Manifestation of polar reaction pathways of 2,3-dichloro-5,6-dicyano-p- benzoquinone

Article abstract of DOI:10.1021/ja405890d

Reactions of 2,3-dichloro-5,6-dicyano-p-benzoquinone (DDQ) with silyl enol ethers, silyl ketene acetals, allylsilanes, enamino esters, and diazomethanes have been studied in CH₃CN and CH₂Cl₂ solutions. The second-order rate constants for C attack at DDQ (log k _C) correlate linearly with the nucleophile-specific parameters N and s_N and are 2-5 orders of magnitude larger than expected for SET processes, which strongly supports the polar mechanism for C-C bond formation. The second-order rate constants for O attack agree well with the calculated rate constants for rate-determining single electron transfer (SET). As a radical clock experiment ruled out outer sphere electron transfer, an inner sphere electron transfer mechanism is suggested for O attack.

Full text of DOI:10.1021/ja405890d

Article
pubs.acs.org/JACS
Manifestation of Polar Reaction Pathways of 2,3-Dichloro-5,6-
dicyano‑p‑benzoquinone
Xingwei Guo and Herbert Mayr*
Department Chemie, Ludwig-Maximilians-Universitat Munchen, Butenandtstraße 5-13 (Haus F), 81377 Munchen, Germany
̈
̈
̈
S
* Supporting Information
ABSTRACT: Reactions of 2,3-dichloro-5,6-dicyano-p-benzo-
quinone (DDQ) with silyl enol ethers, silyl ketene acetals,
allylsilanes, enamino esters, and diazomethanes have been
studied in CH3CN and CH2Cl2 solutions. The second-order
rate constants for C attack at DDQ (log kC) correlate linearly
with the nucleophile-specific parameters N and sN and are 2−5
orders of magnitude larger than expected for SET processes,
which strongly supports the polar mechanism for C−C bond
formation. The second-order rate constants for O attack agree well with the calculated rate constants for rate-determining single
electron transfer (SET). As a radical clock experiment ruled out outer sphere electron transfer, an inner sphere electron transfer
mechanism is suggested for O attack.
been created by using a series of benzhydrylium ions and
structurally related quinone methides as reference electro-
philes.9 We have now used the linear free energy relationship
(1) to elucidate the mechanisms of the reactions of DDQ with
π-systems (Table 1).
INTRODUCTION
■
2,3-Dichloro-5,6-dicyano-p-benzoquinone (DDQ), first synthe-
sized by Thiele and Gunther in 1906,1 has a high redox
̈
potential (0.54 V vs SCE in acetonitrile)2 and is one of the
most important oxidizing reagents in organic chemistry. It
usually reacts as a one-electron acceptor to give a radical anion
in the first step and has been used for the oxidation of steroid
ketones, hydroaromatic compounds, alcohols, phenols, and
heterocycles.3 Recently, DDQ was used as an oxidant for
several oxidative coupling reactions.4
RESULTS
■
Product Studies in Acetonitrile. In line with earlier
investigations,6 the cyclic silyl enol ethers 2a−c were found to
give predominantly 4(a-c)-C, the products of C attack in
acetonitrile, and only a trace (∼4%) of 4b-O, the product of O
Triggered by the observation of DDQ-substrate adducts
during the oxidation of 4-aza-3-ketosteroids by DDQ,5
Bhattacharya and co-workers investigated the reactions of
cyclic silyl enol ethers with DDQ and found the formation of
adducts 3a-C and 3a-O (Scheme 1).6
The dramatic solvent and temperature effects on the product
ratios were considered to be indicative of the operation of two
distinctly different pathways for the formation of 3a-C and 3a-
O. Though both products were suggested to be formed via a
radical ion pair, the possibility of a nucleophilic attack of the
silyl enol ether on DDQ to form the carbon−carbon adduct 3a-
C was explicitly mentioned as an alternative.6 We now report
kinetic investigations of the reactions of DDQ with π-
nucleophiles, which clearly show that polar mechanisms are
rather the rule than the exception for these reactions.
1
attack, was detected by H NMR spectroscopy (Scheme 2).
Compounds 4(a-c)-C were isolated as mixtures of diaster-
eomers after column chromatography. The major diaster-
eomers of 4(a,c)-C were obtained and characterized as pure
compounds by recrystallization from a mixture of dichloro-
methane and pentane. Additionally, the major diastereoisomer
of 4a-C was characterized by X-ray crystallography (Figure 1).
The reactions of the terminal silyl enol ethers 2d and 2e and
silyl ketene acetals 2f and 2g with DDQ yielded the products of
C attack (4(d-g)-C) quantitatively (Scheme 3), which were
purified by column chromatography.
