Reactivity of 2-aryl-1,3-dithiane anions towards neopentyl, neophyl and phenyl iodides. New evidence for an SRN1 mechanism

Article abstract of DOI:10.1002/poc.1837

The reactions of 2-(4-Z-phenyl)-1,3-dithiane anions (Z=H, OMe, Cl, CN) with neopentyl, neophyl and phenyl iodides were studied in DMSO, taking into consideration the effect of the Z substituent on the dithiane anions reactivity as well as on the product distribution. These substitution reactions proceed by an S_RN1 mechanism with radicals and radical anions as intermediates. Two competitive pathways are possible for the radical anion of the substitution product, namely electron transfer (ET) to the substrate giving the substitution product and C-S bond fragmentation to yield a distonic radical anion. ET is the main pathway for the reactions between dithiane anions bearing electron-donor substituents and neopentyl or its analogue iodides affording the substitution products in moderate yields (41-53%). Copyright

Full text of DOI:10.1002/poc.1837

Research Article
Received: 8 October 2010,
Revised: 16 November 2010,
Accepted: 16 December 2010,
Published online in Wiley Online Library: 11 February 2011
(wileyonlinelibrary.com) DOI 10.1002/poc.1837
Reactivity of 2-aryl-1,3-dithiane anions
towards neopentyl, neophyl and phenyl
iodides. New evidence for an SRN1 mechanism
a
a
*
´
˜ ˜
Gabriela Oksdath-Mansilla and Alicia B. Penenory
The reactions of 2-(4-Z-phenyl)-1,3-dithiane anions (Z ¼ H, OMe, Cl, CN) with neopentyl, neophyl and phenyl iodides
were studied in DMSO, taking into consideration the effect of the Z substituent on the dithiane anions reactivity as
well as on the product distribution. These substitution reactions proceed by an SRN1 mechanism with radicals and
radical anions as intermediates. Two competitive pathways are possible for the radical anion of the substitution
product, namely electron transfer (ET) to the substrate giving the substitution product and C–S bond fragmentation
to yield a distonic radical anion. ET is the main pathway for the reactions between dithiane anions bearing
electron-donor substituents and neopentyl or its analogue iodides affording the substitution products in moderate
yields (41–53%). Copyright ß 2011 John Wiley & Sons, Ltd.
Supporting information may be found in the online version of this article.
Keywords: dithiane anion; electron transfer; fragmentation; radical; radical anion
*
In aliphatic systems without a p MO to stabilise the
INTRODUCTION
radical anion (RX).S, Equations (1) and (3) occur simultaneously
The radical nucleophilic substitution concerning electron
transfer (ET) steps or the SRN1 mechanism is a chain process
involving radicals and radical anions as intermediates. In these
reactions, a compound bearing an adequate leaving group is
usually substituted by a nucleophile. Over the years, the scope
of this process has been considerably extended, and nowadays,
it is an important synthetic method to achieve substitution ipso
to the leaving group of different substrates.[1–4] Furthermore,
non-activated aromatic halides, alkyl halides activated by strong
electron withdrawing groups (EWGs; nitroalkyl, nitroallyl, nitro or
cyanobenzyl and their heterocyclic analogues) and other simple
alkyl halides, can be substituted by nucleophiles through ET
mechanisms. These compounds characterised mostly by their
low reactivity towards classical polar nucleophilic substitutions,
mainly due to steric and strain factors. Examples of this type of
compounds include neopentyl halides and derivatives, bridge-
head bicyclo- and polycycloalkyl halides. Several nucleophiles are
known to react by ET processes to form new C–C or C-heteroatom
bonds with good yields, such as carbanions and anions from
Sn, Sb, P, As, S, Se and Te.
(concerted dissociative step),[16] giving the substitution product
.
and the alkyl radical R (which propagates the chain cycle).
Furthermore, since the
SRN1 is a chain mechanism, the
overall reactivity depends on the initiation, propagation, and
termination steps. In order for this process to work efficiently, the
initiation step may be slow. However, the chain propagation
needs to be fast and efficient in order to allow long chains
to build up. Finally, the wide variety of nucleophiles that can be
used, the great functional group tolerance, and the fact
that many C–C and C-heteroatom bonds can be obtained, make
the SRN1 reaction an effective synthetic tool.[17,18]
The 1,3-dithianes derived from aldehydes and ketones
have become highly important in organic synthesis for their
role in protecting the carbonyl groups, a strategy commonly used
in natural product synthesis.[19,20] Additionally, cyclic thioacetals
are versatile synthetic reagents acting as acyl anion equivalents
for C–C bond formation by reaction with a wide range of alkyl
halides, carbonyl compounds and other electrophilic reagents.[21–24]
For example, the reaction between 2-lithio-1,3-dithiane (1) and the
optically active (R)-2-iodo-octane in THF was found to proceed
with complete inversion of configuration.[25] This result implies
The chain reaction of the SRN1 process requires an initiation
step, which involves an ET from the nucleophile or from a suitable
electron source to the substrate. The most frequently used
methods for initiation are chemical initiation by alkali metals in
liquid ammonia,[5] electrochemical initiation at a cathode,[4]
and photoinitiation,[2,3] with other procedures including the use
˜´˜
Correspondence to: A. B. Penenory, Facultad de Ciencias Qu´ımicas, Departa-
´
mento de Qu´ımica Organica, Universidad Nacional de Cordoba, Ciudad
Universitaria, 5000 Cordoba, Argentina.
