Energy and electron transfer reactions of polyenic acids with variable chain lengths

Article abstract of DOI:10.1016/j.tet.2012.10.041

Highly unsaturated conjugated fatty acids (carotenoic acids) with 5-15 conjugated double bonds were irradiated by flash pulses in the presence of the sensitizer riboflavin. The resulting triplet and radical cation transients were analyzed and the underlyi

Full text of DOI:10.1016/j.tet.2012.10.041

Tetrahedron 69 (2013) 219e227
Contents lists available at SciVerse ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Energy and electron transfer reactions of polyenic acids with
variable chain lengths
,
,
, ,
,
,
*
Asma Zaidi a y, He Li a y, Hans-Richard Sliwka a y, Vassilia Partali a y, Hansgeorg Ernst b, Thor B. Melø c
*
a Department of Chemistry, Norwegian University of Science and Technology, 7491 Trondheim, Norway
b BASF SE, Fine Chemical Research, 67056 Ludwigshafen, Germany
c Department of Physics, Norwegian University of Science and Technology, 7491 Trondheim, Norway
a r t i c l e i n f o
a b s t r a c t
Article history:
Highly unsaturated conjugated fatty acids (carotenoic acids) with 5e15 conjugated double bonds were
irradiated by flash pulses in the presence of the sensitizer riboflavin. The resulting triplet and radical
cation transients were analyzed and the underlying energy and electron transfer reactions determined
experimentally and by calculation. The two key antioxidant reactions of carotenoids summarize in ar, the
ratio of the reaction constants for electron and energy transfer. The optimal antioxidant ratio ar cumu-
lated in the polyenic acid with 11 double bonds. The photophysical formation of radical cations was
chemically ascertained by reaction of the acids with FeCl3.
Received 12 June 2012
Received in revised form
25 September 2012
Accepted 15 October 2012
Available online 22 October 2012
Keywords:
Antioxidant
Ó 2012 Elsevier Ltd. All rights reserved.
Carotenoids
Electron transfer
Energy transfer
Carboxylic acids
1. Introduction
for diverse functions, e.g., 11 double bonds for singlet oxygen
quenching, 7e11 or 9 and 11 double bonds for photobleaching pro-
The electron-rich carotenoids are eminent electron donators as
well as distinguished triplet energy acceptors. The combination of
these properties established the reputation of carotenoids as potent
antioxidants. Antioxidation can be characterized (a) by electron
transfer, i.e., the carotenoid delivers an electron to a molecule de-
preciated of electrons, or (b) by energy transfer, i.e., the carotenoid
combats oxygen in its highly reactive singlet electron configuration
taking over oxygen’s excessive energy. Other possible antioxidant
mechanisms operate to a minor degree with carotenoids (radical
addition, H-transfer, chemical oxygen quenching) or occur with
specific radicals ðO2$ꢀÞ.1e3 The antioxidant effect of carotenoids has
been determined (a) by varying the peripherals (rings, substituents)
keeping the polyene chain constant or (b) by keeping the peripherals
constant varying the chain lengths.4,5 The first method ranked
astaxanthin and dihydroxyisorenieratene as superior antioxidants;6,7
the result of the second method is not uniquely identified. However,
it is commonly assumed the longer a polyene chain the better the
antioxidant properties. On the other hand, systematic measurements
with different carotenoids have suggested an optimal chain length
tection, 10 or 9e11 double bonds for light harvesting function, and 7
double bonds for photon-to-current conversion (solar cell).8e14
Since an all-purpose method for determining the total antioxi-
dant activity (TAA),15 total antioxidant capacity (TAC),16 or total
antioxidant potential (TAP)17 is lacking the results depend above all
on the test system.18 The common antioxidant acronyms largely
ignore carotenoids as a singlet oxygen quencher (energy transfer
reaction).19 Our own antioxidant measurements showed to either
energy or electron transfer but did not encompass both actions.20e25
A theoretical study with 28 miscellaneous carotenoids established
the 1O2-quenching constant, the ionization energy and the
HOMOeLUMO energy difference as distinct descriptors for the an-
tioxidant property.26 Another work concentrated only on electron
transfer processes and related the antioxidant properties to the re-
dox potentials.27 We concentrated in this study on both the energy
and electron transfer reaction of a series of carotenoic acids (Car)
specified as Cn (n¼number of carbon atoms) with N C]C bonds
Cn:N, i.e., C20:5eC45:15. The compounds are interrelated; the only
variable is the number of double bonds (Scheme 1).
Generally, Car do not easily react with other organic compounds.
Therefore, Car are not affected by riboflavin (Rib), tocopherol (Toc),
chlorophyll (Chl), nitronaphthalene (NN), or other sensitizers
(Sen).20,28e30 However, when these sensitizers are irradiated, they
* Corresponding authors. E-mail addresses: hrs@nvg.ntnu.no (H.-R. Sliwka),
thor.melo@ntnu.no (T.B. Melø).
y
Fax: þ47 73596155.
0040-4020/$ e see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.tet.2012.10.041

220
A. Zaidi et al. / Tetrahedron 69 (2013) 219e227
Scheme 1. Carotenoic acids (Car) specified as Cn (n¼number of carbon atoms) with N C]C bonds: Cn:N. Particular names: C20:5 retinoic acid, C35:11 neurosporaxanthin, C40:13
torularhodin.
become excited to their high energetic triplet state, e.g., 3Rib, 3Toc, 3Chl,
3. The antioxidant ratio ar¼kel/ken
3NN, 3Sen; these triplet species can now react with Car: Car releases
4. The optimal chain length for ar
either an electron to 3Rib developing the radical ions Carꢁþ and Ribꢁꢀ or
the energy of 3Rib is transferred to Car forming 3Car (Scheme 2).31
2. Results and discussion
Synthesis. Acids C20:5eC35:11 were produced by BASF.38 C40:13
and C45:15 were synthesized by subsequent Wittig reactions from
the relevant aldehyde precursors (Experimental section).
