The effect of deuteration on the keto-enol equilibrium and photostability of the sunscreen agent avobenzone

Article abstract of DOI:10.1039/d0pp00265h

The remarkable properties of deuterium have led to many exciting and favourable results in enhancing material properties, for applications in the physical, medical, and biological sciences. Deuterated isotopologues of avobenzone, a sunscreen active ingredient, were synthesised to examine for any changes to the equilibrium between the diketone and enol isomers, as well as their UV photostability and photoprotective properties. Prior to UV irradiation, deuteration of the diketone methylene/enol moiety (i.e. avobenzone-d2) led to an increase in the % diketone compared to non-deuterated, determined by 1H NMR experiments in CDCl3 and C6D12. This can be rationalised from two angles; mechanistically by a deuterium kinetic isotope effect for the CH vs. CD abstraction step during tautomerisation from the diketone to the enol, and a weaker chelating hydrogen bond for the enol when deuterated allowing increased equilibration to the diketone. Avobenzone-d2 was further examined by solid state 13C NMR. The higher % diketone for avobenzone-d2 was postulated to favour increased photodegradation by a non-reversible pathway. This was investigated by UV irradiation of the avobenzone isotopologues in C6D12, both in real time in situ within the NMR by fibre optic cable as well as ex situ using sunlight. An increase in the relative amount of photoproducts for avobenzone-d2 compared to non-deuterated was observed by 1H NMR upon UV irradiation ex situ. Overall, the study demonstrates that deuteration can be applied to alter complex equilibria, and has potential to be manifested as changes to the properties and behaviour of materials.

Full text of DOI:10.1039/d0pp00265h

Photochemical &
Photobiological Sciences
View Article Online
View Journal
PAPER
The effect of deuteration on the keto–enol
equilibrium and photostability of the sunscreen
Cite this: DOI: 10.1039/d0pp00265h
agent avobenzone†
a
b
c
a
Rhys B. Murphy,
John Staton, Aditya Rawal
and Tamim A. Darwish
*
The remarkable properties of deuterium have led to many exciting and favourable results in enhancing
material properties, for applications in the physical, medical, and biological sciences. Deuterated isotopo-
logues of avobenzone, a sunscreen active ingredient, were synthesised to examine for any changes to the
equilibrium between the diketone and enol isomers, as well as their UV photostability and photoprotective
properties. Prior to UV irradiation, deuteration of the diketone methylene/enol moiety (i.e. avobenzone-
1
d
2
) led to an increase in the % diketone compared to non-deuterated, determined by H NMR experi-
3 6
ments in CDCl and C D12. This can be rationalised from two angles; mechanistically by a deuterium
kinetic isotope effect for the CH vs. CD abstraction step during tautomerisation from the diketone to the
enol, and a weaker chelating hydrogen bond for the enol when deuterated allowing increased equili-
13
bration to the diketone. Avobenzone-d
ketone for avobenzone-d was postulated to favour increased photodegradation by a non-reversible
pathway. This was investigated by UV irradiation of the avobenzone isotopologues in C 12, both in real
2
was further examined by solid state C NMR. The higher % di-
2
6
D
time in situ within the NMR by fibre optic cable as well as ex situ using sunlight. An increase in the relative
1
Received 27th May 2020,
Accepted 14th September 2020
amount of photoproducts for avobenzone-d2 compared to non-deuterated was observed by H NMR
upon UV irradiation ex situ. Overall, the study demonstrates that deuteration can be applied to alter
complex equilibria, and has potential to be manifested as changes to the properties and behaviour of
materials.
DOI: 10.1039/d0pp00265h
rsc.li/pps
Sunscreens are typically comprised of a combination of
reflecting or absorbing agents, including physical mineral par-
ticulates such as titanium dioxide and zinc oxide, as well as
Introduction
Avobenzone – a chemical sunscreen agent
Skin cancers arising from exposure to UV radiation from the combinations of chemicals which absorb UVA (400–320 nm)
sun are highly prevalent in society. Internationally, 2–3 million and UVB (320–290 nm) to provide broad spectrum protection.
non-melanoma diagnoses are reported annually (130 000 cases Avobenzone (4-tert-butyl-4′-methoxydibenzoylmethane) is
a
1
of melanoma), and the rate is considered to be increasing.
common chemical UVA filter approved for use in sunscreens
Australians are significantly over-represented in the diagnosis and cosmetic formulations globally. It exists as two tautomers;
of sun-related skin cancers, with two in three people diag- an enol (major) and a diketone (minor) (Fig. 1). As avobenzone
nosed by age 70, and is the third most common cancer in is asymmetric, there are two enols, existing in the solid state
2
3,4
Australia. The application of sunscreen to skin prior to sun and in solution. Each of the enols also has a chelated E-
exposure significantly reduces damage caused by recreational (cis-) isomer with an intramolecular hydrogen bond, and a
UV exposure.
non-chelated Z- (trans-) isomer (not shown), which arises from
5
–7
both bond rotation and photoisomerisation.
A methylated
analogue has been synthesised to increase the amount of di-
8
,9
ketone to better comprehend its role (not shown). A large
number of avobenzone derivatives have been synthesised and
evaluated for their efficacy as sunscreen agents, but are not
reviewed in this work.
The photostability of sunscreens has been reviewed,
including that of avobenzone. Avobenzone exhibits photoinst-
ability by both photoisomerisation and photodegradation
a
Australian Nuclear Science and Technology Organisation, Locked Bag 2001,
Kirrawee DC, NSW, 2232, Australia. E-mail: tamim.darwish@ansto.gov.au
b
Eurofins Dermatest, 20 King Street, Rockdale, New South Wales, 2216, Australia
c
Nuclear Magnetic Resonance Facility, Mark Wainwright Analytical Centre,
10–12
University of New South Wales, Sydney, Australia
†
Electronic supplementary information (ESI) available: NMR, MS, IR, ISO 24443
testing data, equilibration graphs. See DOI: 10.1039/d0pp00265h
This journal is © The Royal Society of Chemistry and Owner Societies 2020
Photochem. Photobiol. Sci.

View Article Online
Paper
Photochemical & Photobiological Sciences
Each has drawbacks, in practicality, or in the complexity of
multi-component formulations. For example, chemical addi-
tives which quench the triplet excited state of avobenzone can
reduce self-degradation as well as the degradation of other
components in the formulation ordinarily photosensitised by
5
,7,9,10,22,23
avobenzone.‡
Avobenzone has been shown to
photosensitise oxygen to singlet oxygen (e.g. oxygen quenches
the diketone triplet excited state to produce singlet
7
,14,24
oxygen),
however, singlet oxygen can subsequently
2
3
degrade avobenzone. De-oxygenation has been reported to
improve stability in select solvents, and the addition of antiox-
5
idants which quench singlet oxygen improve avobenzone
9
,23
photostability.
Interestingly however, when the supply of
oxygen is continuous, the degradation of avobenzone ana-
logues is reduced, which the authors relate to triplet quench-
1
4
ing by oxygen. Furthermore, antioxidants reduce the inter-
action of avobenzone and cells with reactive oxygen species
(
ROS) which can arise from combination sunscreen formu-
lations containing metal oxides, with further reductions by
coating or doping the metal oxide (ref. 10 and references
within). Finally, avobenzone stability can be improved by
encapsulation
microparticles
within
or as cyclodextrin
micelles/lipospheres/solid
lipid
2
5–27
28,29
30
and metal com-
Fig. 1 Avobenzone is in equilibrium between the diketone and enol.
plexes. Avobenzone has itself provided partial photoprotection
to topical photosensitive drugs such as diclofenac, as a photo-
labile conjugate.
3
1
mechanisms upon irradiation with UV, studied experimentally
Deuteration to alter material properties
5
–7,9,10,13–15
under different conditions,
with recent progress
The exchange of hydrogen with deuterium can lead to marked
differences in the physical and chemical properties of com-
pounds. Long predicted pharmaceutical applications are
involving improved understanding of photochemistry, photo-
physics and hydrogen bonding, both experimentally and
1
6–20
computationally.
The multiple degradation pathways of
3
2–36
gaining momentum,
including approval of the first deute-
avobenzone and ability of species to interconvert leads to a
complex equilibria of species, which can reduce sunscreen
efficacy and quality.
3
7
rated drug (US FDA), clinical trials of several deuterated drug
3
8–42
candidates,
and deuterium-stabilised lipid drugs incorpor-
4
3
ated into the human diet to treat disease. For covalently
deuterated compounds, this can decrease the rate of enzymatic
metabolism at vulnerable sites due to the kinetic isotope
effect, or can lead to a switch in the metabolic pathway. In
relation to this, we and others have incorporated deuterium
Photoisomerisation of the chelated E-enol leads to loss of
absorbance by various mechanisms which have been
5
–7
described in detail and are reported to be solvent dependent
polarity, hydrogen bonding ability, proticity). Avobenzone is
reported in several works to be more photostable in polar sol-
(
4
4–54
6
,14,15
into radiotracers
to increase availability, decrease bone
vents compared to non-polar solvents.
The UV-Vis absor-
uptake of radiometabolites, and improve diagnostic medical
image quality (e.g. positron emission tomography scan).
Deuterium is also known to alter the strength of hydrogen
bance can sometimes recover upon storage in the dark in
6
,15
certain organic or cosmetic solvents.
Photodegradation
leads to irreversible decomposition of avobenzone, and occurs
via α-cleavage of the diketone in the triplet excited state to
generate carbon-centred radical species which undergo proton
abstraction, oxidation, or radical recombination to form pro-
ducts of concern for their potential to harm human
5
5–57
bonds.
Examples of utilising deuterated hydrogen bonds
includes changing the structure and properties of crystalline
5
8–63
materials,
and application of the kinetic isotope effect to a
tautomeric system to extend the otherwise short lifetime of a
6
4
5
–7,9,10,14,15,21
particular molecule and enable its chemical reaction. It has
health.
This is reported to be more common in
5
,6,14,15
been shown that the intramolecular O–D bond in D O is
2
non-polar solvents or concentrated solutions.
While
shorter than O–H in H O, and that the “hydrogen” bond is
2
solvent is an important factor to consider in sunscreen formu-
lations, it has been proposed that it is an over simplification to
consider avobenzone to be photostable in polar solvents and
6
5
2 2
longer in D O than in H O. Interestingly, there is conflicting
5
photounstable in non-polar solvents.
A number of methods other than solvent have been
explored and developed to improve the stability of avobenzone. cal additives with avobenzone.
‡
Accelerated degradation has been observed for certain combinations of chemi-
9
Photochem. Photobiol. Sci.
This journal is © The Royal Society of Chemistry and Owner Societies 2020