While the isobutyraldehyde-derived silyl enol ether 2h
reacted via O attack to give 68% of 3h-O and 32% of the
dehydrogenated product 5h (Scheme 4), according to 1H
NMR analysis of the crude reaction product, the structurally
analogous silyl ketene acetal 2i reacted with DDQ preferentially
at carbon to give 94% of 3i-C and 6% of 3i-O. At elongated
reaction times 3i-C slowly transformed into a mixture of 3i-O
and 5i (Scheme 4). The parallel increase of 3i-O and 5i (Figure
2) shows that both products are formed in parallel reactions
In recent years, we have developed a linear free energy
relationship based model for polar organic reactions,7 which
uses eq 1 to predict rates and selectivities for these reactions:
log k(20°C) = sN(E + N)
(1)
In eq 1 the second-order rate constant (log k) is calculated by
two nucleophile-specific parameters sN and N and one
electrophile-specific parameter E. A comprehensive nucleophil-
icity scale covering more than 30 orders of magnitude8 has
Received: June 12, 2013
Published: July 24, 2013
© 2013 American Chemical Society
12377
dx.doi.org/10.1021/ja405890d | J. Am. Chem. Soc. 2013, 135, 12377−12387

Journal of the American Chemical Society
Article
Scheme 1. Reaction of DDQ (1) with the Silyl Enol Ether 2a as Suggested by Bhattacharya et al.6
Table 1. Nucleophiles 2a−o and Their Reactivity Parameters
N and sN in CH2Cl2
Scheme 2. Reactions of DDQ with Cyclic Silyl Enol Ethers
2a−c
a
b
1
Isolated yield. Not observed by H NMR in the crude material.
Figure 1. Crystal structure of 4a-C (major diastereomer, thermal
ellipsoids are drawn at the 50% probability level).
obtained by column chromatography on silica gel in 45 and
63% yield, respectively (see the Supporting Information).
The rearrangement of 3i-C to 3i-O can be assumed to
proceed via a radical pathway, as shown in Scheme 5. When
(2,2,6,6-tetramethylpiperidin-1-yl)oxyl (TEMPO) was added to
crude 3i-C, which was formed from 1 and 2i in CH3CN within
1 min (Scheme 4), a 45: 55 mixture of 3i-O and 6 was formed
within 72 h at room temperature (1H NMR). Methyl
methylacrylate 5i probably was lost during evaporation of the
solvent. Column chromatography of the mixture provided pure
6 in 49% yield, which was fully characterized.
a
b
From ref 8. This work.
from 3i-C and excludes the generation of 5i by elimination
from 3i-O. In view of the slow conversion of 3i-C into 3i-O
(after 10 min, only 7% of 3i-O are present), the observation of
a small amount of 3i-O even after 1 min indicates that 6% of 3i-
O is directly formed from DDQ and 2i (kinetic product
control). The desilylated products 4h-O and 4i-O were
12378
dx.doi.org/10.1021/ja405890d | J. Am. Chem. Soc. 2013, 135, 12377−12387

Journal of the American Chemical Society
Article
Scheme 3. Reactions of DDQ with Terminal Silyl Enol
Ethers and Ketene Acetals
Scheme 5. Trapping of the Intermediate of the
Rearrangement of 3i-C to 3i-O
Scheme 4. Relative Yields (1H NMR) of the Reactions of
DDQ with 2h and 2i in CD3CN
Scheme 6. Reaction of DDQ with 2j
a
n.d. = not detected.
Figure 2. Time resolved relative 1H NMR yields of 3i-C, 3i-O, and 5i
during the reaction of DDQ with 2i (in CD3CN, at 20 °C).
Figure 3. Crystal structure of 4j-O (thermal ellipsoids are drawn at the
50% probability level).
The reaction of 1,2-disiloxycyclohexene 2j with DDQ gave
exclusively the product of O attack (Scheme 6). The partially
desilylated product 4j-O crystallized from CH2Cl2/pentane
solution and was analyzed by X-ray crystallography (Figure 3).
The reaction of allylsilane 2k with DDQ initially gave 3k-C,
the product of C attack, that slowly rearranged into the allyl
phenyl ether 3k-O (Scheme 7). After hydrolysis, the desilylated
derivative 4k-O was obtained by crystallization in 81% yield
(see the Supporting Information).
Previous investigations have shown that DDQ reacts with the
secondary enaminoester 7 to give the spirane 9 (Scheme 8),
indicating that 7 attacks at C-2 of DDQ.10 In this work, we
observed the exclusive formation of 3l-O from 1 and the
tertiary enaminoesters 2l or 2m after aqueous workup, which
can be explained through O attack and formation of 10 as
illustrated in Scheme 8. Proton migration, cyclization,
hydrolysis, and decarbonylation eventually yield benzofuran
3l-O as the final product. Possibly, 2l and 2m initially also
attack 1 at the carbon, to give a zwitterion analogous to 8,
which cannot cyclize to a spirane because of the absence of an
NH proton. Therefore, it may revert to reactants and
subsequently react via O attack to give 3l-O as the final
product.
While the reaction of phenyldiazomethane with DDQ gave
92% of the bicyclic diketone 3n-C (diastereomer ratio: ca. 55/
45), no reaction between ethyl diazoacetate and DDQ was
observed at room temperature within 1 h (Scheme 9).
Solvent Effect. Scheme 10 shows that the 4a-O/4a-C ratio
generated by the reaction of DDQ with 2a depends strongly on
the concentration of the reactants and on the solvent.
According to entries 1 and 2, the 4a-O/4a-C ratio increases
from 40/60 to 92/8 when reducing the concentration of 2a
from 0.1 to 0.02 mol L−1 in CH2Cl2. In the dilute solution (0.02
12379
dx.doi.org/10.1021/ja405890d | J. Am. Chem. Soc. 2013, 135, 12377−12387

Journal of the American Chemical Society
Article
Scheme 7. Reaction of DDQ with 2k
Scheme 9. Reactions of DDQ with Diazoalkanes
a
Yield and dr (1H NMR) correspond to the crude product; only the
major isomer was isolated and characterized.
and reaches a constant value of 7/93 for >10% of MeOH
(entries 13−17, Scheme 10).
