Externally sensitized mesolytic fragmentations in dithiane-ketone adducts

Article abstract of DOI:10.1016/j.tet.2006.03.065

The apparent activation enthalpies, ΔH^≠, for externally sensitized mesolytic fragmentations in benzophenone-dithiane adducts were obtained in variable temperature photolyses and compared with DFT activation barriers calculated for β-scission in the corresponding oxygen-centered radicals. The results of these experimental and theoretical studies further support the mechanism in which deprotonation of the hydroxy-group, in the transient cation radical, is coupled with intramolecular electron transfer furnishing the O-centered radical, which subsequently fragments. The quantum yields of fragmentation increase for higher alkyl substituted dithiane adducts.

Full text of DOI:10.1016/j.tet.2006.03.065

Tetrahedron 62 (2006) 6574–6580
Externally sensitized mesolytic fragmentations
in dithiane–ketone adducts
*
Tiffany P. Gustafson, Alexei N. Kurchan and Andrei G. Kutateladze
Department of Chemistry and Biochemistry, University of Denver, Denver, CO 80208-2436, USA
Received 11 September 2005; accepted 12 November 2005
Available online 17 April 2006
Abstract—The apparent activation enthalpies, DHs, for externally sensitized mesolytic fragmentations in benzophenone–dithiane adducts
were obtained in variable temperature photolyses and compared with DFT activation barriers calculated for b-scission in the corresponding
oxygen-centered radicals. The results of these experimental and theoretical studies further support the mechanism in which deprotonation of
the hydroxy-group, in the transient cation radical, is coupled with intramolecular electron transfer furnishing the O-centered radical, which
subsequently fragments. The quantum yields of fragmentation increase for higher alkyl substituted dithiane adducts.
Ó 2006 Elsevier Ltd. All rights reserved.
1. Introduction
One of the early examples by Whitten and Ci involved thio-
indigo-sensitized fragmentation in the threo- and erythro-2-
morpholino-1,2-diphenylethanol.3a Assuming that the only
temperature-dependent process in the reaction of the gemi-
nate ion radical pair is kfrag, they determined the Arrhenius
activation energy for the threo- and erythro diastereomers
to be 4.9 and 2.8 kcal/mol, respectively. This demonstrates
a strong conformational dependence of the cleavage and
shows that the activation barrier is very low.
Heteroatom-centered cation radicals, Cb–Ca–Xꢀ+, which are
readily generated by photoinduced electron transfer (ET),
are subject to mesolytic fragmentation. It is well known
that removal of one electron from the heteroatom’s lone
pair in the ground state significantly reduces the bond order
of the geminal bonds, which either increases the acidity of
the alpha-proton (H–Ca) or causes the Ca–Cb bond to cleave.
In our previous work we utilized this increased acidity of the
alpha-proton and developed an efficient electrochemical de-
protection for carboxylates in general, and amino acids in
particular, based on esters of hydroxymethyldithiane.1 In
this paper we will focus primarily on photoinduced C–C
bond fragmentations in adducts of dithianes with ketones.
O-CPh2
SET
H
O
OH
HO-CPh2
S
CHO
S
S
C-C cleavage
Ph2C=O
+
S
S
hν
S
ET-induced fragmentations in vicinal amino alcohols and
diols have been extensively studied.2 The accepted mecha-
nistic rationale includes photoinduced electron transfer to an
ET-sensitizer, e.g., benzophenone, followed by a mesolytic
cleavage of the generated cation radical, assisted by the ben-
zophenone anion radical deprotonating the vicinal hydroxy
group. It was noted that the C–C bond cleavage step is rem-
iniscent of the Grob fragmentation in closed shell systems.3
Similar reactions of sulfido-alcohols are also known. For ex-
ample, Gravel et al. utilized the C–C cleavage in b-phenyl-
thioalkanols as a synthetic method for indirect cleavage of
olefins4 and also in carbohydrate synthesis.5 Some time
ago we found that dithianes are particularly suitable for
this chemistry: their adducts are readily synthesized and
they cleave efficiently upon photoinduced fragmentation.