View Article Online
Photochemical & Photobiological Sciences
Paper
data as to whether hydrogen bond strength is decreased or
increased by deuterium. There is significant discussion regard-
ing nuclear quantum effects and the strength of hydrogen
6
6–70
71
bonding for deuterium.
2
The increased viscosity of D O
has been rationalised according to reports of nuclear quantum
6
6,67
effects and stronger hydrogen bonding in D
2
O.
Further
examples where hydrogen bonds are reported to be stronger in
D O occur in discussions around protein unfolding and
2
rupture force (ref. 72 and references within), as well as in
using D
oral poliovirus vaccine,
2
O formulations to increase the thermal stability of
7
3,74
specifically stronger non-covalent Fig. 2 Synthesised isotopologues of avobenzone; avobenzone-d2 (1),
7
3
D⋯O and N⋯O bonds.§ Contrarily, it has been reported avobenzone-d15 (2), avobenzone-d19 (3). Each can exist as the enol (as
drawn) or the diketone tautomer (i.e. Fig. 1).
2
that the O–D bond is shorter and stronger in D O compared to
H O, while the “hydrogen” bond is longer and weaker (see ref.
2
7
5 and references within). The tensile strength of wetted cell-
ulose paper was decreased in H O compared to D O, and was
attributed to stronger hydrogen bonding of H O with cellulose,
2
2
2
Materials and synthesis
7
6
which disrupts the interfibre cellulose hydrogen bonds
additionally, H/D exchange occurs on the surface hydroxyl The synthesised deuterated isotopologues of avobenzone are
groups of cellulose). Further, D O media significantly shown in Fig. 2. Non-deuterated avobenzone was
(
a
2
decreased the rate of hydrolysis in a fluoroethyl homocysteine Pharmaceutical Secondary Standard and Certified Reference
7
7
radiotracer,
attributed to stronger hydrogen bonding Material (Sigma-Aldrich, PHR1073). Avobenzone-d
2
(com-
1
between H O and F, which promoted the mechanism of loss of pound 1, >92%D by H NMR, ESI S9†) was synthesised from
2
fluorine to generate a thiirane intermediate.
With regard to β-diketones, deuteration of the keto–enol ture by exchange in excess CH
moiety and the effect of D O vs. H O solvent has been explored The hydrogen–deuterium exchange of β-diketones (in
this standard under a nitrogen atmosphere at room tempera-
3
OD as a deuterium source.
2
2
for an impact on the equilibrium between tautomers, given general) in deuterated methanol to afford the 2,4-dione-3,3-
the mechanism of interconversion between the chelated enol
isotopologue is well known and has been studied
and diketone involves the breaking of hydrogen bonds as well kinetically.
as proton abstraction. There remains conjecture as to whether
Synthesis of the phenyl ring deuterated avobenzone isoto-
the equilibrium constant or the rate at which the equilibrium pologues was undertaken according to the literature reported
d
2
9
1
9
2
is established are altered (see ref. 78 and references within), for non-deuterated,
with seemingly conflicting experimental data (see ref. 79–82 NaNH , followed by addition of a benzoic acid ester, with
and 83–86). Deuteration has also enabled the equilibrium acidic workup (Fig. 3). 4-Methoxyacetophenone-d (compound
by treating an acetophenone with
2
4
4
4
between the two cis-enol tautomers in asymmetrical 5) at 95–97%D was synthesised in two steps from phenol-d
6
5
5,87,88
β-diketones to be studied.
In a very different but note- (98%D, Cambridge Isotope Laboratories, DLM-370-5), while 4-
worthy study, the keto–enol equilibrium constant of
a
tert-butylbenzoic acid methyl ester-d13 (compound 7, ESI
β-diketone confined to extended nanospaces in water was S5–8†) at 98%D was synthesised in two steps from hydro-
8
9
altered in D
2
O compared to H
2
O.
thermal hydrogen–deuterium exchange of 4-tert-butylbenzoic
We envisaged that as the photodegradation of acid (Fig. 3). After Claisen condensation of the acetophenone
avobenzone is different for the diketone compared to the enol, and benzoic acid ester derivatives (compounds 8 and 9),
deuteration to alter either the keto–enol equilibrium constant exchange to obtain the additional d
or the rate at which the equilibrium is established (or both) was again undertaken in CH OD, to afford avobenzone-d19
could favour a certain degradation pathway, albeit potentially (compound 3, 97.5 ± 2%D by MS for d17, ESI S20,† >90%D for
2
for the keto–enol moiety
3
1
detrimental to the photoprotective properties of avobenzone. additional d
2
by H NMR, ESI S17†) and avobenzone-d15 (com-
The photostability of deuterated sunscreen agents has not pound 2, 97.3 ± 2%D by MS for d13, ESI S16,† >94%D for
1
been explored to the best of our knowledge, except for the additional d
by H NMR, ESI S13†).
was required for the photostability
2
increased excited state lifetime of bemotrizinol-OD (not a
β-diketone).
Only avobenzone-d
2
9
0
study, however, avobenzone-d15 and avobenzone-d19 are likely
to find uses in analytical applications where the site of deu-
teration is relevant (commercially available labelled avoben-
zone is typically deuterated at the methoxy group), or where
highly deuterated avobenzone could assist with contrast
matching during neutron scattering, such as that recently
examined for non-deuterated avobenzone in a model biologi-
§
Although in this case, other mechanisms could be involved, such as a slower
73
rate of enzymatic autocatalytic degradation.
2 2
D O/H O can also affect protein
flexibility and compactness due to differences in hydrophobic effects and
7
5
93
solvation.
cal system using a deuterated lipid.
This journal is © The Royal Society of Chemistry and Owner Societies 2020
Photochem. Photobiol. Sci.

View Article Online
Paper
Photochemical & Photobiological Sciences
13
Fig. 4 Labels for the carbon atoms of avobenzone for C NMR charac-
terisation (R = 4-methoxyphenyl, R’ = 4-tert-butylphenyl).
2 2 2
avobenzone-d in CD Cl (ESI S11†) for the central enol reso-
1
nance (e.g. position 2a, Fig. 4) when decoupled from H only.
This appeared as a singlet for non-deuterated avobenzone, in
3
–5
line with the literature in several solvents,
and became a
1
2
singlet for avobenzone-d2 when both H and
H were
decoupled (ESI S11†). For the phenyl ring deuterated ana-
logues, avobenzone-d15 and avobenzone-d19, NMR revealed
1
3
characteristic triplets for the deuterated C aromatic reso-
1
nances, and loss of the H aromatic resonances, but gained
2
these same resonances in the H NMR (ESI S13–15 and
S17–19†).
1
The chelated cis-enol OH H NMR chemical shift for non-
deuterated avobenzone appeared downfield due to strong
3
hydrogen bonding, and occurred at 17.01 in CDCl3 in line
3
with the literature. Interestingly, the enol OH resonance
1
remained in the H NMR spectrum for avobenzone-d
2
and was
2
barely detected in the H NMR. A possible explanation was
COCl, 0–RT °C, back-exchange due to trace water in NMR solvents, despite
I, reflux, 3 h, 96%. best efforts. This seemed likely, given the OH resonance dis-
Fig. 3 Synthetic scheme for avobenzone-d15 (2) and avobenzone-d19
8
5
(
1
(
3). (a) (1) TFMSAA, D
h, 80%. (b) (1) K CO
c) D O, NaOD, Pt/C, 180 °C, 24 h, 2 cycles, 95%. (d) (1) K
RT, 15 min, (2) CH I, reflux, 4 h, 98%. (e) (1) acetophenone derivative,
NaNH , toluene, 80 °C, 15 min, (2) 4-tert-butylbenzoic acid methyl
2
O, 130 °C, 48 h, (2) TFMSA-d
1
, CH
3
2
3
, acetone, reflux, 15 min, (2) CH
3
2
2 3
CO , acetone,
appeared when a drop of D O was added to a CDCl solution
of non-deuterated avobenzone, yet remained when H O was
2
added (although slightly broadened). Single atom deuterium
exchange not including the enol OH would be unlikely, given
2
3
3
2
3
ester-d13, 80 °C, 4 h, 9–15% (after purification). (f) CH OD, RT, 3–4 days,
quant.
(
1) both the enol OH and vinylic CH exchange in other
8
8
β-diketones, (2) the enol OH is reported to exchange rapidly
and the vinyl proton slowly,
8
5,94
and (3) the stepwise mecha-
nism of deuteration for β-diketones involves keto–enol
tautomerism.
Results and discussion
91
Isotopologue characterisation
Additionally, NMR of avobenzone was informative for moni-
toring two further processes influenced by deuteration. This
included time-dependent equilibration between the diketone
NMR is well suited to characterising the site of deuteration in
1
isotopologues. H NMR of non-deuterated avobenzone in
and enol isomers in a CDCl solution (over minutes to hours),
3
CD Cl revealed the vinyl resonance at 6.84 ppm with an inte-
2
2
and the generation of photoproducts from UV irradiation,
which are now both discussed.
gration of one proton (6.78 ppm in CDCl ), suggesting the enol
was the predominant species in a solution, and in line with
3
5
,14
the literature (6.65–6.66 ppm in C D ,
7.23 ppm in DMSO-
d6 ). For avobenzone-d2 this resonance is almost removed
from the spectrum in CD Cl (<10% residual proton remains,¶
6
12
5
Equilibration between the diketone and enol
2
2
1
2
The observation of two H NMR peaks for R-groups attached to
ESI S9†), and was instead observed in the H NMR (ESI S10†).
Further evidence for deuterium incorporation was gained from
the multiplet observed at 92.1–91.8 ppm in the C NMR of
β-diketones is informative of the amount of diketone and
8
6,89
1
3
enol.
During the course of characterisation of the avoben-
zone isotopologues, it was observed that avobenzone had two
peaks (major and minor) associated with a number of the
resonances immediately on dissolving, particularly apparent
for the methoxy and tert-butyl groups as well as several of the
aromatic resonances. Interestingly, the ratio of these peaks
¶
2
NMR solvents frequently contain trace H O, and back-exchange occurs readily,
particularly at low avobenzone concentrations. In C
remains for the vinylic enol resonance.
6
D12, <15% residual proton
was observed to change with time in CDCl . A number of
3
Photochem. Photobiol. Sci.
This journal is © The Royal Society of Chemistry and Owner Societies 2020

