity.

Jablonski Diagram for generation of singlet oxygen (ref 1)

From a photochemical point of view, one of the most important dye characteristics is that it will form a high concentration of triplet excited state. Since the mechanism of action is generation of singlet oxygen, a singlet state may not be long-lived enough to allow time for reaction with oxygen (singlet-singlet deactivation via fluorscence or non-radiative means is an allowed process, and therefore very fast). If the triplet forms via intersystem crossing, its deactivation is forbidden, and hence slower, allowing the energy transfer to oxygen to be more competitive. This again provides potential for future research (see below).

Photofrin

Photofrin (R)

The first clinically approved PDT drug was photofrin (R). Photofrin is a porphyrin -based compound and you may wish to examine its structure to identify hydrophilic and lipophilic components alluded to above. It absorbs at 630 nm, which is within the PDT window and has a respectable quantum yield of inter-system crossing of about 25%. However, its absorption is not significant, with a low extinction coefficient. Its worth noting here also that PDT is not limited to cancer based therapy, it has also been used as alternative to antibiotics and for gum disease (There is a good overview article in Chemistry World, ref 2, and some nice pictures for dental treatment at the Periowave site, ref 3)

Current Developments

Absorption spectra of chlorins and bacteriochlorins (Ref 1)

Photofrin’s limitation is primarily its light absorption. To get to a point where PDT can become more versatile, the photosensitiser needs first to absorb in the 700 – 800 nm window (and then subsequently satisfy all other demands re singlet oxygen generation, accumulation in tumours….). Reduction of one (chlorins) and two (bacteriochlorins) of the four pyrroles in porphyrin based compounds have been found to shift the wavelength of absorption to longer wavelengths. In the example shown, the absorption shifts to about 700 nm for chlorin-based molecules and to about 800 nm for bacteriochlorin based molecules. The compound shown has a high extinction coefficient of absorption and good oxygen generation capabilities. More information on these compounds is available in Ref 1.

Intersystem crossing can be enhanced by the heavy atom effect, and this is the subject of another class of boron-based compounds. It was noted for certain sites of substitution of iodine, the singlet oxygen generation capacity increased, attributed to an increased intersystem crossing yield caused by the iodine heavy atoms. More information on these compounds is available in reference 4.

Utilising heavy atom effect to enhance ISC (Ref 4)

Summary

PDT provides huge potential in treatment of cancerous tumours and a range of other antibiotic treatments. It has been called a very selective surgeon’s knife thanks to its ability to isolate the affected area for treatment with little collateral damage. At the core of future developments of PDT is an understanding of the photochemistry at its heart, and now a century after the first PDT action was discovered, it looks like it has a positive future.

References

1. (Primary reference for this article) Chem Soc Rev, 2011, 40, 340: Sections A, B, C.1, C.4, D, F.2, F.3

2. Chemistry World, 2012, April, 52, see also Education in Chemistry, 2004, May, 71.

3. Periowave blog (accessed December 2012)

3. Chem Soc Rev, 2013, 42, 77: pages 77 – 81.

Quenching Mechanisms

Excited states can be deactivated in several ways – they can emit, giving off light energy, deactivate – resulting in a “vibrationally hot” ground state (i.e. energy loss as heat) or be quenched by another molecule. In this section, we will consider the process of quenching, and outline some ideas that use the process of quenching in applications. In addition, we will examine how the process of quenching can be studied to give us information on the nature of the excited state-quencher interaction. It is assumed the reader is familiar with the information presented in the Light Absorption and Fate of Excited State post.

Overview

Quenching of the excited state is a significant process because it is usually a very efficient process. The excited state of many organic compounds, for example, are efficiently quenched by the presence of oxygen, at rate constants several orders of magnitude faster than emission processes from the triplet state. (Emission from the triplet is spin forbidden, and hence has rate constants in the range 10 to 103 dm3 mol-1 s-1, whereas oxygen quenching may take place at rate constants of the order 109 dm3 mol-1 s-1. Therefore, to study the emission from triplets, we need to deaerate the sample (and have it at low temperature – see the experimental section). Quenching processes can occur by two processes – electron transfer or energy transfer. In both cases, the excited state energy of the luminophore (the luminescent species) is deactivated due to the presence of the quencher. There are two scenarios by which quenching is generally modelled, and these are discussed below.

