Investigating the reactivity of radical cations: Experimental and computational insights into the reactions of radical cations with alcohol and p -toluene sulfonamide nucleophiles

Article abstract of DOI:10.1021/ja307046j

The reactivity of electrochemically generated radical cations toward alcohol and p-toluene sulfonamide nucleophiles was directly investigated through competition experiments. Alcohol-trapping of the radical cation is the kinetically favored pathway and is

Full text of DOI:10.1021/ja307046j

Article
pubs.acs.org/JACS
Investigating the Reactivity of Radical Cations: Experimental and
Computational Insights into the Reactions of Radical Cations with
Alcohol and p‑Toluene Sulfonamide Nucleophiles
John M. Campbell, Hai-Chao Xu, and Kevin D. Moeller*
Washington University in Saint Louis, Saint Louis, Missouri 63130, United States
S
* Supporting Information
ABSTRACT: The reactivity of electrochemically generated
radical cations toward alcohol and p-toluene sulfonamide
nucleophiles was directly investigated through competition
experiments. Alcohol-trapping of the radical cation is the
kinetically favored pathway and is reversible. Trapping with the
sulfonamide leads to the thermodynamic product. Both
reaction pathways were investigated computationally with
density functional theory (UB3LYP/6-31G(d)) calculations.
a
Scheme 1. General Design of a Competition Study
INTRODUCTION
■
Radical cations are important intermediates in organic
chemistry and may be employed to initiate and accomplish a
variety of reactions and transformations.1−6 Our work has
focused on the chemistry of radical cations generated by
electrochemical oxidation of electron-rich double bonds. The
reactions are of synthetic interest because they reverse the
polarity of enolate equivalents in a manner that allows them to
function as an electrophile.7,8 The reactions are easily
implemented9 and general in their application.10 However, an
accurate mechanistic and kinetic understanding of these
reactions can be difficult to achieve. Newcomb and co-workers
nicely explored the chemistry of enol ether radical cations and
were able to establish reaction rates.11 We seek to further
investigate the chemoselectivity of radical cation reactions and
explore the reactivity of radical cations formed from a variety of
electron-rich olefins.
To address these questions, a competition experiment was
designed so that the relative reactivity of olefin-based radical
cations toward various intramolecular nucleophiles can be
directly investigated. As shown in Scheme 1, the plan calls for
an electron-rich olefin to be tethered to two different
nucleophiles. Upon oxidation of 1, the radical cation
intermediate 2 can be trapped by either nucleophile. The
ratio of the products ultimately formed (3 and 4) will then yield
information on the relative selectivity of the radical cation for
the two nucleophiles.
Initial efforts along these lines have focused on substrates
with an alcohol and a p-toluene sulfonamide nucleophile (5,
Scheme 2). The studies were motivated by the observation that
oxidative cyclizations between electron-rich olefins and
sulfonamide nucleophiles benefited greatly from the use of
basic reaction conditions.12,13 The base deprotonated the
sulfonamide prior to the electrolysis, a situation that led to
questions about the mechanism of the cyclization. Did the
a
The products obtained (3 and 4) yield information on the relative
selectivity of the radical cation 2 for the two different nucleophiles
(X,Y = electron donating).
addition of base aid the cyclization by enhancing the
nucleophilic addition of the sulfonamide to the radical cation
Received: July 18, 2012
Published: October 13, 2012
© 2012 American Chemical Society
18338
dx.doi.org/10.1021/ja307046j | J. Am. Chem. Soc. 2012, 134, 18338−18344

Journal of the American Chemical Society
Article
reversible. The sulfonamide-based cyclization leads to the
formation of the thermodynamic product.
Scheme 2. Substrate for the Competition Experiment
Presented in This Article
a
RESULTS AND DISCUSSION
■
Our initial understanding of the reaction originating from the
oxidation of 5 is illustrated in Scheme 4. We postulated that if
the reaction was kinetically controlled, it would be governed by
a
X,Y = electron donating.
