Ab initio Study of Potential Ultrafast Internal Conversion Routes in

Oxybenzone, Caffeic Acid and Ferulic Acid: Implications for sunscreens

Tolga N. V. Karsili a,b*, Barbara Marchetti a, Michael N. R. Ashfold a* and Wolfgang

Domcke b

aSchool of Chemistry, University of Bristol, Bristol, BS8 1TS, United Kingdom

bDepartment of Chemistry, Technische Universität München, Lichtenbergstr. 4, 85747

Garching, Germany

Number of Figures: 11

Number of Tables: 2

ABSTRACT

Oxybenzene (OB) and ferulic acid (FA) both find use in commercial sunscreens; caffeic acid

(CA) differs from FA by virtue of an –OH group in place of a –OCH3 group on the aromatic

ring. We report the results of ab initio calculations designed to explore the excited state non-

radiative relaxation pathways that provide photostability to these molecules and the

photoprotection they offer towards UV-A and UV-B radiation. In the case of OB, internal

conversion (IC) is deduced to occur on ultrafast timescales, via a barrierless electron driven H

atom transfer pathway from the S1(11n*) state to a conical intersection (CI) with the ground

(S0) state potential energy surface (PES). The situation with respect to CA and FA is

somewhat less clear cut, with low energy CIs identified linking excited states to the S0 state

following photoexcitation and subsequent evolution along (i) a ring centred out-of-plane

deformation coordinate, (ii) the E/Z isomerism coordinate and, in the case of CA, (iii) an O–

H stretch coordinate. Analogy with catechol suggests that the last of these processes (if

active) would lead to radical formation (and thus potential phototoxicity), encouraging a

suggestion that FA might be superior to CA as a sunscreen ingredient.

1

1. INTRODUCTION

Jablonski diagrams1 invariably show internal conversion (IC) as a probable spin allowed

radiationless pathway for photoexcited molecules but it is only now, with the advent of high

level electronic structure calculations, that we are in a position to understand and to predict

the detailed nuclear dynamics by which IC occurs in any given molecule. Conical

intersections (CIs) between potential energy surfaces (PESs) are now recognised as crucial to

enabling population transfer between different electronic states,2, 3 and this article illustrates

some of the diverse range of nuclear distortions that can drive IC – in the context of

photoprotection mechanisms that prevail in selected molecules with real (or potential)

application in sunscreens.

Sunscreens are designed to attenuate the UV-A and UV-B components within the solar

spectrum (i.e. wavelengths longer than ~280 nm), thereby reducing the probability of skin

cancers like melanoma or squamous cell carcinoma.4, 5 Commercial sunscreens typically

comprise a heavy carrier oil, photostable inorganic particulates (TiO2 and/or ZnO, designed

to reflect, scatter and absorb some of the incident radiation) and conjugated organic

molecules that offer photoprotection without photodegrading or inducing unwanted

phototoxicity. Most organic molecules used in sunscreens comprise aromatic rings

conjugated to carbonyl groups (e.g. avobenzene, oxybenzone, cinnamates6 (including caffeic

and ferulic acid) and salicilates)7. These molecules offer photoprotection by virtue of their

high UV absorption cross-sections and high IC efficiencies; UV absorption results in

electronic excitation, but the dominant (ideally the exclusive) fate of this electronic excitation

is rapid conversion to vibrational energy which is then dissipated (as heat) within the

sunscreen.

The present study explores, theoretically, the detailed anatomy of several possible IC routes

in three molecules that either are, or potentially could be, used in sunscreens: caffeic acid (3-

(3,4-dihydroxyphenyl)-2-propenoic acid, henceforth often written simply as CA), its close

relative ferulic acid (3-(4-hydroxy-3-methoxyphenyl)-2-propenoic acid, henceforth FA) and

oxybenzone ((2-hydroxy-4-methoxyphenyl)-phenylmethanone, henceforth OB). CA and FA

are also intermediates in the synthesis of monolignols (the monomers of lignin).8, 9 The

minimum energy structures of each of these are depicted in fig. 1, with the constituent atoms

numbered for future reference.

2

Figure 1: (a) MP2 ground state minimum energy geometry of the most stable isomer of

caffeic/ferulic acid. (b) CAM-B3LYP minimum energy geometry of oxybenzone. The

constituent atoms are numbered for future reference, and the dotted lines indicate hydrogen

bonds.

CA and FA can both be viewed as substituted phenols or as substituted cinnamic acids.

Adding the hydroxy/methoxy groups to the ring (cf. bare cinnamic acid) has the effect of red-

shifting the UV absorption spectrum so as to match better with that needed for sunscreen

applications. Previous studies of phenol itself have identified possible IC pathways from the

first excited (S1 state) as a result of distortions along both the O–H stretch coordinate, and an

out-of-plane ring deformation.10, 11 The side chain in CA and FA also contains a C=C double

bond, that has the potential to isomerise upon UV photoexcitation. Tuna et al.12 recently

reported a detailed theoretical study of IC enabled by just such a photo-induced isomerism

(henceforth photoisomerism) in urocanic acid – a UV filter that is naturally present in human

3

skin. Other well documented examples of molecules that undergo efficient E/Z

photoisomerism upon UV excitation include stilbenes13 and retinal.14 OB serves to illustrate

another potential IC route (which could also operate in urocanic acid). OB contains an H

atom donor (a hydroxyl group) and an acceptor (the carbonyl oxygen atom) aligned so as to

encourage intramolecular hydrogen transfer. Such electron-driven H atom transfer processes

on excited state PESs have been identified as key relaxation pathways underpinning the

photostability of many biomolecules.15-19 The present work constitutes a computational study

of the energetics associated with each of these various pathways in CA, FA and OB, the

results of which enables prediction of the most probable IC pathway(s) in each case.

2. COMPUTATIONAL METHODOLOGY

The calculations for CA/FA and for OB used different methodologies and in some cases

sought to address different questions. Key points are summarised below, and further details

are provided as Supporting Information (SI).

i) Caffeic/Ferulic acid

Ground state conformers. The ground (S0) state minimum energy geometries of selected low

lying conformational and E/Z isomers of CA and FA were optimised in two ways. One

involved use of Molpro 2010.1,20 and Møller-Plesset second order perturbation theory (MP2)

coupled with the aug-cc-pVDZ 21 basis set with extra sets of even-tempered s and p diffuse

functions added to the O atoms. No symmetry constraints were used during this optimisation.

The second employed Gaussian 09,22 density functional theory (DFT), the Coulomb

Attenuated Model Becke-3rd parameter-Lee-Yang-Parr (CAM-B3LYP) functional and the

aug-cc-pVDZ basis set.