Absorption spectra. The broad absorption bands of the shorter
acids resolve gradually into fine structural features with increasing
Scheme 2. The predominant channels of carotenoid antioxidant reactions; kel and ken
are the reaction constants for electron and energy transfer. The antioxidant ratio is
expressed as ar.
chain length (Fig. 1). A similar spectroscopic pattern has been ob-
served with other long chain carotenoids.39e41 The longer the
conjugation the smaller was the increase in lmax of Car and 3Car,
While the assignment of 3Car is unproblematic,32 the spectro-
whereas lmax of Carꢁþ increased linearly with N (Fig. 2).
scopic characterization of other Car transients is ambiguous. Carꢁ,
The development of chemically generated Carꢁþ was monitored
Carnþ, Carꢁþ, Carꢁꢀ, Carne all appear in the same bathochromic
visually (solutions turned to red, blue and black) and by VIS spec-
troscopy (Figs. 3 and 4, 1SD). Table 1 collects lmax or 0)0 vibra-
wavelength region.33e36 For that reason, the photophysical genera-
ꢁ
tion of Carþ was validated by chemically generated cation radicals:
tional transitions (assigned by deconvolution of the experimental
line shapes) for Car, 3Car and Carꢁþ. The values of lmax largely agree
Car þ Fe3þ/Carꢁþ þ Fe2þ 37
:
for photophysically and chemically created Carꢁþ confirming the
identity of the generated radical cations (Table 1).
We specifically examine in this work:
The flash photolysis of the Rib/Car solution generated four
species (Scheme 2, Fig. 12, Experimental section). Subtracting the
spectra of 3Rib and Ribꢁꢀ from the spectrum of the compound
mixture uncovered the neat spectra of 3Car and Carꢁþ as
1. The absorption spectra of Car, 3Car and Carꢁþ relative to N
2. The energy and electron transfer rate constants kel and ken

A. Zaidi et al. / Tetrahedron 69 (2013) 219e227
221
1,0
0,5
0,0
A
300
400
500
600
λ
/nm
Fig. 4. Formation of C35ꢁþ
845 nm. Spectra recording in minute intervals.
(
lmax¼958 nm) with FeCl3 in CH2Cl2. Secondary product at
Fig. 1. UVevis spectra of carotenoic acids C20 d, C22
C30 , C32 , C35 , C40 and C45
, C25
, C27
,
dissolved in MeCN/H2O¼9:1.
The spectra are normalized to unity at lmax for better comparison. Spectra of C27eC40
acids have been previously measured in petroleum ether.42
Table 1
Wavelength at maximum absorption (lmax) or 0)0 vibrational transition of Car, and
lmax of 3Car, and Carꢁþ (photophysical generation¼light, chemical formation¼Fe);
nd¼not determined due to low compound amounts, decomposition, experimental
or instrumental limitation
Car Cn
N
Car nm
3Car nm
Carꢁþ (light)
MeCN nm
Carꢁþ (Fe) CH2Cl2 nm
20
22
25
27
30
32
35
40
45
5
6
7
8
9
10
11
13
15
360
388
416
440
464
484
503
525
539
434
455
466
485
502
nd
535
557
586
573
642
714
763
821
nd
920
nd
nd
568
633
725
765
850
891
958
nd
nd
Fig. 2. Increase of lmax with the number of double bonds N.
,
,
(light,
MeCN), Carꢁþ (FeCl3, CH2Cl2).
0,1
0,0
A
-0,1
400
600
800
λ
/nm
Fig. 5. Spectra of C30-transients in MeCN/H2O 9:1 after irradiation of the Car/Rib
solution at various delays (5
m
s d, 15
m
s
, 40
m
s
, 80
m
s
) between the
irradiating pump flashes and the probe pulse. 3C30 lmax¼510 nm, C30ꢁþ lmax¼815 nm.
The Rib transients were subtracted from the spectral traces of the compound mixture.
3Rib formed immediately after irradiation and was rapidly con-
Fig. 3. Formation of C20ꢁþ
(
lmax¼568 nm) with FeCl3 in CH2Cl2. The decay of C20ꢁþ is
sumed by reaction with C30:9 to 3C30 and C30ꢁþ. After 30 s 3C30
m
accompanied with the growth of oxidation products at 470 nm and 525 nm. Spectra
had practically disappeared whereas the formation of C30ꢁþ had
recording in minute intervals. Figure on formation of C30ꢁþ in Ref. 37.
reached its maximum, then decaying in about 200
ms to un-
identified products (Fig. 7).
exemplified for C30:9 (Fig. 5): rapid formation of 3C30 at 510 nm
was observed, its fast decay, and the concurrent development of
longer living C30ꢁþ at 815 nm. The spectra of all 3Car and Carꢁþ are
shown in Fig. 6. The peak heights of 3Car were similar, while those
of Carꢁþ increased with polyene chain length.
The lifetime (20 min) of chemically produced C30ꢁþ was estab-
lished by measuring the gradual decrease of its absorption
(Fig. 8).37 Carꢁþ was sufficiently stable for detection by capillary
electrophoresis; shorter chain Carꢁþ migrated faster than longer
chain Carꢁþ. The increased stability of the chemically generated
Carꢁþ may arise from concentration and solvent effects, or the
presence of O2 and counter ions.
Kinetic measurements. Formation and decay of the transient
species after flash photolysis (Scheme 2) are typified with C30:9.

222
A. Zaidi et al. / Tetrahedron 69 (2013) 219e227
2,2
2,0
1,8
1,6
0,4
0,2
0,0
A
E
,4
eV
,2
1,0
0,8
0,6
4
6
8
10 12 14 16 18 20 22
N
400
600
800
λ
/nm
Fig. 9. Energetic relation between 3Car
, and redox potential E(Carꢁþ/Car)
Fig. 6. 3Car and Carꢁþ spectra (MeCN/H2O¼9:1) of C20 d, C22
,
C25
,
in relation to N.
C27
, C30
, C35
and C40
recorded at a delay of 15 ms between the
pump and the probe pulse. The intensity of the pump flashes and the concentrations of
Rib and Cn were identical.