View Article Online
Photochemical & Photobiological Sciences
Paper
Fig. 5 Representation of the proposed NOEs observed between the di-
ketone methylene protons and the phenyl ring ortho-protons.
additional NMR experiments (see later) confirmed this arises
from equilibration of the enol with the diketone, and we have
demonstrated in this study a difference between avobenzone-
2
d and non-deuterated.
1
3
Fig. 6 Equilibration to diketone from enol with time for avobenzone-d
and non-deuterated in CDCl (approx. 25 mM, 298 K), using the relative
integration of the pair of methoxy resonances.
2
Firstly, NMR ( C NMR as well as HSQC and HMBC, ESI
3
S23, S25 and S26†) was used to assign the diketone reso-
nances, visible for sufficient scans in a sample of non-deute-
rated avobenzone which had been equilibrated in CDCl3 for
6
–8 days prior to NMR. The major resonances at 185.91 ppm
The time- and deuteration-dependent amount of
diketone and enol for avobenzone at approx. 25 mM in
CDCl †† was subsequently investigated by kinetic measure-
3
ments at 298 K over 14.5 hours (870 minutes). Fig. 6 shows the
two tautomers take significant time to equilibrate. Relative
integration of the pair of methoxy resonances‡‡ at
and 184.32 ppm are assigned to positions 1a and 3a of the
3
,4
enol (Fig. 4), in line with the literature. The minor reso-
nances at 194.03 ppm and 192.76 ppm are assigned to the
diketone carbonyls 1b and 3b (avobenzone is asymmetrical
and these are non-equivalent), similar to the literature
3
(195.5/194.2 ppm in CH CN ). Additional minor peaks in
3
3
.89/3.87 ppm with time allowed the equilibration between
species to be monitored. At the time of making the solution
after 5 min to prepare the sample and acquire the NMR spec-
trum), the proportion of diketone is greater for avobenzone-d2
1
13
both the H and C NMR spectra can be observed for the
methoxy and tert-butyl groups as well as aromatic reso-
nances, which have 2D correlations between each other, but
(
not with the major resonances. Importantly, a resonance at
(
(
approx. 5.5%, blue circles, Fig. 6) than non-deuterated
<3%,§§ red circles, Fig. 6). The % diketone continues to
1
4
.57 ppm in the H NMR was assigned to the central methyl-
ene group between the diketone, in line with the literature
increase with time (10–11% for avobenzone-d2 and 4% for
non-deuterated at 14.5 h).
The increased % diketone for avobenzone-d compared to
2
non-deuterated can be rationalised from two angles. The keto–
3
5
(4.55 ppm in CDCl
3
,
4.75 ppm in DMSO-d
6
, 4.35 ppm in
1
4
13
C6D12
∥
),** and which had a HSQC correlation to the
C
resonance at 50.5 ppm (position 2b, Fig. 4), and a HMBC
correlation to 194.03/192.76 ppm. This assignment of the
enol tautomerisation mechanism (acid or base catalysed) likely
1
3
C NMR diketone methylene resonance (position 2b, Fig. 4)
at 50.5 ppm was similar to the literature (49.3 ppm in
3
1
3
CH CN ). Integration of the H NMR diketone methylene
resonance at 4.57 ppm relative to the minor peaks for the
methoxy and tert-butyl resonances was approximately 2 to 3
†
†The kinetic profile for Fig. 6 was obtained in anhydrous CDCl
70699), while that obtained in ESI S30† was in a typical sample of CDCl
CO , filtered through alumina, stored
3
(Sigma-Aldrich
5
3
(trace
water remaining following shaking with K
2
3
to 9 respectively, as expected for the proposed structure. on molecular sieves for 48 hours prior to use). In both cases, the % diketone is
Using 1D NOESY with selective irradiation, the methylene greater for avobenzone-d compared to non-deuterated, however, for ESI S30,†
this was accompanied by a gradual decrease in maximum % diketone for avo-
benzone-d with time, likely to be the equilibrium adjusting for a concomitant
2
resonance of the diketone at 4.57 ppm displayed an NOE
interaction with two doublets in the aromatic region
2
decrease of 17%D over 14.5 h (determined from integration of the residual
vinylic enol resonance at the start and end of the experiment). This back-
(8.04–7.94 ppm), presumably the phenyl ring ortho- reso-
nances (Fig. 5 and ESI S24†).
exchange was presumably from residual water in CDCl
3
. Water is crucial in keto–
and their clusters and proton relays are shown com-
putationally to lower the energy of the transition state for accessible keto–enol
64,104–106
enol tautomerism,
104–106
tautomerism.
Further, the methylene C–H bond of the ketone is not
5
∥
A resonance at 5.10 ppm in another work was assigned to the diketone
methylene in C 12. However, in our hands, we found the methylene resonance
to be at 4.35 ppm, as reported by others.
readily cleaved by nucleophiles other than water in
a
hydrogen bonded
104
107
6
D
network.
Water can disrupt the enol intramolecular hydrogen bond
and
14
108
increase the amount of diketone. Thus it is likely that the water content (or a
trace acid/base impurity) could influence the rate at which the equilibrium is
obtained and/or the equilibrium constant between Fig. 6 and ESI S30.†
‡‡The pair of tert-butyl resonances at 1.37/1.34 ppm can also be used (ESI S29†).
§§The value of % diketone for non-deuterated (red circles, Fig. 6) at 0 minutes is
closer to 1% than 3%, with the discrepancy due to the automated batch inte-
gration, but is accurate for the later time points.
*
*It has been reported that the diketone methylene protons are not visible in
5
MeOD-d
4
,
which is likely due to the combination of solvent polarity encoura-
ging the enol, although the gradual exchange of the labile protons with deuter-
ium from the protic solvent with time could also explain this observation.
Avobenzone-d
2 1
in our work was synthesised by exchange with MeOD-d over the
course of several days at room temperature.
This journal is © The Royal Society of Chemistry and Owner Societies 2020
Photochem. Photobiol. Sci.

View Article Online
Paper
Photochemical & Photobiological Sciences
Table 1 % diketone and Keq for avobenzone-d
2
and avobenzone in with UV, being either reversible or non-reversible, and which
C
6
D
12 and CDCl
3
depends on the experimental conditions. Avobenzone can
photorecover in the dark in non-polar solvents such as hexane
a
a
Solvent
Compound
% diketone
Keq = [enol]/[diketo]
at UV-Vis concentrations over 12 hours, observed as a nearly
complete replenishment of the UV-Vis absorption pertaining
C D
6 12
Avobenzone-d
Avobenzone
Avobenzone-d
Avobenzone
2
4.5
2.3
21.4c
b
42.5
8.5
24
15
to the enol compared to the non-irradiated sample. Slow
CDCl
3
2
10.5
4
reversibility with minimal photodegradation has also been
−
5
demonstrated in CH CN by both UV-Vis (10 M) and NMR
3
a
Calculated by relative integration of the two methoxy resonances for
diketone and enol (at 14.5 h for anhydrous CDCl ). Literature value of
1
13
−3
b
( H, C, concentrations lower than 1.6 × 10 M) for short
3
1
4 c
5
3
3.5% diketone.
Literature values of Keq = 46 and Keq = 28 (calcu- irradiation times. Changing these conditions (increased avo-
14
lated from 3.5% diketone and 96.5% enol ).
benzone concentration, longer UV irradiation, or solvents such
as CHCl and MeOH) resulted in multiple photoproducts and
non-reversible photodegradation rather than reversible photoi-
3
3
1
13
displays a deuterium kinetic isotope effect for the CH vs. CD somerisation. NMR ( H, C, DEPT) has also been used to
abstraction step in going from the diketone to the enol. monitor avobenzone photoisomerisation and photodegrada-
5
Additionally, a weaker chelating hydrogen bond for the enol tion in other solvents including C D . These authors quanti-
6
12
when deuterated could allow increased equilibration to the di- fied the photoisomerisation between the diketo and enol
ketone for avobenzone-d
2
.
species from the change in relative integration between the
The equilibration of avobenzone was solvent dependent, vinylic enol and diketone methylene resonances with
and was comparatively fast in C D , remaining relatively con- irradiation time, while photodegradation can be observed
6
12
stant with time after dissolving. Only trace diketone forms in from the multiple resonances for the tert-butyl, methoxy and
5
14
5
6 6
C D12, reported to be 3.5%. In our hands in C D12, the % di- phenyl moieties following irradiation. In our case with deute-
ketone was 2.3% for non-deuterated, but was 4.5% for avoben- rated avobenzone, the vinylic enol and diketone methylene
1
zone-d
resonances at 3.76/3.33 ppm.
The equilibrium constant for the tautomerisation between both photodegradation and photoisomerisation.
enol and diketone (i.e. Keq = [enol]/[diketo]) has previously
Two different methods for the UV irradiation of avobenzone
been calculated for non-deuterated avobenzone in C 12 (as and avobenzone-d2 solutions were explored, including real
2
, calculated by relative integration of the two methoxy resonances are H NMR silent, and so we elected to examine
the methoxy resonance, and using conditions likely to inflict
6
D
well as in DMSO-d ) using the relative integration of the enol time in situ within the NMR (irradiation using a LED source
6
vinylic and diketone methylene resonances to reflect the con- via fibre optic cable, 5.4 mM in C D , continuous 18 hours),
6
12
5
centration of each species. Table 1 summarises Keq for avo- as well as ex situ using sunlight (samples in parallel, 3.2 mM
benzone-d2 and non-deuterated in C D and CDCl3 (equili- in C D , total 9 h ¶¶). A typical NMR of the methoxy resonance
6
12
6 12
bration has not completely stabilised at 14.5 h in CDCl
3
), cal- before and after UV irradiation is shown in Fig. 7.
culated from the relative integration of the two methoxy reso-
Fig. 8a shows the overlayed avobenzone and avobenzone-d2
nances. We acknowledge seemingly conflicting data regarding (same concentration) for the ex situ sunlight experiment
the effect of deuterium on the equilibrium constants for immediately after irradiation, where the intensity of the enol
7
8–86,95,96
β-diketones,
however, with the trend in Fig. 6 in methoxy resonance has been manually adjusted to be equal to
CDCl3 and data in C D , there is no extrapolated time at enable visual comparison of the relative intensity of the new
6
12
which it could be envisaged that the % diketone could be resonances for the methoxy group. In addition to the chelated
equal for deuterated and non-deuterated, and in this case enol (3.76 ppm) and diketone (3.73 ppm), two major photo-
suggests a difference in the equilibrium constant. We did not products can be observed at 3.74 and 3.75 ppm, and which
accelerate equilibration by the addition of 0.01–1% triethyl- appear to be increased for avobenzone-d2 compared to non-
8
5,86
amine as has been undertaken by others.
We postulated that the different proportion of diketone and irradiated sample with an additional 16 h in the dark,
enol for avobenzone-d and non-deuterated had the potential however, there continues to be some change to the relative
deuterated after a total of 9 h irradiation. Fig. 8b is the same
2
to be manifested as a difference in the degradation pathway intensity of the photoproduct at 3.75 ppm during this time,
between the two isotopologues (i.e. non-reversible α-cleavage which would suggest photodegradation and not photoisomeri-
of the diketone in the triplet excited state to generate carbon- sation (i.e. photorecovery was not observed). This result is sup-
centred radical species). As the photodegradation of non- ported by a deuteration-related increase in the % diketone,
deuterated avobenzone has been previously characterised in with this additional diketone undergoing α-cleavage to carbon-
5
C
6
D
12
,
we opted to undertake measurements in the same centred radical species. An attempt was made to quantify the
amount of each methoxy-containing species by baseline resol-
solvent.
UV photostability by NMR
Avobenzone exhibits interesting and at times confusing photoi-
¶
¶Irradiated for three hours, stored in the dark overnight, and then irradiated
somerisation or photodegradation behaviour upon irradiation for a further six hours, for sufficient accumulation of photoproducts.
Photochem. Photobiol. Sci.
This journal is © The Royal Society of Chemistry and Owner Societies 2020

