Coupling and fast decarboxylation of aryloxyl radicals of 4-hydroxycinnamic acids with formation of stable p-quinomethanes

Article abstract of DOI:10.1016/j.tet.2005.11.010

The reaction at room temperature of 3,5-di-tert-butyl- and 3,5-di-methoxy-4-hydroxycinnamic acids 1 and 2 with the dpph^. radical in acetone or other non-hydroxylic polar solvents yields interesting dimeric p-quinomethanes 10-16 characterized by

Full text of DOI:10.1016/j.tet.2005.11.010

Tetrahedron 62 (2006) 1536–1547
Coupling and fast decarboxylation of aryloxyl radicals
of 4-hydroxycinnamic acids with formation of stable
p-quinomethanes
*
Carmelo Daquino and Mario C. Foti
Istituto di Chimica Biomolecolare del CNR - Sez. di Catania, Via del Santuario, 110, I-95028 Valverde (CT), Italy
Received 2 September 2005; revised 14 October 2005; accepted 3 November 2005
Available online 20 December 2005
This paper is dedicated to Keith Usherwood Ingold.
Abstract—The reaction at room temperature of 3,5-di-tert-butyl- and 3,5-di-methoxy-4-hydroxycinnamic acids 1 and 2 with the dpph%
radical in acetone or other non-hydroxylic polar solvents yields interesting dimeric p-quinomethanes 10–16 characterized by a broad and
strong absorption in the visible region. Although the yields appear to be low to moderate (10–40%), this simple synthesis affords quinones not
otherwise obtainable, which contain an unsaturated g-lactone ring (14–16). The structures have been elucidated by interpretation of ESI-MS,
FT-IR and NMR spectral data. In particular, FT-IR spectra in a KBr matrix demonstrate the quinone nature of these compounds because of
the presence of strong absorption bands at 1604–1640 cmK1 and allows excluding the presence of carboxylic acid groups in the molecules.
Kinetic evidence and molecular structures suggest that the formation of these p-quinomethanes is best explained through an 8–8 C–C
coupling of the aryloxyl radicals derived from 1 and 2 and a subsequent fast mono- or di-decarboxylation of the initial dimer by an SE1-type
mechanism. Further oxidation of the phenolic intermediates by dpph% yields the final quinones.
q 2005 Elsevier Ltd. All rights reserved.
1. Introduction
of the carboxylic acid groups strongly influences the
ionization of phenolic OH and modulate the contribution
of Reaction 2 over the direct H-atom transfer (Reaction 3).4
In non-hydroxylic polar solvents, phenols are poorly
ionized5 and the reaction occurs through the direct and
slow (since ArOH are hydrogen bonded to the solvent
molecules)6 transfer of the hydrogen atom from ArOH to
dpph% (Reaction 3).
Phenols (ArOH) are effective antioxidants because of their
ability to react with peroxyl radicals, RO%2, responsible for
the autoxidation processes of organic materials.1 Consider-
able information as to the antioxidant properties of phenols
can be gained from the kinetic parameters and stoichiometry
of reactions with the stable and commercially available
nitrogen-centered radical, 2,2-diphenyl-1-picrylhydrazyl
(dpph%), because these parameters are correlated to those
of ArOH/RO%2.2 This has justified an intense proliferation of
studies on ArOH/dpph% reactions.
ArOH%ArOKCHC
(1)
HC
ArOKCdpph† ꢀꢀꢀ/ꢀ ArO† CdpphKH
(2)
(3)
(4)
The mechanism and rate of ArOH/dpph% reactions depend
largely upon the nature of phenol (vide infra) and the
solvent in which these reactions occur. Recently, Litwi-
nienko and Ingold3 and, independently, Foti et al.4 have
demonstrated that in alcohols these reactions essentially
proceed via an electron transfer (ET) step from the
phenoxide anion ArOK to dpph% (Reactions 1 and 2).
4-Hydroxycinnamic acids (HCA) are demonstrated to be
ideal models to disclose this mechanism since the presence
slow
ArOH Cdpph† ꢀꢀꢀ/ꢀ ArO† CdpphKH
solvents
2ArO† ꢀꢀꢀ/ꢀ various products
Afterwards, additional work has revealed that the solvent
effect plays another intriguing role in the products of the
self-coupling reaction of aryloxyl radicals of HCA’s,
Reaction 4. For instance, in the case of sinapic acid 1/
dpph%, Reaction 4 yielded in methanol or ethanol products
that were totally transparent to the electromagnetic waves of
the spectral region 400–800 nm. However, when this
reaction was carried out in acetone successive scans of the
UV–vis spectrum showed that the course of Reaction 4 was
Keywords: Sinapic acid; Cinnamic acids; dpph% radical; Decarboxylation;
Free radical; p-Quinomethane; g-Lactone.
Corresponding author. Tel.: C39 095 721 2136; fax: C39 095 721 2141;
e-mail: mario.foti@icb.cnr.it
*
0040–4020/$ - see front matter q 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2005.11.010