Scheme 3. Mechanism for Sulfonamide-Cyclization May
Proceed through Either 7 or 8
a
Scheme 4. An Initial Guess at the Energy Profile for This
Competition Experiment
a
X,Y = electron donating.
the nature of the initial cyclization of 10 and/or 11, that is, the
relative energies of the sulfonamide-cyclization and the alcohol-
trapping transition states, 12 and 13, respectively. This
hypothesis was consistent with an application of the Curtin−
Hammett principle, an approach that had served us well in the
past.17 We were eager to see whether or not the Curtin−
Hammett principle would apply in this case.
It was noted that the sulfonamide-cyclization pathway
proceeded through a transition state (12) that was likely to
exhibit very little charge separation, especially in comparison to
the zwitterionic transition state of the alcohol-trapping pathway
(13). With this in mind, it appeared possible to select for
sulfonamide-cyclization with the use of a nonpolar reaction
medium. Such a medium is achieved through the choice of
solvent and electrolyte.
Solvent Experiments. Tables 1−3 show the results of
several solvent polarity studies. In all cases, the use of 100%
methanol as a solvent and lithium perchlorate as an electrolyte
produced products arising from both sulfonamide-cyclization
and alcohol-trapping. Upon the addition of tetrahydrofuran as a
cosolvent, up to 30% MeOH/THF, an increase in sulfonamide-
cyclization relative to alcohol-trapping was observed. Further-
more, using tetraethylammonium tosylate instead of lithium
perchlorate as the electrolyte also served to decrease the
polarity of the reaction medium and led to an increase in
sulfonamide-cyclization. Selectivity for alcohol-trapping was
most successfully achieved by substituting 2,6-lutidine for
lithium methoxide as a base, thereby avoiding deprotonation of
the sulfonamide.
In the case of the enol ether-derived radical cation, the yield
of alcohol-trapping products was low even with the use of 2,6-
lutidine as the base. The low yield appeared to be due to the
instability of 15c and 15d. No such problem arose following the
oxidation of the ketene dithioacetal and vinyl sulfide substrates.
In the case of the vinyl sulfide (Table 3), electrolysis of
initially synthesized substrates led to only sulfonamide-
cyclization products in all cases. Thus, for this particular
(intermediate 8, Scheme 3), or did it change the mechanism of
the reaction so that the cyclization resulted from the addition of
a nitrogen radical to the olefin (intermediate 7, Scheme 3)?
The two mechanisms are potentially related by a fast
intramolecular electron-transfer reaction between the two key
intermediates 7 and 8. Both reaction pathways lead to
intermediate 9.
The alcohol was selected as a competitor for the sulfonamide
because it is known to efficiently trap radical cations.14
Furthermore, unlike many alternative trapping groups, its use
does not add an additional stereocenter to the cyclized
products.
A second goal of this research was to determine how
substituents on the double bond (X and Y in 5) alter the
selectivity of the reaction. Within the context of intramolecular
anodic olefin coupling reactions, it has been observed that the
olefin substituents can have significant effects on the overall
success and chemoselectivity of a cyclization.15 In this study, we
explore the use of enol ethers, ketene dithioacetals, and vinyl
sulfides.16
In the initial study,16 the mechanism of the sulfonamide-
cyclization was probed by examining the effect of solvent
polarity on the oxidation of 5. The experiments confirmed that
the intramolecular electron transfer shown in Scheme 3 did
happen. In addition, cyclization with the sulfonamide
nucleophile in 5 benefited from the use of nonpolar reaction
conditions, while the amount of alcohol-trapping increased with
the use of a more polar medium. These observations were
consistent with a kinetically controlled reaction that was
governed by the Curtin−Hammett principle.
Recent computational studies suggest that this picture of the
mechanism is overly simplistic, a conclusion that is supported
by the effect that both temperature and current density have on
the reactions. We report here that the alcohol-trapping pathway
that results from the oxidation of 5 leads to the kinetic product.