The mechanism of cleavage in dithiane adducts was investi-
gated by utilizing classical physical organic methods such as
the Hammett substituent effect, the kinetic isotope effect,6
and laser flash photolysis studies.7 Our initial mechanistic
findings were in keeping with the universally recognized
‘Grob-like’ mechanism, until we tested the photoinduced
fragmentation in the tert-butyl derivative A, which in addi-
tion to the expected pivalaldehyde B (‘normal’ cleavage)
produced dithiane-2-carboxaldehyde C as a major product
in 1:6.2 ratio.8 It became clear that the quasi-Grob electron
ET
O-CPh2
NR2
Ar
HO-CPh2
kET
*
NR2
Ar
HO
Ar
NR2
Ar
Ph2C=O
kfrag
O
H
O
kBET
Ar
Ar
* Corresponding author. E-mail: akutatel@du.edu
0040–4020/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2006.03.065

T. P. Gustafson et al. / Tetrahedron 62 (2006) 6574–6580
6575
S
S
CHO
B
SET
+
O-CPh2
HO-CPh2
intra-
molecular
ET
H
S
3Ph2C=O
i
S
ii
H
S
O
O
i
O
O
S
S
S
S
S
ii
S
OHC
S
A
+
A
A
?
C
Scheme 1.
pushing rationale needed refinement because adduct A was
primarily cleaving the wrong bond.
challenging, if not impossible, to generate the O-radical Aꢀ
from an alternative precursor, for example, a peroxide, and
study the effect of substituents on the rate of its degradation.
CHO
OH
In this paper we report our experimental and computational
study of the mechanism of photoinduced cleavage in adducts
of 2-alkyl substituted dithianes with benzophenone, which is
intended to further refine the mechanism of fragmentation.
Benzophenone-sensitized mesolytic fragmentations, in ad-
ducts of the sensitizer itself, result in generation of more
benzophenone and, as such, constitute its amplification.
Hence, our particular interest in this system, which we
plan to utilize in various photochemical applications.
B
S
Ph2C=O
1
6.2
hν
S
S
OHC
A
S
C
Further experimental and computational investigation
suggested that deprotonation of the hydroxy group in the
initially generated cation radical does not result in the forma-
tion of a charge separated ‘Grob-like’ precursor ꢁO–C–C–
ꢀ
S+ –R, but rather produces a neutral oxygen-centered radical
2. Results and discussion
(or a species behaving as one) via intramolecular electron
transfer. The O-centered radical undergoes subsequent frag-
mentation in either direction and the partitioning correlates
with stability of the produced radicals, Scheme 1.9 When
methyldithiane derivative D was used in place of A, the ratio
of B:C¼1:6.2 was inverted to become B:E¼1.8:1, all in
keeping with relative stability of tert-butyl, dithian-2-yl,
and 2-methyldithian-2-yl radicals. We also compared the
Benzophenone adducts 2a–e of unsubstituted (1a), 2-
methyl- (1b), 2-ethyl- (1c), 2-hexyl (1d), and 2-decyl (1e)
dithiane were synthesized according to a modified Corey–
Seebach procedure10 (Scheme 2) and their photoinduced
fragmentation was studied over a temperature range from
ꢁ40 to +40 ꢂC in acetonitrile upon benzophenone sensitiza-
tion. The driving force for the oxidative electron transfer
from the dithiane moiety to the triplet state of benzophenone
in acetonitrile is rather large: the one electron reduction po-
tential of triplet benzophenone is ꢁ1.68 V (vs SCE in aceto-
nitrile),11 whereas various 2-substituted dithianes oxidize in
the range of +0.73 to +1.18 V in the same solvent.12 During
the course of a laser flash photolysis study7 we found that the
rate of initial electron transfer quenching of triplet benzo-
phenone with dithiane–benzophenone adduct in dry aceto-
CHO
OH
B
S
1.8
1
Ph2C=O
hν
S
OHC
S
D
S
E
nitrile was near the diffusion limit, 8.4ꢃ0.7ꢄ108 Mꢁ1 sꢁ1
.
results of the fragmentation in methylene chloride and aceto-
nitrile, which constitutes almost an order of magnitude
difference in the dielectric constant, and did not see any dif-
ference in the partitioning within experimental error. All
these results seem to indicate that the charge separated spe-
cies does not exist, or at best it is in fast equilibrium with the
alkoxy radical (Scheme 1).