RO-H/RO-CH3 stretch coordinate: Using Molpro 2010.1, unrelaxed (rigid body) potential

energy functions for the S0 and first few excited singlet states of CA and FA were calculated

using complete active space with second order perturbation (CASPT2) theory, based on a

state-averaged (SA) reference wavefunction (SA5-CASSCF) and assuming CS (i.e. planar)

geometry. The active space for these calculations consisted of twelve electrons in eleven

orbitals (12/11) and comprised the following orbitals: three and three * orbitals, the 2px

orbitals on O(10) and O(12), and the O atom centred 3s Rydberg orbital and the and *

orbitals associated with the O–H/O–CH3 stretch coordinate of interest. An imaginary level

4

shift of 0.5Eh was used in all CASPT2 calculations in this publication to aid convergence and

avoid intruder state problems. These calculations were based on the MP2 optmised geometry

and utilised the aug-cc-pVDZ basis set

Out-of-plane ring deformation and E/Z isomerisation coordinates: The lowest energy CIs

along these two coordinates were optimised at the CASSCF / 6-31G(d) 23 level of theory

using Gaussian 09 and a 6/6 active space involving three and three * orbitals. In each

case, two independent linear interpolations in internal coordinates (LIICs) were then

constructed linking the optimised S0 state geometries of both the E and the Z isomer and the

optimised CI geometry. The energies of the S0 and first few excited singlet states were then

calculated (using MolPro) at various points along each LIIC at the CASPT2/cc-pVDZ level

of theory using a 14/12 (for CA, 12/11 for FA) active space based on a SA4-CASSCF

reference wavefunction. The 12 orbitals used in these calculations for CA were four

occupied and three unoccupied * orbitals, 2px orbitals on O(10) and O(12), and *

orbitals associated with the carbonyl C=O bond and the 2py lone pair on O(21). In the case of

FA, the C=O centred orbital was excluded to aid convergence; subsequent test calculations

for CA using this reduced active space showed its exclusion made negligible difference to the

calculated vertical excitation and CI energies.

ii) Oxybenzone

The ground state minimum energy geometries of selected conformers were optimised using

the CAM-B3LYP and M06-2X functionals embedded within DFT in Gaussian 09, with the

cc-pVDZ basis set. Minimum energy pathways for electron-driven H atom transfer on the S0

and S1 PESs of OB were explored using the O(23)–H(24) stretch as the driving coordinate.

An oxetane-type photoproduct was also identified in these calculations. This was optimised

using the DFT/CAM-B3LYP/cc-pVDZ level of theory, and the variations in the potential

energies of the S1 and S0 states calculated along a LIIC linking this structure to that at which

the S0 and S1 states become near degenerate (RO(23)–H(24) ~2 Å).

Low energy CIs with prefulvenic geometries were optimised in Gaussian 09 using the

CASSCF(10,8)/3-21G level of theory. PECs for the first four singlet excited states along the

LIIC linking the lowest energy CI to the ground state minimum energy geometry were

calculated having first re-optimised the latter at the MP2/cc-pVDZ level of theory. In

Molpro, CASPT2(10,8)/cc-pVDZ was then used for the potential energy calculations. An

5

active space comprising ten electrons in eight orbitals (the four highest the three lowest *

and the O(2py) orbitals) was used in both the CASSCF and CASPT2 calculations.

3. RESULTS and DISCUSSION

3.1 Ground State Conformational Stabilities

Each of the molecules considered in this publication support different conformational and E/Z

isomers, and we here focus just on the sub-set of lower energy isomers depicted in figure 2.

Their respective stabilities were explored with MP2, DFT and, in some cases, CASPT2, and

the (relative) energies so derived are presented in Table 1. The minimum energy isomers of

CA/FA and OB in their respective ground electronic states are identified as structures A and

F in fig. 2, and Cartesian coordinates for the minimum energy geometries of conformers A –

D of both CA and FA are listed in Tables S2 and S3 in the SI.

Figure 2: Selected isomers of caffeic/ferulic acid (A-E) and oxybenzone (F-H), the most

stable of which are, respectively, A and F.

6

Focussing first on CA and FA, we see that rotation about the C(14)=C(16) double bond in the

side chain enables two geometric isomers: E (trans) and Z (cis). The E isomer (A) is more

stable than the Z isomer (B) in both molecules, reflecting the greater steric hindrance in the

latter. Changing the relative orientations of the O-H/O-CH3 moieties attached to the ring (i.e.

rotation about the C(6)-O(12) and/or C(5)-O(10) bonds) gives rise to syn / anti rotational

isomers. As Table 1 shows, the most stable isomers (A and C) are those that benefit from

intramolecular H-bonding between O(10) and H(13). Rotational isomerism can also occur

via rotation around the C(16)C(18) bond. Structure C is calculated to be marginally less

stable in both CA and FA (Table 1). Again, its geometric isomer (D) is significantly less

stable on account of increased steric hindrance. The present results are in good accord with

several previous studies regarding the relative stabilities of the E and Z isomers of CA and

FA,24, 25 though one earlier study of FA 26 (which employed a smaller basis set) placed

conformer C just below A. OB can also exist as two rotational isomers (F and G), the former

of which is favoured by the strong intramolecular hydrogen bond that exists between O(22)

and H(24). Rotation about the C(4)–C(10) bond leads to further (sterically disfavoured)

isomers, the lower energy of which (H) is calculated to be marginally more stable than G.

3.2 Vertical Excitation Energies and Oscillator Strengths

Table 2 lists calculated vertical excitation energies and oscillator strengths of transitions to

the 11*, 11n* and 21* electronic states of different isomers of CA, FA and OB. The

associated orbitals and orbital promotions for CA and OB are shown in figure 3.

7

Figure 3: Orbitals and orbital promotions involved in forming the first three excited singlet

states of E-caffeic acid (left) and oxybenzone (right). Note that the energetic ordering of the

second and third promotions in Z-caffeic acid is reversed (these transitions are labelled in

parentheses, and the corresponding orbitals are displayed in fig. S1). The orbitals and

electronic transitions shown in the left hand column are extremely similar to those for

structures A and B in ferulic acid, which are thus not displayed here but can be found in the

SI (figure S1).

Caffeic/Ferulic Acid

For both molecules, the first three singlet excited states upon vertical excitation are 1*

(two) and 1n* (one) in nature. As in molecules like phenol and catechol,27, 28 the 11* state

is the S1 state for all of the isomers investigated. The participating orbitals (LUMOHOMO

in fig. 3) show good spatial overlap, and the S1–S0 transitions have appreciable oscillator

strengths – as do transitions to the 21* state (see Table 2).29-31 Forming the 11n* state, in

8

contrast, involves electron promotion between orbitals that have little spatial overlap; the

calculated oscillator strength for the 11n*S0 transition is orders of magnitude smaller.

The energetic ordering of these states in the lowest energy E isomer (A) is 11* < 11n* <

21*. The calculated 11* S0 excitation energies are isomer-independent, implying that

the HOMO (and LUMO (orbitals are destabilised by similar amounts upon EZ

isomerisation. Such is not surprising, given that both are predominantly ring centred orbitals.