1,0
a
r
0,10
laser
0,5
0,0
pulse
A
0,05
0,00
5
6
7
8
9
10 11 12 13 14 15
N
Fig. 10. Calculated
lated to the number of double bonds N.
and experimental
antioxidant ratio ar re-
0
30
60
t /μs
Fig. 7. Kinetic traces (average of absorbance vs time after 2 laser pulses into solutions
0.8
of two different concentrations in MeCN/H2O¼9:1) for 3Rib
at 680 nm, 3C30
at 510 nm, and C30ꢁþ
at 815 nm. The traces were fitted to the absorption graphs
A
with a single exponential for 3Rib and a double exponential for 3C30 and C30ꢁþ
.
0.4
0
300
500
λ/nm
Fig. 11. Vis-spectra of C45-acid in benzene
, addition of I2 d, radiation: 10 min
, 20 min
, 30 min
, 70 min
.
0,2
A
0,1
0,0
400
600
800
λ /nm
Fig. 8. Slow decay of C30ꢁþ (from C30þFeCl3 in CH2Cl2).
Fig. 12. Spectra of Rib (dee ), 3Rib (
), and Ribꢁꢀ
(
).

A. Zaidi et al. / Tetrahedron 69 (2013) 219e227
223
2.1. Determination of the reaction constants, the antioxidant
ratio ar and the optimal chain length
seen in Fig. 7. The kinetic data of the other Car were calculated
analogously. 3Rib reacts faster with longer chain Car (increasing
quenching constant K, Table 2); the various quenching constants K
suggest diverse reaction processes depending on the number of
double bonds. The constant kel was determined similarly, taking into
account the lifetime of Carꢁþ. With kel and ken at hand the two major
antioxidant mechanisms of Car (energy and electron transfer) could
be summarized as antioxidant ratio ar¼kel/ken (Table 2).
The reaction constants kel and ken (Scheme 2), exemplified for
C30:9, were established from the sequences:
k
3
unimolecular decay of 3Rib :
Rib / Rib
bimolecular decay by electron transfer :
2.2. Theoretical calculation of the reaction constants, the
antioxidant ratio ar and the optimal chain length
ꢁꢀ
kel
ꢁ
3Rib þ C30 / Rib þ C30 þ
bimolecular decay by energy transfer :
Given that the transfer reactions in polyenes are diffusion-
limited, the rate constant kd was calculated with:
ken
3
3Rib þ C30 / Rib þ C30
kd ¼ 4
p
$RDA$ðDD þ DAÞ$NA$1000
(6)
Kt
3
unimolecular decay of 3C30 :
C30 / C30
RDA, the hydrodynamic radius of the Care3Rib encounter complex,
is the sum of the radii of the donor D and the acceptor A:
RDA¼RDþRA. NA is the Avogadro number and D represents the dif-
fusion constant of the compound, which for a presumed spherical
Kd
unimolecular decay of C30ꢁþ
:
C30ꢁþ / C30:
3Rib decays exponentially in the presence of C30:9 with a quench-
ing constant
molecule of radius R is D¼kBT/6
pRh, where h stands for the vis-
ꢀ
ꢀ
cosity of the solvent. Applying RDz7.5 A (Rib), RAz2.5 A (hydro-
dynamic radius of Car estimated from data given in Ref. 24) and
hMeCN¼0.4ꢃ10ꢀ3 Ns/m2 gave kd¼1.4ꢃ1010 l/mol s.
K ¼ ðKen þ KelÞ$½C30ꢂ þ k:
(1)
The transfer reactions proceed in three steps: diffusion advances
the dispersed molecules to proximity enabling electron or energy
exchange forming tightly located products, which then separate:
Hence, the sum of electron and energy transfer reaction con-
stants (kel and ken) can be found when the decay constant of 3Rib is
measured for different C30 concentrations (Table 2). The time-de-
pendence of 3C30 decay was derived from
kd
#
kt
#
ks
A þ D
A/D
Aꢀ/Dþ / Aꢀ þ Dþ
k
k
ꢀt
ꢀd
ꢀ
ꢁ
k
en$½3Rib0ꢂ$½C30ꢂ
Kt ꢀ K
½3C30ðtÞꢂ ¼
$ eꢀK$t ꢀ eꢀK $t
t
The diffusion- and back diffusion-constants are denoted kd and
ꢀd, the transfer- and back transfer rate constants kt and ket, and
(2)
ꢀ
ꢁ
k
hp$ eꢀK$t ꢀ eꢀK $t
t
the rate constant for product separation ks.
Using the steady state approximation for intermediate concen-
trations, the formation rate F of charged species is expressed as:
Based on this expression the kinetic trace for the absorbance of
3C30 was fitted to a difference of two exponentials multiplied by
a parameter P¼εT$p.
F ¼ ðkd=kt þ k Þ$ðkt=ks þ kꢀtÞ$kszkd:
(7)
ꢀd
ꢀ
ꢁ
3
Að C30ðtÞÞ ¼ εT$½3C30ðtÞꢂ$l ¼ εT$p$ eꢀK$t ꢀ eꢀK $t $l
(3)
t
When the rate constant for electron back transfer k is small
ꢀt
compared to the rate constant for separation ks, and when the rate
constant for back diffusion kꢀd is smaller than that of transfer kt, F is
alike the rate constant kd for a diffusion-limited reaction. Taking the
transfer reactions into consideration F is expressed as:
(optical path length l¼1 cm)
The relation between P and ken was set to
εT$ken$½3Rib0ꢂ$½C30ꢂ
P ¼
(4)
(5)
FðelÞ ¼ ðkel=kel þ kenÞ$kd
and
(8)
Kt ꢀ K
with
FðenÞ ¼ ðken=ðkel þ kenÞ$kd
(9)
1
P$ðKt ꢀ KÞ
εG P$ðKt ꢀ KÞ
ken
¼
$
¼
$
:
½3Ribꢂ$½C30ꢂ
½3Ribꢂ$½AC30
ꢂ
εT
εT
Mataga’s Golden rule is applicable to non-adiabatic radiationless
transitions with electronic interactions (V) smaller than kBT, and
with electron- or energy transfer in the weak coupling category
(V<<kBT):43
The energy transfer constant ken resulted from the fitting routine,
inserting the absorbance of C30 and the initial concentration of 3Rib,
which was determined spectroscopically at 680 nm with the ex-
tinction coefficient ε¼6.0ꢃ103 Mꢀ1 cmꢀ1. The fitted kinetic traces are
ꢂ
ꢃ
ꢂ
ꢂ
ꢃꢃ
1=2
p
Sn
n!