With the reaction conditions used previously, this reaction is
18339
dx.doi.org/10.1021/ja307046j | J. Am. Chem. Soc. 2012, 134, 18338−18344

Journal of the American Chemical Society
Article
Table 1. Results of Solvent Polarity Experiments for an Enol
Ether Substrate
Table 3. Results of Solvent Polarity Experiments for a Vinyl
Sulfide Substrate
a
% yield
% yield
solvent (electrolyte, base)
15a + 15b
15c + 15d
solvent (electrolyte, base)
19a
19b
100% MeOH (LiClO4, LiOMe)
48 (36 + 12)
85 (66 + 19)
79 (62 + 17)
83 (75 + 8)
82 (66 + 16)
91 (71 + 20)
0
10 (0 + 10)
b
100% MeOH (LiClO4, LiOMe)
28
25
29
54
45
34
0
19
16
9
60% MeOH/THF (LiClO4, LiOMe)
30% MeOH/THF (LiClO4, LiOMe)
100% MeOH (Et4NOTs, LiOMe)
60% MeOH/THF (Et4NOTs, LiOMe)
30% MeOH/THF (Et4NOTs, LiOMe)
0
b
60% MeOH/THF (LiClO4, LiOMe)
0
b
30% MeOH/THF (LiClO4, LiOMe)
0
c
100% MeOH (Et4NOTs, LiOMe)
0
0
c
60% MeOH/THF (Et4NOTs, LiOMe)
0
0
c
a
30% MeOH/THF (Et4NOTs, LiOMe)
0
30% MeOH/THF (LiClO4)
20 (0 + 20)
27 (0 + 27)
d
a
30% MeOH/THF (LiClO4)
44
79
30% MeOH/THF (Et4NOTs)
0
d
a
30% MeOH/THF (Et4NOTs)
0
2,6-Lutidine was used as a base.
a
In this case, the sulfonamide tether was extended by one methylene
b
c
unit. 3.0 F/mol passed during electrolysis. 4.0 F/mol passed during
Table 2. Results of Solvent Polarity Experiments for a
Ketene Dithioacetal Substrate
d
electrolysis. 2,6-Lutidine was used as a base.
Table 4. Half-Wave Oxidation Potentials (Ep/2) for Various
Functional Groups
% yield
solvent (electrolyte, base)
17a + 17b
17c + 17d
100% MeOH (LiClO4, LiOMe)
50 (40 + 10)
75 (62 + 13)
83 (71 + 12)
58 (48 + 10)
76 (66 + 16)
87 (70 + 17)
0
27 (27 + 0)
16 (16 + 0)
3 (3 + 0)
20 (20 + 0)
6 (6 + 0)
0
60% MeOH/THF (LiClO4, LiOMe)
30% MeOH/THF (LiClO4, LiOMe)
100% MeOH (Et4NOTs, LiOMe)
60% MeOH/THF (Et4NOTs, LiOMe)
30% MeOH/THF (Et4NOTs, LiOMe)
a
A: Carbon anode, platinum cathode, Ag/AgCl reference electrode,
scan rate 50 mV/s, 0.1 M Et4NOTs, substrate concentration of 0.025
b
M in methanol. B: 0.6 equiv of LiOMe was added. Oxidation
potential exceeded that of the solvent.
a
30% MeOH/THF (LiClO4)
87 (72 + 15)
56 (39 + 17)
a
30% MeOH/THF (Et4NOTs)
0
measured without the addition of lithium methoxide (con-
ditions A). It is worth noting that potentials measured for the
substrates in the presence of lithium methoxide (conditions B)
were lower than that of the sulfonamide anion (entry 4B). This
finding is consistent with a cyclization that occurs at or near the
electrode surface.18
Before interpreting the oxidation potentials, it is important to
recognize that the preparative electrolysis experiments were run
under constant current conditions. This means that the voltage
across the cell was varied dynamically to maintain a preset
current. In such experiments, the potential at the working
electrode adjusts automatically to match the functional group
with the lowest oxidation potential. Hence, given the data in
Table 4, the preparative reactions led to initial oxidation of the
sulfonamide anion and the formation of an olefin/nitrogen
radical pair. This is interesting because the alcohol in our
competition experiments cyclized competitively with the
sulfonamide, a transformation that requires formation of the
radical cation at the olefin. Clearly, an intramolecular electron
transfer between the olefin and the sulfonamide anion does
occur.
a
2,6-Lutidine was used as a base.
substrate, we chose to extend the tether connecting the
sulfonamide to the olefin by one methylene unit. This slowed
the sulfonamide-cyclization and allowed alcohol-trapping to
compete, although it also led to lower overall yields and lower
current efficiency due to side reactions associated with
sulfonamide-cyclizations forming six-membered rings.13
All three electrolysis substrates behaved according to the
same general trend. Nonpolar conditions favored sulfonamide-
cyclization, while polar conditions favored alcohol-trapping.