Dynamic Quenching of an Excited State

If a solution with emitting species is studied, and for every 100 photons absorbed by the solution, 30 are re-emitted, the quantum yield of emission is said to be 0.3. What happens to the other 70? They are translated into radiationless transitions, such as deactivation. As mentioned in the Ruthenium polypyridyl photochemistry post, we can quantify the quantum yield of emission (or any process) as being the rate constant of that process (in this case emission) divided by the sum of all rate constants deactivating the excited state. If we divide the emission quantum yield in the absence of quencher by that in the presence of quencher, we can generate an expression known as the Stern-Volmer equation, as shown below.

SV_quenching

Derivation of the Stern-Volmer Equation based on considering rate constants of deactivation in the absence and presence of quencher

The Stern-Volmer equation models what is called dynamic quenching, quenching which occurs by the quencher diffusing through solution and interacting with luminophore, resulting in a deactivation of the excited state. The emission intensity is reduced, because as well as other deactivation pathways before the presence of quencher, the presence of quencher now adds another deactivation pathway in competition with luminescence. This quenching process is controlled by how fast the quencher can diffuse through solution and “collide” with luminophore, and as diffusion is usually a very fast process in solutions, it can be very efficient.

The Stern-Volmer equation is the equation of a straight line, and hence it allows for  very easy experimental determination of the quenching rate constant, kq. If the emission intensity (or lifetime) in the absence of quencher and then in the presence of incremental amounts of quencher is measured, and the resulting ratio of emission intensities (I(0)/I) is plotted as a function of quencher concentration, the resulting graph (called a Stern-Volmer plot) will have an intercept of 1 and a slope called the Stern-Volmer constant, KSV. KSV is the product of the natural radiative lifetime (the lifetime in the absence of quencher, τ0, and the quenching rate constant, kq. Knowing the slope and the natural radiative lifetime allows easy calculation of the quenching rate constant. An outline of a common experiment – quenching of a ruthenium polypyridyl complex emission with Fe3+ is shown below.

Vodpod videos no longer available.
The fact that quenching can be so efficient means that it can be a useful probe in studying systems with emission properties. For example, ruthenium polypyridyl complexes have been used successfully as oxygen sensors, whereby the complex has been incorporated into a silica matrix and the resulting stub located inside packaging. In the absence of oxygen, emission is observed when the stub is irradiated with light. However, if oxygen leaches into packaging, the emission observed will be substantially reduced, as it will be quenched by the oxygen. By calibrating the reduction in intensity using a Stern-Volmer plot, it is possible to estimate the concentration (partial pressure) of oxygen in the system. The concept has applicability in food packaging and for containers holding oxygen sensitive artefacts (e.g. paintings).

Static Quenching

Dynamic quenching results from collisions between excited state and quencher. However, if the quencher is somehow associated with the luminophore in solution prior to light absorption, the association may mean that the luminophore will not emit, due to induced changes in its properties because of presence of quencher. Therefore the reduction in emission intensity will be affected by the extent to which the quencher associates to the luminophore and the number of quenchers present. The reduction in emission intensity can be quantified as follows. If the luminophore, M, associates with quencher, Q according to an equilibrium constant of association, Ks, then this association constant can be quantified as the ratio of associated luminophore-quenchers luminophore-quenchers moieties ([M-Q]) to the product of unassociated luminophore and quencher; [M][Q]. Since the total concentation of luminophore, [M]0 is equal to the sum of associated and unassociated luminophore, substitution of this into the equilibrium expression, followed by rearrangement results in another equation of a straight line, very similar in form to the Stern-Volmer equation. However, while plotting I0/I (as emission intensity can be said to be proportional to concentration) against [Q] will result in a straight line for static quenching, analogous to dynamic quenching, interpretation of the slope is different. In this case, the slope quantifies the association constant between quencher and luminophore – and therefore is useful in providing information on how these two species interact in the ground state.

static-quenching

Derivation of an expression for static quenching

Dynamic or Static?

The question that immediately arises now is that if plots of emission intensity against quencher concentration both produce straight line graphs, how do we know which type of quenching is occurring? The answer lies in thinking again about the nature of each type of quenching. For dynamic quenching, all luminophores are affected by the quenching process as it is probable that they will all collide with a quencher during their excited state lifetime, so both emission intensity and lifetime reduced on increasing quencher concentration. For static quenching by association, only luminophore-quencher associations result in reduction in emission, unassociated luminophores are free to luminesce as if there was no quencher present. Increasing quencher concentration affects emission intensity, because there are more associations, but not emission lifetime, as the unassociated luminophores can emit in the absence of quencher. (Note that these two scenarios are the extremes, and there are cases where a mixture of both static and dynamic quenching may occur simultaneously.)

dynamic_versus_static_quenching

Schematic of dynamic versus static (association) quenching

Therefore the diagnostic test for assigning whether a quenching mechanism is dynamic or static is to compare how the emission intensity and emission lifetime changes as a function of increasing concentration. In the case of dynamic quenching, plots of relative emission intensities and emission lifetimes will be th same, changing on increasing quencher concentration. For static quenching, only a plot of relative emission intensity will change, the emission lifetime plot will have  slope close to zero.

diagnostic_plots

Model diagnostic plots to distinguish between dynamic and static quenching

Another model of static quenching is where the quencher is in a fixed position close to the luminophore (e.g. in a frozen matrix or a zeolite). This is modelled by the Perrin model of quenching, which will be discussed in the experimental techniques section when discussing phosphorescence.