View Article Online
Photochemical & Photobiological Sciences
Paper
photoproducts could be limited for the in situ UV LED experi-
ment, compared to the ex situ sunlight experiment in Fig. 8
and literature methods which use broad spectrum UV sources/
5
,28
filters/solar simulators.
We further examined photostability by testing avobenzone
isotopologues to ISO 24443 for the determination of sunscreen
9
7
UVA photoprotection in vitro, however, no clear trend in the
photostability data was observed for the isotopologues com-
pared to non-deuterated reference standards (ESI S28†).
Solid state NMR
Solid state NMR of non-deuterated β-diketones (including avo-
benzone) has previously been reported and compared with
solution NMR data, in addition to also examining if a prefer-
4
ence exists between the two enols in asymmetric systems. In
Fig. 7 Typical H NMR spectrum of the methoxy resonance prior (blue our case, solid state NMR was undertaken for avobenzone-d2
1
line) and post (red line) UV irradiation of avobenzone in C
6
D
12 (approx. and non-deuterated avobenzone to determine whether any
3.2–5.4 mM). Initially, only two methoxy resonances can be observed
additional information compared to solution could be
obtained, regarding the % diketone and the effect of deutera-
tion on this equilibrium.
for the enol (major, cropped) and diketone (minor).
1
3
The C NMR spectrum of non-deuterated avobenzone
Fig. 9, blue line) showed two distinct signals at 194.5 ppm
(
and 183.2 ppm, which were assigned to the CvO (I) and
CvCOH (K) carbon sites in the enol, and which had equal
intensity. The CvO signal in the solid state (194.5 ppm)
appeared to be shifted downfield by 8.6 ppm compared to
solution (185.9 ppm). Concomitant to the enol structure, a
signal at 95.8 ppm for the central vinylic CH (J) site was
observed. The diketone was unable to be observed; the methyl-
ene CH signal at ∼50 ppm (value in solution) was absent.
2
In the case of avobenzone-d , all the peaks except the CvO (I),
2
CvCOH (K) and CH (J) were superimposable with non-
deuterated avobenzone. The position of the vinylic CH signal
(
in this case CD) was shifted to a slightly lower value of
1
95.5 ppm, which was attributed to the isotope effect which has
been previously demonstrated in deuterated compounds.
Fig. 8 Overlayed partial H NMR (methoxy region) for ex situ UV (sun-
9
8
2
light) irradiated avobenzone (red) and avobenzone-d (blue) where the
intensity of the enol methoxy resonance has been adjusted to be equal. The overall intensity of the CvO and CvCOH signals was
1
(
a) Immediately after UV exposure, (b) sample (a) after an additional 16 h
reduced due the reduction in the number of H near these
in the dark.
ving resonances in Fig. 8 by using line broadening parameters
(ESI S31†), however the peak integrals are heavily influenced
by the manual baseline correction and showed significant
variability.
For the in situ UV LED experiment (ESI S32†), no difference
could be observed in the relative amount of photoproducts for
2
avobenzone-d compared to non-deuterated. While the irradi-
ance of the UV LED (4 mW) was of the same order as literature
methods using solar simulators (in which the irradiance was
−
2
measured by UV meter to be 2.5–16 mW cm for UVA and
.4–0.54 mW cm− for UVB
2
3,14,28
), the source had a narrow
0
13
Fig. 9
avobenzone-d
same intensity for the tert-butyl signal (N). The spinning sidebands are
C solid state NMR spectra of non-deuterated avobenzone and
wavelength distribution characteristic of LED light and only in
the UVA region (365 ± 5 nm). Increased degradation without fil-
2
overlayed in blue and red respectively and scaled to the
1
4
tering UVB has been observed, and so the generation of marked by *. An enlargement is provided in ESI S33.†
This journal is © The Royal Society of Chemistry and Owner Societies 2020
Photochem. Photobiol. Sci.

View Article Online
Paper
Photochemical & Photobiological Sciences
sites (due to deuteration), and therefore less magnetisation ducts integrated into society could lead to new opportunities
1
3
transfer to the
Interestingly, there was also a slight change in the relative products, and health/medicine/diagnostics.
intensity of the CvO and CvCOH signals for avobenzone-d
C sites in the cross-polarisation step. in high tech materials, advanced manufacturing, personal care
2
,
with the CvO site having a slightly higher intensity than the
CvCOH. This could be explained by a slight increase in the di-
ketone for deuterated avobenzone, in line with solution obser-
vations by H NMR, provided the diketone CvO C signals
are not shifted in the solid state compared to solution (194.03/ Kinetic H NMR experiments involved dissolving approx.
Experimental
Methods
1
13
1
1
92.76 ppm). However, as for non-deuterated avobenzone, the 5.5 mg of avobenzone-d
2
or non-deuterated reference in 700 µL
1
resonance for the central methylene CH of the diketone was CDCl3 (approx. 25 mM) and immediately obtaining the
H
2
2
not observed for avobenzone-d either, although it is possible NMR spectrum. This process took approximately 5 minutes.
that in this case the signal is suppressed due to the lack of any The kinetic experiment was subsequently commenced
directly bonded protons (i.e. it is a CD2 group), with the immediately, with spectra collected automatically at set time
nearest protons being four bonds away.
intervals (initially every 1 minute, then 10 minutes, then
0 minutes) over approximately 14.5 hours. Identical phasing
3
and integration was applied to all timed spectra for the same
sample.
Conclusions
To ensure reasonable comparison of the keto–enol equili-
Deuteration of the keto–enol moiety of avobenzone increased brium between avobenzone-d2 and non-deuterated (solution
the % diketone compared to non-deuterated in both CDCl and solid state), the non-deuterated reference was also dis-
12 as examined by using H NMR. This can be rational- solved in CH OH and the solvent removed after two to three
ised from two angles; mechanistically by a deuterium kinetic days, using the same procedure detailed in the experimental
isotope effect for the CH vs. CD abstraction step during tauto- for the synthesis of avobenzone-d (which instead used
3
1
and C
6
D
3
2
merisation from the diketone to the enol, and by a weaker che- CH
lating hydrogen bond for the enol when deuterated allowing
3
OD).
1D NOESY with selective irradiation was undertaken on
after 24 h equilibration,
increased equilibration to the diketone. The equilibration non-deuterated avobenzone in CDCl
between diketone and enol occurred over minutes to hours in examining the diketone methylene resonance at 4.57 ppm.
CDCl , and was rapid in C D . In the solid state, the enol NMR, HMQC, and HMBC spectra were recorded on a similar
3
1
3
C
3
6 12
remained dominant. The change in the relative intensity of the sample equilibrated over six to eight days prior to
1
3
CvO and CvCOH C solid state NMR signals upon deutera- measurement.
tion could suggest an increase in the amount of diketone for In situ UV NMR degradation experiments utilised a UV LED
avobenzone-d , but was unable to be confirmed due to the source at 365 nm (UVLED365-10E, Roithner Laser Technik,
2
1
3
absence of the central methylene CH
solid state.
2
/CD
2
C signals in the CW output power = 4.0 mW, λmin = 360 nm, λmax = 370 nm,
spectrum half width = 15 nm) fitted to a controller (Ocean
UV irradiation experiments in C
revealed increased photoproducts for avobenzone-d
6
D
12 ex situ using sunlight Optics LLS series module) set to continuous irradiation at
compared maximum power. A fibre optic cable (Thorlabs 0.39 NA TECS
2
to non-deuterated, and suggested photodegradation rather hard cladding, step-index, multimode fibre, part number
than photoisomerisation. This would align with the higher % FT1000UMT with a SMA 905 connection, sheath removed at
diketone for avobenzone-d
α-cleavage to carbon-centred radical species. No difference was sample, and was inserted just above the solution (1–3 mm) in
observed for in situ irradiation experiments using a UVA LED a single walled NMR tube containing avobenzone in C
2
, given the diketone undergoes the tip to expose 5 mm of the core) delivered UV light to the
6 12
D
source, which could be related to the absence of UVB, given (approx. 5.4 mM, 1.0 mg in 600 µL, 45 minutes equilibration
increased avobenzone degradation without filtering UVB has in the dark prior to irradiation). The sample was continuously
been observed by others. Future studies will use analytical irradiated, with spectra collected automatically at set time
techniques such as LC-MS to identify photoproducts and to intervals (initially every 1 minute, then 10 minutes, then
provide relative quantitation of these between deuterated and 30 minutes) over approximately 18 hours.
non-deuterated avobenzone.
The deuteration of other compounds could confer advan- parallel on samples of each of avobenzone-d
tageous (or disadvantageous) properties for their intended rated in C D (approx. 3.2 mM, 1.0 mg in 1 mL, 2.5 h equili-
Ex situ sunlight degradation experiments were performed in
2
and non-deute-
6
12
application, given an appropriate mechanism and which mani- bration in the dark prior to irradiation) by exposing the pair of
fests as a change in the behaviour of the bulk material or upon NMR tubes to full sun on the same day and location. This was
its interaction with external stimuli. We encourage scientists undertaken between November and January in Australia
and industry to investigate the deuteration of materials beyond (Sydney), on days when the UV index was typically on a scale of
current niche applications (e.g. MS, NMR, neutron scattering), 8–11+ (very high to extreme) according to data reported by the
as the adoption of deuteration technology into common pro- Bureau of Meteorology (Australian Government). NMR spectra
Photochem. Photobiol. Sci.
This journal is © The Royal Society of Chemistry and Owner Societies 2020