X
ktrans
¼
$V2
eꢀS
$exp
Z2$
l
s$kBT
n
Table 2
!
(10)
ð
D
G0 þ ls þ nhvÞ2
Quenching constant K, experimental reaction constants kel and ken, and experi-
mental and calculated kel/ken ratio ar
ꢀ
4
lskBT
Car Cn
N
K Rib$109 kel Car$109 ken Car$109 kel/ken exper. kel/ken calcd
C22
C25
C27
C30
C35
C40
C45
6
2.5
2.7
3.9
5.0
0.41
0.50
0.75
0.89
1.20
1.5
2.4
1.8
1.3
1.1
0.9
1.3
2.5
0.17
0.27
0.57
0.81
1.33
1.15
0.42
0.13
0.27
0.52
0.85
1.14
1.08
0.42
(ktrans¼energy or electron transfer)
The solvent reorganization energy ls corresponds to the elec-
trical work needed to reorganize the solvent molecules surround-
ing the two charged species in the electron transfer reaction;
7
8
9
11 10
13 20
15 26
S¼li/h
n represents the inner reorganization energy, expressed as
the crossing point of the vertical line from the minimum of the
1.07

224
A. Zaidi et al. / Tetrahedron 69 (2013) 219e227
reactant’s energy parabola to the product parabola,
n
is the aver-
Car support previous outcomes on optimal number of double bonds
for other carotenoids. It is striking to note that carotenoids with 12
or 13 double bonds occur rarely, the longest natural carotenoid
reaches 14C]C bonds.48
aged frequency for the nuclear motion in the parabolic potential
energy surfaces,
G0 denotes the free energy change, which for
triplet energy transfer is
D
3
3
D
G0 ¼ Eð CarÞ ꢀ Eð RibÞ
(11a)
3. Conclusion
with E(3Rib)¼2.0 eV44
Summarizing reactions, tables, and figures the following pecu-
liarities were found: Carotenoic acids (Car) with 5e15 double bonds
can both reduce triplet sensitizers and FeCl3. The spectroscopic
nature of Carꢁþ, obtained by photophysical or chemical means, is
identical. The two major antioxidant actions of carotenoids, elec-
tron, and energy transfer, sum up in a comprehensive expression,
the antioxidant ratio ar¼kel/ken. Calculated and experimentally
derived ar of the investigated polyene acids series reached a maxi-
mum with 11 conjugated double bonds.
0
3
D
G
¼ E Car ꢀ 2:0 eV
ð Þ
(11b)
The energetic relation between 3Car and N was found with
ꢀ
ꢁ
0:26
3
Eð CarÞ ¼ 7:25 1=N
ꢀ 0:34$ð1 þ 1=NÞ eV
(12)
and reproduced the measured triplet levels as a function of N45
(Fig. 9).
When applyingV¼0.024 eV, li¼0.3 eV, h ¼0.15 eV, and ls¼0.40eV
n
4. Experimental section
4.1. Synthesis
to Eq. 10 the energy transfer rates ken were obtained. The electron
transfer rate kel was calculated from the RehmeWeller equation for
the free energy change in an electron transfer reaction:46
D
G ¼ ½EðCarꢁþ=CarÞ þ f ꢀ EðRibꢁꢀ=RibÞꢂ ꢀ e2=4
p
e$rDA
þ
ls
Acids C20:5eC35:15, C30-aldehyde (C30ald), the C5 Wittig salts
C5aldP and C5esterP were produced by BASF.38 C40- and C45-acid
were synthesized by subsequent Wittig reactions42 from C30ald
(Schemes 3 and 4) as exemplified for C45-acid: C30ald was reacted
with the protected aldehyde Wittig salt C5aldP to C35-aldehyde
(C35ald), which was extended with C5aldP to C40ald; which
again was prolonged with C5esterP to C45-ester; hydrolysis affor-
ded the acid C45:15.
3
ꢀ Eð RibÞ
¼ ½EðCarꢁþ=CarÞ þ 0:24 eV ꢀ ðꢀ0:3 eVÞꢂ ꢀ 0:7 eV þ 0:7 eV
ꢀ 2:0 eV
¼ EðCarꢁþ=CarÞ ꢀ 1:46 eV:
(13)
The calculation was performed with values for redox potentials
4.1.1. C35-aldehyde.40 A solution of C30ald (0.2 g, 0.67 mmol) in
dry MeOH (10 mL), phosphonium salt C5aldP (1.7 g, 3.4 mmol) in
dry MeOH (20 mL), and MeONa (0.2 g, 3.4 mmol) in dry MeOH
(10 mL) reacted at reflux temperature for 24 h. The solvent was
removed and H2O was added followed by repeated extraction of the
aqueous phase with CH2Cl2. The combined extracts were dried over
MgSO4, the solvent was evaporated, the residue dissolved in ace-
tone and p-toluenesulfonic acid (0.01 g) was added. After 2 h stir-
ring the solvent was evaporated, H2O was added, the mixture
extracted with CH2Cl2, and the combined extracts were dried over
MgSO4. Flash chromatography (hexane/ethyl acetate, 3:1) of the
measured versus the standard hydrogen electrode (SHE):
E(Ribꢁꢀ/Rib)¼ꢀ0.3 eV.44 The Coulomb term is the electric energy
gained fromthe two opposite charges e at distance rDA, thedistanceof
the closest approach of the electron donor Carand the acceptor Rib. In
MeCN with a dielectric constant ε¼37.5 the Coulomb term is esti-
mated 0.7 eV; the same value is taken for the reorganization energy
ls. The redox potential E(Carꢁþ/Car) changes with N. Redox potentials
have previously been evaluated with polyenes against the saturated
calomel electrode.47 Comparability was achieved by adding a factor
f¼0.24 eV to the SHE related values, which is the reference in the
redox potential measurements of Rib. The found kel-values were then
divided by the rates of ken; the result is collected in Table 2.
residue gave C35ald (0.198 g, 85%). HRMS: calcd for C35H46
482.3549; found 482.3565.