A rationalization for these observations is given in Scheme
12. Attack of 2a at a carbon of DDQ to give 11 is faster than O
attack to give 12. As silyl shifts to convert 11 into 3a-C and 12
into 3a-O cannot occur intramolecularly, silyl-carriers are
needed. Such a carrier may be the solvent when the reacton is
carried out in acetonitrile solution (→ N-trimethylsilyl-nitrilium
ions). As the product ratio [4a-O]/[4a-C] obtained from 2a is
not affected by addition of the stronger nucleophile methanol
in acetonitrile, we conclude that 11 and 12 are formed
irreversibly in acetonitrile solution due to the fast subsequent
silyl shift, and that the product ratio [4a-O]/[4a-C] = 7/93
corresponds to kO/kC. The independence of the observed rate
constants kobs (for measurements see below) of methanol
additives (Figure 4) and NBu4OTs additives (Tables S1 and
S11, see the Supporting Information) are in line with this
interpretation.
mol L−1), the 4a-O/4a-C ratio decreases dramatically from 92/
8 to 7/93 when the solvent is changed from CH2Cl2 to
CH2Cl2/MeOH, and CH3CN (entries 2−7, Scheme 10). In
acetonitrile solution, the product ratio is neither affected by the
concentration of 2a nor by the presence of methanol (entries
7−10, Scheme 10). As 3a-C does not rearrange into the
thermodynamically more stable isomer 3a-O when dissolved in
CD2Cl2 (Scheme 11), we can rule out isomerization at the
product stage.
When the tert-butyldimethylsilyl substituted enol ether 2a′
was used, more product of O attack was observed than in the
corresponding reactions with the trimethylsilyl substituted
compound 2a in both CH2Cl2 and CH3CN (entries 2/11 and
7/12, Scheme 10). In the reactions of DDQ with 2a′, the 4a-
O/4a-C ratio decreases when methanol is added to acetonitrile
Because of its lower nucleophilicity, dichloromethane cannot
act as a silyl-carrier, and the conversions 11 → 3a-C and 12 →
3a-O can only occur in the presence of other nucleophiles, e.g.,
methanol, or when bimolecular reactions of two molecules of
11 or 12 lead to the formation of 3a-C and 3a-O. In the
Scheme 8. Reactions of DDQ with Enamino Esters
a
Yield isolated after crystallization from AcOH/CH3CN
12380
dx.doi.org/10.1021/ja405890d | J. Am. Chem. Soc. 2013, 135, 12377−12387

Journal of the American Chemical Society
Article
Scheme 10. Solvent Effect on the Reaction of DDQ with 2a and 2a′
a
b
In all cases, 10 equiv of 2a or 2a′ were used, that is, [2a or 2a′]/[1] = 10. Quantitative product formation, ratios of diastereoisomers were
1
determined by H NMR of the crude product mixture and are given in parentheses (see the Supporting Information for details).
at DDQ becomes completely reversible, and 4a-O is the only
product (entry 11, Scheme 10). In pure acetonitrile, C attack
(which is irreversible with 2a, see above) becomes partially
reversible in the reaction with 2a′, and addition of the more
nucleophilic methanol now leads to a decrease of the ratio [4a-
O]/[4a-C] (entries 13−17, Scheme 10).
Scheme 11. Examination of Product Stability of 3a-C in
CD2Cl2
Charge Transfer (CT) Complexes. The formation of CT
complexes between electron-deficient and electron-rich π-
systems is well-known.11 As the association constants between
the nucleophiles 2 and DDQ (1) are small, at room
temperature the corresponding CT complexes could only be
observed in few cases as short-lived intermediates when high
concentrations of the reactants were employed. Since lowering
of the temperature increases the association constants and
retards the reactions of 1 with 2, we have been able to observe
several CT complexes, however, when mixing the nucleophiles
2 with DDQ in CH2Cl2 at −80 °C (Scheme 13). The new
absorption bands observed under these conditions can be
assigned to charge transfer complexes, because neither DDQ
nor the nucleophiles 2 examined absorb beyond 450 nm.
Though the CT complexes are likely to correspond to the
encounter complexes preceding the formation of the reaction
products from 1 and 2, the kinetic data do not allow us to
differentiate whether the CT complexes are intermediates on
absence of methanol or in highly dilute solutions in
dichloromethane, silyl transfers will be slow, and the formation
of 11 that is thermodynamically less stable than 12 will be
reversible. As a consequence, in highly dilute dichloromethane
solutions (where also the concentrations of 11 and 12 will be
small), the ratio [3a-O]/[3a-C] will be large because of the
reversible formation of 11. Already small amounts of methanol
are sufficient to accelerate the silyl transfer, which results in an
increase of the observed rate constant (factor of 2 in a 0.05 mol
L−1 solution, Figure 4) as well as the yield of 3a-C.