View Article Online
Photochemical & Photobiological Sciences
Paper
were collected before and immediately after exposure to sun- (250 mL) with rapid stirring, and extracted into EtOAc (3 ×
light, with samples stored in the dark overnight for 16 h 70 mL). The EtOAc was washed with 1 M HCl (1 × 200 mL), sat.
between exposures spanning multiple days.
3
NaHCO (2 × 200 mL), sat. brine (1 × 200 mL), dried with
Solid state NMR experiments were run on a Bruker Avance MgSO
4
, filtered, and concentrated to dryness by rotary evapor-
III 700 MHz instrument with a 16.4 tesla superconducting ation. The brown solid was redissolved in EtOAc and passed
magnet operating at frequencies of 700 MHz and 175 MHz for through a silica plug (25% EtOAc in n-hexane), which upon
1
13
1
the H and C nuclei respectively. Approximately 10 mg of evaporation afforded a beige powder (5.5 g, 80%). H NMR
sample was packed in 2.5 mm zirconia rotors fitted with (400 MHz, CDCl , 298 K): 7.92 (s, 0.03H/2H, residual proton),
3
2
Vespel® caps and spun to 15 kHz at the magic angle in a 6.93 (s, 0.09H/2H, residual proton), 2.58 (s, 3H). H NMR
1
3
13
1
2
.5 mm HXY triple resonance probehead. The C spectra were (61 MHz, CDCl
3
, 298 K): 7.97 (s, 2D), 7.00 (s, 2D). C NMR ( H
acquired with the CPMAS pulse sequence with a 1 ms contact power-gated decoupling, 100 MHz, CDCl , 298 K): 198.39,
3
1
13
1
2
time and 5 s recycle delay. The H 90 pulse length of 3 μs, 161.12, 130.92*, 129.79, 115.27*, 26.46. C NMR ( H and H
SPINAL-64 decoupling of 83 kHz field strength and 20k transi- gated decoupling, 100 MHz, CDCl , 298 K): 198.99, 161.63,
ents co-added for sufficient signal to noise. The C chemical 131.02, 129.40, 115.36, 26.42. MS (ESI) for C H D O [M −
3
1
3
−
8
3 4 2
−
shift was referenced to the methylene peak of adamantane at H] : calc: 139.1. Found: 139.1. Average % deuterium incorpor-
3
8.48 ppm.
3 4
ation (d + d ): 10.5 + 86 = 96.5 ± 2% (by MS). Average % deu-
FT-IR was undertaken on a Thermo Scientific Nicolet™ terium incorporation (phenyl ring): 94–96% (by integration of
1
3
98
1
iSTM10 spectrometer fitted with the ATR accessory, and refer-
enced to background (air).
C NMR phenyl resonances and from residual H NMR
phenyl resonances).
-Methoxyacetophenone-d4 (5).
procedure, 4-hydroxyacetophenone-d
4
4,101,102
4
Using a modified
(5.5 g, 39.2 mmol) was
General synthesis
4
Silica gel (Sigma-Aldrich, 60737) was used for column chrom- dissolved in acetone (200 mL) under a nitrogen atmosphere.
atography. Kieselgel silica gel 60 F254 aluminium sheets Potassium carbonate (5.7 eq., 30.9 g, 224 mmol) was added,
(
Merck) were used for thin later chromatography (TLC). and the mixture heated under reflux for 15 minutes. Methyl
Compounds were visualised using a UV lamp (short wave iodide (9 eq., 22 mL, 353 mmol) was added cautiously down
54 nm and long wave 365 nm) or permanganate stain. the condenser in four portions over 3 minutes, and the
Percentage yields were calculated based on the purified mixture allowed to reflux for 3 hours, after which no starting
product mass. material was detected by TLC. The mixture was cooled, filtered,
2
NMR spectra were recorded on a Bruker Avance III 400 MHz the solids washed with further acetone (3 × 50 mL), and the fil-
spectrometer at 298 K, equipped with a 5 mm PABBO BB H/D trate concentrated to dryness by rotary evaporation (in a fume-
z-gradient probe. Spectra were referenced to residual deute- hood). The concentrate (oil and white solid) was redissolved in
9
9
rated solvent. The % deuterium incorporation for com- Et O (100 mL), filtered, the solids washed with further Et O (2
2
2
pounds was calculated by averaging of MS peak areas for the × 50 mL), and the filtrate again concentrated to dryness by
different deuterated isotopologues using an in-house devel- rotary evaporation (in a fumehood). The oil was dissolved in a
oped spreadsheet, with the exception of avobenzone d , d
and d19 where integration of the residual H NMR resonance through a plug of silica (1 : 1 EtOAc/n-hexane), to recover a pale
small quantity of EtOAc (10 mL), and purified by passage
2
15
1
at 6.84 ppm for the vinylic enol was measured relative to a yellow oil which solidified on cooling. The material was dried
1
non-deuterated resonance in avobenzone (methoxy group) as in vacuo to afford a pale yellow solid (5.8 g, 96%). H NMR
2
an internal standard. H NMR were recorded as detailed in a (400 MHz, CDCl
3
, 298 K): 7.93 (s, 0.03H/2H, residual proton),
9
8 13
previous publication.
attached to deuterium; these signals are triplets when only H
C NMR resonances marked * are 6.92 (s, 0.07H/2H, residual proton), 3.86 (s, 3H), 2.55 (s, 3H).
1
2
13
H NMR (61 MHz, CDCl , 298 K): 7.98 (s, 2D), 6.98 (s, 2D).
C
3
1
2
1
is decoupled, and resolve to singlets when both H and H are NMR ( H power-gated decoupling, 100 MHz, CDCl
3
, 298 K):
decoupled.
196.88, 163.55, 130.3* (two overlapping resonances in this
13
1
2
multiplet), 113.42*, 55.58, 26.45. C NMR ( H and H gated
decoupling, 100 MHz, CDCl , 298 K): 196.89, 163.55, 130.31,
Trifluoromethane sulfo- 130.29, 113.42, 55.58, 26.45. MS (ESI) for C
nic anhydride (51.6 g, 0.183 mol) was heated with D O (1 eq., calc: 155.1. Found: 155.1. Average % deuterium incorporation
Compound synthesis
3
4
4,100
+
+
4
-Hydroxyacetophenone-d
4
(4).
9 7 4 2
H D O [M + H] :
2
3
.30 mL, 0.182 mol) under a nitrogen atmosphere for 48 h (oil (d
3 4
+ d ): 10 + 87 = 97 ± 2% (by MS). Average % deuterium
1
3
bath temperature set to 130 °C, vigorous reflux observed for incorporation (phenyl ring): 93–95% (by integration of
the first 6 h) to obtain trifluoromethane sulfonic acid-d1. NMR phenyl resonances and from residual H NMR phenyl
C
9
8
1
Using a modified procedure, the trifluoromethane sulfonic resonances).
acid-d
1
was cooled on ice to 0–5 °C, and phenol-d
6
(4.89 g,
4-tert-Butylbenzoic acid-d13 (6). 4-tert-Butylbenzoic acid
4
8.8 mmol) and acetyl chloride (1 eq., 3.47 mL, 48.8 mol) were (10 g, 56.1 mmol) was dissolved in D O (100 mL) containing
2
2
added in quick succession. The mixture was allowed to warm 1.1 eq. of NaOD (6.3 mL of a 40 wt% NaOD in D O). Pt/C
to room temperature over 1 h, after which no starting material (1.0 g) was added, and the solution sealed in a Parr pressure
was detected by TLC. The mixture was poured onto cooled D O vessel reactor (450 mL capacity). The solution was purged with
2
This journal is © The Royal Society of Chemistry and Owner Societies 2020
Photochem. Photobiol. Sci.