In 10% aqueous acetonitrile the quenching rate was even
higher, 1.31ꢃ0.06ꢄ109 Mꢁ1 sꢁ1. In the present study we
have determined the dependence of the quantum efficiency
of fragmentation on temperature and compared the experi-
mental (apparent) activation enthalpy with the calculated
activation barrier for the b-scission in the corresponding
O
The case of tert-butyl derivatives, such as A and D, is unique
in a sense that it allowed us to discover this new channel in
the mechanism of fragmentation simply by product analysis.
For dithiane adducts of aromatic aldehydes and ketones the
barrier for the aryl radical departure is prohibitively high,
i.e., it is the dithiane radical, which is always departing.
Therefore, it is much more difficult to ascertain whether or
not the original Grob-like mechanism for cleavage competes
with the anomalous O-centered radical mechanism (or either
one of them is operating exclusively) in the case of aromatic
adducts. Because of the presence of sulfur it is synthetically
S
HO
S
R
R
S
BuLi/hexanes
S
THF, r.t.
(5-10 min)
1
2
a, R = H
b, R = Me
c, R = Et
d, R = n-hexyl
e, R = n-decyl
Scheme 2.

6576
T. P. Gustafson et al. / Tetrahedron 62 (2006) 6574–6580
Table 1. Quantum efficiencies and dithiane recovery rates at +20 ꢂC for
adducts 2b–e
increases for adducts of dithianes substituted with longer
alkyl chains (Table 1).
Adduct
f
Dithiane recovery (%)
The adduct of the parent compound, unsubstituted dithiane,
showed poor efficiency of cleavage. This deficiency is not
due to the impediment of the initial electron transfer: our
previous LFP experiments show that the initial electron
transfer quenching rate increases with the decreased substi-
tution.7 It is difficult to argue definitively about the partition-
ing of the initially formed dithiane cation radical—
benzophenone anion radical pair (i.e., the partition between
the back electron transfer and the productive deprotonation
of the adduct by the anion radical leading to fragmentation).
However, following Whitten’s assumption3a that the only
temperature-dependent process in a reaction of the geminate
radical ion pair is fragmentation, we obtained the enthalpy of
activation by plotting the data shown in Table 2 as log(f/T)
versus 1/T for the reaction of 2b and 2c (Fig. 1).
Methyl- (2b)
Ethyl- (2c)
Hexyl- (2d)
Decyl- (2e)
0.20
0.23
0.25
0.27
32.8
35.3
41.0
47.8
O-centered radicals. The computations were run at the DFT,
B3LYP/6-31G(d), level using Gaussian 03, revision C.02.13
Photolyses were carried out in a carousel Rayonet photo-
reactor with a Pyrex heat exchange jacket cooled by ethanol
circulating through a FTS Systems Multi-Cool refrigeration
unit. The quantum yield determinations were done using the
classic benzophenone—benzhydrol actinometer system as
a standard.14 Having benzophenone as a common sensitizer,
in both the photoinduced fragmentation under study and the
actinometer system, significantly improves the accuracy of
quantum yield determinations. In addition to the overall
quantum yield of adduct disappearance another important
parameter is the percent recovery of the corresponding
dithiane. Both the Grob-like and the O-centered radical
mechanisms have a dithian-2-yl radical as the immediate
fragmentation product. Its fate depends on multiple factors,
including the reactivity of such a radical towards oxygen, di-
merization, disproportionation etc. The recovery of dithiane
in turn depends on all of the above and also on the rate of its
degradation as a result of secondary photooxidation by
benzophenone. We found that more substituted dithianes
showed better recovery, which is in agreement with the over-
all trends in reactivity of the dithian-2-yl radicals (i.e., un-
substituted dithianyl radicals react/degrade faster) and of
the final products—dithianes (i.e., less substituted dithianes
undergo benzophenone-sensitized oxidative photodegrada-
tion faster). The quantum efficiency of cleavage also steadily
The activation enthalpies, DHs, are obtained from the
slopes: 4.7 kcal/mol for the methyl derivative 2b and
1.5 kcal/mol for the ethyl derivative 2c (Fig. 1a). These
values are very similar to the activation energies obtained
by Whitten for the thioindigo-sensitized fragmentation in
vicinal amino alcohols.3a
We suggest that the effect of substitution at position 2 of di-
thiane reflects largely the acceleration of the fragmentation
in the deprotonated species, the oxygen-centered radical.