*
´
´
of Fe2þ [6–8]
,
SmI2,[9] or Na(Hg).[10] In a few systems, a thermal
E-mail: penenory@fcq.unc.edu.ar
(spontaneous) initiation is observed.[11–14] For example, the SRN
1
˜´˜
G. Oksdath-Mansilla, A. B. Penenory
Facultad de Ciencias Qu´ımicas, Departamento de Qu´ımica Organica,
´
Universidad Nacional de Cordoba, Ciudad Universitaria, 5000 Cordoba,
a
substitution of aliphatic systems with EWGs is in most cases
initiated by a thermal ET that can be accelerated by light.[15]
Scheme 1 presents the propagation steps of an SRN1 mechanism.
´
´
Argentina
J. Phys. Org. Chem. 2011, 24 1136–1143
Copyright ß 2011 John Wiley & Sons, Ltd.

SRN1 REACTIONS OF 2-ARYL-1,3-DITHIANE ANIONS
Chart 1. Dithiane anions and iodides studied.
Scheme 1. SRN1 mechanism.
anion (3) with neopentyl (4) and neophyl (5) iodides, under a
variety of conditions in DMSO at 20 8C (Chart 1). For comparison
purposes, the reactivity towards phenyl iodide has also been
investigated. New evidence for the operating mechanism is
presented and discussed.
that the SN2 mechanism is the preferred pathway between 1
and unhindered alkyl halides. However, when the neopentyl type
probe (2) was used, both uncyclised and cyclised substitution
products were found at low yields, suggesting that this occurred,
at least in part, by an ET reaction (Scheme 2).[25] For these
reactions a chain mechanism with radical and radical anions as
intermediates, such as the SRN1 mechanism, was discarded by the
authors; they proposed instead the occurrence of an ET process
with cage collapse of the radical pair.
EXPERIMENTAL
Chemicals
The reactivity of neopentyl and neophyl iodides by the
RN1 process has been well established, for carbanions as well
tert-BuOK, neopentyl iodide (4), neophyl iodide (5), MeI,
DTBN, thiophenol, iodobenzene, p-dinitrobenzene, benzaldehyde,
4-methoxybenzaldehyde, 4-chlorobenzaldehyde, 4-formylbenzo-
nitrile and 1,3-propanedithiol were all high-purity commercial
samples, used without further purification. DMSO was distilled
S
as for sulphur, phosphorous, selenide and telluride anions.[26–28]
A kinetic study of the reaction of neopentyl bromide with PhSꢀ
and Ph2Pꢀ ions was performed in liquid ammonia and MeCN,
with the rate constants for the coupling of neopentyl radicals
˚
under vacuum and stored over molecular sieves (4 A). Anions
with PhSꢀ ions determined to be about 1.2 ꢁ 108 Mꢀ1 sꢀ1 [29]
.
3a–d were generated in situ by acid–base deprotonation using
tert-BuOK. Dithiane precursors of anions 3a–d were synthesised
by standard procedures from the reaction between commercial
benzaldehydes and 1,3-propanedithiol in CH3Cl, in the presence
of a catalytic amount of iodine.[33] 5-(Iodomethyl)-5-methyl-
1-hexene 2 was synthesised as previously reported.[25,32]
More recently, detailed photochemical studies were conducted
to investigate the reactions of neophyl iodide with carbanions
such as the enolate anion of acetophenone (ꢀCH2COPh) and the
nitromethane anion (ꢀCH2NO2) in DMSO, with the rate constant
for the coupling of ꢀCH2COPh anion with the neophyl radical
found to be 1.2 ꢁ 105 Mꢀ1 sꢀ1. In relation to this, on the basis of
determinations of quantum yields for the initiation step, these
reactions were proposed to occur by the SRN1 mechanism, with
chain lengths of 127 (ꢀCH2COPh) and 2 (ꢀCH2NO2), although the
overall quantum yields were below 1.[30] Conversely, in the
reaction of iodide 2 with lithium propiophenone in HMPA
(lab. light), O-alkylation was the principal reaction (66%),
accompanied by a small yield of C-alkylation (11%) and a small
amount of the cyclic hydrocarbon. It was suggested that this
reaction takes place, at least partly, with radicals as intermedi-
ates.[31] The effects of leaving groups, solvents, and hydrogen
atom donors have been investigated in the reaction of probe 2
with i-PrSLi, with quantitative straightforward substitution being
obtained in THF. However, the same reaction, in the presence of
the radical trap dicyclohexylphosphine (DCPH), yielded up to
19% of the cyclic and 3% of the straightforward hydrocarbon
products, respectively. These results were also attributed to an ET
pathway with geminate radical coupling within the solvent cage,
to some extent, for the reaction of probe 2 with i-PrSLi in THF.[32]
The aim of this study is to explore further the reactivity of
dithiane anions as electron donors in radical nucleophilic
substitution by the SRN1 mechanism. Thus, we herein report a
comprehensive study of the reactions of 2-aryl-1,3-dithiane
General methods
Irradiation was conducted with two 400 W medium pressure
Hg lamps emitting maximally at 365 nm. All the reaction
products were isolated by radial chromatography from the
reaction mixture, and characterised by 1H and 13C NMR and
mass spectrometry. All the dithiane[34] precursors of anions 3a–d
and 4-benzylbenzonitrile (24)[35] are well known and exhibited
physical properties identical to those reported in the literature.