Differences in selectivity between the three substrates were
attributed to the olefin substituents.
Oxidation Potentials. To gain further insight into the
mechanism of sulfonamide-cyclization, the oxidation potentials
of the functional groups involved in the cyclization were
investigated. Interestingly, the deprotonated sulfonamide
exhibited an oxidation potential lower than either the enol
ether, the ketene dithioacetal, or the vinyl sulfide (Ep/2 = 0.9 V
vs Ag/AgCl, Table 4). The potentials for the olefins were
18340
dx.doi.org/10.1021/ja307046j | J. Am. Chem. Soc. 2012, 134, 18338−18344

Journal of the American Chemical Society
Article
Computational Results.19 Pushing for a deeper under-
standing of the mechanisms of sulfonamide-cyclization and
alcohol-trapping, we turned to a computational approach.
Density functional theory (DFT) has been utilized by
others20−25 to investigate the chemistry of radical cations due
to its low computational cost and ability to avoid problems of
spin contamination, a serious consideration when modeling
open-shell molecules.26 However, the limitations of DFT,
including its tendency to underestimate dissociation barriers27
and to unduly delocalize spin and charge,22,28 are well
documented. Hence, all calculations reported herein are
presented as a supplement to the experimental data. We
elected to use the unrestricted B3LYP functional due to its
successful application to other radical cation systems. Recently,
other functionals have enjoyed successful application to radical
cation systems, especially the M06 family of functionals.29
Transition structures for sulfonamide-cyclization with an
enol ether, a ketene dithioacetal, and a vinyl sulfide were
located using DFT (UB3LYP/6-31G(d)) and are depicted in
Figures 1−3. The alcohol tether was excluded from the
calculation to save computation time. Each transition structure
exhibited only one vibration with an imaginary frequency. The
length of the forming nitrogen−carbon bond was between 2.07
and 2.21 Å, depending on the olefin substituents.
Figure 4. Intrinsic reaction coordinate (IRC) around the transition
structure for the coupling of a sulfonamide with a ketene dithioacetal.
The color maps shown represent the unpaired electron density
(expressed in units of unpaired electrons per cubic bohr (eα − eβ)/a3o)
mapped onto the total electron density surface.
The intrinsic reaction coordinate (IRC) around each
transition structure was then computed. In all sulfonamide-
cyclizations, the unpaired electron spin in the intermediates
leading from the reactant to the transition structure was
primarily localized at the nitrogen of the sulfonamide. For a
representative example, see Figure 4. These results indicated
that sulfonamide-cyclization is best described as a radical-like
cyclization. This finding contradicts the conclusions reported
from our earlier investigation of this competition experiment16
in which we postulated that the reaction involved an olefin-
localized radical cation and an anionic sulfonamide nucleophile.
To quantify the energetics associated with sulfonamide-
cyclization, energies of activation and net changes in energy
were calculated and are presented in Table 5. Our conclusions
Figure 1. Transition structure for the coupling of a sulfonamide to an
enol ether.
Table 5. Calculated Activation Energies (Eact) and Net
Changes in Energy (ΔE) Given in Terms of Electronic
Energy (ESCF) and Gibbs Free Energy (G)
Figure 2. Transition structure for the coupling of a sulfonamide to a
ketene dithioacetal.
from these data are that sulfonamide-cyclization is exothermic
and has a low-to-modest energy of activation that is on the
order of a few kcal/mol depending on the identity of the olefin
substituents.
Similar attempts to explore the alcohol-trapping pathway
computationally were not as successful. Unable to find a
transition structure, we submitted the simplified product of an
Figure 3. Transition structure for the coupling of a sulfonamide to a
vinyl sulfide.
18341
dx.doi.org/10.1021/ja307046j | J. Am. Chem. Soc. 2012, 134, 18338−18344

Journal of the American Chemical Society
Article
Scheme 5. Distonic Radical Cation 21 Not at a Local Energy
Minima
Table 6. Results of Temperature Experiments for an Enol
Ether Substrate
a
a
No transition structure for the cyclization or the deprotonation could
be found.