View Article Online
Paper
Photochemical & Photobiological Sciences
nitrogen gas and then hydrogen gas, and heated to 180 °C 200 mL), saturated NaCl solution (2 × 200 mL), dried with
10 bar) for 24 h. The solution was cooled and filtered through Na SO , filtered, and concentrated in vacuo at 50 °C to afford a
(
2
4
Celite, rinsing the Celite with water. The aqueous layer (prior yellow/orange paste. The material was purified by semi-auto-
to acidification) was washed with CH Cl (2 × 100 mL), and the mated flash chromatography (Reveleris, silica, loaded CH Cl
2
2
2
2
,
organic layer discarded. The aqueous layer was acidified with 5 eluted 10% EtOAc/n-hexane), and the middle of the band with
M HCl to pH = 1 and extracted into EtOAc (2 × 200 mL). The the same TLC as the non-deuterated reference was collected.
organic layer was washed with water (2 × 200 mL), saturated This material was recrystallised from a minimum amount of
NaCl solution (2 × 200 mL), dried with MgSO , filtered, and hot MeOH (cooled to room temperature, then at −20 °C for
4
concentrated in vacuo to afford a white powder (9.5 g, 95%, 2 hours), filtered, and dried in vacuo to afford the product as a
1
2
cycles). H NMR (400 MHz, CDCl
3
, 298 K): 8.07 (s, residual white powder.
2
proton), 7.49 (s, residual proton), 1.32 (s, residual proton). H
NMR (61 MHz, CDCl , 298 K): 8.12 (s, 2D), 7.55 (s, 2D), 1.33 (s, recovered if necessary). H NMR (400 MHz, CD
D). C NMR ( H power-gated decoupling, 100 MHz, CDCl 7.99 (d, J = 9.0 Hz, 2H), 7.92 (s, 0.03H/2H, residual proton),
98 K): 172.67, 157.65, 129.88*, 126.53, 125.21*, 34.63, 29.96*. 7.52 (s, 0.05/2H, residual proton), 7.00 (d, J = 9.0, 2H), 6.82 (s,
C NMR ( H and H gated decoupling, 100 MHz, CDCl
98 K): 172.68, 157.65, 129.89*, 126.53, 125.22*, 34.63, 30.13*. (61 MHz, CD
Avobenezone-d13 (8). 0.5 g, 15% (further material could be
1
3
2 2
Cl , 298 K):
13
1
9
2
3
,
1
3
1
2
2
3
,
1H), 3.88 (s, 3H), 1.31 (s, 0.11H/9H, residual proton). H NMR
Cl , 298 K): 7.97 (s, 2H), 7.56 (s, 2H), 1.32 (s, 9H).
C NMR ( H power-gated decoupling, 100 MHz, CDCl ,
2
2
1
2
−
−
13
MS (ESI) for C D O [M − H] : calc: 190.2. Found: 190.2.
1
1
13
2
3
Average % deuterium incorporation (d12 + d13): 24.5 + 73.5 = 98 298 K): 185.95, 184.30, 163.26, 155.97, 132.73, 129.36, 128.47,
1
3
±
2% (by MS).
-tert-Butylbenzoic acid methyl ester-d13 (7).
modified procedure, 4-tert-butylbenzoic acid-d13 (5.0 g, 184.30, 163.26, 155.97, 132.72, 129.35, 128.46, 126.63*,
6.1 mmol) was dissolved in acetone (200 mL) under a nitro- 125.33*, 114.09, 92.22, 55.59, 34.47, 30.16*. MS (ESI) for
126.63*, 125.33*, 114.09, 92.22, 55.60, 34.47, 30.17*. C NMR
1
03
1
2
4
Using a ( H and H gated decoupling, 100 MHz, CDCl , 298 K): 185.95,
3
2
+
+
gen atmosphere with high stirring and suspended with anhy- C H D O [M + H] : calc: 324.25. Found: 324.16. Average %
2
0
10 13 3
drous potassium carbonate (3 eq., 10.85 g, 78.4 mmol). After deuterium incorporation (d11 + d12 + d13): 5.0 + 21.0 + 71.3 =
5 minutes, methyl iodide (1.1 eq. 1.8 mL, 28.7 mmol) was 97.3 ± 2% (by MS).
added. The suspension was heated to reflux for 4 hours. The Avobenzone-d17 (9). 0.3 g, 9% (further material could be
mixture was cooled, filtered, and the solids rinsed with further recovered if necessary). H NMR (400 MHz, CD
1
1
2
2
Cl , 298 K):
acetone. The acetone was removed in vacuo and the material 7.99 (s, 0.03H/2H, residual proton), 7.92 (s, 0.03H/2H, residual
passed down a silica plug to remove baseline material, eluting proton), 7.51 (s, 0.05H/2H, residual proton), 7.00 (s, 0.07H/1H,
a single front band with CH
2 2 2 2
Cl . The CH Cl was removed in residual proton), 6.82 (s, 1H), 3.88 (s, 3H), 1.31 (s, 0.10H/9H,
2
vacuo to afford a colourless oil (5.27 g, 98%), which solidified residual proton). H NMR (61 MHz, CD
2
Cl
2
, 298 K): 8.22–7.83
1
13
1
upon cooling. H NMR (400 MHz, CDCl , 298 K): 7.97 (s, (m, 4H), 7.56 (s, 2H). 7.05 (s, 2H), 1.32 (s, 9H). C NMR ( H
3
0
.03H/2H, residual proton), 7.44 (s, 0.07H/2H, residual power-gated decoupling, 100 MHz, CDCl
3
, 298 K): 185.92,
2
proton), 3.90 (s, 3H), 1.29 (s, 0.14H/9H, residual proton). H 184.32, 163.20, 155.97, 132.71, 128.97*, 128.26, 126.62*,
NMR (61 MHz, CDCl , 298 K): 8.02 (s, 2D), 7.49 (s, 2D), 1.31 (s, 125.34*, 113.71*, 92.22, 55.61, 34.48, 30.17*. C NMR ( H and
9
1
3
1
3
13
1
2
D). C NMR ( H power-gated decoupling, 100 MHz, CDCl
3
,
3
H gated decoupling, 100 MHz, CDCl , 298 K): 185.92, 184.31,
2
3
98 K): 167.28, 156.56, 129.18*, 127.34, 125.05*, 52.05, 34.49, 163.19, 155.97, 132.70, 128.97*, 128.25, 126.62*, 125.34*,
1
3
1
2
0.16*. C NMR ( H and H gated decoupling, 100 MHz, 113.71*, 92.22, 55.62, 34.48, 30.16*. MS (ESI) for
+
+
CDCl
3
, 298 K): 167.28, 156.57, 129.18*, 127.34, 125.05*, 52.05,
C
20
H
5
D
17
O
3
Na [M + Na] : calc: 350.25. Found: 350.24. Average
+
+
3
2
d
4.49, 30.14*. MS (ESI) for C12H D O Na [M + Na] : calc: % deuterium incorporation (d15 + d16 + d17): 6.4 + 25.6 + 65.5 =
3 13 2
28.2. Found: 228.4. Average % deuterium incorporation (d12
+
97.5 ± 2% (by MS).
13): 21.6 + 76.6 = 98.2 ± 2% (by MS).
General procedure for deuteration of the keto–enol moiety
General procedure for avobenzone synthesis (8 and 9). (1–3). Avobenzone (non-deuterated reference material, d13, or
9
2
Using a literature procedure for non-deuterated avobenzone,
d ) was dissolved in MeOD-d (i.e. CH OD, 12 mL per 100 mg
17 1 3
either 4-methoxyacetophenone-d
oxyacetophenone (1.46 g, 9.7 mmol) was dissolved in toluene sphere at room temperature, and protected from light. The
12–15 mL) and NaNH (1.25 eq., 0.475 g, 12.2 mmol) added. depletion of the enol vinylic CH resonance at 6.84 ppm in the
4
(1.5 g, 9.7 mmol) or 4-meth- of avobenzone) and stirred for 3–4 days under a nitrogen atmo-
(
2
1
The mixture was heated to 80 °C for 15 minutes, after which 4-
H NMR was monitored, and when exchange was complete,
tert-butylbenzoic acid methyl ester-d13 (2.0 g, 9.7 mmol) was the MeOD was removed in vacuo at room temperature.
added, and heating continued for a further 4 h (TLC was moni- Occasionally, a second cycle with fresh MeOD was required.
1
tored for significant depletion of the 4-tertbutylbenzoic acid
Avobenzone-d
2
(1). H NMR (400 MHz, CD
2
Cl
2
, 298 K): 7.99
methyl ester). Occasionally the mixture formed a thick paste (d, J = 8.9 Hz, 2H), 7.93 (d, J = 8.6 Hz, 2H), 7.53 (d, J = 8.6 Hz,
and required additional toluene. The mixture was cooled, 2H), 7.00 (d, J = 8.9 Hz, 2H), 6.83 (s, 0.08H/1H, residual
2
diluted with toluene (100 mL) and an aqueous 1 M HCl solu- proton), 3.88 (s, 3H), 1.37 (s, 9H). H NMR (61 MHz, CD
tion (200 mL) was added with vigorous shaking for 3 minutes. 298 K): 6.87 (s).
2
Cl
2
,
1
3
1
C NMR ( H power-gated decoupling,
The product was extracted into toluene, washed with H O (2 × 100 MHz, CD Cl , 298 K): 185.88, 184.17, 163.41, 156.23,
2
2
2
Photochem. Photobiol. Sci.
This journal is © The Royal Society of Chemistry and Owner Societies 2020