1H and 13C NMR spectra were recorded at 400.16 and 100.62 MHz
respectively on a 400 spectrometer, and all spectra were reported
in d (ppm) relative to Me4Si, with CDCl3 as solvent at 30 8C. Gas
chromatographic analyses were performed with
a flame-
ionisation detector, on 30 m capillary column of a 0.32 mm ꢁ
0.25 mm film thickness, with a 5% phenylpolysiloxane phase.
GS-MS analyses were performed employing a 25 m ꢁ 0.2 mm ꢁ
0.33 mm with a 5% phenylpolysiloxane phase column.
Representative experimental procedure
These reactions were carried out in a 10 mL three-necked Schlenk
tube, equipped with a nitrogen gas inlet and a magnetic stirrer.
Scheme 2. ET reaction between 2-lithio-1,3-dithiane (1) and 5-(iodomethyl)-5-methyl-1-hexene (2).
J. Phys. Org. Chem. 2011, 24 1136–1143
Copyright ß 2011 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/poc

˜ ´ ˜
G. OKSDATH-MANSILLA AND A. B. PENENORY
The tube was dried under vacuum, filled with nitrogen and then
with dried DMSO (5 mL). tert-BuOK (336.7 mg, 3.0 mmol),
dithiane precursor of 3a–d (294.0 mg, 1.5 mmol), and neopentyl
iodide (99.0 mg, 0.5 mmol), were added to the degassed solvent
under nitrogen. After 1 h of irradiation with a medium pressure
Hg lamp emitting maximally at 365 nm, the reaction was
quenched by the addition of ammonium nitrate in excess and
10 mL of water, and the mixture was extracted with methylene
chloride (3 mL ꢁ 20 mL). The organic extract was washed twice
with water and dried over anhydrous MgSO4, and the products
were quantified by GC using the internal standard method, or
isolated by radial chromatography from the crude product
reaction mixture.
2-(2-Methyl-2-phenylpropyl)-2-phenyl-1,3-dithiane (15)
Solid, m.p. 57.3–57.8 8C. 1H NMR: d ¼ 1.15 (s, 6 H), 1.80–1.83 (m, 2
H)ppm. 2.45–2.63 (m, 6 H), 7.15–7.25 (m, 8 H), 7.77–7.78 (m, 2 H).
13C NMR: d ¼ 24.9, 27.9, 28.1, 30.5, 39.19, 58.7, 58.99, 125.5, 125.86,
126.05, 126.18, 126.64, 127.67, 128.11, 129.81, 141.12, 148.7 ppm.
MS (EIþ), m/z (relative intensity), 328 (5, Mþ); 209 (8); 121 (20); 119
(64); 103 (70); 91 (100); 77 (41); 51 (27). HRMS, calcd for C20H24S2
328.1319, found 328.1319.
2-(2,2-Dimethylhex-5-enyl)-2-phenyl-1,3-dithiane (16)
Solid, 1H NMR: d ¼ 0.70 (s, 6 H), 1.08–1.12 (m, 2 H), 1.86–1.93 (m, 4
H), 2.15 (s, 2 H), 2.56–2.76 (m, 4 H), 4.83–4.90 (m, 2 H), 5.60–5.69 (m,
1 H), 7.26–7.38 (m, 3 H), 7.97–7.99 (m, 2 H) ppm. 13C NMR: d ¼ 25.0,
27.9, 28.5, 28.6, 35.9, 42.9, 55.9, 58.96, 113.7, 126.8, 128.3, 129.9,
139.5, 141.4 ppm. MS (EIþ), m/z (relative intensity), 306 (4, Mþ);
231 (14); 209 (24); 195 (25); 121 (42); 106 (100); 103 (68); 91 (24); 77
(27); 55 (70). HRMS (MHþ), calcd for C18H27S2 307.1554, found
307.1553.
2-Neopentyl-2-phenyl-1,3-dithiane (6a)
White solid, m.p.: 147.8–148.1 8C. 1H NMR: d ¼ 0.722 (s, 9 H),
1.90–1.93 (m, 2 H), 2.16 (s, 2 H), 2.58–2.73 (m, 4 H), 7.24–7.37 (m, 3
H) ppm, 7.96–7.98 (m, 2 H). 13C NMR: d ¼ 25.05, 27.86, 31.29, 33.44,
58.11, 58.95, 126.80, 128.24, 129.89, 141.35 ppm. MS (EIþ), m/z
(relative intensity), 266 (26, Mþ); 209 (6); 195 (48); 136 (68); 121
(37); 103 (100); 91 (11); 77 (37); 57 (74). HRMS, calcd for C15H22S2
266.1163, found 266.1164.