References

MK Seery, N Fay, T McCormac, E Dempsey, RJ Forster, TE Keyes, Photophysics of Ruthenium Polypyridyl Complexes formed with lacunary polyoxotungstates with iron addenda, Phys. Chem. Chem. Phys., (2005), 19(7), 3426 – 3433. An example showing unusual static quenching between a quencher (large polyoxometallate clusters) and a luminophore (a ruthenium complex).

B Valeur, Molecular Fluorescence: Principles and Applications, Wiley: Weinheim, 2002. Discusses the principles of dynamic and static quenching well.

Light Absorption and Fate of Excited State

Photochemistry is the study of what happens to molecules when they absorb light. Therefore it is important to consider the factors affecting whether and how efficiently molecules absorb. In addition, in the very short time-frame after a molecule has absorbed light, it can undergo a variety of processes. In applications, we may desire a particular process, so again an understanding of what pathways are available to excited states is important so that systems can be optimised as required (e.g. by changing solvent, modifying the molecule).

Students should note that this topic is traditionally approached from a quantum chemical background. All textbooks on photochemistry cover this well (for example see Turro or Gilbert and Baggott) so it is not necessary to relay it in too much detail here. Instead, a qualitative overview is presented for the purposes of providing a background to the material elsewhere on this site.

1. Light Absorption – Formation of the Excited State

Photochemistry is based on the reaction/reactivity of molecules in their excited state after they have absorbed light. By “light”, we mean that part of the electromagnetic spectrum that can promote electrons in the outer atomic orbitals to unoccupied orbitals – i.e. electrons near or at the highest occupied molecular orbital (HOMO) to orbitals near or including the lowest unoccupied molecular orbital (LUMO). To do this, the light must be of sufficient energy to promote electrons between electronic energy levels, and this is found to be light in the UV/visible region of the electromagnetic spectrum. For this reason, the region of the spectrum 200 nm < λ < 800 nm is sometimes referred to as the “photochemical window”. The range of wavelengths in the spectrum and the result of absorption by the atom/molecule is shown below.

Regions of the electromagnetic spectrum and their impact on atom structure

Regions of the electromagnetic spectrum and their impact on atom structure

Therefore, absorption of a photon of light of wavelength 200 – 800 nm may result in a HOMO-LUMO transition (dependent on other factors which we will discuss later). A very clear indication of this is observed in d-block complexes. For example, a ruthenium (II) complex has six d-electrons and has a low spin octahedral configuration t2g6. On absorption of visible light (λ ~450 nm), an electron is promoted to an eg orbital, giving the complex its red-orange colour. This transition is in the visible region. For d0 complexes such as TiO2, a d-d transition is not possible, and a transition from the oxide ligand to the metal centre – a ligand – to metal charge transfer (LMCT) transition occurs, but only if the molecule is irradiated by UV light (λ < 390 nm). Hence TiO2 is white, as it does not absorb any visible light.

Absorption of a photon of visible light causes a d-d transition in Ru(II) giving the molecule a visible colour

Absorption of a photon of visible light causes a d-d transition in Ru(II) giving the molecule a visible colour

As well as the type of transitions possible, a second factor to consider is the intensity of absorption as a function of wavelength. These absorptions, measured by UV/visible absorption spectroscopy for gases or solutions and diffuse reflectance spectroscopy (DRS) for solids will vary depending on the extinction coefficient, ε, of the molecule at that wavelength. The extinction coefficient is a measure of the probability of an electronic transition from ground to excited state, at a given wavelength. This probability is calculated via quantum chemical parameters that are beyond the scope of this course. However, in simple terms, the value of ε gives an indication of how “allowed” a transition is, where “allowed” is a meant strictly as a quantum chemical term. If ε is measured to be greater than 105 dm3 mol-1 cm-1, then the transition is “fully allowed” – all quantum chemical rules are passed. For transitions below ~100 [dm3 mol-1 cm-1 ,units implied from hereon], the transition is “forbidden”, indicating that all quantum rules are not passed, and the probability of transition is very low – i.e. the molecule does not absorb well at this wavelength.