C. Daquino, M. C. Foti / Tetrahedron 62 (2006) 1536–1547
1537
different since an intensely coloured product was formed.
O
Chromatographic purification on silica gel of the reaction
mixture gave a reasonable amount of this compound (yield
37%), which surprisingly proved to be nearly NMR-silent in
many deuterated polar solvents (acetone, methanol, DMSO)
(vide infra). In two solvents (CD2Cl2 and CDCl3), however,
we succeeded in obtaining excellent 1H and 13C NMR
spectra, which revealed that this compound was the ethylene
bis(p-quinomethane) 10.
R'
R'
A
R"
R"
C
B
O
O
O
14: R'= R"= OCH3
R''
15: R'= R"= C(CH3)3
16: R'= C(CH3)3
R"= OCH3
R'
O
O
Formation of C–C coupled dimers is not surprising in view of
the fact that aryloxyl radicals undergo many of the reactions
typical of oxygen and carbon radicals, such as C–C and/or
C–O dimerization, isomerization, disproportionation, hydro-
gen abstraction and addition.8 In fact, the chemistry of
aryloxyl radicals has been investigated intensely during 1960s
and early 1970s.8,9 Various products of C–C and C–O
coupling of the aryloxyl radicals derived from 1 and its
methyl ester 8 have already been isolated10,11 among which
thomasidioic acid12 (a phenolic lignan) is one of the major
products. However, tothe bestofour knowledge innocase has
formation of decarboxylated dimers been reported in mild
oxidative reactions of cinnamic acids. Therefore, the
p-quinomethanes 10–16 formed under our experimental
conditions are even more interesting because they show that
a process of mono- or di-decarboxylation has occurred
spontaneously at room temperature at some stage of the
reaction sequence (vide infra).
R''
R'
10: R' = R'' = OMe
11: R' = R'' = tert-Bu
12: R' = OMe; R'' = tert-Bu
The undoubted importanceof thiscompound7 and the peculiar
mechanism by which it is formed (vide infra) led us to explore
thebehaviour ofother HCA’s(1–8, seeChart 1) withdpph% (or
MnO2) in various non-hydroxylic polar solvents. We found
that HCA 2 gave the corresponding dimeric p-quinomethane
11 as well, whereas the others did not except for 3,5-di-bromo-
p-coumaric acid 3 and ferulic acid 6, which gave traces of their
correspondingp-quinomethanes. UsingHCA’s1and2ina1:2
ratio (mol/mol) it was also possible to prepare the
asymmetrical p-quinomethane 12.
These reactions also afforded other oxidation products, which
were isolated and characterized as another interesting group of
stablequinonesbearinganunsaturatedg-lactonering(14–16).
We therefore, report herein on the synthesis and spectral
characterization of all these new compounds and propose a
mechanistic rationale for their formation compatible with
our kinetic data.
OH
R1
R2
1: R1 = R2 = OCH3
2: R1 = R2 = C(CH3)3
2. Results and discussion
3: R1 = R2 = Br
4: R1 = R2 = H
2.1. Isolation and structure determination
5: R1 = H; R2 = OH
6: R1 = H; R2 = OCH3
In Figure 1 is reported the time evolution at 25 8C of the UV–
vis spectrum of an acetone solution of dpph% (0.12 mM) in
which sinapic acid 1 was added to a final concentration of
1.28 mM. For comparison, the evolution of this reaction in
methanol is also shown in the Inset (Fig. 1). The spectra in
acetone show the formation of a compound with broad
absorption bands in the visible region (lmax at 508 nm) and
which is responsible for the purplish colour of the solution.
The UV–vis spectrum of this compound, isolated from the
reaction mixture (see Section 4), is shown in Figure 2a. The
large values of the molar extinction coefficient and lmax (see
Table 1) suggested a highly conjugated structure for 10. The
FT-IR spectrum in a KBr matrix also showed the presence of
a,b-unsaturated carbonyl groups (1621.9, 1563.4 and
CO2H
OH
OH
H3CO
OCH3
OCH3
CO2H
CO2CH3
7
8
O2N
Ph
Ph
1548.5 cm )
K1 13 and unexpectedly the absence of carboxylic
N
N
NO2
acid groups. HPLC-ESI-MS spectrumof this compoundin the
positive ion-mode showed three peaks at m/z: 357, 379 (base
peak) and 735. Since many compounds form adducts with
adventitious alkali metals14 and some of the above peaks
could be due to such adducts, we deliberately used water/
acetonitrile containing 0.1 mM LiCl as eluents in the HPLC-
O2N
dpph
Chart 1. 4-Hydroxy cinnamic acids and relative derivatives employed in
the present study.

1538
C. Daquino, M. C. Foti / Tetrahedron 62 (2006) 1536–1547
Table 1. FT-IR spectra in KBr and UV–vis spectra in CH2Cl2 of quinones
10–16
2.0
1.5
1.0
0.5
0.0
2.0
1.5
1.0
0.5
0.0
Quinone
nCO
(cmK1
nC]C
(cmK1
)
lmax (nm) (3/103 MK1 cmK1
)
)
10
11
12
1621.9
1605.3
1563.4
1548.5
1570.9
1557.6
1567.7
1551.7
519 (110); 484 (56); 453 (21)
486 (110); 454 (72); 427 (31)
502 (98); 468 (57); 440 (26)
400
500
600
700
800
Wavelength, nm
1629.9
1607.7
1602.9
1782.6
1639.4
1621.6
1781.2
1612.0
1768.7
1640.7
1614.0
13
14
455 (54); 428 (36)
531 (40); 355 (13)
1565.1
15
16
1572.6
1566.2
507 (45); 477 (41); 324 (19)
503 (41); 344 (13)
400
500
600
700
800
Wavelength, nm
Figure 1. Spectral evolution of the reaction between sinapic acid 1 (1.28!
10K3 M) and dpph% (1.20!10K4 M) in acetone at 25 8C. Spectra were
recorded at: 0; 6; 11; 21; 31; 46; and 76 s. Inset: spectral evolution of
sinapic acid 1 (1.28!10K3 M) Cdpph% (1.25!10K4 M) in methanol at
25 8C at: 0; 1; 2; 3; 4; and 5 s.
the difference, 379K363Z735K719Z16 was equal to the
difference in the masses of NaC and LiC, the above signals
were given by the following pseudo-molecular ions:
[MCH]C (m/z 357); [MCNa]C (m/z 379); [MCLi]C (m/z
363); [2MCNa]C (m/z 735) and [2MCLi]C (m/z 719) and
hence, we determined that the molecular weight of 10 was
356 (C20H20O6 356.38).
1.2
1.0
0.8
0.6
0.4
0.2
0.0
(a)
1112
13
10
When we tried to record the NMR spectra of 10 for the final
structure elucidation we unexpectedly observed that the H
1
NMR spectrum showed little information in many solvents
(acetone, methanol, DMSO), that is, this quinone was
demonstrated to be NMR-silent.15 Addition of ascorbic acid
into the NMR tube caused the bleaching ofthe solutionand the
appearanceofthe 1H and 13C NMR spectra reportedin Section
4. These spectra, the pseudo-molecular peak [MKH]K at m/z
357, the appearance in the FT-IR spectrum in CH2Cl2 of a
fairly sharp band at ca. 3529 cmK1 attributable to intra-
molecular H-bonded OH’s16 and finally the UV-spectrum,
which was similar to that of 1,4-diphenyl-1,3-butadiene17–19
prompted us to assign structure 10a to this compound and,
consequently, structure 10 to the oxidized form (Reaction 5).
Contrary to the behaviour described above, we successively
observed that in CD2Cl2 and CDCl3 the p-quinomethane 10
gave both the 1H and 13C NMR spectra (see Table 2), which
provided a further evidence of its structure.
300 350 400 450 500 550 600 650 700
Wavelength, nm
1.2
(b)
16
14
1.0
0.8
0.6
0.4
0.2
0.0
15
O
OH
OMe
MeO
MeO
OMe
ascorbic acid
300
400
500
600
700
(5)
Wavelength, nm
Figure 2. (a) Normalized UV–vis spectra of 10–13 obtained from the
HPLC UV–DAD. The eluent composition during the readings was 20:80
water/acetonitrile for 10; 15:85 water/acetonitrile for 12; 100% acetonitrile
for 11 and 13. The values of lmax and 3 in CH2Cl2 are reported in Table 1.
(b) Normalized UV–vis spectra of 14–16 obtained from the HPLC
UV–DAD. The eluent composition during the readings was 15:85 water/
acetonitrile for 14; 10:90 water/acetonitrile for 16; 100% acetonitrile for 15.
The values of lmax and 3 in CH2Cl2 are reported in Table 1.
MeO
OMe
MeO
OMe
O
OH
10a
10
1
One interesting aspect of the H NMR spectrum of 10 in
CD2Cl2 was represented by the splitting of the signals
corresponding to the methoxy groups (dH 3.80 and 3.86) and
MS analyses. In the presence of LiCl, the spectrum essentially
consisted of two peaks at m/z: 363 (base peak) and 719. Since