Walling and Padwa15 studied the substitution effect in de-
composition of hypochlorites of alkyldimethylcarbinols,
R–C(Me)2–O–Cl, by comparing the differences in activation
energies for decomposition and hydrogen abstraction. They
reported that while the difference was 10 kcal/mol for tert-
butoxy radical (i.e., R¼Me), it decreased to 1.7 for R¼
i-Pr, 0.7 for R¼benzyl, and was negligibly small for R¼
t-Bu. The absolute value for the activation energy of frag-
mentation in tert-butoxy radical is 11–13 kcal/mol.16 As-
suming that the rate of hydrogen abstraction does not
change much, introduction of the tert-butyl group in place
of methyl should lower the activation energy of fragmenta-
tion in these alkoxy radicals by about 10 kcal/mol, to approx.
1–3 kcal/mol. 2-Methyl-1,3-dithian-2-yl and 2-ethyl-1,3-di-
thian-2-yl are very stable (and bulky) radicals. Judging by
our previous observations of the competitive cleavage, the
2-methyldithianyl radical is more stable than tert-butyl,
which in turn is more stable than the unsubstituted di-
thiane-2-yl radical.8 Summarizing these arguments, it is
Table 2. Quantum yields of fragmentation and % dithiane recovery as a
function of temperature
T (ꢂC)
Ethyl adduct 2c
Methyl adduct 2b
f
Dithiane
recovery (%)
f
Dithiane
recovery (%)
ꢁ40
ꢁ20
0
20
40
0.04
0.13
0.18
0.21
0.32
0.7
6.2
17.8
14.8
37.6
0.01
0.05
0.10
0.15
0.25
1.5
4.3
6.2
8.7
12.5
-6
-7
-6.75
ln(φ/T) = -2155.8/T - 0.097
r2 = 0.97
ln(φ/T) = -745.8/T - 4.619
r2 = 0.94
-7.25
-8
-9
1/T
1/T
-7.75
0.003
0.0035
2c
0.004
0.003
0.0035
2b
0.004
Figure 1. 1/T dependence of fragmentation quantum yields for ethyl (2c) and methyl (2b) adducts.

T. P. Gustafson et al. / Tetrahedron 62 (2006) 6574–6580
Table 4. DFT relative energies for the conformers of alcohols 2a–c
6577
Table 3
SS (kcal/mol)
Ethyl-derivative 2cꢀ (O-radical)
TS (kcal/mol)
10.2
22.6
DHs (kcal/mol)
Conformers
Rel energy (kcal/mol)
aMaO-2b
aMgO-2b
eMaO-2b
eMgO-2b
3.15
3.45
2.90
0.00
aEaO-2cꢀ
aEgO-2cꢀ
eEaO-2cꢀ
eEgO-2cꢀ
5.4
4.5
1.1
0.0
4.77
18.15
—
a
1.7
1.70
aEaO-2c
aEgO-2c
eEaO-2c
eEgO-2c
5.52
1.91
3.18
0.00
Methyl derivative 2bꢀ (O-radical)
aMaO-2bꢀ
aMgO-2bꢀ
eMaO-2bꢀ
eMgO-2bꢀ
3.4
2.8
0.9
0.0
8.5
5.1
5.05
2.24
—
a
aHaO-2a
aHgO-2a
eHaO-2a
eHgO-2a
0.01
0.00
3.31
2.80
1.7
1.72
a
Computations did not converge.
not unreasonable to assume that the apparent DHs of 4.7
and 1.5 kcal/mol, obtained in this study, correspond to
fragmentation in the oxygen-centered radicals, expelling
methyldithianyl and ethyldithianyl radicals, respectively.
equatorial position, whereas the bulky benzhydryl group was
axial, with oxy-radical being gauche to methyl/ethyl (i.e.,
anti to one of sulfur atoms). The transition state geometries
were obtained at the same B3LYP/6-31G(d) level of theory.