% yield
temp (°C)
15a + 15b
15c + 15d
alcohol-trapping cyclization to Gaussian for geometric opti-
mization (21, Scheme 5). However, this structure was not at an
energy minimum. Instead, the optimized geometry was that of
the starting material (20), with the unpaired electron spin
distributed over both carbons of the olefin. Hence, it was
postulated that a deprotonation of the alcohol occurs in the
transition state of the cyclization, leading directly to 22.
Such a hypothesis is not without precedent. Zipse reported a
computational investigation of the reaction of water and
ethylene radical cation.30 It was found that the formation of a
distonic radical cation had no barrier, and that the resultant
product was highly acidic. Furthermore, Okazaki and co-
workers published a kinetics study of reactions of alcohols with
9,10-diphenylanthracene radical cations, concluding that
deprotonation of the alcohol was the rate-determining step of
the reaction.31 More recently, Arnold and Newcomb have both
proposed that a deprotonation is involved in the alcohol-
trapping of olefin-based radical cations.5,32
Despite this mechanistic insight, a transition structure for
alcohol-trapping could not be found, and the search for one was
abandoned. Furthermore, attempts to quantify the energetics
involved in alcohol-trapping were also unsuccessful in that the
calculation results were not consistent with experiment.
Alcohol-trapping involves the neutralization of a charged
species and a deprotonation, both of which are likely to
occur at or near the electrode, that is, within the electrical
double layer and surrounded by electrolyte. It is likely that
limitations of the commonly available solvation models prohibit
an accurate calculation of the relevant species.
45
25
65 (9 + 56)
46 (6 + 40)
3 (3 + 0)
0
7 (0 + 7)
37 (0 + 37)
33 (0 + 33)
33 (0 + 33)
0
−15
−42
2 (2 + 0)
4 (4 + 0)
Table 7. Results of Temperature Experiments for a Ketene
Dithioacetal Substrate
% yield
a
temp (°C)
17a + 17b
17c + 17d
45
25
71 (62 + 9)
50 (40 + 10)
31 (31 + 0)
18 (18 + 0)
9 (9 + 0)
26 (24 + 2)
27 (27 + 0)
38 (34 + 4)
37 (33 + 4)
15 (15 + 0)
0
−23
−42
a
The starting material became insoluble at lower temperatures.
Table 8. Results of Temperature Experiments for a Vinyl
Sulfide Substrate
While sulfonamide-cyclizations also occur at or near the
electrode surface, they are entirely intramolecular, involve no
formal charges, and are more isodesmic in nature. For these
reasons, we believe that a computational investigation of
sulfonamide-cyclization is less prone to error.
At this point, we turned to an experimental approach to
explore the energetics of alcohol-trapping relative to
sulfonamide-cyclization. With the energetics of sulfonamide-
cyclization in hand, we were eager to see which of the two
pathways was the kinetically favored one. Thus far, the
competition experiments appeared to indicate that sulfona-
mide-cyclization was kinetically favored under most reaction
conditions (see Tables 1−3). However, if indeed the barrier to
alcohol-trapping involved only an intramolecular electron
transfer and a deprotonation, how high could the alcohol-
trapping barrier be relative to the energetics presented in Table
5? We decided to begin by exploring the effects of reaction
temperature on our competition experiments.
% yield
temp (°C)
24a + 24b
24c + 24d
25
0
82 (30 + 52)
53 (28 + 25)
32 (13 + 19)
28 (6 + 22)
1−3 (0 + 1−3)
21 (0 + 21)
38 (3 + 35)
46 (4 + 42)
−23
−42
temperatures are lowered, alcohol-trapping eventually becomes
the major pathway. In the case of the enol ether (Table 6),
isolation of the alcohol-trapping products remains problematic,
and as a result the yields reported are low. In the case of the
ketene dithioacetal (Table 7), the starting material becomes
insoluble in the solvent at low temperatures, leading to lower
overall yields at lower temperatures.