C. Daquino, M. C. Foti / Tetrahedron 62 (2006) 1536–1547
Table 2. H and 13C NMR spectra of quinones 10–12 in CD2Cl2 and quinone 13 in (CD3)2CO at room temperature with respect to TMS
1539
1
O
O
O
O
11
C(CH3)3
9
12
12
11
C(CH3)3
9
H3CO
10
OCH3
9
(H3C)3C
H3CO
22
OCH3
21
(H3C)3C
10
1
1
1
1
10
6
5
6
5
6
5
6
5
2
3
2
3
2
3
2
3
4
4
4
4
H
H
H
H
H
H
H
H
H
H
H
7
7
7
7
H
H
H
H
8
8
8
H
9
H
H
10
H
(H3C)3C
C(CH3)3
11
16
12
O
18
(H3C)137C
20
C(CH3)3
14
15
13
H3CO
OCH3
(H3C)3C
C(CH3)3
19
O
O
O
10
11
12
13
a
a
a
a
No.
dC
dH
dC
dH
dC
dH
dC
dH
1
2
3
4
5
6
7
8
174.4
152.7
111.3
132.8
102.5
152.8
138.0
133.6
55.6
186.6
149.2
134.3
134.6
125.3
149.4
140.3
134.8
35.4
174.4
186.6
149.3
135.0
136.9
125.5
149.8
135.4
152.6
111.2
132.9
102.5
152.8
137.8
133.4
134.4
139.9
133.9
133.7
148.6
186.0
148.7
124.7
34.9
6.38d 1.8
6.81d 1.8
6.99d 2.3
7.50d 2.3
6.37d 1.8
6.79d 1.8
7.29d 2.4
7.80d 2.4
7.86s
6.90dd 2.6, 8.6
7.33dd 2.6, 8.6
3.91
6.93dd 2.8, 8.4
7.37dd 2.9, 8.4
6.93m
7.35m
7.35m
6.93m
9
35.4
35.8
10
11
12
13
14
15
16
17
18
19
20
21
22
55.6
3.85
35.9
29.6
29.7
b
1.36
1.32
1.34s
1.31s
b
6.99d 2.4
7.50d 2.4
35.3
29.0
29.0
55.5
1.32
1.36
3.90
3.84
55.5
a The values of dH are followed by multiplicity and coupling constants (Hz).
b The methyl carbon signals of the tert-butyl groups fall into the solvent signal (acetone).
theringprotons(dH 6.33and6.76).Alesspronouncedsplitting
was also observed in the 13C NMR spectrum for the pertinent
carbon atoms (see Table 2). Similarly, the 1H and 13C NMR
spectra of the other three p-quinomethanes 11–13 showed an
analogous splitting (see Table 2).
and only one for the four methoxy groups at about 3.88 ppm
down to a temperature of 190 K.
All p-quinomethanes 10–12 can also be prepared (yields
30–40%) by oxidation of HCA’s at room temperature with
activated MnO2 in various solvents (see Section 4). The
oxidation pattern with MnO2 (evaluated by TLC analysis)
was similar to that of dpph% but the slightly higher yields
allowed an easier isolation and characterization of the major
reaction products (Reactions 6–8).
This molecular asymmetry has previously been observed20
with compound 13 and rationalized.21 It is worth mention-
ing that in the case of 10a the H NMR spectrum showed
one signal only for the four ring protons at about 6.67 ppm
1
O
H3CO
OCH3
O
OH
H3CO
OCH3
H3CO
OCH3
MnO2 or dpph
acetone
OCH3
O
+
(6)
O
O
CO2H
OCH3
H3CO
OCH3
O
1
10 (37%)
14 (10%)

1540
C. Daquino, M. C. Foti / Tetrahedron 62 (2006) 1536–1547
O
(H3C)3C
C(CH3)3
O
O
O
(H3C)3C
C(CH3)3
OH
(H3C)3C
C(CH3)3
(H3C)3C
C(CH3)3
C(CH3)3
O
MnO2 or dpph
acetone
+
+
(7)
O
CO2H
(H3C)3C
C(CH3)3
C(CH3)3
(H3C)3C
C(CH3)3
O
O
15 (20%)
2
11 (40%)
13 (<5%)
O
H3CO
OCH3
O
O
OH
OH
(H3C)3C
C(CH3)3
H3CO
OCH3
(H3C)3C
C(CH3)3
MnO2 or dpph
acetone
OCH3
+
+
+ 10 + 11
(8)
O
O
CO2H
CO2H
OCH3
(H3C)3C
C(CH3)3
O
1
2
12 (22%)
16 (10%)
signals of the tert-butyl and methoxyl groups, the 13C NMR
spectrum of 12 in CD2Cl2 showed 16 distinct carbon atoms
versus the 8 carbon signals of the symmetric quinones 10
and 11. HMQC, HMBC and NOESY 2D experiments
(see Fig. 3) were done to establish the direct C–H bonds and
the C–C connectivity.
The resemblance of the UV–vis and NMR spectra of
quinone 11 to those of 10 (see Figure 2a and Table 2)
allowed a straightforward recognition of its structure. In
agreement, the ESI-MS spectrum in the positive ion-mode
and in the presence of 1 mM LiCl in the acetonitrile phase
showed the following peaks at m/z 483 [MCNa]C; 467
[MCLi]C and 461 [MCH]C (C32H44O2 460.71).
1
The H NMR spectrum of the asymmetric p-quinomethane
H
H
C(CH3)3
O
12 in CD2Cl2 solution revealed resonances of two non-
equivalent MeO groups at 3.89 and 3.84 ppm and two non-
equivalent tert-butyl groups at 1.36 and 1.32 ppm.
Additional proton signals were observed at d 6.93 (2H, m)
and at 7.35 (2H, m), which could be assigned to the couples
of protons H-7/H-10 and H-8/H-9, respectively, (see
Table 2), on the basis of the correspondence with the
resonance frequencies of H-7 and H-8 in the symmetrical
quinones 10 and 11 (Table 2). The ring protons resonated as
doublets at d 6.37 (1H, JZ1.8 Hz), 6.79 (1H, 1.8 Hz), 6.99
(1H, 2.4 Hz) and 7.50 (1H, 2.4 Hz). In this case, the
correspondence in terms of coupling constants and chemical
shifts with the ring protons of quinones 10 and 11 (see
Table 2) allowed to assign the upfield signals, that is, 6.37
and 6.79 ppm, to the protons of the ring bearing the MeO
groups and the downfield signals, that is, 6.99 and 7.50 ppm,
to the ring with the tert-butyl groups.
H
H3CO
O
H
H
H
C(CH3)3
H
H
H3CO
Figure 3. Selected HMBC (dotted line) and NOESY (solid line)
correlations observed in the asymmetric quinone 12.
Finally, the ESI-MS spectrum of 12 in the positive ion-mode
and in the presence of 0.1 mM LiCl confirmed the structure
assignment because of the presence of peaks at m/z 431
[MCNa]C; 415 [MCLi]C and 409 [MCH]C (C26H32O4
408.54).
The two carbonyl groups of 12 at positions 1 and 14
resonated in the 13C NMR spectrum in CD2Cl2 at 174.4 and
186.0 ppm, respectively, as in the symmetric quinones 10
(174.4 ppm) and 11 (186.6 ppm) (see Table 2). This
correspondence was also observed in the FT-IR spectra as
the frequencies of CO stretching of the three quinones were
1621.9 cmK1 in 10, 1606.3 cmK1 in 11, and 1629.9 and
1607.7 cmK1 in 12 (see Table 1). Aside from the carbon
Quinones 14–16 were isolated from the reaction mixtures of
1, 2 and 1C2 (1:2 mol/mol) with either dpph% or MnO2 in
acetone at room temperature (yields ca. 10–20%, Reactions
6–8). Their UV–vis spectra are reported in Figure 2b and
1
Table 1. The H and 13C NMR spectra reported in Table 3
indicated that compounds 14–16 shared the same gross
molecular structure. Indeed, the proton spectrum of 14 in