2-((3,3-Dimethylcyclopentyl)methyl)-2-phenyl-1,3-dithiane
(17)
1
Solid, H NMR: d ¼ 0.85 (s, 3 H), 0.92 (s, 3 H), 1.03–1.43 (m, 6 H),
2-(4-Methoxyphenyl)-2-neopentyl-1,3-dithiane (6b)
1.94–2.12 (m, 5 H), 2.61–2.76 (m, 4 H), 7.25–7.29 (m, 1 H), 7.37–7.41
(m, 2 H), 7.94–7.96 (m, 2 H) ppm. 13C NMR: d ¼ 25.2, 27.67, 27.69,
29.5, 30.3, 30.7, 35.2, 38.5, 40.5, 49.7, 52.7, 59.4, 126.8, 128.4, 129.1,
141.9 ppm. MS (EIþ), m/z (relative intensity), 306 (19, Mþ); 231 (16);
217 (18); 195 (100); 136 (73); 121 (38); 103 (18); 91 (15); 69 (26); 55
(32). HRMS (MHþ), calcd for C18H27S2 307.1554, found 307.1553.
1
Solid, m.p. 69.5–69.8 8C. H NMR: d ¼ 0.75 (s, 9 H), 1.88–1.95 (m,
2 H), 2.15 (s, 2 H), 2.57–2.63 (m, 2 H), 2.72–2.79 (m, 2 H), 3.85 (s, 3 H),
6.89–6.91 (d, 2 H, J ¼ 8.8), 7.86–7.88 (d, 2 H, J ¼ 8.8) ppm. 13C NMR:
d ¼ 25.1, 27.8, 31.3, 33.4, 55.3, 58.2, 58.7, 113.4, 131.1, 133.2,
158.4 ppm. HRMS (MHþ), calcd for C16H25OS2 297.1347, found
297.1344.
1,3-bis(2,2-dimethylhex-5-enylthio)propane (18)
Liquid, 1H NMR: d ¼ 0.98 (s, 12 H), 1.38–1.42 (m, 4 H), 1.85–1.89 (q,
2 H, J ¼ 7.2), 2.00–2.06 (m, 4 H), 2.48 (s, 4 H), 2.60–2.64 (t, 4 H,
J ¼ 7.2), 4.94–5.06 (dd, 4 H), 5.79–5.87 (m, 2 H) ppm. 13C NMR:
d ¼ 26.7, 28.6, 29.9, 32.97, 34.6, 40.4, 45.8, 114.0, 139.3 ppm. MS
(EIþ), m/z (relative intensity), 328 (5, Mþ); 217 (65); 121 (18); 107
(26); 95 (27); 69 (100); 55 (76). HRMS (MHþ), calcd for C19H37S2
329.2337, found 329.2341.
2-(4-Clorophenyl)-2-neopentyl-1,3-dithiane (6c)
Solid, m.p. 83.9–84.2 8C. 1H NMR: d ¼ 0.75 (s, 9 H), 1.92–1.95 (m, 2
H), 2.15 (s, 2 H), 2.59–2.71 (m, 4 H), 7.33–7.35 (d, 2 H, J ¼ 8.8),
7.92–7.94 (d, 2 H, J ¼ 8.8) ppm. 13C NMR: d ¼ 24.9, 27.8, 31.4, 33.5,
58.2, 58.3, 128.3, 131.5, 132.8, 140.3 ppm. MS (EIþ), m/z (relative
intensity), 302 (16, Mþ2); 300 (39 Mþ); 243 (51); 229 (67); 191 (40);
170 (94); 155 (41); 139 (41); 137 (100); 107 (34); 57 (86). HRMS
(MHþ), calcd for C15H22ClS2 301.0851, found 301.0846.
4-(2-Phenyl-1,3-dithian-2-yl)benzonitrile (23)
Solid, 1H NMR: d ¼ 2.03–2.06 (m, 2 H), 2.72–2.81 (m, 4 H), 7.30–7.38
(m, 3 H), 7.63–7.65 (d, 4 H, J ¼ 8.4), 7.84–7.86 (d, 2 H, J ¼ 8.4) ppm.
13C NMR: d ¼ 24.1, 29.3, 111.4, 118.6, 128.1, 128.7, 129.1, 130.25,
132.2 ppm. MS (EIþ), m/z (relative intensity), 297 (28, Mþ); 223
(100); 190 (95); 146 (18); 121 (24); 105 (5). HRMS (MHþ), calcd for
C17H16NS2 298.0724, found 298.0725.
4-(2-Neopentyl-1,3-dithian-2-yl)benzonitrile (6d)
Solid, m.p. 98.8–99.2 8C. 1H NMR: d ¼ 0.73 (s, 9 H), 1.96–1.99 (m, 2
H), 2.18 (s, 2 H), 2.63–2.665 (m, 4 H), 7.67–7.70 (d, 2 H, J ¼ 8.8),
8.13–8.16 (d, 2 H, J ¼ 8.8) ppm. 13C NMR: d ¼ 24.6, 27.8, 31.4, 33.6,
58.1, 130.8, 132.0 ppm. MS (EIþ), m/z (relative intensity), 293 (4,
M
þ2); 291 (38 Mþ); 234 (100); 220 (91); 161 (83); 146 (51); 128 (55);
107 (98); 57 (96); 41 (72). HRMS, calcd for C16H21NS2 291.1115,
found 291.1105.