The 11n* state arises from a LUMOHOMO1 excitation. The HOMO–1 is dominated by

the 2py orbital on O(21) (fig. 3). Table 1 shows that EZ isomerisation reverses the relative

ordering of the 11n* and 21* states in both rotational isomers, that the 11n*S0

excitation energies are similar in A and C, and that the corresponding transition in D is

substantially red-shifted compared with B. These differences can be rationalised by

considering the likely consequences of steric crowding between the COOH moiety and the

ring. In the E isomers (A and C), the distance between these groups is sufficiently large that

the relative orientation of the COOH group is unimportant whereas, in the Z isomers, a

logical starting premise might be that steric interaction between the ring (particularly H(8))

and the COOH group would substantially destabilise the HOMO–1 and thus reduce the

11n*S0 gap. Such expectations are largely borne out in the case of D, but not by B. These

isomers are distinguished by having the C=O group oriented towards (B) or away from (D)

the ring. As fig. S1 (in the SI) shows, in the specific case of isomer B, this interaction

polarises the LUMO–1 (the 2py orbital on O(21)), which increases the strength of the H-bond

between O(21) and H(20) and provides some counteracting stabilisation of the n orbital.

The 21*S0 excitations show obvious red-shifts upon EZ isomerisation. As fig. 3 shows,

these are best viewed as LUMO+1HOMO promotions and, as noted previously, EZ

isomerisation destabilises the HOMO. Thus the calculated energies imply such isomerisation

also induces a (relative) stabilisation of the LUMO+1. These trends can also be attributed to

steric crowding. A plausible starting premise in this case would be that the steric crowding

accompanying EZ isomerisation destabilises the valence orbitals (including the * orbital

of interest in this transition). That being the case, the reduced excitation energies of the Z (cf.

E) isomers must reflect some preferential stabilisation of the LUMO+1, most notably in the

case of isomer D. This can be explained as follows: the terminal COOH group is electron

withdrawing, and thus attracts π density from the adjacent C=C double bond, thereby

destabilising the LUMO+1 (since it is antibonding in this region). So any geometric

9

rearrangement that reduces the electron withdrawing ability of the COOH group (such as

unfavourable electrostatic interaction between the ring π system and the ‘acceptor’ C=O π*

orbital in the case of the Z isomers) will act to stabilise the LUMO+1 and reduce the

21*S0 transition energy (most particularly in the case that the C=O group is directed

towards the ring (i.e. in isomer B)).

Oxybenzone

Table 2 also shows calculated vertical excitation energies to the first three excited singlet

states of our chosen isomers of OB. In all cases, the 11n* state formed by electron

promotion from the O(2py) orbital to the LUMO lies lowest in energy. The calculated

oscillator strength of this S1–S0 transition is fairly low (reflecting the indifferent spatial

overlap of the combining orbitals). This state develops increasing charge transfer (CT)

character upon extending the O(23)H(24) bond, and the n* description progressively

evolves towards n* with the * orbital localised on this bond. All higher excited states of

OB investigated in this work are of * character.

3.3 Possible Internal Conversion Pathways

3.3.1 Caffeic/Ferulic Acid

OH/OCH3 Bond Extension

Much recent interest has focussed on the RO-H stretch coordinate in photoexcited phenolic

molecules.10, 32 Figure 4(a) displays PECs for the S0, 11*, 21* and 11* states of the

most stable isomer of CA along the lowest energy (O(10)H) bond extension coordinate.

Note that the 11n* state was excluded from these calculations in order to aid convergence,

and that the corresponding PECs along RO-H for other isomers of CA are shown in fig. S2 of

the SI. The PEC for the 11* state shows a CI with that of the dissociative 11* state at

short RO-H (~1.2 Å) and, as in phenol,27 catechol28 and related systems,30, 33 it is reasonable to

assume that molecules excited to low vibrational levels of the 11* state of CA could tunnel

through the barrier under this 11*/11* CI.27, 28, 34, 35 The 11* PEC experiences another

CI, with the diabatic S0 PEC, at extended RO-H (~1.8 Å). As noted previously,36 non-adiabatic

coupling at this 11*/S0 CI could provide a route for IC to the ground state PES. Again,

10

analogy with phenol, catechol, etc, 27, 28, 35, 37 suggests that excitation at energies above the

11*/11* CI is likely to result in prompt O(10)–H bond fission.

The lowest energy 1* state in FA, in contrast, arises as a result of a *O(10)–CH3← excitation

and, as fig. 4(b) shows, is dissociative (at long range) along RO–CH3. (Readers are referred to

fig. S3 for the corresponding PECs for other isomers of FA and for PECs along the lowest

energy RO-H stretch coordinate). The asymptotic limit of the 11*O(10)–CH3 PEC in FA lies ~1

eV lower than the corresponding limit for the 11*O(10)–H PEC in CA; the implied difference

in O–H and O–CH3 bond strengths accords well with that found in, for example, 4-

methoxyphenol, for which D0(H3COPhOH) = 3.55 eV 38 cf. D0(H3C–OPhOH) = 2.7 eV.39

The 11*O(10)–CH3 PEC in FA also exhibits CIs with the 11* and S0 states (at RO–CH3~1.7 Å

and 2.5 Å, respectively), but shows clear differences with the 11*O(10)–H PEC in CA. The

obvious minimum at RO–CH3~1.7 Å is attributable to increased Rydberg/valence mixing.31, 40

The resulting barrier in this diabatic PEC, the magnitude of the barrier under the 11*O(10)–

H/11* CI and the mass of the departing CH3 group all encourage the view that tunnelling

along RO–CH3 is an improbable loss mechanism for FA molecules following photoexcitation to

the 11* state. Such a view accords with the results of a recent study of the near UV

photochemistry of guaiacol (2-methoxyphenol), which found no translationally excited CH3

fragments such as would be expected in the event of O–CH3 bond fission on the 11*

potential following excitation to the analogous 11* state.9

11

Figure 4: Potential energy profiles for the S0, 11*, 21* and 11* states of (a) caffeic

acid along RO-H and (b) ferulic acid along ROCH3. The filled black circle shows the CASPT2

energy calculated at the CS symmetry constrained MP2 optimised ground state geometry.

Ring Centred Out-of-Plane Deformations

Another much studied excited state deactivation coordinate in aromatic molecules involves

out-of-plane deformation that leads to CIs with prefulvenic geometries.41 Such CIs were also

identified and optimised in CA and FA, and fig. 5 shows the potential energy profiles of the

S0, 11*, 11n* and 21* states for the LIIC from the MP2 optimised ground state

equilibrium geometry of (a) CA and (b) FA to the corresponding 1*/S0 CI. In both cases,

we identify an energy barrier en route to the CI, but we also note that the barrier height

indicated by a LIIC calculation must be an upper bound to the true value. As fig. 5 also

shows, the 11n* and 21* excited states are both bound in this coordinate and do not

correlate with this low energy CI.