Temperature Experiments. The electrolysis reactions
were repeated, varying the reaction temperature from as high
as 45 °C to as low as −42 °C. As shown in Tables 6−8, lower
temperatures led to an increase in the yield of alcohol-trapping
products relative to sulfonamide-cyclization products. As
The fact that lower temperatures lead to an increase in
alcohol-trapping products indicates that the enthalpic energy
18342
dx.doi.org/10.1021/ja307046j | J. Am. Chem. Soc. 2012, 134, 18338−18344

Journal of the American Chemical Society
Article
barrier (ΔH⧧) to alcohol-trapping is lower than that of
sulfonamide-cyclization. Yet, at high temperatures, sulfona-
mide-cyclization becomes the major product. There are two
possible explanations for this. First, the entropic energy barrier
(ΔS⧧) may be smaller for sulfonamide-cyclization. If this were
the case, then alcohol-trapping would be kinetically favored at
lower temperatures and sulfonamide-cyclization kinetically
favored at higher temperatures. This seemed very plausible,
given the evidence for a deprotonation in the alcohol-trapping
transition state.
Table 9. Results of Current Experiments for Anodic
Coupling Reactions
The second possibility is that the reactions at higher
temperatures are under thermodynamic control. This would
mean that higher temperatures allow for equilibration to the
thermodynamic product, while lower temperatures select for
the formation of the kinetic product. For this to be the case,
alcohol-trapping of the radical cation would need to be
reversible.
a
Reaction temperature kept at 0 °C.
Current Density Experiments. What was needed to
distinguish the two mechanistic possibilities was a method that
would alter the reversibility of the cyclization without changing
the temperature (and hence TΔS⧧) of the reaction. In an
electrolysis reaction, this can be accomplished by varying the
current density. Consider the working model for the
cyclizations illustrated in Scheme 6.
CONCLUSION
■
Our current understanding of the competition experiment
reported herein, based upon our experimental and computa-
tional investigations, is illustrated in Scheme 7. Our
Scheme 7. Revised Understanding of the Energy Profile for
This Competition Experiment
Scheme 6. Kinetics of the Cyclization of a Radical Cation
with a Nucleophile
In this scenario, the reversibility of the cyclization of 25
(whose forward and reverse rates are indicated by k1 and k−1,
respectively) is determined by k2. When k2 < k−1, the cyclization
reaction is under thermodynamic control, and the cyclization
that leads to the lowest-energy intermediate will give rise to the
major product. When k2 ≈ k−1, the cyclization is governed by
steady-state kinetics. When k2 > k−1, the cyclization reaction is
under kinetic control, and it is the cyclization with the lowest
energy of activation that leads to the major product of the
reaction. So, if we can accelerate the second oxidation (make k2
larger), then a thermodynamically controlled process can be
converted to a kinetically controlled one. One of the advantages
of anodic oxidation is that the rate of oxidation can be directly
controlled. By increasing or decreasing the flow of current, we
can alter the rate of electron transfer and hence the rate at
which both the substrate and the radical 26 are oxidized. A fast
rate of oxidation can be used to select the kinetic cyclization.
Applying this method to the oxidation of substrate 5 led to
the results reported in Table 9. For all three substrates,
sulfonamide-cyclization was favored when the current was set
to 6 mA. However, when the current was increased to 46 mA,
alcohol-trapping was favored. These results showed that
alcohol-trapping leads to the kinetic product of the electrolysis
reactions. Furthermore, it demonstrated that alcohol-trapping
of the radical cation is reversible. This is evidenced by the fact
that if the current is low enough, the intermediates will
equilibrate to the thermodynamically favored intermediate
product.
experimental results, taken all together, indicate that the
reaction is not under Curtin−Hammett control as we
previously reported.16 Instead, it is a competition between
the kinetic pathway and the thermodynamic one.
The anodic coupling of a sulfonamide to an electron-rich
olefin under basic conditions exhibited a radical-like mecha-
nism, with the initial oxidation occurring at the deprotonated
sulfonamide. However, an intramolecular electron transfer
occurs that affords an olefinic radical cation. This radical cation
can be trapped by a competing alcohol nucleophile. When in
direct competition with alcohol-trapping, sulfonamide-cycliza-
tion was found to afford the thermodynamically favored
product and may be enhanced with nonpolar reaction
conditions, elevated temperatures, and low current. Alcohol-
trapping of the radical cation was found to be facile, reversible,
and afforded the kinetically favored product. Alcohol-trapping
may be promoted with polar reaction conditions, low
temperatures, and high current.
18343
dx.doi.org/10.1021/ja307046j | J. Am. Chem. Soc. 2012, 134, 18338−18344

Journal of the American Chemical Society
Article
(15) Tang, F.; Moeller, K. D. Tetrahedron 2009, 65, 10863−10875.