4-((3-Mercaptopropylthio)(phenyl)methyl)benzonitrile (25)
Solid, m.p: 139.5–140.0 8C. 1H NMR: d ¼ 1.88–1.92 (m, 1 H),
2.08–2.16 (m, 1 H), 2.80–2.90 (m, 4 H), 5.12 (s, 1 H), 7.47–7.48 (m, 5
H), 7.53–7.54 (m, 1 H), 7.62–7.65 (m, 2 H), 7.67–7.89 (m, 1 H) ppm.
13C NMR: d ¼ 24.8, 32.0, 48.0, 112.0, 118.4, 128.4, 128.7, 128,8,
129.9, 131.4, 133.8, 138.1, 141.7, 141.9 ppm. MS (EIþ), m/z
(relative intensity), 299 (3, Mþ); 297 (27); 223 (24); 222 (83); 191
(100); 106 (38). HRMS (MHþ), calcd for C17H18NS2 300.0875, found
300.0856.
1,3-bis(Neopentylthio)propane (7)
Liquid, 1H NMR: d ¼ 1.00 (s, 18 H), 1.83–1.90 (q, 2 H, J ¼ 7.2), 2.46 (s,
4 H), 2.61–2.64 (t, 4 H J ¼ 7.2) ppm. 13C NMR: d ¼ 28.94, 29.92,
32.32, 33.11, 47.60 ppm. MS (EIþ), m/z (relative intensity), 248 (24,
Mþ); 177 (79); 121 (20); 107 (100); 57 (25); 43 (71). HRMS, calcd for
C13H28S2 248.1632, found 248.1633.
wileyonlinelibrary.com/journal/poc
Copyright ß 2011 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2011, 24 1136–1143

SRN1 REACTIONS OF 2-ARYL-1,3-DITHIANE ANIONS
RESULTS AND DISCUSSION
Table 1. Dark reactions of anion 3a with neopentyl iodide (4)
in DMSOa
Reaction of 2-phenyl-1,3-dithiane anion (3a) with neopentyl
iodide (4) in DMSO
Product yield
(%)b
2-Phenyl-1,3-dithiane anion (3a) can be generated in
DMSO under nitrogen atmosphere by deprotonation of the
dithiane precursor (pKa ¼ 30.5)[36] with tert-BuOK (pKa ¼ 32.2).[33]
Accordingly, the addition of an excess of MeI to a solution of
anion 3a (produced with 3 equiv of base) afforded 2-methyl-2-
phenyl-1,3-dithiane in a 71% yield (determined by GC quanti-
fication). In the presence of only 1.5 equiv of tert-BuOK, a 60%
yield of the methylated derivative was obtained.
When a solution[37] of 2-phenyl-1,3-dithiane anion (3a) and
4 was stirred for 1 h under irradiation at lmax > 350 nm, the
substitution product 2-neopentyl-2-phenyl-1,3-dithiane (6a) was
obtained in 42% yield, together with small amounts of the
neopentyl substituted product 7 (7% yield) and traces of the
hydrocarbons 8a and 9a (Eqn 4, Table 1, entry 1). This reaction
proceeded to a similar conversion in the dark to afford 6a as the
major product, with a minor amount of product 7 (Table 1, entry
2). Both experiments were performed using a 3a:4 ratio of 3:1 and
1 equiv of tert-BuOK.
Conditions
3a:4:t-BuOK (eq)
Entry
Time (min)
Iꢀ
6a
7c
1d
2
3:1:6
3:1:6
3:1:3
1:1:1
3:1:6
3:1:6
1:1:1
1:1:1
1:1:1
3:1:6
3:1:6
1:—:1
—:—:1
—:—:3
3:1:6
60
60
60
60
60
60
5
82
81
76
60
84
9
42
42
51
50
32
0
7
4
10
3
14
0
3
4
5e
6f
7
8e
62
20
49
79
81
—
—
—
35
—
30
9
8
3
7
9
7
—
—
k
5
5
9d
10g
11h
12i
13d,i
14i,j
15l
16l
43
43
32
92
90
24
10
—
60
60
60
180
180
180
180
—
—
3:1:6
a Under nitrogen atmosphere, at room temperature;
[3a] ¼ 0.3 M (ratio 3a:4 ¼ 3:1 or 1:1 as indicated). This reaction
was quenched with NH4NO3.
b Product yield determined by CG using the internal method,
error <5%, and iodide ions potentiometrically.
c Together with 8a and 9a, detected by GC-MS but not
quantified (<5%).
d Under irradiation, lmax > 350 nm.
When equimolar amounts of dithiane and tert-BuOK
were employed, a slight improvement in the yields of the
substitution products 6a and 7 was observed after 60 min (51%
and 10% yields, respectively; Table 1, entry 3). A comparable
performance of this reaction was achieved by using a ratio
dithiane:tert-BuOK:4 of 1:1:1, giving a 50% yield of 6a after
60 min. In order to assess the participation of radicals and radical
anions in these dark reactions, the following experiments
were conducted under different conditions. The addition of
10 mol.% of p-dinitrobenzene (p-DNB), a well-known electron
acceptor, caused a small inhibition of the reaction. However, in
the presence of 1 equiv of p-DNB, the reaction was completely
inhibited (Table 1, entries 5 and 6). Finally, when the reaction was
performed to a minor conversion after only 5 min of stirring,
the yield of 6a was reduced from 30% to 9% by adding 10 mol.%
of p-DNB, and was slightly catalysed by light (Table 1, entries 7–9).