C. Daquino, M. C. Foti / Tetrahedron 62 (2006) 1536–1547
1541
CD2Cl2 showed the presence of four distinct signals in the
range 3.86–3.94 ppm corresponding to four non-equivalent
methoxy groups. In the case of compound 15, three upfield
signals at 1.33, 1.36 and 1.38 ppm (ratio 1:2:1, respectively)
were observed and assigned to four tert-butyl groups. Two
distinct signals of methoxy groups at 3.91 and 3.92 ppm and
two distinct signals for two tert-butyl groups at 1.33 and
1.38 ppm were present in the spectrum of 16. These spectral
data therefore indicated the presence of two di-substituted
benzene rings in each molecule. In addition, all 1H
NMR spectra in CD2Cl2 exhibited 4 sharp doublets (4H,
Jw1.8–2.4 Hz), suggestive of two couples of meta benzene
protons, and 2 sharp singlets (2H) in the spectral region
6.42–8.07 ppm, see Table 3.
in the 1H NMR spectra in CD2Cl2 and the extended
p-electron system of these molecules, which were intensely
coloured suggested an unsaturation a to the C]O and an
additional exocyclic double bond on the carbon a to the O.
H
H
A
A
B
B
O
O
O
O
The above spectral data allowed the partial structures, that
is, two p-quinomethane units and the g-lactone unit, to be
assembled into the structures given for 14–16. The number
of quaternary and C–H carbons observed in the 13C NMR
and DEPT spectra of 14–16 (i.e., 11 and 6, respectively) was
consistent with the proposed structures as well as the
HMQC, HMBC and NOESY 2D spectra. Figure 4 shows a
few selected correlations observed in the HMBC and
NOESY spectra of 15 and 16.
The presence of the g-lactone moiety in 14–16 was deduced
by a combination of NMR and IR spectroscopy. The FT-IR
spectra in KBr suggested the presence of two different
carbonyl groups (nCOw1640–1612 and nCOw1780–
1770 cmK1) one of which could be attributed to a quinone
moiety (1640–1612 cmK1) and the other one to a g-lactone
(1780–1770 cmK1).22 Resonance peaks at about 167 ppm in
the 13C NMR spectra in CD2Cl2 of 14–16 confirmed the
presence of a g-lactone moiety (see Table 3). Further, the
presence of a considerably deshielded proton at about 8 ppm
In the case of 16, the specific substitution of the rings A and B
was deduced from the observed identity of the chemical shifts
of the carbons and protons 3 and 5 with those of 15, and 13 and
17 with those of 14, see Table 3. This supported the conclusion
1
Table 3. H and 13C NMR spectra of quinones 14–16 in CD2Cl2 with respect to TMS
O
O
O
20 21
23
22
21
20
25
24
22
23
(H3C)3C
C(CH3)3
H3CO
OCH3
(H3C)3C
C(CH3)3
1
1
1
2 A 6
19
18
18
2 A 6
2 A 6
18
OCH3
OCH3
C(CH3)3
5
3
3
5
3
5
4
4
4
H
H
H
H
H
H
10
H
H
10
H
H
H
10
H
14
14
14
O
O
O
13
12
17
7
13
13
7
7
11
11
B 15
12 B 15
12 B15
11
16
16
17
16
17
H
H
H
8
8
8
C
C
C
O
OCH3
19
O
OCH3
19
C(CH3)3
O
9
9
9
H
21
20
H
H
O
O
O
14
15
16
a
a
dH
a
No.
dC
dH
dC
dC
dH
1
2
3
4
5
6
7
8
174.6
151.2
111.7
129.4
104.4
152.6
129.0
136.3
167.1
129.7
153.6
118.7
102.4
153.7
174.6
153.5
102.6
56.3
186.2
149.5
134.0
129.6
126.0
150.8
126.2
138.4
167.2
131.1
152.9
118.9
124.6
151.5
185.9
150.5
124.6
35.4
186.2
149.3
134.1
128.7
126.1
150.6
126.7
137.8
167.1
130.4
151.1
118.9
102.2
153.7
174.6
153.4
102.5
56.2
6.42d 1.8
7.14d 1.8
6.80s
7.06d 2.2
7.67d 2.2
6.86s
7.06d 2.3
7.68d 2.3
6.87s
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
7.92s
8.07s
8.00s
6.58d 1.8
7.29d 2.4
6.56d 1.8
6.97d 1.8
3.93s
3.94s
3.86s
3.92s
7.61d 2.4
1.36s
6.95d 1.8
3.91s
3.92s
56.3
56.1
55.6
29.0
35.6
29.0
35.0
28.9
35.7
29.1
56.1
35.0
29.0
35.5
1.36s
1.38s
1.33s
1.38s
29.1
1.33s
a The values of dH are followed by multiplicity and coupling constants (Hz).