The electronic energies for all the species in Table 3 are
zpe-corrected. As follows from the table, the eMgO/eEgO
conformers have the lowest energies both at the minima
and the transition states (see also Fig. 2). This is also in keep-
ing with the computed conformational energies of the parent
alcohols that showed at least a 2 kcal/mol preference for the
eRgO conformers (Table 4).
To support this hypothesis we carried out DFT computations
at a B3LYP/6-31G(d) level, using the whole untruncated
structures of the methyl- and ethyldithiane adducts of benzo-
phenone (2b and 2c). The initial geometries of the respective
oxygen-centered radicals (2bꢀ and 2cꢀ) were generated with
Chem3D and pre-optimized at the AM1 level. Vibrational
analysis of the DFT optimized geometries show no imagi-
nary frequencies for the computed minima (ground states)
and only one imaginary frequency for the transition states
corresponding to the reaction coordinate (i.e., C–C stretch).
Given the importance of conformational considerations, we
scanned the conformational space and analyzed the relative
energies of the starting alkoxy radicals and their respective
fragmentation transition states for the four major con-
formers: (i) equatorial 2-alkyl with the CO bond in anti
conformation to this alkyl, denoted eMaO, i.e., equatorial
Methyl anti Oxygen (the oxygen is anti to the methyl group),
and eEaO for Ethyl; (ii) equatorial 2-alkyl—gauche CO
bond, eMgO/eEgO; (iii) axial 2-alkyl—anti CO bond;
aMaO/aEaO; (iv) axial 2-alkyl—gauche CO bond, aMgO/
aEgO. The other two sets of the gauche conformers are
enantiomers of eMgO/eEgO and aMgO/aEgO. The relative
energies of the starting oxy-radicals and their respective
transition states are listed in Table 3 with the calculated
DHs summarized in the third column.
On the contrary, for the C-2 unsubstituted dithiane adduct it
was the axial-H (i.e., the equatorial benzhydrol) conformers
aHaO and aHgO that were expectedly more stable by
approx. 3 kcal/mol.
NMR spectroscopic study of the parent alcohols supported
the DFT findings. The initial analysis of the 1D proton
NMR spectra of the three adducts shows systematic upfield
shift of the dithiane’s H2C(3) and H2C(5) protons upon intro-
duction of the methyl and then ethyl group. The downfield
multiplet corresponds to two axial H2C(3) and H2C(5) protons
and the upfield—equatorials (Fig. 3).
Low temperature NMR data show dramatic differences in
conformational behavior of the substituted (2b and 2c)
versus unsubstituted (2a) derivatives. The temperature-
1
dependent changes in one-dimensional H NMR spectra
are shown in Figure 4. Below ꢁ40 ꢂC the multiplets for both
axial and equatorial protons H2C(3) and H2C(5) in the Me-
and Et-derivatives split into two sets. It appears that the
Strikingly, the lowest energy conformations for both methyl
and ethyl derivatives had the smaller alkyl substituent in the
Figure 2. The lowest energy conformers of the respective O-centered radicals: eMgO-2bꢀ (left) and eEgO-2cꢀ (right).

6578
T. P. Gustafson et al. / Tetrahedron 62 (2006) 6574–6580
Figure 3. 1D proton NMR spectra of 2a–c.
Figure 4. VT NMR data for 2a–c. The inset shows low temperature COSY spectrum for 2c at ꢁ95 ꢂC.