The observations with the tBuMe2Si substituted enol ether
2a′ are in line with these interpretations. As the transfer of the
sterically more shielded tBuMe2Si group is slower than that of
the Me3Si group, in dilute dichloromethane solution, C attack
12381
dx.doi.org/10.1021/ja405890d | J. Am. Chem. Soc. 2013, 135, 12377−12387

Journal of the American Chemical Society
Article
Scheme 12. Possible Mechanism of the Reaction of DDQ with 2a
absorption maxima of DDQ (278 nm, Figure 5). For some
nucleophiles (2d, 2l, and 2m), the formation of the products
Figure 4. Effect of methanol on the rate of the reaction of DDQ (5.0
× 10−3 mol L−1) with 2a (5.0 × 10−2 mol L−1) in CH2Cl2 and in
CH3CN (monitoring the CT complex at 570 nm).
Scheme 13. Characteristic Absorption Bands of the CT
Complexes of Nucleophiles 2 with DDQ in CH2Cl2 at −80
°C
Figure 5. Time-dependent UV−vis absorbance during the reaction of
DDQ (1.0 × 10−4 mol L−1) with 2e (3.9 × 10−3 mol L−1) in CH3CN
at 20 °C.
was monitored (at 350 nm) because of the intensive
absorbance of nucleophiles at lower wavelength (<300 nm).
Pseudo first-order rate constants kobs were obtained by least-
squares fitting of the absorbances to the monoexponential
obst
function At = A0e−k t + C or At = A0 (1 − e−k ) + C. Second-
obs
order rate constants k2 were derived from the linear correlation
of kobs (s−1) with the concentrations of the nucleophiles (Figure
6).
In line with the fast reversible formation of the CT
complexes, identical rate constants for the reaction of DDQ
with 2a were derived by following the decay of the absorption
band of the CT complex (589 nm) and the absorption band of
DDQ (286 nm) as well as the increase of the absorption of the
product (350 nm) in CH2Cl2 at −60 °C (Figure 7).
The observation of second-order kinetics for the reactions of
DDQ with nucleophiles 2 in CH2Cl2 is surprising in view of the
discussion of Schemes 10 and 12. Possibly, the perfect linearity
of the kobs vs [2] correlations (Tables S15−S21) indicates that
the way to the products or correspond to dead ends of a
nonproductive side channel (Curtin-Hammett principle12).
Kinetic Studies. All kinetic investigations of the reactions of
DDQ with the nucleophiles 2 were performed in acetonitrile or
dichloromethane solution at 20 °C. The reactions were
monitored by UV−vis spectroscopy at or close to the
12382
dx.doi.org/10.1021/ja405890d | J. Am. Chem. Soc. 2013, 135, 12377−12387

Journal of the American Chemical Society
Article
Ab Initio Calculation of Adiabatic Ionization Poten-
tials. In order to elucidate the feasibility of single electron
transfer (SET) processes, we have determined the ionization
potentials of the π-nucleophiles 2(a-q) by quantum chemical
calculations using the Gaussian 09 program package.13
Adiabatic ionization potentials were calculated at the
G3(MP2) level of theory, which is a high-level composite ab
initio molecular orbital theory method as the sum of the
CCSD(T) calculations with a double-ζ basis set and basis set
corrections carried out at the MP2 level of theory to
approximate CCSD(T) calculations with a large triple-ζ basis.
G3(MP2) has been shown to reproduce a large test set of gas-
phase experimental data within chemical accuracy.14 The
adiabatic ionization potentials listed in Table 3 have been
calculated as the difference of the enthalpies of the radical
cations and those of the neutral molecules in the gas phase.
Figure 6. UV−vis spectroscopic monitoring of the reaction of DDQ
(1.0 × 10−4 mol L−1) with 2e (3.9 × 10−3 mol L−1) at 278 nm in
CH3CN at 20 °C. Insert: Determination of the second-order rate
constant k2 = 23 L mol−1 s−1 from the dependence of the first-order
rate constant kobs on the concentration of 2e.
Table 3. Calculated Adiabatic Ionization Potentials
[G3(MP2)] of Nucleophiles 2 and the Corresponding
Electrochemical Oxidation Potentials
nucleophile
IPa/eV
Eox (vs SCE)/V
a
2a
2b
2c
2d
2e
2f
7.77
7.83
7.71
8.10
8.24
7.89
7.65
7.77
7.22
7.05
8.26
7.71
7.70
7.87
8.48
7.18
1.30
b
1.21
b
1.15
a
1.32
b
1.43
b
1.24
b
2g
2h
2i
1.12
b
1.18
a
0.90
b
2j
0.80
b
2k
2l
1.44
b
1.15
Figure 7. UV−vis spectra of a mixture of DDQ (1.0 × 10−4 mol L−1)
and 2a (4.0 × 10−3 mol L−1) in CH2Cl2 at −60 °C and the time
dependent UV−vis absorbance at 286, 350, and 589 nm.
b
2m
2n
1.14
b
1.23
c
a
2p
1.50
c
a
2q
0.83
a
b
also 2a-f,k may act as silyl carriers under the conditions of the
kinetic experiments.
In CH3CN from ref 15. Calculated by substitution of calculated
c
ionization potentials into eq 2. See Figure 8 for the structures of these
Table 2 shows that the ratios of the second-order rate
constants for the reactions of DDQ with nucleophiles 2 in
compounds.
CH2Cl2 and CH3CN solutions, kC2 H Cl /k2MeCN, vary from 0.44 to
3.1 indicating that the kinetics are only slightly affected by
solvent polarity.