The effect of light and the inhibition results clearly indicate
the participation of ET process in these reactions. On the other
hand, in the presence of a very efficient radical trap such as
di-tert-butylnitroxide (DTBN), there was no inhibition in these
reactions (Table 1, entries 10 and 11). A possible oxidation of the
dithiane anion 3 by DTBN may have taken place, thus avoiding
inhibition.[38,39]
e In the presence of 10 mol.% of p-DNB.
f In the presence of 1 equiv of p-DNB.
g In the presence of 14 mol.% of DTBN.
h In the presence of 1 equiv of DTBN.
i Control experiment of 6a, quenched by addition of MeI after
irradiation.
j Under irradiation, lmax ¼ 300 nm using a quartz tube.
k Hydrocarbons 8 and 9 were quantified by GC, showing yields
of 38% and 26%, respectively.
l Performed in NH3 (liq) at ꢀ33 8C with [3a] ¼ 0.012 M, and
quenched by the addition of MeI after the indicated time.
From this mixture, 6a was recovered at a 92% yield, showing no
evidence of 1,3-bis(methylthio)propane being detected. Finally,
when product 6a was irradiated at lmax > 350 nm for 3 h in
the presence of 1 equiv of tert-BuOK and then quenched with
MeI, 6a was recovered at a 90% yield revealing no trace of other
products being detected by GC-MS (Table 1, entries 12 and 13).
These control experiments clearly point out the stability of 6a
in the presence of an excess of anion 3a, even after irradiation
with tert-BuOK under the photoinduced reaction conditions
used in the present study. In order to force the reduction of 6a, a
solution of this dithiane in the presence of 3 equiv of tert-BuOK
was irradiated at l ¼ 300 nm, using a quartz tube. After 3 h,
hydrocarbons 8a and 9a were obtained at 38% and 26% yields,
respectively; dithiane 6a recovered at a 24% yield (Table 1, entry
In view of the absence of the corresponding aldehyde,
the possible formation of 1,3-propane dithiolate anion (10)
by deprotection of the dithiane precursor of anion 3a can be
disregarded, during photochemically and thermally induced
reactions. Furthermore, for the control experiment, the substi-
tution product 6a was added to a solution of anion 3a in DMSO.
Then, after stirring for 1 h the mixture was quenched with MeI.
J. Phys. Org. Chem. 2011, 24 1136–1143
Copyright ß 2011 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/poc

˜ ´ ˜
G. OKSDATH-MANSILLA AND A. B. PENENORY
14). Finally, liquid ammonia (ꢀ33 8C) was tested as a solvent. After
3 h of irradiation, a mixture of anion 3a with iodide 4 was
quenched with MeI, resulting in only the substitution product 6a
being obtained (10% yield), where derivative 7 was not observed.
This reaction was completely inhibited in the dark (Table 1,
entries 15 and 16).
The formation of the secondary substitution product 7 by
the reaction of iodide 4 with the 1,3-propane dithiolate anion
(10), generated in the reaction medium (see results above) was
unequivocally confirmed by a series of independent experiments
(Eqn 5, supplementary material, Table S1). The reaction between
4 and 10 was thermally induced, occurred very fast (within
<5 min) and was slightly catalysed by light. The ET ability of
lithium thiolate ions towards alkyl iodide has been previously
established; from the above results and the low reactivity of
neopentyl halides in the polar process, the participation of
an ET reaction is the main plausible mechanism to account for
the formation of the derivative 7.[32]
On the other hand, the reaction between 3a and iodide 2 gave
a 53% yield of the un-cyclised substitution product 16, while the
cyclised derivative 17 was not detected. The secondary product
18, produced from substitution of iodide 2 with the 1,3-propane
dithiolate anion (10) that was generated in the reaction medium,
was obtained at a 10% yield (Eqn 7, Table 2, entry 3).[44] However,
when the reaction was performed with a ratio substrate:nucleo-
phile of 2:1 in a 10-fold more diluted solution, it was possible to
observe, both the un-cyclised and cyclised substitution products
16 and 17, respectively, together with 18 (Table 2, entry 4).