T. P. Gustafson et al. / Tetrahedron 62 (2006) 6574–6580
6579
rotation of the benzhydryl group is stopped at this tempera-
ture and that the most stable isomer is unsymmetrical.
instrument in CDCl3 with TMS as an internal standard (un-
less noted otherwise). Low temperature NMR was carried
out in CD3OD using a Varian Mercury VT system. Temper-
ature was controlled using dry nitrogen flow through a liquid
nitrogen Dewar. Column chromatography was performed on
silica gel, 70–230 mesh ASTM, using ethyl acetate–hexane
mixtures as eluent. Photoreactions were carried out in the
carousel Rayonet photoreactor outfitted with a jacketed
Pyrex reaction vessel connected to a FTS Systems Multi-
Cool refrigeration unit with a peristaltic pump, using ethanol
as a coolant.
On the contrary, NMR spectrum of unsubstituted 2a shows
negligible temperature-dependent changes, which could
indicate that the most stable conformer is symmetric (i.e.,
anti). An alternative interpretation is that the rotation of
the benzhydrol group is not stopped even at ꢁ95 ꢂC in
unsubstituted 2a.
Equal integrated intensities of the split signals of the equato-
rial and axial protons indirectly confirm that they belong to
the same conformer, not to two frozen populations of con-
formers, e.g., chair-to-chair, or chair–twist conformers of
dithiane, in which case they would be expected to have
different integrated values. For more definitive assignment,
we ran a COSYexperiment of 2c at ꢁ95 ꢂC, which unambig-
uously showed strong cross-peaks for the respective split
pairs of signals. We therefore conclude, that below ꢁ40 ꢂC
both equatorial protons Heq–C(3) and Heq–C(5) of the dithiane
ring are not equivalent and separate into two signals. The
same is true for the pair of axial protons Hax–C(3) and
Hax–C(5). These data can only be explained in terms of asym-
metric conformation of the frozen benzhydrol group, which
is in keeping with our computational finding that the asym-
metric eRgO conformers are the most stable. The low tem-
perature data for the Me derivative 2b in Figure 4 show
some additional line broadening for each of the upfield
(right) signals of the split peaks, indicating that another de-
gree of freedom is being frozen at ꢁ60 ꢂC and below. We do
not have an immediate explanation for this behavior—there
can be a number of conformational processes that can cause
the observed coalescence of signals at this temperature.
Ab initio computations were performed on a Linux worksta-
tion using Gaussian 03, Revision C.02.13 Input geometries
were created and pre-optimized using a force field geometry
optimization as implemented in Chem3D (Cambridgesoft).
The geometries were further pre-optimized at the AM1 level.
Full geometry optimizations were performed using density
functional theory (DFT) at the B3LYP/6-31G(d) level of the-
ory (the Becke three-parameter hybrid functional combined
with Lee, Yang, and Parr correlation functional17).
3.2. General method for dithiane preparation
1,3-Propanedithiol (0.07 mol) and the appropriate aldehyde
(0.06 mol) were dissolved in 250 mL CH2Cl2. BF3$Et2O
(0.26 mol) was added to the solution. The reaction mixture
was then stirred overnight at room temperature. The mixture
was washed with NaOH (5% aq soln) and water. The organic
layer was dried over anhydrous NaSO4 and the solvent was
removed by rotary evaporator and the resulting product was
distilled under vacuum.
3.2.1. 2-Ethyl-1,3-dithiane (1c). Yield 8.7 g, 59 mmol,
86%; bp 38 ꢂC/63 mTorr; 1H NMR (CDCl3, 400 MHz):
d 4.0 (t, 1H, J¼6.8 Hz), 2.80–2.92 (m, 4H), 2.09–2.16 (m,
1H), 1.84–1.92 (m, 1H), 1.77–1.92 (m, 2H), 1.09 (t, 3H).
In conclusion, the measured apparent enthalpies of activa-
tion for photoinduced fragmentation in the alkyl dithiane–
benzophenone adducts are in keeping with the kinetic
barriers computed at the B3LYP/6-31G(d) level of theory
for the mechanism involving formation of an alkoxy radical
and its subsequent fragmentation. The findings do not rule
out the charge separated Grob-like mechanism, but rather
provide new evidence to support an alternative nonpolar
mechanism. As to the design and development of the
dithiane-based photolabile latches: utilization of higher
alkyls at position 2 of dithiane increases both the quantum
yield of fragmentation and recovery of dithiane, attesting
to the potential of longer alkyl chains as a promising alterna-
tive for the molecular design of efficient photolabile tethers.