We note that, in both cases, the CASPT2/cc-pVDZ energy difference between the S0 and

11* states at the CASSCF/6-31G(d) optimised geometry of the 1*/S0 CI is ~0.4 eV,

whereas the corresponding energy difference in the CASSCF/6-31G(d) calculation is <0.001

eV. Similar (and larger) energy differences have been noted elsewhere12, 18 and, as in those

cases, are simply a consequence of differences between CASSCF vs. CASPT2, of cc-pVDZ

(spherical harmonics) vs. 6-31G(d) (Cartesian harmonics) and of SA4 vs. SA2.

12

Figure 5: Diabatic potential energy profiles along the out-of-plane deformation coordinate in

(a) caffeic acid and (b) ferulic acid, along with views of the optimised geometry at the

respective 1*/S0 CIs. The filled circles show the CASPT2 energy at the (MP2 optimised)

ground state equilibrium geometry and at each (CASSCF optimised) minimum energy CI.

E/Z Photoisomerism

Figures 6(a) and 6(b) show potential energy profiles for the S0, 11*, 11n* and 21* states

of CA and FA, respectively, along the E/Z photoisomerism coordinate. Two independent

LIICs were constructed, from the most stable E and Z isomers (structures A and B,

respectively) to the lowest energy 1*/S0 CI, which is found at a C3-C14-C16-C18 torsional

angle of ~90°.

13

Figure 6: Potential energy profiles of the S0, 11*, 11n* and 21* states of the most stable

isomer of (a) caffeic acid and (b) ferulic acid calculated at the CASPT2 level. In each case,

two independent LIICs were constructed, from the MP2 optimised geometry of the most

stable rotational conformer of the E and the Z isomer, to the minimum energy 1*/S0 CI.

The filled circles show the CASPT2 energies at the optimised MP2 ground state equilibrium

geometries and at the CASSCF optimised minimum energy CI for the E/Z isomerism. The

inset structures show the calculated minimum energy geometries of the ground state E (left)

and Z (right) isomers and of the 1*/S0 CI.

Figure 7 displays the derivative coupling (h) and gradient difference (g) vectors of this

minimum energy CI in the E/Z isomerism coordinate for CA. The equivalent CI was also

optimised for FA, and the corresponding g and h vectors are shown in fig. S4 of the SI.

Cartesian coordinates for this minimum energy CI in both CA and FA are listed in Table S4,

while Table S5 shows the corresponding coordinates for the minimum energy CI in the

equivalent E/Z isomerism coordinate linking isomers C and D. As in the recent study of

urocanic acid,12 the calculated gradient difference vector in both cases has a large component

associated with E → Z isomerism. The h vector in both CA and FA has a component of C=C

stretch, which is intuitively understandable as the LUMO+1 orbital that correlates with the

E/Z photoisomerism CI has significant * orbital density localised around the C=C bond.

Figure 7: Nuclear displacements associated with the gradient difference (g) and derivative

coupling (h) vectors of the 1*/S0 CI that enables excited state E/Z photoisomerism in

caffeic acid.

14

We now consider the various excited state potentials along the E→Z isomerism path in turn.

Starting from the E isomer, the 11* PEC in both CA and FA rises rapidly, maximises at a

torsional angle ϕ~90°, then declines again to the planar minimum energy geometry of the Z

isomer. The 11n* potential exhibits a similar high energy maximum at ϕ~90°and then

declines again to a (higher energy) minimum associated with the (more sterically hindered) Z

isomer though as fig. 6 shows, it first exhibits a local minimum at ϕ~150°.

The 21* PEC, in contrast, shows a steady decrease from the E isomer to ϕ~90°, where it

forms a CI with the S0 potential. The barrier-less nature of the 21* PEC can be understood

by inspecting the orbitals involved in forming this state; photo-excitation promotes electron

density from an orbital (the HOMO) that contributes bonding density in the region of the

C=C bond into an antibonding * orbital (the LUMO+1) localised on this same C=C bond.

This predicted reduction in bond order is reflected in the lengthening of the C(14)–C(16)

bond (from 1.35 Å at the S0 minimum to 1.48 Å at the 21*/S0 CI). Continuing on, the

21* PEC for both CA and FA displays a small barrier at ϕ~70° en route to the Z isomer.

The present calculations thus suggest that 21*S0 excitation of the majority E isomer of

CA or FA will induce torsional motion around the C(14)–C(16) bond and subsequent IC via

the 21*/S0 CI on an ultrafast timescale. Notwithstanding the caveat that potential barriers

indicated by a LIIC calculation necessarily represent upper bounds to the true barrier height,

it is tempting to suggest that the tilted nature of this CI along the E/Z photoisomerism

coordinate will favour reformation of the E isomer in the S0 state.

Inspecting fig. 6 suggests that this need not be the sole decay pathway for CA and FA

molecules following excitation to the 21* state. Molecules that avoid the 21*/S0 CI

could fully isomerise to the Z conformer on the excited state PEC if the predicted barrier at

ϕ~70° is not prohibitive. Focussing on the E isomer, the 21* PEC also forms CIs with both

the 11* and 11n* states at ϕ~140°, and there is also a minimum energy CI between these

latter two states at ϕ~140°. (All attempts to optimise the latter 11*/11n* CI were

unsuccessful, on account of its extremely shallow gradient.) The 21*/11* and

21*/11n* CIs are noteworthy as they constitute possible routes for IC to lower lying

excited states – somewhat reminiscent of those proposed previously as the first step in the

deactivation of photoexcited DNA/RNA bases to their respective S0 states.17, 42,43

15

3.3.2 Oxybenzone

Like CA and FA, OB shows a low energy CI along the out-of-plane deformation coordinate

(fig. 8). In this case, the 11nπ* state has the lowest vertical excitation energy, and the higher

lying 11ππ* state exhibits the lowest energy CI with the S0 state. As fig. 8 shows, the 11ππ*

PEC also shows a CI with that of the 11nπ* at smaller out-of-plane distortions. OB, like FA,

mequinol, guaiacol, etc has a CH3O group pendant to the benzene ring and, as in these other

molecules, we can anticipate that the 11nπ* and 11ππ* states will also both show a CI with the

first 1π* PEC along RO–CH3, and that this latter PEC will show a CI with the S0 state at longer

RO–CH3. However, analogy with these other molecules also suggests that the probability of O–

CH3 bond fission through the Rydberg/valence barrier at short RO–CH3 is likely to be very

small.

Figure 8: Diabatic potential energy profiles along the out-of-plane deformation coordinate in

oxybenzone, along with a view of the optimised geometry at the 1*/S0 CI. The filled

circles show the CASPT2 energy at the (MP2 optimised) ground state equilibrium geometry

and at the (CASSCF optimised) minimum energy CI.

Electron-Driven Hydrogen Atom Transfer

OB also has the possibility of undergoing photoinduced H atom transfer. As fig. 9 illustrates,

this mechanism requires that, upon photoexcitation of OB (structure F, an enol when viewed

as a decorated phenol), H(24) migrates from O(23) to O(22), thereby forming the keto-

tautomer F'. Similar photoinduced H atom transfer processes have been proposed to account

16

for the ultrafast deactivation of 3-hydroxy-picolinic acid and salicylic acid following UV

excitation.44, 45

Figure 9: Proposed UV photoinduced excited state hydrogen transfer from oxybenzone (F)

to its keto-tautomer (F').