2
2
Figure 8 shows that the calculated ionization potentials
correlate with the experimental oxidation potentials15 of 2a, 2d,
Table 2. Second-Order Rate Constants k for the Reaction of DDQ with Nucleophiles (2a−n) at 20 °C
kC2 H Cl /L mol−1 s−1
kM2 eCN/L mol−1 s−1
kC2 H Cl /kM2 eCN
kO/L mol−1 s−1 (in CH3CN)
kC/L mol−1 s−1 (in CH3CN)
2
2
2
2
nucleophile
2a
2b
2c
2d
2e
2f
1.9 × 101
3.4 × 102
9.6 × 102
2.0 × 102
4.5 × 101
1.1 × 104
2.5 × 101
5.5 × 102
7.1 × 102
4.5 × 102
2.3 × 101
3.5 × 103
8.5 × 105
1.8 × 10−1
1.1 × 107
7.0 × 104
3.9 × 10−1
6.8
0.76
0.62
1.4
1.9
2.3 × 101
5.5 × 102
7.1 × 102
4.5 × 102
2.3 × 101
3.5 × 103
8.5 × 105
0.44
2.0
3.1
2g
2h
2i
1.8 × 10−1
6.6 × 105
7.0 × 104
1.0 × 107
2j
2k
2l
3.9 × 10−1
1.0
3.9 × 10−1
6.8
2.3 × 101
2m
2n
2.3 × 101
8.6 × 103
8.6 × 103
12383
dx.doi.org/10.1021/ja405890d | J. Am. Chem. Soc. 2013, 135, 12377−12387

Journal of the American Chemical Society
Article
quinone methides correlate linearly with their reduction
potentials (Figure 10).17 When the data for DDQ are added
Figure 8. Correlation between calculated adiabatic ionization
potentials IPa and experimental oxidation potentials Eox (vs SCE in
CH3CN).
Figure 10. Correlation between the electrophilicity parameters E and
the reduction potentials Ered of reference electrophiles17 and DDQ.
2i, 2p, and 2q as expressed by eq 2, which is similar to the
correlation previously reported for aromatic compounds (IPa =
1.5Eox + 5.8).16 Using correlation (2) we have calculated
experimentally not available oxidation potentials from their
adiabatic ionization potentials (Table 3).
to this correlation, one can see that DDQ reacts much more
slowly with π-nucleophiles than expected from its reduction
potential. The position of DDQ in Figure 10 implies that the
transition states of the reactions of DDQ with πCC systems
profit much less from the formation of the new CC bond
(product stabilizing factor) than the corresponding reactions of
benzhydrylium ions and quinone methides. Yu and co-workers
have demonstrated recently that a relationship between
electrophilicities E and LUMO energies holds only within
species that share similar substitution patterns.18 Thus,
arylidene Meldrum’s acids and arylidene barbituric acids were
found to be 5 orders of magnitude more electrophilic than
quinone methides of comparable LUMO energies, indicating
that the product stabilizing factor must be more important in
the transition states of the former species. The conclusion
drawn from Figure 10 that product stabilizing factors are less
important in the reactions of DDQ with π-nucleophiles implies
that SET processes can be expected to be more likely in
reactions of DDQ with nucleophiles than in the corresponding
reactions of benzhydrylium ions having similar electrophilicity
E.
IP = 1.88Eox + 5.55
(2)
a
DISCUSSION
■
Correlation Analysis. In order to examine the validity of eq
1 for the reactions of DDQ (C attack) with π-nucleophiles, a
plot of (log kC)/sN against the nucleophilicity parameters N of
nucleophiles is presented in Figure 9. A fair linear correlation
(R2 = 0.91) was obtained.
The kinetics of one-electron oxidations of silylated enol
ethers and ketene acetals as well as of allylsilanes with one
electron oxidants have previously been studied by Fukuzumi
and co-workers.15 The oxidation potentials of these π-systems
as well as the intrinsic barriers for the electron transfer
oxidation of these compounds by one-electron oxidants were
derived from experimental rate constants using the Rehm−
Weller Gibbs relationship (4), which derives the Gibbs energy
of activation for the electron transfer (ΔG‡et) from the Gibbs
energy of electron transfer (ΔG0et given by eq 5) and the
intrinsic barrier (ΔG‡0,et). The latter quantity represents the
Gibbs activation free energy for a process where the driving
force of electron transfer is zero, i.e. ΔG‡et = ΔG0‡,et for ΔGe0t = 0.
Due to the significant rearrangements of structure accompany-
ing electron transfer, large intrinsic barriers ΔG‡0,et (15−20 kJ
mol−1) were observed for the one-electron oxidation of a large
variety of silylated π-nucleophiles, including several compounds
shown in Figure 8.
Figure 9. Plot of (log kC)/sN vs N for the reactions of DDQ (C attack)
with π-nucleophiles in CH3CN at 20 °C.
According to eq 1, the electrophilicity parameter of DDQ
(reactivity at C-2) was determined as E = −3.66 by least-
squares minimization of Δ2 = Σ(log kC − sN(N + E))2 using the
second-order rate constants kC given in Table 2 and the N and
sN parameters of the nucleophiles 2a-n from Table 1.