Similar cyclised and un-cyclised derivatives were previously
observed in the dark reaction of dithiane anion 1 with iodide 2,
(Scheme 2).[25]
Reaction of 2-phenyl-1,3-dithiane anion (3a) with radical
probe iodides 2 and 5 in DMSO
The rearrangement of the neophyl radical (11) to a tertiary
radical 12 (kr ¼ 4.02 ꢁ 102 sꢀ1 [40]
)
2,2-dimethyl-5-hexenyl radical (13) to the cyclopentylmethyl
and also the cyclisation of the
radical 14 (kc ¼ 3.6 ꢁ 106 sꢀ1 [41]
)
are well-known processes,
Reaction with other 2-aryl-1,3-dithiane anions (3b–d) in
DMSO
with the Arrhenius functions for both reactions determined
previously,[42,43] In order to investigate the presence of free
radicals, the photoinduced reactions between anion 3a and
neophyl iodide (5) and the neopentyl-type probe 2 were carried
out in a nitrogen atmosphere. The reactions of iodide 5 with
anion 3a were performed under different concentrations and
nucleophile:substrate ratios, but afforded only the unrearranged
product 15 (Eqn 6, Table 2, entries 1 and 2). The fact that no
rearranged product was observed could arise from a nucleophilic
addition rate to radical 11 occurring faster than that of radical
rearrangement.[30]
The reactivity of iodide 4 towards the anions from 2-(4-Z-
phenyl)-1,3-dithiane (3b–d, Z ¼OMe, Cl, CN) was also studied. In
the reaction between iodide 4 and the para-OMe substituted
anion 3b, the major product found was the substitution one (6b)
in the dark and under irradiation (Table 3, entries 1 and 2).
The behaviour of anion 3b was comparable to that of anion 3a.
However, the photoinduced reaction of 4 with anion 3d (Z ¼CN)
after 3 h afforded a 19% yield of the substitution product (6d) and
a 20% yield of 7, together with 4-(3,3-dimethylbutyl)benzonitrile
Table 2. Reactions of anion 3a with neophyl iodide (5) and 5-(iodomethyl)-5-methyl-1-hexene (2) in DMSOa
Product yield (%)b
Entry
Substrate
3a/substrate (M)
15
16
17
18c
1
2
3
4
5
5
2
2
0.3/0.1
0.1/0.1
0.15/0.1
0.021/0.012
10
26
d
53
19
10
15
5
a Irradiation for 3 h under nitrogen atmosphere. This reaction was quenched with NH4NO3.
b Determined by CG using the internal method, error <5%.
c Together with the reduced product analogous to 8 and 9, which were detected by GC-MS but not quantified.
d Not detected.
wileyonlinelibrary.com/journal/poc
Copyright ß 2011 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2011, 24 1136–1143

SRN1 REACTIONS OF 2-ARYL-1,3-DITHIANE ANIONS
propagation of the reaction; and pathway C, the radical anion
.
Table 3. Reactions with other derivatives 2-aryl-1,3-dithiane
anions with substrate 4 in DMSOa
(6a–d ꢀ) intermediate can cleave by fragmentation of the C–S
bond to provide a distonic radical anion 20. By subsequent
hydrogen abstraction, reduction and fragmentation, 20 gives the
dianion 10 and the alkyl radical 21. Disproportionation of 21
affords hydrocarbons 8 and 9, while the secondary substitution
reaction between anion 10 and iodide 4 yields the disubstitution
product 7 (Scheme 3).
Product yield (%)b
Entry
Anion
6
7
6/7
1
3b (OMe)
3b
3c (Cl)
3c
3d (CN)
3d
42 (6b)
41 (6b)
22 (6c)
13 (6c)
19 (6d)
9 (6d)
4
10
—
2
—
1
The different reactivity observed in the reactions of iodide 4
with anions 3a–d gives further support to the mechanism
proposed. The ET from anions 3a–d to 4 depends mainly on the
oxidation potential of the anion, which determines its ability as an
electron donor. Bordwell et al. developed a linear correlation of
Eox (Aꢀ) versus pKHA, with slopes near unity, which established
the presence of an intrinsic relationship between the oxidation
potential of carbanions and their equilibrium acidities in DMSO.
These authors showed that for an increase in the ion basicity,
there was an Eox (Aꢀ) shift to a more negative potential. In
general, it has been observed that the higher the pKHA, the
higher the electron donor capability of its conjugate base, and
consequently the higher the probability of a spontaneous ET
reaction.[45–47] Our experimental data reveal that the electron
donor ability of anions 3 followed the order: 3a ¼ 3b > 3c > 3d.
Unfortunately, there have been no pKa values reported for
dithiane 3b–d, the electron donor ability order observed was in
good agreement with the Hammett s substituent constant: OMe:
ꢀ0.28; Cl: 0.24 and CN: 0.65, as expected.[48]
2c
3
<5
10d
—
20e
—
4c
5
6c
—
a Irradiation for 3 h under nitrogen atmosphere, at room
temperature; [3] ¼ 0.3 M (ratio 3:4 ¼ 3:1). This reaction was
quenched with NH4NO3.
b Determined by CG using the internal method, error <5%.
c Dark reaction.
d Together with 8c, detected by GC-MS but not quantified.
e Together with 8d, detected by GC-MS but not quantified.
(8d), which was detected by GC-MS but not quantified.
This reaction was strongly inhibited in the dark, with only 9%
yield of 6d being quantified after 3 h. For the para-Cl substituted
anion 3c, the substitution 6c and the fragmentation product 7
were obtained under irradiation, while this reaction was partially
inhibited in the dark (Table 3, entries 3–6).