O
For better comparison the previously obtained experimental
data were reproduced by expressing redox potentials (Fig. 9):
4.1.2. C40-aldehyde (torularhodinaldehyde).40 The aldehyde was
synthesized analogously from C30ald (0.1 g, 0.21 mmol) in dry
MeOH (10 mL), phosphonium salt C5aldP (1.0 g, 1.9 mmol) in dry
MeOH (20 mL), and MeONa (0.1 g, 1.7 mmol) in dry MeOH (10 mL).
Chromatography work-up afforded C40ald (60 mg, 53%). HRMS
calcd for C40H52O 548.4018; found 548.4015.
EðCarꢁþ=CarÞ ¼ 0:51 eV þ 0:70 eV$eꢀ0:13Nꢀ3
(14)
When the redox potential values were applied in Eq. 14, the
resulting electron to energy transfers ratios were smaller by a factor
of two compared with the experimental data. The deviation may
originate from differences in ls for the individual Car. The shorter
Car have a slightly larger ls than the longer chain Car, a linear re-
lation between ls and the length of the molecule was assumed.
4.1.3. C45:15 acid. C40ald (23.5 mg, 0.043 mmol) in dry THF
(20 mL) was reacted with C5esterP (74.5 mg, 0.145 mmol) in dry
MeOH (20 mL) and MeONa (7.89 mg, 0.145 mmol) in dry MeOH
(10 mL). The mixture was refluxed for 48 h and gave after work-up
C45-ester, which was dissolved in methanol (30 mL). A solution of
25% KOH/MeOH (10 mL) was added and the mixture refluxed for
2 h.49 The solvent was removed and the residue washed with 15%
H2SO4. The aqueous layer was extracted with CH2Cl2 and the
combined organic phases were washed repeatedly with H2O.
Concentration under vacuum and purification by flash chroma-
tography on silica gel (hexane/acetone 6:1) gave C45:15 (12 mg,
When the expression for
DG was revised by inserting a term-
modulating solvent reorganization energy
Dls ¼ ꢀ0:1ðN ꢀ 7:5Þ eV
(15)
a satisfactory agreement concerning the calculated and experi-
mental transfer rates kel and ken was achieved (Fig. 10).
2.3. Comparison of experimental and calculated data
44%). 1H NMR (400 MHz, DMSO-d6):
(m, 2H, C34), 1.60 (m, 2H, C33), 1.74 (s, 3H, CH3, C31), 1.79 (s, 6H,
2CH3, C23, C27), 1.87 (m, 12H, 4CH3, in chain), 2.0 (m, 2H, C32), 2.05
(s, 3H, CH3, C2), 6.16e7.26 (m, 21H, olefin protons), 12.24 (acid
d
¼1.07 (s, 6H, 2CH3, C35), 1.48
The calculated antioxidant ratio ar¼kel/ken (Table 2) is in accor-
dance with the experimentally determined ar, both reaching the
greatest magnitude with 11C]C bonds (Fig. 10). Our results with

A. Zaidi et al. / Tetrahedron 69 (2013) 219e227
225
Scheme 3.
corresponding aldehydes with Wittig salt C2esterP (representative
synthesis in Scheme 4):
C20ald þ C2esterP6C22 : 6
C20ald þ C5esterP6C25 : 7
C25ald þ C2esterP6C27 : 8
C30ald þ C2esterP6C32 : 10
C30ald þ C5esterP6C35 : 11:
The acids C22, C25, C27, C35, and C40 have been previously
synthesized by analogous syntheses.8
Scheme 4.
proton), for convenience the common carotenoid numbering has
been abandoned (Scheme 3), IR: 1671 cm-1, HRMS: calcd for
C45H58O2 630.4433; found 630.4437.
4.2. cis/trans-Isomers
The occurrence of cis-isomers, common for short chain poly-
enes, becomes unlikely with longer chain lengths (the longer the
chain the more trans).51e53 The regular increase of lmax with N
supports the formation of all trans-isomers (Fig. 2, 2SD). The ab-
sence of spectral changes during the classical cis/trans isomeri-
zation experiment of C45:15 with I2 and light was an additional
indication for the predominance of the all-trans isomer: C45:15 was
dissolved in benzene, a minute amount of I2 in benzene was added
and the solution exposed to bright sunshine; spectra were recorded
until decomposition after 70 min (Fig. 11).54,55
4.1.4. C40:13 acid (torularhodin).42 The acid was synthesized
analogously from C30ald with C5aldP followed by hydrolysis of the
obtained C40:13 ester (8.4 mg, 75%). HRMS: calcd for C40H52O2
564.3967; found 564.3964. The NMR-data of the acids were com-
parable to the values given in Refs. 49 and 50.
4.1.5. Other acids. C25:7, C35:11, and C40:13 were synthesized
similarly. C22:6, C27:8, and C32:11 were obtained from the

226
A. Zaidi et al. / Tetrahedron 69 (2013) 219e227
4.3. Chemical formation of CarꢁD
(triplet minus singlet spectra, TmS-Rib) and the spectrum of 3Rib
arise from similar concentrations, the extinction coefficient ε at
lmax of 3Rib was estimated in relation to the known extinction
coefficient of Rib ε¼1.10ꢃ104 Mꢀ1 cmꢀ1 at 442 nm.58 The ratio
between the heights of the 3Rib absorption at 700 nm and the Rib
absorption at 442 nm is 0.50 and hence εz0.55ꢃ10ꢀ4 Mꢀ1 cmꢀ1 for
3Rib. These ε-values were needed in the calculation of the rate
consꢀtants kel and ken. The Ribꢁꢀ spectrum was verified after adding
NO2 to the Rib-solution. Irradiation formed 3Rib, which immedi-
ately reacted to Ribꢁꢀ by an electron transfer reaction:
A drop of a FeCl3-solution (1.5% in CH2Cl2/acetone 50:1) was
added to a cuvette filled with a carotenoic acid dissolved in
CH2Cl2.37 The formation of C20ꢁþ, C22ꢁþ, and C25ꢁþ was seen by an
immediate color change; the absorption of longer chain Carꢁþ oc-
curred in the invisible wavelength range and the reaction solution
appeared black. Capillary electrophoresis of C25ꢁþ, C27ꢁþ, C30ꢁþ, and
C32ꢁþ was performed by co-injection with the corresponding Car
and acetophenone as neutral markers.37 Since the charge was
constant, shorter Carꢁþ migrated faster than longer Carꢁþ
.