1542
C. Daquino, M. C. Foti / Tetrahedron 62 (2006) 1536–1547
O
O
A
(CH3)3C
H
C(CH3)3
(CH3)3C
H
C(CH3)3
A
H
H
H
H
H
H
C(CH3)3
O
H
C(CH3)3
O
H
C
O
C
O
O
O
(a)
(b)
B
B
H
H
C(CH3)3
C(CH3)3
O
O
A
(CH3)3C
H
C(CH3)3
(CH3)3C
H
C(CH3)3
A
H
H
H
H
H
H
OCH3
O
H
OCH3
H
C
O
(a)
(b)
C
O
B
O
B
O
O
H
H
OCH3
OCH3
Figure 4. (a) Selected NOESY correlations observed in 15 and 16; (b) selected HMBC correlations observed in 15 and 16.
that the t-Bu groups were located on the ring A whereas ring B
contained two MeO groups. Consistently, the NOESY
spectrum of 16 (see Fig. 4) showed a correlation of H-7
(dHZ6.87) with the nearest proton of the ring bearing the t-Bu
groups, that is, H-3, dHZ7.06. The transoid geometry of the
protons 7 and 10 was deduced from the NOESY spectrum,
which showed a marked correlation of H-10 with H-5 whereas
there was no noticeable correlation between H-7 and H-10.
surprising because of the mild conditions employed in our
experiments. Decarboxylation of a,b-unsaturated carboxylic
acids is known to take place with difficulty and usually
requires vigorous experimental conditions23 although for
cinnamic acids the presence of p-OH seems to accelerate the
process, which in any case requires comparatively high
temperatures.24 On the contrary, the intermediates involved in
our reactions apparently undergo decarboxylation fairly
readily. Under our experimental conditions, we observed
that the free carboxylic acid group in the reacting HCA’s1 and
2 is of pivotal importance. Indeed, when the methyl ester 8 of
sinapic acid 1 was allowed to react with dpph% in acetone at
room temperature no traces of 10 or 14 were detected in
solution.25 The spectral changes observed during the reaction
corresponded exclusively to the occurrence of Reaction 3 with
a rate constant of 16G2 MK1 sK1 and a stoichiometric
factor26 of ca. 1.0. The latter value demonstrates that once the
aryloxyl radicals from 8 were formed in Reaction 3 they
exclusively self-quenched (Reaction 4) without reacting
further with dpph%. It is interesting to observe that in the
case of the reactions 1Cdpph% or 2Cdpph% carried out in
acetone the stoichiometry was 1:2.
The ESI-MS spectra of quinones 14–16 were done in the
positive ion-mode. For the analyses of 14 and 16 the
eluents (water/acetonitrile) were added of 0.1 mM LiCl
whereas in the case of 15, which ionized with difficulty
the LiCl concentration in the acetonitrile phase was
increased to 1 mM. The MS spectrum of 14 showed
peaks at m/z 437 [MCK]C; 421 [MCNa]C; 405 [MC
Li]C and 399 [MCH]C, which confirmed the structure
assignment (C21H18O8 398.37). Analogously, the main
peaks in the MS spectra of quinones 15 and 16 were at
m/z 525 [MCNa]C; 509 [MCLi]C, 503 [MCH]C
(C33H42O4 502.70) and 473 [MCNa]C; 457 [MCLi]C
and 451 [MCH]C (C27H30O6 450.54), respectively.
2.2. Kinetic aspects and mechanism
Very recently, Bietti and Capone27 have provided spectro-
scopic evidence that arylethanoic acids, in the presence of
SO%4K radicals (the oxidizing agent), lose CO2 in water via an
aromatic radical-cation formed after an ET process to SO4%K
(with formation of SO24K ions). Subsequently, the aromatic
radical-cation undergoes fast intramolecular ET from the
carboxylate anion to the ring followed by decarboxylation
with formation of a resonance-stabilized benzyl radical either
by a concerted or stepwise mechanism.28
The molecular structures of 10–16 suggest that the
formation of these p-quinomethanes proceeds through a
C–C coupling of ArO% radicals at the positions 8 of the
side chains (one exception is represented by 13). The
ArO% radicals are produced by H-atom abstraction from
the phenolic OH of 1 and 2 by dpph% (or MnO2),
Reaction 3. The absence in the final products of one
(14–16) or two (10–12) carboxylic groups is particularly

C. Daquino, M. C. Foti / Tetrahedron 62 (2006) 1536–1547
1543
Unlike arylethanoic acids, in our case similar reactions
cannot be invoked for the process of decarboxylation of
the precursors of quinones 10–16 for at least two
reasons. Firstly, once the radical-cations 1%Cor 2%C
were formed by ET from 1 or 2 to dpph% they would
rapidly lose a proton from the phenolic OH, affording the
aryloxyl radical ArO%, long before the process of
ionization of the carboxylic acid group since phenol
solution (kw109 MK1!sK1).32 If the actual mechanism of
formation of 10–16 involved coupling of Ar–CH]CH% in the
presenceofdioxygentherewouldnotbeanyformationofC–C
dimers.33 Actually, the experiment showed that the yields of
quinones10and11wereunaffectedbythepresenceorabsence
of oxygen in solution. Thereby, we can conclude that
decarboxylation must occur after the 8–8 C–C dimerization
of aryloxyls (Reaction 9).35
OH
R1
R1
R1
R1
R1
CO2H
CO2H
dpph
acetone
O
O
R1
CO2H
C-C coupling
R1
(9)
OH
R2
R2
O
O
O
R1
O
HO
radical-cations are strong acids30 (in any case, the methyl
ether of 1 demonstrated to be inert to dpph% or MnO2).
Secondly, if a fraction of 1%C or 2%C were able to give
the acyloxyl radical, that is, Ar–CH]CH–CO%2, by ET
from the carboxylate anion to the aromatic ring, this
The initial C–C dimer formed in Reaction 9 may lose
CO2 via an SE1 mechanism36,24 (Reaction 10) followed
by a fast oxidation of the phenolic intermediates
(Reaction 11).
R1
R1
O
R2
R2
R2
R2
O
O
O
-CO2
(10)
O
O
R1
R1
O
HO
R1
R1
R2
R2
O
R2
R2
O
2 dpph
O
R1
(11)
HO
R1
R1 = R2 = OMe; tert-Bu
R1 = OMe
R2 = tert-Bu
would likely react with the H-atom donors present in our
system, (i.e., 1 or 2, dpph-H and acetone) thus
regenerating the parent phenol31 since the process of
decarboxylation of such a vinylacyloxyl radical is
expected to be comparatively slow (lifetime of the
order of microseconds).31b–e
Reaction 9 is the rate-determining step for the formation of
the quinones 10–16 in aprotic solvents and its rate is
proportional to [ArO%]2 (the rate constant can be close to the
diffusion limit37). The steady-state concentration of ArO% is
essentially determined by the rates of Reaction 3 and of the
overall processes of ArO% quenching. The substituents
present on phenols exert strong effects on the rate constant
of Reaction 3. Electron-donating (ED) groups in the ortho
and para positions of the phenol ring decrease the bond
dissociation enthalpy (BDE) of OH and increase the rate of
Reaction 3.2,38 On the contrary, electron-withdrawing
groups increase the OH BDE and decrease the rate of
In connection with the above arguments, it is important to
point out that experiments carried out with dioxygen did not
highlight formation of carbon-centered radicals, that is, Ar–
CH]CH%, in our system. Generally, carbon radicals are
known to react very quickly with the dioxygen dissolved in