Potential energy profiles for the S0, 11n*, 11* and 21* states of OB along the O(23)–

H(24) bond extension coordinate, calculated with DFT using the CAM-B3LYP functional,

are presented in fig. 10. Equivalent PECs calculated with the M06-2X functional are shown

in the SI (fig. S5). In each set of calculations, the S0 state was initially scanned along RO(23)–

H(24) (henceforth simply RO–H), fixing the OH bond length at selected values and, at each step,

allowing all other internal degrees of freedom to relax to their minimum energy

configuration. The S0 energies so derived are shown as filled black circles in fig. 10; while

the open coloured points show the respective excited state energies calculated at the various

relaxed S0 geometries. This process was then repeated for the S1 state (filled red circles in fig.

10), after which the S0 energies at the various relaxed S1 geometries were also computed

(shown as open black circles).

With all levels of theory employed here, the S0 potential energy rises with increasing RO–H

while the PEC for the 11n* state declines steadily and becomes near degenerate with the S0

PEC at ROH ~2 Å. The S1 state is appropriately described as 11n* in the vertical excitation

region, but acquires ever greater charge transfer (CT) character at longer RO–H. This is

evident from inspecting the evolution of the participating orbitals (illustrated in fig. 10(b))

and of the dipole moment vector of OB molecules at the S0 state equilibrium geometry and in

the S1 excited state at ROH = 1 Å ( = 4.8 D) and at the optimised geometry at ROH = 2 Å ( =

8.6 D) – shown in fig. S6 in the SI. Analogous CT states have also been identified in DNA

base residues, DNA base pairs and kynurenines.16-18 Based on the PECs displayed in fig. 10,

therefore, it appears that electron driven hydrogen transfer in the S1 state of OB is a very

plausible ultrafast relaxation pathway back to the S0 state – in much the same way as has been

17

identified for several other photoexcited biomolecules (e.g. adenosine and the adenine-

thymine and guanine-cytosine base pairs).15-17

Figure 10: (a) Potential energy profiles along ROH for the S0 and S1 states of oxybenzone

calculated at the CAM-B3LYP level of theory (together with more restricted scans for the S2

and S3 states). The use of filled and open symbols is explained in the accompanying text. (b)

Pictorial representation of the HOMO and LUMO at the optimised S1 geometry when ROH =

1 Å (b1) and 2 Å (b2) to illustrate the CT character of the S1 state upon extending the O(23)–

H(24) bond.

Oxetane formation

In addition to IC to reform OB(S0) molecules (structure F), the present calculations identify

another possible diabatic product arising from electron driven H atom transfer. The rationale

for this alternative product is evident from inspecting the evolving minimum energy

geometry of OB(S1) along the electron-driven hydrogen transfer coordinate. As fig. 11

shows, elongation of the O(23)–H(24) bond is accompanied by a progressive twisting of the

molecular frame about the C(4)–C(10) bond, such that by ROH = 2 Å the C5-C4-C10-O22

dihedral angle is ~70°. Similar behaviour along the analogous excited state hydrogen transfer

coordinate has been noted previously in the case of salicylic acid 44 and, as in that case, the

driving force for the distortion can be traced to the 2+2 cycloaddition reaction (yielding a

18

stable ground state oxetane) that is accessible to the keto form of OB once RO-H ~ 2 Å. As fig.

11(b) shows, the lowest energy PEC along the LIIC linking the S1 optimised geometry at RO-H

= 2.0 Å and the ground state oxetane is barrierless.

Figure 11: (a) Reprise of the calculated S1 and S0 PECs of oxybenzone as a function of RO–H,

together with (b) the corresponding PECs along the LIIC connecting the minimum energy

geometries of the oxetane formed by intramolecular 2+2 cycloaddition and the optimised

geometry at RO-H = 2.0 Å. The structures below provide two different views of the

progressive twisting of the molecular geometry at four points along the electron driven

hydrogen transfer coordinate

4 GENERAL DISCUSSION AND CONCLUSIONS

The primary aim of this study was to characterise the most probable nuclear motions that

drive IC in two classes of molecule that are (or could be) used in commercial sunscreens. Ab

19

initio electronic structure calculations are arguably the current method of choice for providing

detailed insight into such radiationless transitions in molecules such as those selected for

study – oxybenzene, and caffeic and ferulic acids – though we appreciate that the present

study makes no allowance for possible intersystem crossings to triplet excited states. We also

recognise that the present calculations relate to the isolated molecules, and are thus blind to

additional complexities (e.g. solvent interactions, possible photocatalysis by virtue of the

proximal metal oxide nanoparticles, bimolecular reactions with molecules in the air

(principally O2) or in the cells of the skin they are designed to protect) that would form part

of any forensic mechanistic study of sunscreen formulations.

Nonetheless, the current findings are revealing. As noted in the introduction, the primary role

of the conjugated organic molecules in a sunscreen is to absorb solar radiation in the UV-A

and UV-B regions, and to convert that electronic excitation into vibrational energy (which is

then dissipated by heating the surrounding matrix) without detriment to itself. Thus the

molecule is required to provide photoprotection and to be photostable.

We first consider oxybenzone. The calculated vertical excitation energies of the optically

‘bright’ 11* and 21* states accord sensibly with the experimental UV absorption

spectrum (which spans the requisite wavelength region and shows absorption maxima at

~320 and ~285 nm (~3.75 and 4.35 eV)).46 Such an assignment would imply that the much

weaker 11n*S0 transition would contribute to the long wavelength tail of the spectrum. An

early report suggesting that OB was very susceptible to photooxidation 47 was subsequently

refuted.48 Controlled studies employing commercial UV-A and UV-B sources and

environmentally relevant irradiances have confirmed the impressive photostability of this

molecule,46 though other health concerns (notably its photoallergic potential and its potential

as an endocrine disruptor) continue to generate debate.5 The present calculations which

reveal a barrierless (and thus potentially ultrafast) electron driven H atom transfer pathway

from the S1(11n*) state to a CI with the ground state PES provide a clear explanation for the

impressive photostability of OB. What the current calculations cannot assess is the product

branching once molecules have accessed the 11n*/S0 CI (i.e. the relative probabilities of

reforming the starting enol isomer (F) vs the (less stable) keto isomer (F), or of forming the

oxetane product identified in fig. 11). Given its demonstrable photostability, we conclude

that almost all of the evolving population must be funnelled towards the S0(enol) structure,

which then vibrationally cools by energy transfer to the bulk. Given the (high) probability of

(as yet unidentified) CIs between the higher lying, strongly absorbing 1* state PESs and

20

that of the S1 state, it is likely that this electron driven H atom transfer process is the dominant

IC route for OB molecules at all excitation wavelengths relevant to sunscreen function. Thus

OB adds to the ever-growing list of excited state molecules whose IC is driven by

intramolecular electron driven hydrogen atom transfer.12, 16-19, 49

The mechanism(s) by which caffeic and ferulic acid provide photoprotection, and their

photostability, are less clear cut. The reported UV absorption spectra span the wavelength

range required for sunscreen applications, but the observed absorption maxima (which, in CA

for example, appear at 312 and 287 nm (3.97 and 4.35 eV)) 47 are well below the calculated

vertical excitation energies (Table 2). As noted previously, the probability of direct

photoexcitation to the 11n* (or the 11*) states in CA and FA is small, but we can

anticipate many CIs between the PESs for these states and those of the optically bright 11*

and 21* states (some of which are identified in the present work) that can promote facile

transfer of population between these various states.