(16) For a preliminary account of this work, see: Xu, H.-C.; Moeller,
K. D. Org. Lett. 2010, 12, 1720−1723.
(17) Redden, A.; Moeller, K. D. Org. Lett. 2011, 13, 1678−1681.
(18) Xu, H.-C.; Moeller, K. D. Angew. Chem., Int. Ed. 2010, 49,
8004−8007.
(19) Density functional theory (DFT) was used to calculate
optimized geometries, find transition structures, and compute intrinsic
reaction coordinate pathways using the unrestricted B3LYP functional,
a 6-31G(d) basis set, and the PCM methanol solvation model. Initial
molecular geometries were built and optimized using semi-emperical
methods (AM1) in Spartan (Spartan ’10; Wavefunction, Inc.: Irvine,
CA). These geometries were then imported into the Gaussian 03 suite
of programs (Frisch, M. J.; et al. Gaussian 03, revision C.02; Gaussian,
Inc.: Wallingford, CT, 2004) for all subsequent calculations.
Calculations were performed at the Washington University in Saint
Louis Computational Chemistry Facility. Images of transition
structures were generated using CYLview (Legault, C. Y. CYLview,
These conclusions are consistent with earlier observations of
radical cation/alcohol-trapping reactions and may well have
broad implications for a variety of radical cation-induced
cyclizations. Effort to extend these findings to additional
cyclizations is continuing with the use of new competition
substrates.
ASSOCIATED CONTENT
* Supporting Information
■
S
A general procedure for the electrolysis reaction, procedures for
the synthesis of electrolysis substrates, characterization of
electrolysis substrates, and a description of the computational
methods for calculating optimized geometries, transition
structures, and intrinsic reaction coordinates. This material is
available free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
moeller@wuchem.edu
́
1.0b; Universite de Sherbrooke, 2009 (http://www.cylview.org)).
■
(20) (a) Wiest, O. J. Am. Chem. Soc. 1997, 119, 5713−5719.
(b) Haberl, U.; Wiest, O.; Steckhan, E. J. Am. Chem. Soc. 1999, 121,
6730−6736. (c) Swinarski, D. J.; Wiest, O. J. Org. Chem. 2000, 65,
6708−6714. (d) Radosevich, A. T.; Wiest, O. J. Org. Chem. 2001, 66,
5808−5813. (e) Oxgaard, J.; Wiest, O. J. Phys. Chem. A 2002, 106,
3967−3974. (f) Donoghue, P. J.; Wiest, O. Chem.-Eur. J. 2006, 12,
7018−7026. (g) Valley, N. A.; Wiest, O. J. Org. Chem. 2007, 72, 559−
566.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work is supported by the National Science Foundation,
grant CHE-0809142. The Washington University Computa-
tional Chemistry Facility was constructed with support from
the National Science Foundation, grant CHE-0443501.
(21) (a) Fry, A. J. Tetrahedron 2008, 64, 2101−2103. (b) Wu, X.;
Davis, A. P.; Lambert, P. C.; Steffen, L. K.; Toy, O.; Fry, A. J.
Tetrahedron 2009, 65, 2408−2414.
(22) Sodupe, M.; Rodríguez-Santiago, J. B. L.; Baerends, E. J. J. Phys.
Chem. A 1999, 103, 166−170.
REFERENCES
■
́
(23) Sastry, G. N.; Bally, T.; Hrouda, V.; Carsky, P. J. Am. Chem. Soc.
(1) Schmittel, M.; Burghart, A. Angew. Chem., Int. Ed. Engl. 1997, 36,
2550−2589.
1998, 120, 9323−9334.
(24) Herbertz, T.; Roth, H. D. J. Am. Chem. Soc. 1998, 120, 11904−
(2) (a) Jang, H.-Y.; Hong, J.-B.; MacMillan, D. W. C. J. Am. Chem.
Soc. 2007, 129, 7004−7005. (b) Beeson, T. D.; Mastracchio, A.; Hong,
J.-B.; Ashton, K.; MacMillan, D. W. C. Science 2007, 27, 582−585.
(c) J. J. D., III; Conrad, J. C.; MacMillan, D. W. C.; R. A. F., II. Angew.