1544
C. Daquino, M. C. Foti / Tetrahedron 62 (2006) 1536–1547
ArO% formation.2,38 Therefore, the low yields observed in
the reactions of HCA’s 3 (bromine present in the ortho
positions), 4 (no ED groups in the ortho positions), 6 (one
only ortho ED group and intramolecular hydrogen-bond)6
and 7 (one ED group in meta position) with dpph% are
readily explained. In the case of caffeic acid 5, the rate
constant of Reaction 3 is comparatively large because of the
presence of two ortho OHs.2 However, the main stabiliz-
ation pathway of the semiquinone radical of 5 consists of an
additional H-atom transfer to the dpph% radical with
formation of ortho-quinone therefore precluding the
dimerization and subsequent reactions.
that hydroxylic solvents support the ionization of Brønsted
acids better than non-hydroxylic solvents of similar
dielectric constant.3–5,40 It seems likely, therefore, that the
solvent plays an important role in the process of
decarboxylation of the initial dimer (Reaction 10). In fact,
similar solvent effects were reported for the unimolecular
decarboxylation of substituted benzisoxazole-3-carboxylate
anion.41 Dramatic rate accelerations resulted if protic
solvents (water, methanol or ethanol) were replaced by
aprotic solvents. For instance, the rate of decarboxylation at
30 8C of 6-nitrobenzisoxazole-3-carboxylate ion in acetone
resulted to be ca. 100,000 times larger than in methanol and
ca. 3,300,000 times larger than in water!41
The reaction mechanism suggested above for the formation
of quinones 10–12 justifies the presence of a few minor
compounds found in solution. Traces of phenol 10a were
detected, for instance, during the oxidation of 1 with dpph%
or MnO2 in acetone at room temperature although this
phenol reacts quickly with dpph%, being k3(10a)Z1130G
80 MK1 sK1 and nZ2.0. The presence of 10a can be
considered a convincing evidence of the occurrence of
Reaction 10. Other minor compounds include quinones
14–16 bearing the g-lactone moiety, which may originate
from an intramolecular nucleophilic addition of the
carboxylate anion to the opposite terminal methylene of
the quinone system (Reaction 12). A subsequent SE1
mechanism36 of decarboxylation of the intermediate
carboxylated g-lactone and a stepwise oxidation by dpph
radicals may lead to the final quinones bearing the
unsaturated g-lactone ring (Reaction 13).
There are a number of potential explanations42 for the
inhibitory effect of alcohols but we consider it most probable
that in these media the strong solvation of negative ions by
hydrogen bonding is retarding decarboxylation of the initial
dimer (Reaction 10) by stabilizing the carboxylate ion.41
3. Conclusion
We have described the synthesis and spectral identification
of highly conjugated dimeric quinones 10–16 some of
which bearing a peculiar unsaturated g-lactone ring
(14–16). The synthesis of these quinones consists of
oxidizing 4-hydroxy cinnamic acids with dpph% (or with
MnO2) in an appropriate solvent at room temperature, the
process being most successful when: (i) the solvent is non-
O
R1 = R2= OMe; tert-Bu
R1
R1
O
O
O
R1 = OMe
R1
R1
R1
R1
R2 = tert-Bu
O
O
OH
OH
O
dpph
(12)
R2
R2
HO
O
O
O
O
O
O
O
R2
R2
R2
R2
O
O
O
O
R1
R1
O
R1
dpph
R1
-CO2
R1
R1
O
(13)
R2
R2
R2
O
O
O
O
O
O
O
O
O
R2
R2
R2
Finally, the polar nature of the intermediates involved in the
mechanisms outlined above may also explain the solvent
effects observed on the yields and course of Reactions 6–8.
We found that non-hydroxylic polar solvents with higher
dielectric constants generally promoted the formation of
quinones better than solvents with low dielectric con-
stants.39 In alcohols, however, no formation of quinones
10–16 was observed39 at room temperature despite the fact
hydroxylic and of high dielectric constant; and (ii) the ortho
positions to the phenolic OH of HCA are both occupied by
bulky electron-donating groups.
The yields are low to moderate (10–40%) because of the
many side reactions, however, the complex structure of
these compounds makes these yields acceptable.

C. Daquino, M. C. Foti / Tetrahedron 62 (2006) 1536–1547
1545
Kinetic data along with molecular structures and the
presence in solution of a few intermediates suggest that
the mechanism of formation of 10–16 with dpph% proceeds
through four different steps: (1) formation of ArO% by
H-atom transfer from HCA to dpph%; (2) dimerization by 8–
8 C–C coupling of two ArO% radicals; (3) fast decarboxyla-
tion at room temperature of the intermediate dimer by an
SE1-type mechanism; and finally, (4) oxidation of the
intermediate phenolate anions by dpph%.
(200 mg, ca. 0.9 mmol) were allowed to react with 1.06 g
of dpph% (2.69 mmol) in 300 ml of acetone at 25 8C in the
dark for 2 h. After solvent removal, the crude product was
purified by column chromatography on silica gel using ethyl
acetate–hexane (80/20, v/v), acetone and methanol as
eluents to give ca. 60 mg of 10 (final yield 37%) and ca.
1
20 mg of 14 (final yield 10%) as a dark violet powder. H
and 13C NMR spectra were performed in dilute CD2Cl2
solutions and are reported in Tables 2 and 3. The UV–vis
spectra are shown in Figures 2a and b whereas the FT-IR
spectra are given in Table 1. The ESI-MS spectrum is
discussed in Section 2.
4. Experimental
4.1. General
4.2.2. Preparation of 4-[(3,5-di-tert-butyl-4-oxocyclohexa-
2,5-dienylidene)but-2-enylidene]-2,6-dimethoxycyclo-
hexa-2,5-dienone 12 and 3-[(3,5-di-tert-butyl-4-oxocyclo-
hexa-2,5-dienylidene)methyl]-5-(3,5-dimethoxy-4-oxo-
cyclohexa-2,5-dienylidene)furan-2(5H)-one 16 by using
Cinnamic acids 1, 4, 6 were purchased from Fluka; ascorbic
acid, dpph%, cinnamic acid 2 were obtained from Aldrich and
`
cinnamic acids 5 and 7 from Extrasynthese. All compounds
were used as received. The methyl ester 8 of sinapic acid was
available from a previous work, its synthesis is described in
Ref. 4. 3,5-Di-bromo-4-hydroxycinnamic acid 3 was donated
MnO2. 3,5-Di-tert-butyl-4-hydroxycinnamic acid
2
(180 mg, 0.65 mmol) and sinapic acid 1 (75 mg,
0.33 mmol) were solubilized in 10 ml of acetone. The
solution was then added with a syringe-pump (200 ml/min)
to an initial suspension of 200 mg of MnO2 in 1 ml of
acetone (in the dark and under stirring) to which aliquots of
100 mg each of MnO2 per 2 ml of phenol solution pumped
were successively added (700 mg in total corresponding to
8 mmol). After 2 h the suspension was filtered and the
solvent removed. The crude residue was then purified on a
preparative TLC plate (20!20 silica gel, 1 mm thick) using
hexane–acetone (90/10) as eluent to give 30 mg of 12 (yield
22%) and 12 mg of 16 (yield 10%). The UV–vis and FT-IR
spectra are reported in Figures 2a and b and in Table 1.
1H and 13C NMR spectra performed in dilute CD2Cl2
solutions are reported in Tables 2 and 3. The ESI-MS
spectra are reported in Section 2.
`
by Dr. Paolo Bovicelli (ICB-CNR, Universita La Sapienza,
Roma) and its synthesis and spectral characterization will be
reported in a separate paper. All solvents (Carlo Erba and
Merck) were of the highest commercially available quality
and were used without further purification (except for diethyl
ether and THF, which were distilled prior to their use). NMR
spectra were recorded at 400.13 MHz (1H) and 100.62 MHz
(13C) in CD2Cl2 solutions at 298 K on a Bruker AvanceTM
400 spectrometer. Chemical shifts were referenced to the
residual signal of CD2Cl2. HPLC analyses were done on an
instrument (Waters 1525) equipped with ESI-MS (Waters
Micromass ZQ) and UV–DAD (Waters 996) detectors
(column: Phenomenexw Luna, C18, 250!4.6 mm (5 mm)
at 20 8C using as eluent system H2O/CH3CN containing 0.1
or 1 mM LiCl). A double-ray Perkin Elmer Lambda 25
spectrophotometer was used for the kinetics and to record the
UV–vis spectra whereas the FT-IR spectra were obtained
with a Perkin Elmer Spectrum BX FT-IR System spectro-
photometer. Analytical and preparative (silica gel, 20!
20 cm, 0.5–1 mm thick) TLC plates and silica gel
(63–200 mm) were purchased from Merck. The syntheses
and purification reported in the following paragraphs for 10
and 14 with dpph% and 12 and 16 with MnO2 are general and
apply to all quinones. The purity of quinones 10–16
determined by HPLC analysis was not inferior to 95%.
4.2.3. Purification of 4,40-(2-butene-1,4-diylidene)-
bis(2,6-di-tert-butyl-2,5-cyclohexadien-1-one) 11, 4,40-
(1,2-ethanediylidene)-bis(2,6-di-tert-butyl-2,5-
cyclohexadien-1-one) 13 and 5-(3,5-di-tert-butyl-4-oxo-
cyclohexa-2,5-dienylidene)-3-[(3,5-di-tert-butyl-4-oxocy-
clohexa-2,5-dienylidene)methyl]furan-2(5H)-one 15.
These quinones can be obtained with both methods reported
above. Their purification can be accomplished by column
chromatography on silica gel or preparative TLC (silica gel)
using hexane–THF (95/5) (final yields, 40% for 11 and 20%
for 15). Quinone 13 was obtained in very low yield (!5%);
however, it was possible to characterize this molecule and
its UV–vis and FT-IR spectra are reported in Figure 2a and
Table 1. The NMR spectra are reported in Table 2; the ESI-
MS spectrum obtained in the presence of 1 mM LiCl in the
acetonitrile phase showed the following peaks at m/z 441
[MCLi]C and 435 [MCH]C (C30H42O2 434.67).
4.2. Preparation of activated MnO2
MnSO4$4H2O (11 g) were dissolved in 15 ml of distilled
water and the solution treated with 11.7 ml of 40% NaOH
(solution A); 9.6 g of KMnO4 were dissolved in 60 ml of hot
distilled water (solution B). Then, the two solutions A and B
were slowly mixed together in about 1 h under vigorous
stirring. The final solution was centrifuged and the precipitate
of MnO2 washed with distilled water until the wash waters
were colourless. The solid was then dried at 100–120 8C.
4.2.4. Synthesis of 1,4-di(4-hydroxy-3,5-dimethoxyphe-
nyl)-1,3-butadiene 10a. This phenol was obtained by
reduction of 10 with ascorbic acid using the following
procedure: 4 mg of 10 (0.011 mmol) were dissolved in 2 ml
of CH2Cl2–CH3OH (1/1 v/v) then 3.9 mg of ascorbic acid
(0.022 mmol) were added. The solution was shaken at
35–45 8C for ca. 1 h in the dark. After solvent removal,
compound 10a was extracted from the residue with CH2Cl2
(3!1 ml).Theevaporationofthesolventyielded10aasapale
4.2.1. Preparation of 4,40-(2-butene-1,4-diylidene)-
bis(2,6-dimethoxy-2,5-cyclohexadien-1-one) 10 and 5-
(3,5-dimethoxy-4-oxocyclohexa-2,5-dienylidene)-3-[(3,
5-dimethoxy-4-oxocyclohexa-2,5-dienylidene)methyl]-
furan-2(5H)-one 14 by using dpph%. Sinapic acid 1