3Rib þ NO2ꢀ/Ribꢁꢀ þ NO2
4.4. Absorption extinction coefficient, oxidation potential
ꢁ
ꢀ
As all 3Rib was converted with NO2 to Rib ꢀ its ε was derived
analogously to the calculation of 3Rib.
Absorption spectra were recorded using a Shimadzu UV-1601 PC
spectrophotometer. The cuvettes with the Car-solutions were
flushed with argon for 30 min and irradiated in an argon atmo-
sphere. The relative error in determining ε, mostly due to
the weighing, is less than 10%: εꢃ104 [Mꢀ1 cmꢀ1] C20:5 ε¼4.97,
C22:6 ε¼5.41, C25:7 ε¼6.87, C27:8 ε¼8.53, C30:9 ε¼10.20, C32:10
ε¼11.30, C35:11 ε¼11.90. The calculated oxidation potentials of
C20:5eC45:15 occur in the range Eox¼1.2e0.63 V (vs SHE).56,57
3C30 and C30ꢁþ were acquired by subtraction of the 3Rib and
Ribꢁþ spectra (TmS-Rib and DmS-Rib; Singlet S, doublet D and
triplet T designate the eigenstates adopted by the electron spin of
the molecule) from the spectral traces of the mixture at various
delays between the irradiating pump flashes and the probe pulse
(Fig. 5). The 3C30 absorption corresponds to the difference
spectrum recorded after 15e40
m
s, while the C30ꢁþ spectrum
is almost identical to the absorption after 80
ms. The ε-ratio
3C30/C30 is 1.5 with the corresponding ratio C30ꢁþ/C30¼1.8.
These ratios were virtually identical to the other 3Car and Carꢁþ
,
4.5. Transient spectra
were, therefore, likewise attributed to 3C32, C32ꢁþ, C40ꢁþ, and
C45ꢁþ
lmax not detectable, Table 1, Fig. 6), and were used for
calculating kel and ken
A miniaturized flash photolysis equipment, composed of com-
mercial and purpose built units, was used. The pump beam (ap-
proximately 5 ms) was generated by discharging a 2 mF capacitor
through a linear Xenon flash lamp (EG&G Model QXA 17) supplied
with 3 kV. The discharge of the pump flash was monitored by
a photomultiplier (PM) and the signal from the PM-tube was sent to
a delay generator (Princeton Instruments, Model PG200). The
output from the delay generator was triggering the probe lamp
(
.
The comparison of the 3Car and Carꢁþ spectra in Fig. 6 was
possible using similar intensities of the pump beams and equal
concentrations of Rib. Since the reactions in Scheme 2 are bi-
molecular, the rate of formation depends upon the product of 3Rib
and Car concentrations; therefore, small variations in quantities of
Car were adjusted.
E (eV) of Car (S0/S2), 3Car (T/T) and Carꢁþ (D/D) were
(EG&G Model FX-409U), which has a flash duration of 2 ms. Both
D
flashes were monitored by a storage oscilloscope (Le Croy 9410).
The probe light is fixed at a right angle to the pump beam and is
directed, after its emergence from the sample cuvette, by an optical
fiber to an optical multichannel analyzer (Zeiss, MCS 224). The
transient absorption spectrum was computed with a homemade
calculated from lmax for C20eC27 and from the 0)0 vibrational
transitions for C30eC45. When E is plotted against the inverse
number of double bonds 1/N the slopes (B factor) of the linear
graphs for Car and Carꢁþ
(D
E¼AþB/N) were found similar (Fig. 2SD).
software program: A(t,
l
i)¼lg[(RiꢀDi/SiꢀEi)] is the delay time be-
tween the pump and the probe flash set by the delay generator and
measured by the storage oscilloscope. Di is the dark current in di-
ode i, Ri is the intensity of the ith diode in the absence of pump
flashes, and Si in the presence of a pump flash. Ei is the spectrum
from the pump flashes only, and it contains essentially the resulting
4.6. Kinetic measurements
The time-dependence of the signal DA(l,t) at a fixed wavelength
was recorded in a kinetic spectrometer, in which the pump source
was the third harmonic of a pulsed Nd:YAG laser (BM Industries);
a 150 W tungstenehalogen lamp provided the monitoring light,
and a prism monochromator (Zeiss M4 QIII), placed behind the
sample compartment, served as the wavelength selector. The pump
and probe beams were perpendicular to each other. A Hamamatsu
photomultiplier (R928) with a 50-U terminating resistor was used
for detecting the light emerging from the exit slit of the mono-
chromator, and the output of the detector was recorded by
a 600 MHz digital storage oscilloscope (Agilent, Infinium 54830B).
The transients shown here are averages of two recordings.
Formation and decay of the C20eC45 transients are typified
with C30. C30ꢁþ appeared at 815 nm, the main contribution at
680 nm to the overall absorbance (A) is caused by 3Rib, at 510 nm
both 3C30 and 3Rib were absorbing (Figs. 6 and 12). The absorbance
of the transients were determined by:
fluorescence spectrum of the substance;
responding to diode number i.
li is the wavelength cor-
A transient absorption spectrum is defined as:
ꢄ
ꢅ
ꢆ ꢇ
DAðl; tÞ ¼ Aðl; tÞ ꢀ AðlÞ ¼ ep$cp$l þ eg$ c0 ꢀ cp $l ꢀ eg$c0$l
ꢅ
ꢆ
¼
ep ꢀ eg $cp$l
where A(
l
,t) is the absorbance at time t after the pump flash, and
A(l) is the absorbance before the flash. A certain concentration of
photoproducts cp is formed by the flash with extinction denoted εp,
at the expense of ground state molecules. The extinction co-
efficients of the photoproduct and ground state molecules are
denoted εp and εg, respectively, and the ground state concentration
is c0. Inserting the fraction a¼cp/c0 of the ground state absorption
spectrum to the transient absorption spectrum allowed to obtain
the spectrum of the photoproduct: Ap(l,t)¼DA(l,t)þa$A(l)¼ep$c.