The present calculations suggest that CA/FA molecules in low vibrational levels of the

respective first excited singlet (11*) states should be relatively stable with respect to all of

the distortions investigated (i.e. O–H/Me stretching, out-of-plane deformation and E/Z

photoisomerism). Such predictions accord with experimental reports of fluorescence

following near UV excitation of both molecules;52 indeed, FA has been suggested as a major

contributor to the blue-green fluorescence observed during 337 nm excitation of the leaves of

green plants.53 Photodegradation was clearly observed during prolonged irradiation of CA (in

deionised water) at 31210 nm, however, only part of which could be attributed to O2-

assisted photooxidation. The identification of esculetin amongst the products was taken as

evidence for EZ photoisomerism followed by cyclization.50

This finding reinforces the earlier conclusion (from comparison with the absorption data) that

the various excited state energies are overestimated by the present calculations which, for

both CA and FA, locate the minimum energy 11*/S0 CI in the ring centred out-of-plane

deformation coordinate at ~4.7 eV and the minimum energy 21*/S0 CI in the E/Z

isomerism coordinate at ~4.9 eV (cf. the ~4.4 eV energy of the shortest wavelength UV

photons relevant when designing a commercial sunscreen). As noted above, however, the

present calculations are for isolated gas phase and it is reasonable to expect that the energies

of the 1* and 1n* states and of */S0 CI to red-shift upon solvation. For the purpose of

this discussion we assume that the calculations systematically overestimate the energies of

21

both 1* potentials. CA (and FA) would thus be predicted to offer photoprotection and

demonstrate photostability as follows: Energy absorbed when exciting to the lowest levels of

the S1 state (i.e. levels that would be populated by absorbing the longest wavelength photons)

is dissipated by re-emission. Scanning to shorter wavelengths, the calculations suggest that

the S1(1*) state lifetime will decrease rapidly as IC, driven by out of plane distortion to the

lowest energy 11*/S0 CI and, at somewhat higher energies, by E/Z isomerisation via the

lowest energy 21*/S0 CI, offer progressively faster decay routes. The slopes of both CIs

(figs. 5 and 6) suggest that both will be efficient at funnelling population back to the starting

E isomer.

Recalling fig. 4(a) and by analogy with recent studies of catechol photolysis following

excitation in the region of its S1 origin,28, 35 we might also anticipate some contribution from

O(10)–H bond fission following excitation to the S1 state of CA, by tunnelling under the

11*/11* CI in the RO–H stretch coordinate. The timescale for this process in catechol is

~10 ps.35 Whether this would constitute an IC route (and thus offer any photoprotection) is a

moot point, as the prior studies of catechol photolysis (in both the gas and condensed phase)

show clear evidence for bond fission, i.e. radical formation, which would generally be

deemed undesirable in a sunscreen. If this process is at all competitive when exciting CA in

the UV-A/UV-B regions, the present study points to the benefits of using FA (with a heavy

CH3 group on O(10)) in place of CA in sunscreen formulations.

The present work adds to the still small but fast growing literature reporting detailed

theoretical 12, 18 and experimental 54 studies of (often subtle) dynamical aspects of the excited

state photochemistry and photophysics that help determine whether a given molecular

building block affords photostability and photoprotection in response to UV excitation, or

photodegrades or exhibits phototoxicity. We can anticipate many more such studies in the

next few years.

ASSOCIATED CONTENT

Supporting Information

Fuller details of the computational methodology used; molecular orbitals involved in the

various electronic transitions of Z-caffeic acid and of both E- and Z- ferulic acid; PECs (along

RO–H) for the first five singlet states of syn- and anti-conformers of E-caffeic acid; PECs

(along RO–CH3 and/or RO–H) for selected conformers of ferulic acid; Cartesian coordinates for

22

the ground state minimum energy geometries of isomers A to D of both CA and FA and for

the optimised 1ππ*/S0 CI along the E/Z isomerism coordinates linking conformers A and B,

and C and D, in both CA and FA; g and h vectors associated with the minimum energy CI in

the E/Z isomerism coordinate for ferulic acid; PECs along the electron driven H atom transfer

coordinate for oxybenzone calculated at the DFT/M06-2X/cc-pVDZ level of theory; dipole

moment vectors for oxybenzone at its minimum energy geometry in the S0 and S1 states and

at the optimised geometry of the S1/S0 CI (ROH = 2 Å). This material is available free of

charge via the internet at http://pubs.acs.org.

AUTHOR INFORMATION

Corresponding authors

E-mail:

tolga.karsili@bristol.ac.uk

mike.ashfold@bris.ac.uk

NOTES

The authors declare no competing financial interests.

ACKNOWLEDGEMENTS

The Bristol authors are grateful to the EPSRC (Programme Grant EP/L005913) for funding

and to Prof. J.N. Harvey and Dr. G.M. Roberts for helpful discussions.

23

Table 1: Ground state energies (in cm-1) of the isomers of caffeic/ferulic acid and

oxybenzone depicted in fig. 2, referenced to that of the most stable conformers (structures A

and F, respectively). The CASPT2 calculations were based on the MP2/aug-cc-pVDZ

optimised geometry.

Structure MP2 CAM-B3LYP CASPT2

Caffeic Acid

A 0 0 0

B 1650 1770 1690

C 140 220 230

D 2740 2930 3030

E - 1450 -

Ferulic Acid

A 0 0 0

B 1690 1790 1630

C 140 230 -

D 2760 2940 -

E - 1600 -

Oxybenzone

F - 0 -

G - 4270 -

H - 3720 -

24

Table 2: Vertical excitation energies (in eV) of first three excited states of the conformational

and E/Z isomers of caffeic/ferulic acid and of oxybenzone calculated using the different

levels of theory indicated in the footnotes, with the respective oscillator strengths from the S0

state in parenthesis calculated at the EOM-CCSD level. The CASPT2 energies are calculated

at the MP2/aug-cc-pVDZ optimised ground state geometry for each isomer.