3.2.2. 2-Hexyl-1,3-dithiane (1d). Yield 5.4 g, 26 mmol,
1
54%; bp 125 ꢂC/72 mTorr; H NMR (CDCl3, 400 MHz):
d 4.03 (t, 1H, J¼6.9 Hz), 2.76–2.90 (m, 4H), 2.06–2.13
(m, 1H), 1.83–1.92 (m, 1H), 1.68–1.74 (m, 2H), 1.44–1.52
(m, 2H), 1.22–1.35 (m, 6H), 0.87 (t, 3H).
3.2.3. 2-Decyl-1,3-dithiane (1e). Yield 12.56 g, 48 mmol,
1
63%; bp 142 ꢂC/69 mTorr; H NMR (CDCl3, 400 MHz):
d 4.07 (t, 1H, J¼6.9 Hz), 2.81–2.94 (m, 4H), 2.10–2.17
(m, 1H), 1.85–1.94 (m, 1H), 1.73–1.79 (m, 2H), 1.48–1.55
(m, 2H), 1.25–1.33 (m, 14H), 0.90 (t, 3H).
3. Experimental
3.1. General
3.3. General method for adduct preparation
A generic method by Corey and Seebach was modified
and used to prepare the desired dithiane–benzophenone
adducts.10 Dithiane (5.1 mmol) was dissolved in freshly
distilled THF (30 mL) and placed under nitrogen. n-Butyl-
lithium (4.3 mL, 6.8 mmol) was added at room temperature
upon stirring and the resulting mixture was stirred for 10
more minutes. Benzophenone (3.4 mmol) was dissolved in
freshly distilled THF (10 mL) and added to the anion mix-
ture with stirring. The reaction was left overnight. The reac-
tion mixture was quenched with a saturated solution of
ammonium chloride and the aqueous layer was extracted
Common solvents were purchased from Pharmco and used
as is, except for THF, which was refluxed over and distilled
from potassium benzophenone ketyl prior to use. n-BuLi
(as a 1.6 M solution in hexane), 1,3-dithiane, 2-methyl-1,3-
dithiane, benzophenone, and benzhydrol were purchased
from Aldrich. 1,3-Dithiol was purchased from Acros. Propa-
nal, decanal, and heptanal were purchased from Alfa Aesar.
All reagents were used without purification. 1H NMR spec-
tra were recorded at 25 ꢂC on a Varian Mercury 400 MHz

6580
T. P. Gustafson et al. / Tetrahedron 62 (2006) 6574–6580
3. (a) Ci, X.; Whitten, D. G. J. Am. Chem. Soc. 1987, 109, 7215;
(b) Ci, X.; Whitten, D. G. J. Am. Chem. Soc. 1989, 111, 3459.
4. Gravel, D.; Farmer, L.; Ayotte, C. Tetrahedron Lett. 1990,
31, 63.
5. Gravel, D.; Farmer, L.; Denis, R. C.; Schultz, E. Tetrahedron
Lett. 1994, 35, 8981.
twice with ethyl acetate. The combined organic layer was
dried over anhydrous sodium sulfate. The solvent was re-
moved with a rotary evaporator, and the residue was purified
by column chromatography (silica gel, ethyl acetate–
hexane) or recrystallization (methanol).
6. McHale, W. A.; Kutateladze, A. G. J. Org. Chem. 1998, 63,
9924.
7. Vath, P.; Falvey, D. E.; Barnhurst, L. A.; Kutateladze, A. G.
J. Org. Chem. 2001, 66, 2886.
8. Li, Z.; Kutateladze, A. G. J. Org. Chem. 2003, 68, 8236.
9. b-Alkylthio-ethoxy radicals were implicated in the dissociation
of hydroxyl radical adducts of (alkylthio)ethanol derivatives:
3.3.1. (2-Methyl-[1,3]-dithian-2-yl)-diphenyl-methanol
(2b). Yield 0.80 g, 2.7 mmol, 80%; 1H NMR (CDCl3,
400 MHz): d 7.85–7.87 (m, 4H), 7.21–7.30 (m, 6H), 2.82
(ddd, 2H, J¼3.5, 10.5, 14.6 Hz), 2.53 (ddd, 2H, J¼3.9,
5.7, 14.7 Hz), 1.80–1.97 (m, 2H), 1.894 (s, 3H).