In previous work we have shown that the electrophilicity
parameters of benzhydrylium ions and structurally related
By applying the same methodology, we have now used the
oxidation potentials in Table 3 to calculate the rate constants
for the outer-sphere electron transfer of the reactions of DDQ
12384
dx.doi.org/10.1021/ja405890d | J. Am. Chem. Soc. 2013, 135, 12377−12387

Journal of the American Chemical Society
Article
with the π-nucleophiles 2a−n according to eqs 3−5, assuming
an intrinsic barrier of ΔG‡0,et = 18 kJ mol−1, a typical value for
such systems according to ref 15.
Scheme 14. Synthesis of the Radical Clock Substrate 13
ket = (kbT/h) exp(−ΔGe‡t/RT)
(3)
ΔGe‡t = ΔGe0t/2 + [(ΔGe0t/2)2 + (ΔG0‡,et)2]1/2
(4)
ΔGe0t = F(Eox − Ered
)
(5)
A plot of experimental rate constants log kC (●) and log kO
(○) for the reactions of DDQ with various π-nucleophiles
versus the calculated rates for SET (log ket, from eqs 3−5) is
presented in Figure 11. The open circles for systems which
Scheme 15. Exclusion of OSET by a Radical Clock
Experiment (CH3CN, 20 °C)
Figure 11. Correlation of experimental rate constants (log k2, in
CH3CN) with calculated rate constants for SET (log ket).
react via O attack are close to the diagonal, indicating that these
reactions proceed with rates as expected for SET processes,
while C attack at DDQ (filled circles) is 2 to 5 orders of
magnitude faster than calculated for SET processes. We,
therefore, conclude that C attack at DDQ usually occurs by a
polar mechanism, which may be reversible if there is no fast
subsequent reaction.
It should be noted that all experimental rate constants for the
reactions of benzhydrylium ions with π-nucleophiles have been
reported to be at least 8 orders of magnitude larger than the
calculated rate constants for one-electron transfer processes.17
Therefore, the reactions of DDQ with π-nucleophiles are
generally closer to the borderline between polar reactions and
electron transfer than the corresponding reactions of
benzhydrylium ions (with π-nucleophiles).
14 allows us to exclude the operation of an OSET mechanism
because Mattay’s report on the formation of carbocycles via
photoinduced electron transfer oxidative cyclization of silyl enol
ethers carrying side chains with olefinic double bonds indicates
cyclizations of the intermediate radical cations to be faster than
competing intermolecular processes.21 Therefore, O attack is
suggested to proceed via an ISET mechanism.
CONCLUSION
■
π-nucleophiles attack DDQ either at C-2 to give 4-
hydroxycyclohexadienones or at oxygen to give O-substituted
hydroquinones. In several cases initial C attack and subsequent
rearrangement to the thermodynamically more stable products
of O attack has been observed. Kinetics for the reactions of π-
systems with DDQ have been determined photometrically, and
it was found that the rate constants kC for the attack of the π-
nucleophiles 2 at C-2 of DDQ can be described by the linear
free energy relationship (1), which allowed us to derive the
electrophilicity parameter E = −3.66 for the C-2 position of
DDQ. It thus possesses an electrophilic reactivity comparable
to the flavylium and tropylium ion, and is considerably more
reactive than the bis(dimethylamino)-substituted benzhydry-
lium ion or other highly reactive Michael acceptors (Figure
12).8,9a
Outer Sphere Electron Transfer (OSET) vs Inner
Sphere Electron Transfer (ISET): Radical Clock Experi-
ments. According to Figure 11, the observed rate constant for
the reaction of 2j with DDQ corresponds to that expected for a
SET process. In order to examine the occurrence of an OSET
mechanism, a radical clock19 substrate 13 was designed based
on the structure of 2j (Scheme 14). When 13 was combined
with DDQ, only 14, the product of O attack, was observed in
1
the H NMR spectrum of the crude material (>95%), and 15
was isolated in 60% yield after column chromatography, while
no cyclization product was observed (Scheme 15). Though
cyclization of the radical cation 16 may be slower than
analogous cyclizations of radicals, the quantitative formation of
12385
dx.doi.org/10.1021/ja405890d | J. Am. Chem. Soc. 2013, 135, 12377−12387

Journal of the American Chemical Society
Article
EXPERIMENTAL SECTION
■
Materials. DDQ was purchased from ABCR and recrystallized
from chloroform or dichloromethane before use. Silyl enol ethers 2(a-
e, a′, h) were prepared as reported in the literature.22 Silyl ketene
acetals 2(f, g, i) were prepared according to ref 23. The synthesis of
disiloxy hexane 2j was achieved by Ruhlmann-type acyloin
̈
condensation.24 Enamino esters 2l and 2m were prepared from
amines and methyl propiolate. Diazomethanes 2n and 2o were
prepared according to refs 25 and 26, respectively.
Kinetics. All of the rates of reactions were determined photo-
metrically in dry acetonitrile or in freshly distilled dry CH2Cl2. Fast
reactions were determined by using the stopped-flow technique. Slow
reactions were measured by conventional photodiode array UV−vis
spectrometers. The temperature was kept constant at 20 0.1 °C by
using a circulating bath thermostat. In all runs, excess of nucleophiles
(at least 8-fold) over DDQ was used to achieve pseudo first-order
kinetics.