Caffeic Acid

StateStructure

A B C D

11* 4.19* (0.0924†) 4.15* (0.3294†) 4.19* (0.1635†) 4.12* (0.2308†)11n* 4.81* (0.0003†) 4.99* (0.0008†) 4.91* (0.0002†) 4.59* (0.0001†)21* 4.97* (0.4566†) 4.66* (0.2344†) 4.92* (0.5124†) 4.33* (0.4088†)

Ferulic Acid

StateStructure

A B C D11* 4.19* (0.2746†) 4.20* (0.3831†) 4.20* (0.2966†) 4.21* (0.0687†)11n* 4.80* (0.0003†) 4.82* (0.0001†) 4.78* (0.0001†) 4.70* (0.0001†)21* 4.90* (0.3816†) 4.80* (0.1727†) 4.88* (0.3288†) 4.43* (0.4848†)

Oxybenzone

StateStructure

F G H11n* 3.36*

4.19‡ (0.0195)4.00± (0.0016)

3.85‡ (0.0015) 3.89‡ (0.0044)

11* 3.72*4.26‡ (0.1858)4.32± (0.2084)

4.72‡ (0.1939) 4.69‡ (0.2081)

21* 4.30*4.82‡ (0.2791)4.87± (0.2525)

5.12‡ (0.0512) 5.11‡ (0.0369)

* CASPT2/cc-pVDZ † EOM-CCSD/cc-pVDZ ‡ TD-DFT/CAM-B3LYP/cc-pVDZ± TD-DFT/M06-2X/cc-pVDZ

References

25

1. Jablonski, A., Efficiency of Anti-Stokes Fluorescence in Dyes. Nature 1933, 131, 839.

2. Yarkony, D. R., Conical Intersections:  The New Conventional Wisdom. J.Phys. Chem. A 2001, 105, 6277.

3. Domcke, W.; Yarkony, D. R.; Koppel, H., Conical Intersections: Theory, Computation and Experiment. Wiley 2011, 17.

4. Kanavy, H. E.; Gerstenblith, M. R., Ultraviolet radiation and melanoma. Seminars in cutaneous medicine and surgery 2011, 30 , 222.

5. Burnett, M. E.; Wang, S. Q., Current sunscreen controversies: a critical review. Photodermatol. photo. 2011, 27, 58.

6. Tan, E. M. M.; Hilbers, M.; Buma, W. J., Excited-State Dynamics of Isolated and Microsolvated Cinnamate-Based UV-B Sunscreens. J. Phys. Chem. Lett. 2014, 5, 2464.

7. Kockler, J.; Oelgemöller, M.; Robertson, S.; Glass, B. D., Photostability of sunscreens. J. Photochem. Photobiol. C: Photochem. Revs. 2012, 13, 91.

8. Boerjan, W.; Ralph, J.; Baucher, M., Lignin biosynthesis. Annual review of plant biology 2003, 54, 519.

9. Young, J. D.; Staniforth, M.; Dean, J. C.; Roberts, G. M.; Mazzoni, F.; Karsili, T. N. V.; Ashfold, M. N. R.; Zwier, T. S.; Stavros, V. G., Towards Understanding Photodegradation Pathways in Lignins: The Role of Intramolecular Hydrogen Bonding in Excited States. J. Phys. Chem. Lett. 2014, 5, 2138.

10. Vieuxmaire, O. P. J.; Lan, Z.; Sobolewski, A. L.; Domcke, W., Ab initio characterization of the conical intersections involved in the photochemistry of phenol. J. Chem. Phys. 2008, 129, 224307.

11. Tseng, C.-M.; Lee, Y. T.; Lin, M.-F.; Ni, C.-K.; Liu, S.-Y.; Lee, Y.-P.; Xu, Z. F.; Lin, M. C., Photodissociation Dynamics of Phenol. J. Phys. Chem. A 2007, 111, 9463.

12. Tuna, D.; Sobolewski, A. L.; Domcke, W., Photochemical Mechanisms of Radiationless Deactivation Processes in Urocanic Acid. J. Phys. Chem. B 2014, 118, 976

13. Quenneville, J.; Martínez, T. J., Ab Initio Study of Cis−Trans Photoisomerization in Stilbene and Ethylene. J. Phys. Chem. A 2003, 107, 829.

14. Abe, M.; Ohtsuki, Y.; Fujimura, Y.; Domcke, W., Optimal control of ultrafast cis-trans photoisomerization of retinal in rhodopsin via a conical intersection. The J. Chem. Phys. 2005, 123, 144508.

15. Tuna, D.; Sobolewski, A. L.; Domcke, W., Mechanisms of Ultrafast Excited-State Deactivation in Adenosine. J. Phys. Chem. A 2013, 118, 122.

16. Perun, S.; Sobolewski, A. L.; Domcke, W., Role of Electron-Driven Proton-Transfer Processes in the Excited-State Deactivation of the Adenine−Thymine Base Pair. The J. Phys. Chem. A 2006, 110, 9031.

17. Sobolewski, A. L.; Domcke, W.; Hättig, C., Tautomeric selectivity of the excited-state lifetime of guanine/cytosine base pairs: The role of electron-driven proton-transfer processes. Proc. Nat. Acad. Sci. 2005, 102, 17903.

18. Tuna, D.; Došlić, N.; Mališ, M.; Sobolewski, A. L.; Domcke, W., How Kynurenines Protect the Retina from Sunburn: A Joint Electronic-Structure and Dynamics Study. J. Phys. Chem. B 2014, Submitted.

19. Sobolewski, A. L.; Domcke, W., Computational studies of the photophysics of hydrogen-bonded molecular systems. J. Phys. Chem. A 2007, 111, 11725.

20. Werner, H. J.; Knowles, P. J.; Knizia, G.; Manby, F. R.; Schütz, M.; Celani, P.; Korona, T.; Lindh, R.; Mitrushenkov, A.; Rauhut, G.; et al., MOLPRO, University of Cardiff: Cardiff, U.K, 2010.

26

21. Dunning, T. H., Jr., J. Chem. Phys. 1989, 90, 1007.22. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.;

Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennuchi, B.; Petersson, G.; et al., Gaussian 09, revision B.01; Gaussian Inc.: Wallingford, CT, 2010.

23. Hehre, W. J.; R.F., S.; Pople, J. A., J. Chem. Phys. 1969, 2657 24. Świsłocka, R., Spectroscopic (FT-IR, FT-Raman, UV absorption, 1H and 13C NMR)

and theoretical (in B3LYP/6-311++G** level) studies on alkali metal salts of caffeic acid. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 2013, 100, 21.

25. Urbaniak, A.; Szeląg, M.; Molski, M., Theoretical investigation of stereochemistry and solvent influence on antioxidant activity of ferulic acid. Comp. Theor. Chem. 2013, 1012, 33.

26. Martens, S. M.; Marta, R. A.; Martens, J. K.; McMahon, T. B., Consecutive fragmentation mechanisms of protonated ferulic acid probed by infrared multiple photon dissociation spectroscopy and electronic structure calculations. J. Am. Soc. Mass. Spectrom. 2012, 23, 1697

27. Nix, M. G. D.; Devine, A. L.; Cronin, B.; Dixon, R. N.; Ashfold, M. N. R., High resolution photofragment translational spectroscopy studies of the near ultraviolet photolysis of phenol. J. Chem. Phys. 2006, 125, 133318.