3.3.2. (2-Ethyl-[1,3]-dithian-2-yl)-diphenyl-methanol
(2c). Yield 0.94 g, 2.8 mmol, 93%; 1H NMR (CDCl3,
400 MHz): d 7.93–7.96 (m, 4H), 7.28–7.33 (m, 4H), 7.22–
7.26 (m, 2H), 2.52 (ddd, 2H, J¼4.8, 7.0, 14.2 Hz), 2.25
(ddd, 2H, J¼4.8, 8.1, 13.0 Hz), 1.97–2.04 (m, 2H), 1.67–
1.73 (m, 2H), 1.12–1.16 (m, 3H).
Schoneich, C.; Bobrowski, K. J. Am. Chem. Soc. 1993, 115,
¨
6538.
10. Seebach, D.; Corey, E. J. J. Org. Chem. 1975, 30, 231.
11. Roth, H. D.; Lamola, A. A. J. Am. Chem. Soc. 1974, 96,
6270.
12. Glass, R. S.; Petsom, A.; Wilson, G. S. J. Org. Chem. 1986, 51,
4337.
3.3.3. (2-Hexyl-[1,3]-dithian-2-yl)-diphenyl-methanol
(2d). Yield 0.61 g, 1.6 mmol, 49%; 1H NMR (CDCl3,
400 MHz): d 7.91–7.94 (d, 4H), 7.20–7.31 (m, 6H), 2.49
(td, 1H, J¼4.8, 10.3 Hz), 2.22 (ddd, 1H, J¼5.9, 8.0,
14.1 Hz), 1.91–1.96 (m, 2H), 1.60–1.73 (m, 4H), 1.10–
1.31 (m, 6H), 0.81–0.85 (t, 3H).
13. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven,
T.; Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.;
Tomasi, J.; Barone, V.; Mennucci, B.; Cossi, M.; Scalmani,
G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara,
M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Naka-
jima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li, X.;
Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Adamo, C.; Jara-
millo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin,
A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.;
Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.;
Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain,
M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari,
K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clif-
ford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko,
A.; Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith,
T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challa-
combe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong,
M. W.; Gonzalez, C.; Pople, J. A. Gaussian 03, Revision
C.02; Gaussian: Wallingford, CT, 2004.
3.3.4. (2-Decyl-[1,3]-dithian-2-yl)-diphenyl-methanol
(2e). Yield 0.62 g, 1.4 mmol, 52%; 1H NMR (CDCl3,
400 MHz): d 7.92–7.94 (dd, 4H), 7.20–7.33 (m, 6H), 2.49
(td, 2H, J¼5.4, 14.1 Hz), 2.23 (ddd, 1H, J¼5.7, 8.2,
13.9 Hz), 1.91–1.96 (m, 2H), 1.60–1.77 (m, 6H), 1.10–
1.34 (m, 12H), 0.81–0.85 (t, 3H).
Acknowledgements
Support for this research by the National Science Foundation
(CHE-314344) is gratefully acknowledged.
References and notes
14. Moore, W. M.; Hammond, G. S.; Foss, R. P. J. Am. Chem. Soc.
1961, 83, 2789.
1. Barnhurst, L. A.; Wan, Y.; Kutateladze, A. G. Org. Lett. 2000,
2, 799.
2. For review see: Gaillard, E. R.; Whitten, D. G. Acc. Chem. Res.
1996, 29, 292.
15. Walling, C.; Padwa, A. J. Am. Chem. Soc. 1963, 85, 1593.
16. Gray, P.; Williams, A. Chem. Rev. 1959, 59, 239.
17. (a) Becke, A. D. Phys. Rev. A 1988, 38, 3098; (b) Lee, C.; Yang,
W.; Parr, R. G. Phys. Rev. B 1988, 37, 785.