Chem., Int. Ed. 2010, 49, 6106−6110.
(3) (a) Liu, L.; Floreancig, P. E. Org. Lett. 2009, 11, 3152−3155.
(b) Liu, L.; Floreancig, P. E. Angew. Chem., Int. Ed. 2010, 49, 5894−
5897. (c) Tu, W.; Liu, L.; Floreancig, P. E. Angew. Chem., Int. Ed. 2008,
47, 41844187. (d) Green, M. E.; Rech, J. C.; Floreancig, P. E. Angew.
Chem., Int. Ed. 2008, 47, 7317−7320. (e) Jung, H. H.; J. R. S., II;
Floreancig, P. E. Angew. Chem., Int. Ed. 2007, 46, 8464−8467.
(f) Floreancig, P. E. Synlett 2007, 191−203.
(4) (a) Horner, J. H.; Newcomb, M. J. Org. Chem. 2007, 72, 1609−
1616. (b) Horner, J. H.; Lal, M.; Newcomb, M. Org. Lett. 2006, 8,
5497−5500. (c) Bagnol, L.; Horner, J. H.; Newcomb, M. Org. Lett.
2003, 5, 5055−5058.
(5) Horner, J. H.; Bagnol, L.; Newcomb, M. J. Am. Chem. Soc. 2004,
126, 14979−14987.
(6) (a) Crich, D.; Sirai, M.; Brebion, F.; Rumthao, S. Tetrahedron
2006, 62, 6501−6518. (b) Crich, D.; Sirai, M.; Brebion, F.; Rumthao,
S. Org. Lett. 2003, 5, 3767−3769.
11911.
(25) Um, J. M.; Gutierrez, O.; Schoenebeck, F.; Houk, K. N.;
MacMillan, D. W. C. J. Am. Chem. Soc. 2010, 132, 6001−6005.
(26) Bally, T.; Borden, W. T. In Reviews in Computational Chemistry;
Lipkowitz, K. B., Boyd, D. B., Eds.; John Wiley and Sons, Inc.: New
York, 1999; Vol. 13, pp 1−97.
(27) Bally, T.; Sastry, G. N. J. Phys. Chem. A 1997, 101, 7923−7925.
(28) Braïda, B.; Hiberty, P. C. J. Phys. Chem. A 1998, 102, 7872−
7877.
(29) (a) Zhao, Y.; Truhlar, D. G. J. Chem. Phys. 2006, 125, 194101.
(b) Zhao, Y.; Truhlar, D. G. J. Phys. Chem. A 2006, 110, 13126−13130.
(c) Zhao, Y.; Truhlar, D. G. Theor. Chem. Acc. 2008, 120, 215−241.
(d) Zhao, Y.; Truhlar, D. G. Acc. Chem. Res. 2008, 41, 157−167.
(30) Zipse, H. J. Am. Chem. Soc. 1995, 117, 11798−11806.
(31) Oyama, M.; Nozaki, K. N.; Nagaoka, T.; Okazaki, S. Bull. Chem.
Soc. Jpn. 1990, 63, 33−41.
(32) Mangion, D.; Arnold, D. R. Acc. Chem. Res. 2002, 35, 297−304.
(7) Little, R. D.; Moeller, K. D. Electrochem. Soc. Interface 2002, 11,
36−42.
(8) Tang, F.; Chen, C.; Moeller, K. D. Synthesis 2007, 3411−3420.
(9) Frey, D. A.; Wu, N.; Moeller, K. D. Tetrahedron Lett. 1996, 37,
8317−8320.
(10) Moeller, K. D. Synlett 2009, 8, 1208−1218.
(11) Horner, J. H.; Taxil, E.; Newcomb, M. J. Am. Chem. Soc. 2002,
124, 5402−5410.
(12) Xu, H.-C.; Moeller, K. D. J. Am. Chem. Soc. 2008, 130, 13542−
13543.
(13) Xu, H.-C.; Moeller, K. D. J. Am. Chem. Soc. 2010, 132, 2839−
2844.
(14) Xu, G.; Moeller, K. D. Org. Lett. 2010, 12, 2590−2593 and
references therein.
18344
dx.doi.org/10.1021/ja307046j | J. Am. Chem. Soc. 2012, 134, 18338−18344