28. King, G. A.; Oliver, T. A. A.; Dixon, R. N.; Ashfold, M. N. R., Vibrational energy redistribution in catechol during ultraviolet photolysis. Phys. Chem. Chem. Phys. 2012, 14, 3338.

29. Lim, J. S.; Kim, S. K., Experimental probing of conical intersection dynamics in the photodissociation of thioanisole. Nat. Chem. 2010, 2, 627.

30. Devine, A. L.; Nix, M. G. D.; Dixon, R. N.; Ashfold, M. N. R., Near-ultraviolet photodissociation of thiophenol. J. Phys. Chem. A 2008, 112, 9563.

31. Ashfold, M. N. R.; King, G. A.; Murdock, D.; Nix, M. G. D.; Oliver, T. A. A.; Sage, A. G., * states in molecular photocheistry. Phys. Chem. Chem. Phys. 2010, 12, 1218.

32. Vallet, V.; Lan, Z. G.; Mahapatra, S.; Sobolewski, A. L.; Domcke, W., Photochemistry of pyrrole: time-dependent quantum wave-packet description of the dynamics at the 1*-S0 conical intersections J. Chem. Phys. 2005, 123, 144307.

33. Sage, A. G.; Nix, M. G. D.; Ashfold, M. N. R., UV photodissociation of N-methylpyrrole: The role of 1* states in non-hydride heteroaromatic systems. Chem. Phys. 2008, 347, 300.

34. Pino, G. A.; Oldani, A. N.; Marceca, E.; Fujii, M.; Ishiuchi, S. I.; Miyazaki, M.; Broquier, M.; Dedonder, C.; Jouvet, C., Excited state hydrogen transfer dynamics in substituted phenols and their complexes with ammonia: 1*-1* energy gap propensity and ortho-substitution effect. J. Chem. Phys. 2010, 133, 124313.

35. Chatterley, A. S.; Young, J. D.; Townsend, D.; Zurek, J. M.; Paterson, M. J.; Roberts, G. M.; Stavros, V. G., Manipulating dynamics with chemical structure: probing vibrationally-enhanced tunnelling in photoexcited catechol. Phys. Chem. Chem, Phys. 2013, 15, 6879.

36. Sobolewski, A. L.; Domcke, W.; Dedonder-Lardeux, C.; Jouvet, C., Excited-state hydrogen detachment and hydrogen transfer driven by repulsive 1* states: A new paradigm for nonradiative decay in aromatic biomolecules. Phys. Chem. Chem. Phys. 2002, 4, 1093.

37. Iqbal, A.; Pegg, L.-J.; Stavros, V. G., Direct versus indirect H atom elimination from photoexcited phenol molecules. J. Phys. Chem. A 2008, 112, 9531.

27

38. Hadden, D. J.; Roberts, G. M.; Karsili, T. N. V.; Ashfold, M. N. R.; Stavros, V. G., Competing 1* mediated dynamics in mequinol: O-H versus O-CH3

photodissociation pathways. Phys. Chem. Chem. Phys. 2012, 14, 13415.39. M. M. Suryan; Karafi, S. A.; Stein, S. E., J. Am. Chem. Soc. 1989, 111, 1423.40. Reisler, H.; Krylov, A. I., Interacting Rydberg and valence states in radicals and

molecules: experimental and theoretical studies. Int. Revs. Phys. Chem. 2009, 28, 267.41. Palmer, I. J.; Ragazos, I. N.; Bernardi, F.; Olivucci, M.; Robb, M. A., An MC-SCF

study of the S1 and S2 photochemical reactions of benzene. J. Am. Chem. Soc. 1993, 115, 673.

42. Perun, S.; Sobolewski, A. L.; Domcke, W., Photostability of 9H-adenine: mechanisms of the radiationless deactivation of the lowest excited singlet states. Chem. Phys. 2005, 313, 107.

43. Barbatti, M.; Aquino, A. J. A.; Szymczak, J. J.; Nachtigallová, D.; Hobza, P.; Lischka, H., Relaxation mechanisms of UV-photoexcited DNA and RNA nucleobases. Proc. Nat. Acad. Sci. 2010, 107, 21453.

44. Sobolewski, A. L.; Domcke, W., Photophysics of intramolecularly hydrogen-bonded aromatic systems: ab initio exploration of the excited-state deactivation mechanisms of salicylic acid. Phys.Chem. Chem. Phys. 2006, 8, 3410.

45. Rode, M. F.; Sobolewski, A. L., Ab initio study on the excited state proton transfer mediated photophysics of 3-hydroxy-picolinic acid. Chem. Phys. 2012, 409, 41.

46. Tarras-Wahlberg, N.; Stenhagen, G.; Larko, O.; Rosen, A.; Wennberg, A. M.; Wennerstrom, O., Changes in ultraviolet absorption of sunscreens after ultraviolet irradiation. J. Invest. Dermatol. 1999, 113, 547.

47. Schallreuter, K. U.; Wood, J. M.; Farwell, D. W.; Moore, J.; Edwards, H. G., Oxybenzone oxidation following solar irradiation of skin: photoprotection versus antioxidant inactivation. J. Invest. Dermatol. 1996, 106, 583.

48. Santoro, E., On photo-stability of oxybenzone. J. Invest. Dermatol. 1998, 110, 95.49. Ding, L.; Chen, X.; Fang, W.-H., Ultrafast Asynchronous Concerted Excited-State

Intramolecular Proton Transfer and Photodecarboxylation of o-Acetylphenylacetic Acid Explored by Combined CASPT2 and CASSCF Studies. Organic Lett. 2009, 11, 1495.

50. Le Person, A.; Lacoste, A.-S.; Cornard, J.-P., Photo-degradation of trans-caffeic acid in aqueous solution and influence of complexation by metal ions. J. Photochem. Photobiol. A: Chem. 2013, 265, 10.

51. Belay, A., Spectrophotometric Method for the Determination of Caffeic Acid Complexation and Thermodynamic Properties Int. J. Biophys. 2012, 2, 12.

52. Li, S.; Huang, K.; Zhong, M.; Guo, J.; Wang, W. Z.; Zhu, R., Comparative studies on the interaction of caffeic acid, chlorogenic acid and ferulic acid with bovine serum albumin. Spectrochimica acta. Part A, Molecular and biomolecular spectroscopy 2010, 77, 680.

53. Lichtenthaler, H. K.; Schweiger, J., Cell Wall Bound Ferulic Acid, the Major Substance of the Blue-green Fluorescence Emission of Plants. J. Plant. Physiol. 1997, 152, 272.

54. El Nahhas, A.; Pascher, T.; Leone, L.; Panzella, L.; Napolitano, A.; Sundström, V., Photochemistry of Pheomelanin Building Blocks and Model Chromophores: Excited-State Intra- and Intermolecular Proton Transfer. J. Phys. Chem. Lett. 2014, 5, 2094.

28