View Article Online
Photochemical & Photobiological Sciences
Paper
1
3
CD
32.71, 129.26, 128.21, 126.87, 125.75, 114.03, 92.08*, 55.60,
6 J. Vallejo, M. Mesa and C. Gallardo, Vitae, 2011, vol. 18,
pp. 63–71.
7 A. Cantrell and D. J. McGarvey, J. Photochem. Photobiol., B,
2001, 64, 117–122.
8 C. Paris, V. Lhiaubet-Vallet, O. Jiménez, C. Trullas and
M. Á. Miranda, Photochem. Photobiol., 2009, 85, 178–184.
9 V. Lhiaubet-Vallet, M. Marin, O. Jimenez, O. Gorchs,
C. Trullas and M. A. Miranda, Photochem. Photobiol. Sci.,
2010, 9, 552–558.
1
3
1
2
5.04, 30.93. C NMR ( H and H gated decoupling, 100 MHz,
Cl , 298 K): 185.89, 184.17, 163.41, 156.23, 132.71, 129.27,
28.22, 126.88, 125.75, 114.03, 91.82*, 55.60, 35.05, 30.93.
2
2
1
1
Average % deuterium incorporation (d ): >92% (by H NMR
residual integration of the vinylic CH at 6.83 ppm).
2
1
Avobenzone-d15 (2). H NMR (400 MHz, CD
2
Cl
2
, 298 K): 7.98
(d, J = 8.9 Hz, 2H), 7.92 (s, 0.03H/2H, residual proton), 7.52 (s,
0
.05/2H, residual proton), 7.00 (d, J = 8.9 Hz, 2H), 6.82 (s,
0
.06H/1H, residual proton), 3.88 (s, 3H), 1.31 (s, 0.11H/9H,
10 J. Kockler, M. Oelgemöller, S. Robertson and B. D. Glass,
J. Photochem. Photobiol., C, 2012, 13, 91–110.
11 C. A. Bonda, Sunscreens : Regulations and Commercial
Development, Taylor & Francis, 3rd edn, 2005.
12 B. Brewster, Sun Science : Formulating for Protection,
Allured Books, Carol Stream, IL, 2009.
13 N. J. Lowe, Dermatol. Clin., 2006, 24, 9–17.
14 W. Schwack and T. Rudolph, J. Photochem. Photobiol., B,
1995, 28, 229–234.
2
residual proton). H NMR (61 MHz, CD Cl , 298 K): 7.97 (s,
2
ium incorporation (additional d
integration of the vinylic CH at 6.82 ppm).
2
2
H), 7.56 (s, 2H), 6.87 (s, 1H), 1.32 (s, 9H). Average % deuter-
1
2
): >94% (by H NMR residual
1
Avobenzone-d19 (3). H NMR (400 MHz, CD
.02H/2H, residual proton), 7.92 (s, 0.03H/2H, residual proton), 7.51
s, 0.07H/2H, residual proton), 7.00 (s, 0.07H/1H, residual proton),
.82 (s, 0.1H/1H, residual proton), 3.88 (s, 3H), 1.31 (s, 0.12H/9H,
2 2
Cl , 298 K): 7.98 (s,
0
(
6
2
residual proton). H NMR (61 MHz, CD
2 2
Cl , 298 K): 8.22–7.81 (m,
15 S. P. Huong, E. Rocher, J.-D. Fourneron, L. Charles,
V. Monnier, H. Bun and V. Andrieu, J. Photochem.
Photobiol., A, 2008, 196, 106–112.
4H), 7.56 (s, 2H), 6.87 (s, 1H), 1.32 (s, 9H). Average % deuterium
1
2
incorporation (additional d ): >90%D (by H NMR residual inte-
gration of the vinylic CH at 6.82 ppm).
16 G. H. G. Trossini, V. G. Maltarollo, R. D. A. Garcia,
C. A. S. O. Pinto, M. V. R. Velasco, K. M. Honorio and
A. R. Baby, J. Mol. Model., 2015, 21, 319.
1
7 J. A. Berenbeim, N. G. K. Wong, M. C. R. Cockett,
G. Berden, J. Oomens, A. M. Rijs and C. E. H. Dessent,
J. Phys. Chem. A, 2020, 124, 2919–2930.
Conflicts of interest
JS is the Scientific Director: Solar of Eurofins Cosmetics &
Personal Care. ISO24443 stability tests were undertaken at
Eurofins/Dermatest (Sydney, Australia).
18 A. D. Dunkelberger, R. D. Kieda, B. M. Marsh and
F. F. Crim, J. Phys. Chem. A, 2015, 119, 6155–6161.
1
2
9 M. Kojić, M. Petković and M. Etinski, Phys. Chem. Chem.
Phys., 2016, 18, 22168–22178.
0 N. D. N. Rodrigues and V. G. Stavros, Sci. Prog., 2018, 101, 8–
Acknowledgements
31.
The National Deuteration Facility thanks the NSW Government
for a Research Attraction and Acceleration Program grant to
the Australian Nuclear Science and Technology Organisation.
The National Deuteration Facility is partly supported by the
National Collaborative Research Infrastructure Strategy – an
initiative of the Australian Government.
21 T. Armeni, E. Damiani, M. Battino, L. Greci and
G. Principato, Toxicology, 2004, 203, 165–178.
22 C. Bonda, Research Pathways to New Additives for Broad
Spectrum Sunscreens, The HallStar Company, 2007.
23 R. K. Chaudhuri, Z. Lascu, G. Puccetti, A. A. Deshpande and
S. K. Paknikar, Photochem. Photobiol., 2006, 82, 823–
8
28.
4 L. de Sola, A. J. Banzo and S. Nonell, Afinidad, 2007, 64,
51–256.
2
2
2
Notes and references
5 V. Iannuccelli, N. Sala, R. Tursilli, G. Coppi and S. Scalia,
1
World Health Organization (WHO), FAQ – Skin cancers – How
common is skin cancer (page 2), https://www.who.int/uv/faq/
skincancer/en/index1.html, (accessed 18 November 2019,
Eur. J. Pharm. Biopharm., 2006, 63, 140–145.
26 S. Scalia and M. Mezzena, Drug Dev. Ind. Pharm., 2009, 35,
192–198.
2
019).
27 K. M. Hanson, M. Cutuli, T. Rivas, M. Antuna, J. Saoub,
N. T. Tierce and C. J. Bardeen, Photochem. Photobiol. Sci.,
2020, 19, 390–398.
28 S. Scalia, S. Simeoni, A. Barbieri and S. Sostero, J. Pharm.
Pharmacol., 2002, 54, 1553–1558.
29 J. Yang, C. J. Wiley, D. A. Godwin and L. A. Felton,
Eur. J. Pharm. Biopharm., 2008, 69, 605–612.
30 M. H. Chisholm, C. B. Durr, T. L. Gustafson,
W. T. Kender, T. F. Spilker and P. J. Young, J. Am. Chem.
Soc., 2015, 137, 5155–5162.
2
Cancer Council Australia, Skin cancer – What is skin cancer?,
https://www.cancer.org.au/about-cancer/types-of-cancer/skin-
cancer.html, (accessed 18 November 2019, 2019).
M. Dubois, P. Gilard, P. Tiercet, A. Deflandre and
M. A. Lefebvre, J. Chim. Phys., 1998, 95, 388–394.
A. J. Vila, C. M. Lagier and A. C. Olivieri, J. Phys. Chem.,
1
G. J. Mturi and B. S. Martincigh, J. Photochem. Photobiol.,
A, 2008, 200, 410–420.
3
4
5
991, 95, 5069–5073.
This journal is © The Royal Society of Chemistry and Owner Societies 2020
Photochem. Photobiol. Sci.