The in-between grey area, for ε values between ~102 and ~104, are where the transitions are “partially allowed”. The quantum mechanical rules are based primarily on two components – spin and symmetry. The spin component says that if a transition involves a change of spin (e.g. singlet to triplet) then the transition is forbidden. The symmetry component examines the symmetry of the ground and excited state, and depending on these symmetries the transition will be allowed or forbidden. But these symmetry calculations are based on a molecule idealised conditions, so the symmetry of the real molecule may be distorted by the presence of solvent or of a heavy atom on the molecule (the so-called “heavy atom effect” – we will return to later). Hence if a transition is spin-forbidden, symmetry allowed, then the probability is very low, and ε will be <100. But if it is spin-allowed, symmetry forbidden, then appreciable absorption may be observed (102 – 104) because of the symmetry distortions mentioned above.

The final factor to consider about light absorption, having discussed types of transition and intensity of absorption above is the shape of absorption spectra. Again, these relate to the discussions above on the value of ε at each wavelength, but for an individual electronic transition (e.g. HOMO – LUMO), transitions between vibrational levels of each orbital may be more intense than others. These transitions are governed by the Franck-Condon Principle, which states that:

the electronic transition in a molecule takes place so rapidly compared to nuclear motion, that immediately afterwards the nuclei have still very much the same nuclear geometry (position and velocity) as before the transition.

In simple terms, this means that electronic transitions take place so quickly the nuclear geometry differences between ground and excited states do not have time to adjust, or in even simpler terms, these transitions are vertical. Consider the potential energy diagram for a HOMO and LUMO shown below. Each electronic orbital has some of its vibrational levels shown. The probability of an electron being in one of these orbitals can be calculated, and are “mapped” using wavefunctions, as shown.

Looking at these qualitatively, we can say that the most probable transition between a vibrational level in the ground state (HOMO) and one in the excited state (LUMO) will be the one where the wavefunctions overlap the most in the vertical line above the ground state groud vibrational level. (in either a positive or negative direction). On the left hand side of the diagram, the greatest overlap is (hypothetically) the ground state vibrational level 1 and the excited state vibrational level 1, so we have a 0 – 1 transition (spectroscopists will get annoyed at this notation, but it is used here just to illustrate the principle). On the right hand side, the excited state geometry is different to the ground state (the potential energy diagram is shifted to the right a little), so in this case the hypothetical best overlap is between 0 in the ground state and 4 in the excited state. Tehrefore the shapes of the two absorption spectra in each of these scenarios is different. Of course, in practice we don’t see this fine structure, the absorption spectra are essentially a line drawn over the tops of the individual transition peaks, resulting in the broad, generally featureless absorption spectra we are used to. But if we were to do it in the gas phase (eg iodine vapour experiment) we would see this fine structure. If you’re wondering, the reason we don’t see fine structure in solution is because the molecules absorbing light are being battered around by solvent molecules, so the energy levels are constantly moving up and down a little, therefore blurring the transitions a little. Each electronic transition will have a suite of different vibrational transitions, so a molecule with, for example, three bands in the experimental absorption spectrum consists three of these processes happening. Because electronic transitions also vary in intensity, some of the bands may be more intense than others.

Ground and excited state potential energy curves with vibrational energy level wavefunctions shown. Left: The most probable transition is the 0 - 1 transition; Right: the different nuclear configuration of the excited state means that in this case, 0 - 4 transition is most intense. Curves over the PE diagrams show the resulting absorption spectra (Based on images presented in Gilbert and Baggott, which covers this area excellently)

Ground and excited state potential energy curves with vibrational energy level wavefunctions shown. Left: The most probable transition is the 0 - 1 transition; Right: the different nuclear configuration of the excited state means that in this case, 0 - 4 transition is most intense. Curves over the PE diagrams show the resulting absorption spectra (Based on images presented in Gilbert and Baggott, which covers this area excellently)

2. The Excited State

If a molecule absorbs light and forms an excited state, then it is in a very different state to one it was in the previous few sub-picoseconds. Excited states have been called “electronic isomers”, which rather underestimates their relevance. To emphasise the point, excited states are chemically different species to their corresponding ground states. This statement reflects the true beauty and power of photochemistry. For every photoactive molecule a second different molecule can be “created” by literally, the flick of a switch – this gives an inkling of the true potential of photochemistry as a discipline. Very often, these states are not accessible by thermal means because of the great differences in energy levels.