ASSOCIATED CONTENT
* Supporting Information
■
S
Details of the kinetic experiments, synthetic procedures, X-ray
crystal structure determination, and quantum chemical
calculations as well as NMR spectra of all characterized
compounds. This material is available free of charge via the
Internet at http://pubs.acs.org.
Figure 12. Comparison of the electrophilic reactivity of DDQ with
that of other electrophiles.8
AUTHOR INFORMATION
Corresponding Author
herbert.mayr@cup.uni-muenchen.de
Notes
The experimental rate constants for the attack of π-systems at
C-2 of DDQ are larger than those calculated for SET processes
from the electrochemical oxidation potentials of the π-systems
and the reduction potential of DDQ, indicating that C attack at
DDQ usually occurs via polar processes. However, DDQ reacts
about 12 orders of magnitude more slowly than benzhydrylium
ions of comparable reduction potential showing that the
transition states of the reactions of DDQ with πCC systems
profit much less from the formation of the new CC-bond
(product stabilizing factor) than the corresponding reactions of
benzhydrylium ions, which can be explained by the high
stabilization of the DDQ radical anion. O-Attack is probably
restricted to reactions with sterically shielded nucleophiles or to
cases, where the faster C attack is highly reversible because the
resulting intermediate cannot undergo fast subsequent
reactions. The observed rate constants for O attack are smaller
than calculated by eq 1 and thus are similar to those for SET
processes (Scheme 16).
■
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
Dedicated to Professor Helmut Schwarz on the occasion of his
70th birthday. We thank the Deutsche Forschungsgemeinschaft
(SFB 749, Project B1) and the China Scholarship Council
(fellowship to X.G.) for financial support. We are grateful to
Dr. Peter Mayer for the X-ray diffraction experiments.
Furthermore, we thank Dr. Armin R. Ofial for help during
preparation of this manuscript.
REFERENCES
■
(1) Thiele, J.; Gunther, F. Liebigs Ann. Chem. 1906, 349, 45−66.
̈
(2) Andrieux, C. P.; Merz, A.; Saveant, J.-M.; Tomahogh, R. J. Am.
Chem. Soc. 1984, 106, 1957−1962.
Since the radical-clock experiment did not provide evidence
for the intermediacy of radical cations of the silylated enol
ethers, these reactions are interpreted as inner sphere electron
transfer processes.
(3) (a) Walker, D.; Hiebert, J. D. Chem. Rev. 1967, 67, 153−195.
(b) Becker, H.-D. In The Chemistry of the Quinoid Compounds, Part I;
Patai, S., Ed.; Wiley: Chichester, U.K., 1974; pp 335−423. (c) Becker,
H.-D.; Turner, A. B. In The Chemistry of the Quinoid Compounds Vol. 2;
Patai, S., Rappoport, Z., Eds.; Wiley: Chichester, U.K., 1988; pp 1351−
1384. (d) Buckle, D. R. In Encyclopedia of Reagents for Organic Synthesis
Vol. 3; Paquette, L. A., Ed.; Wiley: Chichester, U.K., 1995; p 1699.
(4) (a) Zhang, Y.; Li, C.-J. Angew. Chem., Int. Ed. 2006, 45, 1949−
1952. (b) Zhang, Y.; Li, C.-J. J. Am. Chem. Soc. 2006, 128, 4242−4243.
(c) Tu, W.; Floreancig, P. E. Angew. Chem., Int. Ed. 2009, 48, 4567−
4571. (d) Liu, L.; Floreancig, P. E. Angew. Chem., Int. Ed. 2010, 49,
3069−3072. (e) Guo, C.; Song, J.; Luo, S.-W.; Gong, L.-Z. Angew.
Chem., Int. Ed. 2010, 49, 5558−5562. (f) Liu, L.; Floreancig, P. E.
Angew. Chem., Int. Ed. 2010, 49, 5894−5897. (g) Hayashi, Y.; Itoh, T.;
Ishikawa, H. Angew. Chem., Int. Ed. 2011, 50, 3920−3924. (h) Tsang,
A. S.-K.; Jensen, P.; Hook, J. M.; Hashmi, A. S. K.; Todd, M. H. Pure
Appl. Chem. 2011, 83, 655−665. (i) Alagiri, K.; Devadig, P.; Prabhu, K.
R. Chem.Eur. J. 2012, 18, 5160−5164. (j) Rohlmann, R.; Garcia
Scheme 16. C versus O Attack in Reactions of DDQ with π-
Nucleophiles
Mancheno, O. Synlett 2013, 24, 6−10.
̃
(5) Bhattacharya, A.; DiMichele, L. M.; Dolling, U.-H.; Douglas, A.
W.; Grabowski, E. J. J. J. Am. Chem. Soc. 1988, 110, 3318−3319.
12386
dx.doi.org/10.1021/ja405890d | J. Am. Chem. Soc. 2013, 135, 12377−12387

Journal of the American Chemical Society
Article
(6) (a) Bhattacharya, A.; DiMichele, L. M.; Dolling, U.-H.;
Grabowski, E. J. J.; Grenda, V. J. J. Org. Chem. 1989, 54, 6118−
6120. (b) For an earlier report on the formation of enones by
oxidation of silyl enol ethers with DDQ, see: Ryu, I.; Murai, S.;
Hatayama, Y.; Sonoda, N. Tetrahedron Lett. 1978, 19, 3455−3458.