The flash photolysis of the Rib/Car solution results in four spe-
cies (Scheme 2, Fig. 12): Initially, the pump flash generates 3Rib;
then the collision of 3Rib with Car forms Ribꢁꢀ, 3Car, and Carꢁþ. 3Rib
and Ribꢁꢀ spectra were acquired by exposing only Rib to pump
flashes. Since the ground state depletion spectrum of irradiated Rib
AðC30ꢁþð815ÞÞ ¼ Að815Þ ꢀ
$Að680Þ
a
3
Að Ribð680ÞÞ ¼ Að680Þ ꢀ
b
$AðC30ꢁþð815ÞÞ
3
3
Að C30ð510ÞÞ ¼ Að510Þ ꢀ
g
$Að Ribð680ÞÞ

A. Zaidi et al. / Tetrahedron 69 (2013) 219e227
227
23. Yanishlieva, N. V.; Marinova, E. M.; Raneva, V. G.; Partali, V.; Sliwka, H. R. J. Am.
Oil Chem. Soc. 2001, 78, 641e644.
24. Foss, B. J.; Sliwka, H. R.; Partali, V.; Cardounel, A. J.; Zweier, J. L.; Lockwood, S. F.
Bioorg. Med. Chem. Lett. 2004, 14, 2807e2812.
25. Naalsund, T.; Malterud, K. E.; Partali, V.; Sliwka, H. R. Chem. Phys. Lipids 2001,
112, 59e65.
Factor
relative to that at 680 nm (
DmS-Car spectrum at 680 nm relative to 815 nm (
the fraction of the TmS-Rib spectrum at 510 nm relative to 680 nm
a
is the fraction of the TmS-Rib spectrum at 815 nm
¼0.08); similarly, is the fraction of the
¼0.14) and is
a
b
b
g
(g
¼0.76). This procedure allowed the plotting of the kinetic traces
26. Sun, Y. J.; Pang-Jie; Ye, X. Q.; Lu-Yuan; Li-Jun. Chin. J. Struct. Chem. 2009, 28,
163e170.
for 3Rib, 3C30, and C30ꢁþ demonstrating the varying lifetime of the
species (Fig. 7).
ꢁ
ꢁ
27. Kleinova, M.; Hewitt, M.; Brezova, V.; Madden, J. C.; Cronin, M. T. D.; Valko, M.
Gen. Physiol. Biophys. 2007, 26, 97e103.
28. Pogson, B. J.; Rissler, H. M.; Frank, H. A. Adv. Photosynth. Respir. 2005, 22,
515e537.
Acknowledgements
29. Naqvi, K. R.; Melø, T. B.; Sliwka, H. R.; Mohamad, S. B. B.; Partali, V. Photochem.
Photobiol. Sci. 2003, 2, 381e385.
We thank Dr. S.V. Gonzales for recording the HRMS spectra of
C40:13 and C45:15.
30. Ramamurthy, V.; Liu, R. S. H. J. Am. Chem. Soc. 1976, 98, 2935e2942.
31. Tinkler, J. H.; Tavender, S. M.; Parker, A. W.; McGarvey, D. J.; Mulroy, L.; Truscott,
T. G. J. Am. Chem. Soc. 1996, 118, 1756e1761.
32. Becker, R. S.; Bensasson, R. V.; Lafferty, J.; Truscott, T. G.; Land, E. J. J. Chem. Soc.,
Supplementary data
Faraday Trans. 2 1978, 74, 2246e2255.
33. Naqvi, K. R.; Melø, T. B.; Javorfi, T.; Gonzalez-Perez, S.; Arellano, J. B. Phys. Chem.
Chem. Phys. 2009, 11, 6401e6405.
34. Gao, Y. L.; Shinopoulos, K. E.; Tracewell, C. A.; Focsan, A. L.; Brudvig, G. W.;
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.tet.2012.10.041.
Kispert, L. D. J. Phys. Chem. B 2009, 113, 9901e9908.
35. Kispert, L. D.; Konovalova, T.; Gao, Y. Arch. Biochem. Biophys. 2004, 430, 49e60.
36. Lafferty, J.; Roach, A. C.; Sinclair, R. S.; Truscott, T. G.; Land, E. J. J. Chem. Soc.,
Faraday Trans. 1 1977, 73, 416e429.
References and notes
37. Li, H.; Rebmann, E.; Øpstad, C. L.; Schmid, R.; Sliwka, H. R.; Partali, V. Eur. J. Org.
1. Edge, R.; Truscott, T. G. In The Photochemistry of Carotenoids; Frank, A. A., Young,
A. J., Britton, G., Cogdell, R. J., Eds.; Kluwer: New York, NY, 1999.
2. Fiedor, J.; Fiedor, L.; Haessner, R.; Scheer, H. Biochim. Biophys. Acta 2005, 1709, 1e4.
3. Galano, A.; Vargas, R.; Martinez, A. Phys. Chem. Chem. Phys. 2010, 12, 193e200.
4. El-Agamey, A.; Lowe, G. M.; McGarvey, D. J.; Mortensen, A.; Phillip, D. M.;
Truscott, T. G.; Young, A. J. Arch. Biochem. Biophys. 2004, 430, 37e48.
Chem. 2010, 4630e4636.
€
38. Krause, W.; Munster, P.; Ernst, H.; Paust, J.; Schulz, B.; Kochner, A.; Etzrodt, H.
DE19506999, BASF SE, 1996.
39. Karrer, P.; Eugster, C. H. Helv. Chim. Acta 1951, 34, 1805e1814.
40. Schwieter, U.; Gutmann, H.; Lindlar, H.; Marbet, R.; Rigassi, N.; Ruegg, R.;
€
Schaeren, S. F.; Isler, O. Helv. Chim. Acta 1966, 49, 369e390.
41. Andersson, P. O.; Gillbro, T. J. Chem. Phys. 1995, 103, 2509e2519.
42. Isler, O.; Guex, W.; Ruegg, R.; Ryser, G.; Saucy, G.; Schwieter, U.; Walter, M.;
Winterstein, A. Helv. Chim. Acta 1959, 42, 864e871.