Excited states are energetically unstable and very short-lived. “Short” in this context means from sub-nano and nanosecond (if a process is allowed) to milliseconds and seconds, if a process is forbidden, such as phosphorescence. To put these numbers in context, the German photochemist and educator Michael Tausch has pointed out that the positive equivalent of a nanosecond (10-9 s), which is 10+9 s (or 1 gigasecond), is about the equivalent of a human lifetime.

Therefore the equipment and scientists which experimentally determine the processes which are discussed below should not be overlooked, and we will look at some of these in various articles (see Experimental). For now, it can be said that since the discovery of microsecond (x 10-6 sec) flash photolysis by Norrish and Porter in the 1950’s, each decade has seen another power of ten on the limit of time that can be studied culminating in Zewail’s development of femtosecond (x 10-15 sec) spectroscopy in the 1990’s. This is at the limit of atomic vibrations and indeed electron transfer, and so is probably a “true” limit, as beyond this the Heisenberg Uncertainty Principle becomes significant. Scientists at either end of the timescale, Norrish and Porter, and Zewail, won Nobel prizes for their efforts. these developments will be covered in more detail in a future article.

So what is the fate of the excited state? When a molecule absorbs light, it is a very fast process – on the order of picoseconds or lower. Depending on the wavelength of light used, and the Franck-Condon principle, above, the vibrational levels of some upper excited state will be populated with electron density. The various processes which occur can be represented on a Jablonski diagram, a sketch of the electronic energy levels in an atom together with their vibrational levels.

A Jablonksi diagram for an organic molecule. Radiative processes (those which are "vertical" in energy transfer) are shown in solid lines whereas non-radiative processes ("horizontal" energy transfer) are shown using dotted lines

A Jablonksi diagram for an organic molecule. Radiative processes (those which are "vertical" in energy transfer) are shown in solid lines whereas non-radiative processes ("horizontal" energy transfer) are shown using dotted lines. Indicative timescales are shown, although are molecule dependant.

In principle the Jablonski diagram is similar to the transitions in the potential energy curves, shown above, except the potential energy curves are usually not represented. A simple Jablonski diagram for an organic molecule is shown above. Note that a similar diagram for an inorganic compound will also include metal orbitals, so will be different in style. The processes which occur when a molecule absorbs light are below. We will discuss the kinetics of these processes in a separate post, looking at how they can be measured.

  1. Molecule absorbs light and populates upper excited state S* with electrons
  2. Electrons in upper vibrational levels of S* undergo vibrational relaxation and the electrons move to the lowest vibrational level of S*.
  3. The molecules very quickly dissipate this very high energy by internal conversion – the electron density moves to the lowest excited state, S1. Internal conversion occurs by the electron density transferring from the vibrational levels of the upper excited state to vibrational levels of a lower excited state which they are overlapping. Hence this is a “horizontal energy” transition, or a radiationless transition – it does not give off a photon of energy (light) as the electron density has not moved in one “big jump”.
  4. Vibrational relaxation again occurs, and the electron is now in the lowest vibrational state of S1. This is a statement of Kasha’s rule, which says that photochemical processes (fluorescence, quenching) happen from the lowest vibrational state of the lowest excited state (S1). The reason for this is that the processes described above leading to this situation all occur in a matter of picoseconds. The electron now has a choice of what to do next
  5. It may undergo fluorescence, giving off a photon of energy.
  6. It may undergo internal conversion as above.
  7. The electron may undergo intersystem crossing (ISC) to the triplet state. Once here, the molecule can undergo phosphorescence or deactivation. These processes are shown in the Jablonski diagram. Note the timescales involved in the various processes.

3. Conclusion

Light absorption can result in the formation of an (electronically) excited state, which has different chemical properties to the groud state. The intensity and shape of absorption spectra are a result of the nature of excitation between ground and excited states. Various processes result in the deactivation of the excited state.  The timescales of these indicate their efficiency, and we will look at these in more detail in future posts.

4. References

All general photochemistry texts discuss the principles of light absorption and deactivation of the excited state in good detail. some are given below, but any will give pretty much the same information.

Gilbert, A. and Baggott, J. E., Essentials of molecular photochemistry, Blackwell Scientific: London, 1991.

Turro, N. J., Ramamurthy, V. and Scaiano, J. C., Principles of molecular photochemistry: an introduction, University Science Books:Sausalito, 2009. Despite the title, a detailed text with lots on the various photophysical processes that occur on light absorption. These three authors are among the best known photochemists today. Turro’s classic, Modern Molecular Photochemistry, was for a long time the bible for photochemistry.