1546
C. Daquino, M. C. Foti / Tetrahedron 62 (2006) 1536–1547
yellow solid in a quantitative yield. 1H NMR (CD2Cl2)
(numbering system: OH on C-1 and C-7 and C-8 on the
butadiene chain), dZ3.88 (s, 12H, OCH3), 5.54 (s, 2H, OH),
6.54(dd, JZ11.6, 2.4 Hz, 2H, H7), 6.67 (s, 4H, H3,5), 6.83(dd,
JZ11.6, 2.4 Hz, 2H, H8). 13C NMR (CDCl3) dZ56.08
(OCH3), 103.14 (C3,5), 127.31 (C8), 128.88 (C4), 131.80 (C7),
134.78 (C1), 147.13 (C2,6). FT-IR (CH2Cl2, cmK1) 3528.97
(m, OH), 1616.10 (m), 1601.0 (m), 1511.79 (s). The UV–vis
spectrum in methanol shows a maximum at 354 nm 3Z(5.4G
0.2)!104 MK1 cmK1. The HPLC-ESI-MS (water/aceto-
nitrile) spectrum of 10a in the negative ion-mode showed
peaks at m/z 737 [2MKH]K; 357 [MKH]K (base peak); 342
and 327 (C20H22O6 358.39).
11. Wallis, A. F. A. Aust. J. Chem. 1973, 26, 1571–1576.
12. (a) Niwa, T.; Doi, U.; Osawa, T. Bioorg. Med. Chem. Lett. 2002,
12, 963–965. (b) Rubino, M. I.; Arntfield, S. D.; Charlton, J. L.
J. Agric. Food Chem. 1996, 44, 1399–1402. (c) Charlton, J. L.;
Lee, K.-A. Tetrahedron Lett. 1997, 38, 7311–7312.
13. Similar stretching frequencies are reported in the literature for
other a,b-unsaturated carbonyl groups, see for instance: (a)
Breton, J.; Boullais, C.; Burie, J.-R.; Nabedryk, E.;
Mioskowski, C. Biochemistry 1994, 33, 14378–14386. (b)
Filar, L. J.; Winstein, S. Tetrahedron Lett. 1960, 25, 9–16.
14. Murphy, R. C.; Fiedler, J.; Hevko, J. Chem. Rev. 2001, 101,
479–526.
15. Other cases of NMR-silent quinones have been reported, see
for instance: (a) Williams, W.; Sun, X.; Jebaratnam, D. J. Org.
Chem. 1997, 62, 4364–4369. (b) Hilton, B. D.; Misra, R.
Biochemistry 1986, 25, 5533–5539. (c) Gould, S. J.; Melville,
C. R. Tetrahedron Lett. 1997, 38, 1473–1476.
Acknowledgements
The authors are grateful to Dr. Keith U. Ingold and
Dr. Giuseppe Ruberto for helpful discussions. They also
thank Mr. Emanuele Mirabella and Mrs. Cettina Rocco for
technical assistance. This work has been carried out under
the ‘12709 project’ of the Italian MIUR.
16. Foti, M. C.; Barclay, L. R. C.; Ingold, K. U. J. Am. Chem. Soc.
2002, 124, 12881–12888.
17. Allen, M. T.; Whitten, D. G. Chem. Rev. 1989, 89,
1691–1702.
18. Sun,Y.-P.;Bunker,C.E.;Wickremesinghe,P.L.;Rollins,H.W.;
Lawson, G. E. J. Phys. Chem. 1995, 99, 3423–3429.
19. Bunker, C. E.; Lytle, C. A.; Rollins, H. W.; Sun, Y.-P. J. Phys.
Chem. 1997, 101, 3214–3221.
References and notes
20. Brownstein, S.; Ingold, K. U. J. Am. Chem. Soc. 1962, 84,
2258–2259.
1. Howard, J. A. In Peroxyl Radicals; Alfassi, Z., Ed.; Wiley:
Chichester, UK, 1997; Chapter 10.
21. Dyall, L. K.; Winstein, S. J. Am. Chem. Soc. 1972, 94,
2196–2199.
2. Foti, M. C.; Johnson, E. R.; Vinqvist, M. R.; Wright, J. S.;
Barcaly, L. R. C.; Ingold, K. U. J. Org. Chem. 2002, 67,
5190–5196.
22. Silverstein, R. M.; Bassler, G. C.; Morrill, T. C. Spectrometric
Identification of Organic Compounds, 4th ed.; Wiley: New
York, 1981.
3. Litwinienko, G.; Ingold, K. U. J. Org. Chem. 2003, 68,
3433–3438.
23. (a) Graven, A.; Jorgensen, K. A.; Dahl, S.; Stanczak, A. J. Org.
Chem. 1994, 59, 3543–3546. (b) Matsubara, S.; Yokota, Y.;
Oshima, K. Org. Lett. 2004, 6, 2071–2073.
4. Foti, M. C.; Daquino, C.; Geraci, C. J. Org. Chem. 2004, 69,
2309–2314.
5. Litwinienko, G.; Ingold, K. U. J. Org. Chem. 2004, 69,
5888–5896.
24. Cohen, L. A.; Jones, W. M. J. Am. Chem. Soc. 1960, 82,
1907–1910.
6. Snelgrove, D. W.; Lusztyk, J.; Banks, J. T.; Mulder, P.; Ingold,
K. U. J. Am. Chem. Soc. 2001, 123, 469–477.
25. We also verified that sinapic aldehyde reacts with dpph% in
acetone at room temperature without producing the p-quino-
methane 10.
7. We ran a search in the literature with SciFinder and found
about ten patents on this molecule, which can be isolated from
a cell culture of safflower. See: (a) Wakayama, S.; Kusaka, K.;
Kanehira, T.; Yamada, Y.; Kawazu, K.; Kobayashi, A.
J. Biosci. 1994, 49, 1–5. (b) Kanehira, T.; Takekoshi, S.;
Nagata, H.; Matsuzaki, K.; Kambayashi, Y.; Osamura, R. Y.;
Homma, T. Life Sci. 2003, 74, 87–97. (c) Kanehira, T.;
Takekoshi, S.; Nagata, H.; Osamura, R. Y.; Homma, T. Planta
Med. 2003, 69, 457–459.
26. The stoichiometric factor n is the number of dpph% moles
quenched by 1 mol of phenol. The procedure for its
determination is described in Ref. 4.
27. Bietti, M.; Capone, A. J. Org. Chem. 2004, 69, 482–486 and
references therein.
28. However, with a tungsten oxidizing agent a concerted
mechanism of electron-transfer and decarboxylation without
passing through the radical-cation intermediate has been
invoked.29.
8. Altwicker, E. R. Chem. Rev. 1967, 67, 475–531.
9. See also: (a) Mahoney, L. R.; Weiner, S. A. J. Am. Chem. Soc.
1972, 94, 585–590. (b) Weiner, S. A.; Mahoney, L. R. J. Am.
Chem. Soc. 1972, 94, 5029–5033. (c) Kharasch, M. S.;
Joshi, B. S. J. Org. Chem. 1957, 22, 1435–1438. (d)
Cook, C. D.; Nash, N. G.; Flanagan, H. R. J. Am. Chem.
Soc. 1955, 77, 1783–1785. (e) Cook, C. D.; Norcross, B. E.
J. Am. Chem. Soc. 1956, 78, 3797–3799. (f) Tashiro, M.;
Fukata, G. J. Org. Chem. 1977, 42, 428–431. (g) van der
Boom, M. E.; Zubkov, T.; Shukla, A. D.; Rybtchinski, B.;
Shimon, L. J. W.; Rozenberg, H.; Ben-David, Y.; Milstein, D.
Angew. Chem. 2004, 116, 6087–6089.
29. Baciocchi, E.; Bietti, M. J. Chem. Soc., Perkin Trans. 2 2002,
720–722.
30. Ganapathi, M. R.; Hermann, R.; Naumov, S.; Brede, O. Phys.
Chem. Chem. Phys. 2000, 2, 4947–4955 and references
therein.
31. (a) Chateauneuf, J.; Lusztyk, J.; Ingold, K. U. J. Am. Chem.
Soc. 1987, 109, 897–899. (b) Chateauneuf, J.; Lusztyk, J.;
Ingold, K. U. J. Am. Chem. Soc. 1988, 110, 2886–2893. (c)
Aveline, B. M.; Kochevar, I. R.; Redmond, R. W. J. Am.
Chem. Soc. 1995, 117, 9699–9708. (d) Bohne, C.; Boch, R.;
Scaiano, J. C. J. Org. Chem. 1990, 55, 5414. (e) Newcomb, M.
Tetrahedron 1993, 49, 1151. (f) May, D. D.; Skell, P. S. J. Am.
Chem. Soc. 1982, 104, 4500–4502.
10. Bunzel, M.; Ralph, J.; Kim, H.; Lu, F.; Ralph, S. A.;
Marita, J. M.; Hatfield, R. D.; Steinhart, H. J. Agric. Food
Chem. 2003, 51, 1427–1434.