5. Xiang, J. F.; Rondonuwu, F. S.; Kakitani, Y.; Fujii, R.; Watanabe, Y.; Koyama, Y.;
Nagae, H.; Yamano, Y.; Ito, M. J. Phys. Chem. B 2005, 109, 17066e17077.
€
6. Martin, H. D.; Kock, S.; Scherrers, R.; Lutter, K.; Wagener, T.; Hundsdorfer, C.; Frixel,
€
S.; Schaper, K.; Ernst, H.; Schrader, W.; Gorner, H.; Stahl, W. Angew. Chem., Int. Ed.
43. Mataga, N.; Chosrowjan, H.; Shibata, Y.; Yoshida, N.; Osuka, A.; Kikuzawa, T.;
2009, 48, 400e403.
Okada, T. J. Am. Chem. Soc. 2001, 123, 12422e12423.
44. Lu, C. Y.; Wang, W. F.; Lin, W. Z.; Han, Z. H.; Yao, S. D.; Lin, N. Y. J. Photochem.
Photobiol., B 1999, 52, 111e116.
7. Lockwood, S. F.; Gross, G. J. Cardiovasc. Drug Rev. 2005, 23, 199e216.
8. Wang, X. F.; Fujii, R.; Ito, S.; Koyama, Y.; Yamano, Y.; Ito, M.; Kitamura, T.;
Yanagida, S. Chem. Phys. Lett. 2005, 416, 1e6.
45. Kurashige, Y.; Nakano, H.; Nakao, Y.; Hirao, K. Chem. Phys. Lett. 2004, 400,
9. Akahane, J.; Rondonuwu, F. S.; Fiedor, L.; Watanabe, Y.; Koyama, Y. Chem. Phys.
Lett. 2004, 393, 184e191.
425e429.
46. Fournier, T.; Tavender, S. M.; Parker, A. W.; Scholes, G. D.; Phillips, D. J. Phys.
Chem. A 1997, 101, 5320e5326.
10. Ritz, T.; Damjanovic, A.; Schulten, K.; Zhang, J. P.; Koyama, Y. Photosynth. Res.
2000, 66, 125e144.
47. Heinze, J.; Tschuncky, P.; Smie, A. J. Solid State Electrochem. 1998, 2, 102e109.
48. Kushwaha, S. C.; Gochnauer, M. B.; Kushner, D. J.; Kates, M. Can. J. Microbiol.
1974, 20, 241e245.
49. Larsen, E.; Abendroth, J.; Partali, V.; Schulz, B.; Sliwka, H. R.; Quartey, E. G. K.
Chem.dEur. J. 1998, 4, 113e117.
11. Farhoosh, R.; Chynwat, V.; Gebhard, R.; Lugtenburg, J.; Frank, H. A. Photochem.
Photobiol. 1997, 66, 97e104.
12. Claes, H. Biochem. Biophys. Res. Commun. 1960, 3, 585e590.
13. Foote, C. S.; Chang, Y. C.; Denny, R. W. J. Am. Chem. Soc. 1970, 92, 5216e5218.
€
14. Krinsky, N. I. In Carotenoids; Isler, O., Ed.; Birkhauser: Basel, 1971; pp 676e678.
50. Englert, G. In Carotenoids; Britton, G., Liaaen-Jensen, S., Pfander, Eds.; Spec-
15. Rice-Evans, C. A. Free Radical Res. 2000, 33, S59eS66.
16. Young, I. S. J. Clin. Pathol. 2001, 54, 334e339.
€
troscopy; Birkhauser: Basel, 1995; Vol. 1B.
51. Kuhn, R.; Winterstein, A. Helv. Chim. Acta 1928, 11, 87e116.
17. Lissi, E.; Salimhanna, M.; Pascual, C.; Delcastillo, M. D. Free Radical Biol. Med.
52. Koyama, Y.; Fujii, R. In The Photochemistry of Carotenoids; Frank, H. A., Young, A.
J., Britton, G., Codgell, R. J., Eds.; Kluwer: Dordrecht, 1999; p 172.
53. Christensen, R. L.; Barney, E. A.; Broene, R. D.; Galinato, M. G. I.; Frank, H. A.
Arch. Biochem. Biophys. 2004, 430, 30e36.
54. Kon, M.; Britton, G. Food Sci. Technol. Res. 1999, 5, 347e349.
55. Hepperle, S. S.; Li, Q. B.; East, A. L. L. J. Phys. Chem. A 2005, 109, 10975e10981.
1995, 18, 153e158.
18. Huang, D. J.; Ou, B. X.; Prior, R. L. J. Agric. Food Chem. 2005, 53, 1841e1856.
19. Baltschun, D.; Beutner, S.; Briviba, K.; Martin, H. D.; Paust, J.; Peters, M.; Rover,
S.; Sies, H.; Stahl, W.; Steigel, A.; Stenhorst, F. Liebigs Ann./Recl. 1997, 1887e1893.
20. Sliwka, H. R.; Melø, T. B.; Foss, B. J.; Abdel-Hafez, S. H.; Partali, V.; Nadolski, G.;
Jackson, H.; Lockwood, S. E. Chem.dEur. J. 2007, 13, 4458e4466.
21. Mohamad, S. B. B.; Yousef, Y. A.; Melø, T. B.; Javorfi, T.; Partali, V.; Sliwka, H. R.;
Naqvi, K. R. J. Photochem. Photobiol., B 2006, 84, 135e140.
56. Bobrowski, K.; Das, P. K. J. Phys. Chem. 1986, 90, 927e931.
57. Fungo, F.; Otero, L.; Durantini, E.; Thompson, W. J.; Silber, J. J.; Moore, T. A.;
Moore, A. L.; Gust, D.; Sereno, L. Phys. Chem. Chem. Phys. 2003, 5, 469e475.
58. Heelis, P. F.; Phillips, G. O. Photobiochem. Photobiophys. 1979, 1, 63e70.
22. Karagiannidou, E.; Størseth, T. R.; Sliwka, H. R.; Partali, V.; Malterud, K. E.;
Tsimidou, M. Eur. J. Lipid Sci. Technol. 2003, 105, 419e426.