View Article Online
Paper
Photochemical & Photobiological Sciences
3
3
1 I. Aparici-Espert, M. A. Miranda and V. Lhiaubet-Vallet,
Molecules, 2018, 23, 673.
2 V. Uttamsingh, R. Gallegos, J. F. Liu, S. L. Harbeson,
G. W. Bridson, C. Cheng, D. S. Wells, P. B. Graham,
R. Zelle and R. Tung, J. Pharmacol. Exp. Ther., 2015, 354,
49 Y.-S. Ding, J. S. Fowler, S. J. Gatley, J. Logan, N. D. Volkow
and C. Shea, J. Neurochem., 1995, 65, 682–690.
50 T. H. Witney, I. S. Alam, D. R. Turton, G. Smith, L. Carroll,
D. Brickute, F. J. Twyman, Q.-D. Nguyen, G. Tomasi,
R. O. Awais and E. O. Aboagye, Clin. Cancer Res., 2012,
18(4), 1063–1072.
4
3–54.
3
3 S. L. Harbeson, A. J. Morgan, J. F. Liu, A. M. Aslanian,
S. Nguyen, G. W. Bridson, C. L. Brummel, L. Wu,
R. D. Tung, L. Pilja, V. Braman and V. Uttamsingh,
J. Pharmacol. Exp. Ther., 2017, 362, 359–367.
51 M. Jahan, O. Eriksson, P. Johnström, O. Korsgren, A. Sundin,
L. Johansson and C. Halldin, EJNMMI Res., 2011, 1, 33.
52 S. Nag, P. Fazio, L. Lehmann, G. Kettschau, T. Heinrich,
A. Thiele, M. Svedberg, N. Amini, S. Leesch and
A. M. Catafau, J. Nucl. Med., 2016, 57, 315–320.
53 A. Kohen and H.-H. Limbach, Isotope effects in chemistry
and biology, cRc Press, 2005.
54 D. Roston, Z. Islam and A. Kohen, Molecules, 2013, 18,
5543–5567.
55 P. E. Hansen, Magn. Reson. Chem., 1986, 24, 903–910.
56 P. E. Hansen, Prog. Nucl. Magn. Reson. Spectrosc., 1988, 20,
207–255.
57 S. Bolvig, PhD: Deuterium isotope effects on C-13 chemical
shifts as a tool to determine tautomerism and structural fea-
tures in intramolecular hydrogen bonded systems, Roskilde
University, 1997.
58 C. Shi, X. Zhang, C.-H. Yu, Y.-F. Yao and W. Zhang, Nat.
Commun., 2018, 9, 481.
59 A. R. Ubbelohde and K. J. Gallagher, Acta Crystallogr.,
1955, 8, 71–83.
60 S. Crawford, M. T. Kirchner, D. Bläser, R. Boese,
W. I. F. David, A. Dawson, A. Gehrke, R. M. Ibberson,
W. G. Marshall, S. Parsons and O. Yamamuro, Angew.
Chem., Int. Ed., 2009, 48, 755–757.
3
3
3
3
4 G. S. Timmins, Expert Opin. Ther. Pat., 2014, 24, 1067–
1
075.
5 G. S. Timmins, Expert Opin. Ther. Pat., 2017, 27, 1353–
361.
6 R. H. Howland, J. Psychosoc. Nurs. Ment. Health Serv.,
015, 53, 13–16.
7 U.S. Food & Drug Administration, Drugs@FDA: FDA Approved
Drug Products, AUSTEDO (DEUTETRABENAZINE), https://
www.accessdata.fda.gov/scripts/cder/daf/index.cfm?event=over-
view.process&varApplNo=208082.
1
2
3
8 C. Schmidt, Nat. Biotechnol., 2017, 35, 493.
3
9 L. Nguyen, A. L. Scandinaro and R. R. Matsumoto,
Pharmacol., Biochem. Behav., 2017, 161, 30–37.
4
0 Efficacy, Safety, and Tolerability of AVP-786 for the
Treatment of Agitation in Patients With Dementia of the
Alzheimer’s Type, https://www.clinicaltrials.gov/ct2/show/
NCT02442778.
41 Assessment of the Efficacy, Safety, and Tolerability of AVP-786
(Deudextromethorphan Hydrobromide [d6-DM]/Quinidine
Sulfate [Q]) for the Treatment of Agitation in Patients With
Dementia of the Alzheimer’s Type, https://clinicaltrials.gov/ct2/
show/NCT03393520?term=avp-786&rank=5.
61 J. Zhou, Y.-S. Kye and G. S. Harbison, J. Am. Chem. Soc.,
2004, 126, 8392–8393.
4
2 A Phase 2 Study to Evaluate the Safety and Efficacy of CTP-499
in Type 2 Diabetic Nephropathy Patients, https://clinicaltrials.
gov/ct2/show/NCT01487109?term=CTP-499&rank=3.
3 E. C. R. Retrotope, Suite 201, Los Altos, CA 94022,
Retrotope, https://www.retrotope.com/.
4 R. B. Murphy, N. A. Wyatt, B. H. Fraser, N. R. Yepuri,
P. J. Holden, A. T. L. Wotherspoon and T. A. Darwish,
Anal. Chim. Acta, 2019, 1064, 65–70.
62 I. Majerz, Z. Malarski and T. Lis, J. Mol. Struct., 1990, 240,
47–58.
63 J. Zhou, Y.-S. Kye, A. I. Kolesnikov and G. S. Harbison,
Isot. Environ. Health Stud., 2006, 42, 271–277.
64 A. K. Cederstav and B. M. Novak, J. Am. Chem. Soc., 1994,
116, 4073–4074.
4
4
65 A. K. Soper and C. J. Benmore, Phys. Rev. Lett., 2008, 101,
065502.
4
4
4
5 J. Leyton, G. Smith, Y. Zhao, M. Perumal, Q.-D. Nguyen,
E. Robins, E. Årstad and E. O. Aboagye, Cancer Res., 2009,
66 A. Berger, G. Ciardi, D. Sidler, P. Hamm and A. Shalit,
Proc. Natl. Acad. Sci. U. S. A., 2019, 116, 2458–2463.
67 M. Ceriotti, W. Fang, P. G. Kusalik, R. H. McKenzie,
A. Michaelides, M. A. Morales and T. E. Markland, Chem.
Rev., 2016, 116, 7529–7550.
68 T. Ishimoto, M. Tachikawa and U. Nagashima,
Int. J. Quantum Chem., 2009, 109, 2677–2694.
69 K. Sugiura, M. Tachikawa and T. Udagawa, RSC Adv., 2018,
8, 17191–17201.
6
9, 7721–7728.
6 A. Challapalli, R. Sharma, W. A. Hallett, K. Kozlowski,
L. Carroll, D. Brickute, F. Twyman, A. Al-Nahhas and
E. O. Aboagye, J. Nucl. Med., 2014, 55, 256–263.
7 L. Cai, F. T. Chin, V. W. Pike, H. Toyama, J.-S. Liow,
S. S. Zoghbi, K. Modell, E. Briard, H. U. Shetty, K. Sinclair,
S. Donohue, D. Tipre, M.-P. Kung, C. Dagostin,
D. A. Widdowson, M. Green, W. Gao, M. M. Herman,
M. Ichise and R. B. Innis, J. Med. Chem., 2004, 47, 2208–
70 B. Pamuk, P. B. Allen and M. V. Fernández-Serra, J. Phys.
Chem. B, 2018, 122, 5694–5706.
2
218.
71 J. Dedic, H. I. Okur and S. Roke, Sci. Adv., 2019, 5,
eaay1443.
72 K. Tych and L. Dougan, in Proteins in Solution and at
Interfaces: Methods and Applications in Biotechnology and
4
8 J. S. Fowler, G. Wang, J. Logan, S. Xie,
N. Volkow, R. R. MacGregor, D. Schlyer, N. Pappas,
D. Alexoff and C. Patlak, J. Nucl. Med., 1995, 36, 1255–
1
262.
Photochem. Photobiol. Sci.
This journal is © The Royal Society of Chemistry and Owner Societies 2020

View Article Online
Photochemical & Photobiological Sciences
Paper
Materials Science, ed. J. M. Ruso and Á. Piñeiro, 2013, pp.
91 M. A. Nichols and M. J. Waner, J. Chem. Educ., 2010, 87,
952–955.
92 D.-J. Wang, Y.-F. Kang, C.-Y. Zheng and X.-H. Wei, Res.
Chem. Intermed., 2013, 39, 2311–2320.
3
15–334, DOI: 10.1002/9781118523063.ch15.
3 J. B. Milstien, S. M. Lemon and P. F. Wright, J. Infect. Dis.,
997, 175, S247–S253.
7
7
1
4 R. Wu, M.-M. Georgescu, F. Delpeyroux, S. Guillot,
J. Balanant, K. Simpson and R. Crainic, Vaccine, 1995, 13,
93 H. Kumari, A. Eisenhart, J. Pajoubpong, F. Heinrich and
T. L. Beck, RSC Adv., 2020, 10, 15148–15153.
94 D. C. Nonhebel, Tetrahedron, 1968, 24, 1869–1874.
95 N. N. Shapet’ko, Y. S. Bogachev and S. A. Khatipov, Zh. Fiz.
Khim., 1988, 62, 2632–2648.
96 R. P. Bell, The proton in chemistry, Chapman and Hall,
London, 2d edn, 1973.
97 ISO 24443:2012, Determination of sunscreen UVA photo-
protection in vitro, 2012.
1
058–1063.
7
7
7
5 Y. M. Efimova, S. Haemers, B. Wierczinski, W. Norde and
A. A. V. Well, Biopolymers, 2007, 85, 264–273.
6 H. Schott, J. Macromol. Sci., Part B: Phys., 1988, 27, 119–
1
23.
7 T. Bourdier, C. J. R. Fookes, T. Q. Pham, I. Greguric and
A. Katsifis, J. Labelled Compd. Radiopharm., 2008, 51, 369–
3
73.
98 T. A. Darwish, N. R. Yepuri, P. J. Holden and M. James,
Anal. Chim. Acta, 2016, 927, 89–98.
99 G. R. Fulmer, A. J. M. Miller, N. H. Sherden,
H. E. Gottlieb, A. Nudelman, B. M. Stoltz, J. E. Bercaw and
K. I. Goldberg, Organometallics, 2010, 29, 2176–2179.
7
7
8
8
8 J. Emsley, Complex Chemistry: The composition, structure
and hydrogen bonding of the β-diketones, 1984, pp. 147–191.
9 A. L. Allred and D. W. Thompson, J. Phys. Chem., 1971, 75,
4
33–435.
0 D. B. Dahlberg and F. A. Long, J. Am. Chem. Soc., 1973, 95, 100 R. Murashige, Y. Hayashi, S. Ohmori, A. Torii, Y. Aizu,
3
825–3831.
1 T. Riley and F. A. Long, J. Am. Chem. Soc., 1962, 84, 522–
26.
Y. Muto, Y. Murai, Y. Oda and M. Hashimoto, Tetrahedron,
2011, 67, 641–649.
101 P. Gopalan, H. E. Katz, D. J. McGee, C. Erben, T. Zielinski,
D. Bousquet, D. Muller, J. Grazul and Y. Olsson, J. Am.
Chem. Soc., 2004, 126, 1741–1747.
5
8
8
2 T. K. Leipert, Org. Magn. Reson., 1977, 9, 157–159.
3 B. L. Zhang, F. Mabon and M. L. Martin, J. Phys. Org.
Chem., 1993, 6, 367–373.
102 F. Tinto, D. Pagano and E. Manzo, Tetrahedron, 2015, 71,
4379–4384.
103 C. Bosset, R. Coffinier, P. A. Peixoto, M. El Assal,
K. Miqueu, J.-M. Sotiropoulos, L. Pouységu and
S. Quideau, Angew. Chem., Int. Ed., 2014, 53, 9860–9864.
8
8
8
8
8
8
4 Please see the supporting information of: M. A. Nichols
and M. J. Waner, J. Chem. Educ., 2010, 87, 952–955.
5 M. J. T. Robinson, K. M. Rosen and J. D. B. Workman,
Tetrahedron, 1977, 33, 1655–1661.
6 A. I. Kol’tsov and B. A. Ershov, J. Org. Chem. USSR, 1975, 104 S. Yamabe, N. Tsuchida and K. Miyajima, J. Phys. Chem. A,
1, 440. 2004, 108, 2750–2757.
7 S. Bolvig and P. E. Hansen, Magn. Reson. Chem., 1996, 34, 105 B. Casier, N. Sisourat, S. Carniato and N. Capron, Theor.
67–478. Chem. Acc., 2018, 137, 107.
8 P. E. Hansen, E. V. Borisov and J. C. Lindon, Spectrochim. 106 S. R. Domingos and M. Schnell, J. Phys. Chem. Lett., 2018,
Acta, Part A, 2015, 136, 107–112. 9, 4963–4968.
9 T. Tsukahara, K. Nagaoka, K. Morikawa, K. Mawatari and 107 J. N. Spencer, E. S. Holmboe, M. R. Kirshenbaum,
1
4
T. Kitamori, J. Phys. Chem. B, 2015, 119, 14750–
4755.
D. W. Firth and P. B. Pinto, Can. J. Chem., 1982, 60, 1178–
1182.
1
9
0 D. Prettypaul, PhD: Photophysics and skin penetration of 108 Y. Manolova, V. Deneva, L. Antonov, E. Drakalska,
active agents in a commercial sunscreen and insect repellent,
Rutgers, The State University of New Jersey, 2018.
D. Momekova and N. Lambov, Spectrochim. Acta, Part A,
2014, 132, 815–820.
This journal is © The Royal Society of Chemistry and Owner Societies 2020
Photochem. Photobiol. Sci.