(7) (a) Mayr, H.; Patz, M. Angew. Chem., Int. Ed. Engl. 1994, 33,
938−957. (b) Mayr, H.; Ofial, A. R. Pure Appl. Chem. 2005, 77, 1807−
1821. (c) Mayr, H.; Ofial, A. R. J. Phys. Org. Chem. 2008, 21, 584−595.
(d) Mayr, H.; Kempf, B.; Ofial, A. R. Acc. Chem. Res. 2003, 36, 66−77.
(8) For a comprehensive database of nucleophilicity parameters N
and sN as well as electrophilicity parameters E, see: http://www.cup.
lmu.de/oc/mayr/DBintro.html.
(23) Oisaki, K.; Suto, Y.; Kanai, M.; Shibasaki, M. J. Am. Chem. Soc.
2003, 125, 5644−5645.
(24) Ruhlmann, K. Synthesis 1971, 236−253.
̈
(25) Wulfman, D. S.; Yousefian, S.; White, J. M. Synth. Commun.
1988, 18, 2349−2352.
(26) Searle, N. E. Org. Synth. 1956, 36, 25−27.
(9) (a) Mayr, H.; Bug, T.; Gotta, M. F.; Hering, N.; Irrgang, B.;
Janker, B.; Kempf, B.; Loos, R.; Ofial, A. R.; Remennikov, G.;
Schimmel, H. J. Am. Chem. Soc. 2001, 123, 9500−9512. (b) Lucius, R.;
Loos, R.; Mayr, H. Angew. Chem., Int. Ed. 2002, 41, 91−95. (c) Richter,
D.; Hampel, N.; Singer, T.; Ofial, A. R.; Mayr, H. Eur. J. Org. Chem.
2009, 3203−3211. (d) Ammer, J.; Nolte, C.; Mayr, H. J. Am. Chem.
Soc. 2012, 134, 13902−13911.
(10) Kucklaender, U.; Bollig, R.; Frank, W.; Gratz, A.; Jose, J. Bioorg.
Med. Chem. 2011, 19, 2666−2674.
(11) (a) Mulliken, R. S.; Person, W. B. Molecular Complexes; Wiley-
Interscience: New York, 1969. (b) Mataga, N.; Kubota, T. Molecular
Interactions and Electronic Spectra; Marcel Dekker: New York, 1970.
(12) (a) Seeman, J. I. Chem. Rev. 1983, 83, 83−134. (b) Curtin, D. Y.
Rec. Chem. Prog. 1954, 15, 111. (c) Hammett, L. P. Physical Organic
Chemistry; McGraw-Hill: New York, 1970.
(13) Frisch, M. J.; Trucks, G. W.; Schlegel, H.; Scuseria, G. E.; Robb,
M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.;
Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.;
Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.;
Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima,
T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A.;
Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin,
K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.;
Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega,
N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.;
Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.;
Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.;
Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.;
Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas; Foresman, J.
B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09, revision A.02;
Gaussian, Inc.: Wallingford, CT, 2009.
(14) Henry, D. J.; Sullivan, M. B.; Radom, L. J. Chem. Phys. 2003,
118, 4849−4860.
(15) Fukuzumi, S.; Fujita, M.; Otera, J.; Fujita, Y. J. Am. Chem. Soc.
1992, 114, 10271−10278.
(16) Pysh, E. S.; Yang, N. C. J. Am. Chem. Soc. 1963, 85, 2124−2130.
(17) Ofial, A. R.; Ohkubo, K.; Fukuzumi, S.; Lucius, R.; Mayr, H. J.
Am. Chem. Soc. 2003, 125, 10906−10912.
(18) Zhou, L.-G.; Liao, W.; Yu, Z.-X. Asian J. Org. Chem. 2012, 1,
336−345.
(19) (a) Newcomb, M.; Curran, D. P. Acc. Chem. Res. 1988, 21, 206−
214. (b) Tolbert, L. M.; Sun, X.-J.; Ashby, E. C. J. Am. Chem. Soc.
1995, 117, 2681−2685. (c) Zard, S. Z. Radical Reactions in Organic
Synthesis; Oxford University Press: New York, 2003.
(20) Liu, Z.; Peng, L.; Zhang, T.; Li, Y. L. Synth. Commun. 2001, 31,
2549−2555.
(21) (a) Bunte, J. O.; Rinne, S.; Schafer, C.; Neumann, B.; Stammler,
̈
H.-G.; Mattay, J. Tetrahedron Lett. 2003, 44, 45−48. (b) MacMillan’s
observation of the intramolecular attack of an enamine radical cation at
a 1,2-dialkylated ethylene with formation of a five-membered ring,
which has a much lower driving-force confirms this conclusion:
Beeson, T. D.; Mastracchio, A.; Hong, J.-B.; Ashton, K.; MacMillan, D.
W. C. Science 2007, 316, 582−585.
(22) (a) Cazeau, P.; Duboudin, F.; Moulines, F.; Babot, O.;
Dunogues, J. Tetrahedron 1987, 43, 2075−2088. (b) Cazeau, P.;
Duboudin, F.; Moulines, F.; Babot, O.; Dunogues, J. Tetrahedron 1987,
43, 2089−2100.
12387
dx.doi.org/10.1021/ja405890d | J. Am. Chem. Soc. 2013, 135, 12377−12387