C. Daquino, M. C. Foti / Tetrahedron 62 (2006) 1536–1547
1547
32. (a) Maillard, B.; Ingold, K. U.; Scaiano, J. C. J. Am. Chem.
Soc. 1983, 105, 5095–5099. (b) Sortino, S.; Petralia, S.; Foti,
M. C. New J. Chem. 2003, 27, 1563–1567.
40. Foti, M. C.; Sortino, S.; Ingold, K. U. Chem. Eur. J. 2005, 11,
1942–1948.
41. Kemp, D. S.; Paul, K. J. Am. Chem. Soc. 1970, 92, 2553–2554.
Moore, J. W.; Pearson, R. G. In Kinetics and Mechanism;
Wiley, 1981; pp 269–270.
33. Dimerization of carbon-centered radicals, R%, is a diffusion-
controlled process as well, that is, kw109 MK1 sK1 34
.
However, the rate of reaction of R% with O2 would largely be
predominant because the concentration of oxygen in solution
is exceedingly high (w10K3 M).
42. For instance, it is possible that in the alcohols the following
addition reaction to the dimer takes place:
MeO
34. Laufer, A. H.; Fahr, A. Chem. Rev. 2004, 104, 2813–2832.
35. Traces of the dimer represented in Reaction 9 have been found
in solution by HPLC-MS analysis. See also Ref. 11.
36. March, J. Advanced Organic Chemistry: Reactions, Mechan-
isms and Structure 3rd ed.; Wiley, 1985; p 507 and 562.
37. Denisov, E. T.; Khudyakov, I. V. Chem. Rev. 1987, 87,
1313–1357.
MeO
HO
MeO
HO2C
HO2C
ROH
O
OR
MeO
This reaction is, however, evaluated to be comparatively slow.
In the case of a methyl p-quinomethane, the rate constant
relative to the addition of neutral methanol has been reported
to be 0.031 sK1 at 23 8C and ca. 500 times lower for the
addition to a more stable p-quinomethane.21 Other evidence in
support of the poor importance of this reaction is given by the
fact that the yield of formation of 10 is not significantly
enhanced in the sterically hindered tert-butyl alcohol.39
38. Pratt, D. A.; DiLabio, G. A.; Mulder, P.; Ingold, K. U. Acc.
Chem. Res. 2004, 37, 334–340.
39. The yields (deduced from the absorbance at lmax) for the
reaction 1Cdpph% were: 0% (n-hexane and CH2Cl2); 12%
(EtAc); 20% (Et2O); 24% (acetonitrile); 40% (THF); 50%
(DMSO and DMF); 1% (tert-butanol); 0% (MeOH, EtOH,
n-PrOH, 2-PrOH, n-BuOH).