The competition between ET and the C–S bond fragmentation
pathways (B and C, respectively), could be measured by the 6:7
ratio, and followed the order: CN: 1; Cl: 2; OMe: 10. This
trend agrees with the stability order of the distonic radical anions
20a–d, the Cl and CN derivatives being further stabilised by a
captodative effect.[49,50]
Mechanistic Consideration
The occurrence of the reaction between anions 3 and 4 in
the dark, and the inhibition by p-DNB strongly suggest the
participation of a spontaneous or thermal ET as a possible
initiation for a chain reaction, which is light catalysed, and imply
the involvement of radical anions as intermediates. The latter is
undoubtedly supported by the presence of the desulphurised
hydrocarbons 8a,b and 9a and the disubstitution product 7 from
reaction between iodide 4 and 1,3-propanedithiolate dianion
(10), as minor products. These secondary products (8–10)
arose from fragmentation of the radical anion intermediate of the
The results obtained for the photostimulated reaction of anion
3d with the aryl substrates are mechanistically interesting.[51] In
these reactions, the fragmentation step was the most favoured
pathway. For example, when the anion 3d (3 equiv) reacted with
IPh (22) under irradiation, the substitution product
4-(2-phenyl-1,3-dithian-2-yl) benzonitrile (23) was found at a 7%
yield with the fragmentation products 4-benzylbenzonitrile (24)
(17%), thiophenol[52] (4%), and 3-mercaptopropylthio derivative 25
(4%), and also with other minor products that could not be
separated or identified satisfactorily (Eqn 8). This result can be
explained by considering the stability of the distonic radical anion
.
primary substitution product (6a ꢀ). The stability of the primary
product 6a under the reaction conditions has been shown by
different control experiments. Finally, C–S bond fragmentation
was avoided when the reaction was performed in liquid ammonia
(ꢀ33 8C) under photostimulation, attributed to stabilisation of
the radical anion intermediate by lowering 608 the reaction
temperature. Although the radical trap DTBN showed no
inhibition of these reactions,[35] the participation of free
radicals as intermediates was established by the observation
of the cyclised substitution product (17) in the reaction of the
radical-type probe 2 and anion 3a. Thus, a radical substitution
.
arising from the fragmentation step of 23 ꢀ (analogous to 20,
pathway C, Scheme 3, and confirmed by isolation of product 25).
There was no reaction in the dark, but the photoinduced reaction in
the presence of substrate 22 in excess (5 equiv) afforded the
substitution product 23 and the fragmentation product 24 with
yields of 14% and 4%, respectively. As expected, under the latter
conditions, a rise in the concentration of the electron acceptor (PhI)
increased the rate of the intermolecular ET and competed
efficiently with the fragmentation step.
SRN1 mechanism is proposed to account for all the results
described above and is presented in Scheme 3. After initial
interaction between anion 3a–d and iodide 4, an EToccurs as the
initial step A of the process, with formation of the alkyl radical 19.
These free alkyl radicals (19) add to the anion to afford the radical
.
anion of the primary substitution product (6a–d ꢀ), this following
two competitive reaction channels: pathway B, ET to the
substrate affords the substitution product (6a–d) and the alkyl
radical 19, the latter being able to continue with the chain
J. Phys. Org. Chem. 2011, 24 1136–1143
Copyright ß 2011 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/poc

˜ ´ ˜
G. OKSDATH-MANSILLA AND A. B. PENENORY
Scheme 3. SRN1 reaction between 2-aryl-1,3-dithiane anions (3a–d) and neopentyl iodide (4).
´ ˜
[2] R. A. Rossi, A. B. Pierini, A. B. Pen˜enory, Chem. Rev. 2003, 103, 71.
CONCLUSIONS
˜ ´ ˜
[3] R. A. Rossi, A. B. Penenory, in CRC Handbook of Organic Photochem-
istry and Photobiology, 2nd edn, Chapter 47 (Eds: W. M. Horspool, F.
Lenci), CRC Press Inc., Boca Raton, 2004, pp. 47-1–47-24.
Neopentyl iodide (4) reacted with dithiane anions 3a–d by a SRN
1
mechanism with radicals and radical anions as intermediates. The
occurrence of this mechanism was suggested mainly from
observation of the products from the C–S bond fragmentation of
the radical anion of the substitution product. Due to the fact that
the fragmentation process of the intermediate radical anion was
a competitive reaction, the chain may have been short or even
non-existent. As a consequence, the yields of the substitution
product for the reactions between dithianes anions substituted
with electron-donor substituents 3a,b with neopentyl and their
analogue iodides were moderate (41–53%) in DMSO. On the
other hand, in liquid ammonia the fragmentation was avoided,
but as the initiation of the reaction was not effective enough, the
performance of the reaction was poor.
´
[4] J. M. Saveant, Acc. Chem. Res. 1993, 26, 455.
[5] S. M. Palacios, S. E. Asis, R. A. Rossi, Bull. Soc. Chim. Fr. 1993, 130, 111.
[6] C. Galli, J. F. Bunnett, J. Org. Chem. 1984, 49, 3041.
[7] C. Galli, P. Gentili, J. Chem. Soc., Perkin Trans. 2, 1993, 1135.
[8] M. A. Nazareno, R. A. Rossi, J. Org. Chem. 1996, 61, 1645.
[9] M. A. Nazareno, R. A. Rossi, Tetrahedron Lett. 1994, 35, 5185.
[10] E. Austin, C. G. Ferrayoli, R. A. Alonso, R. A. Rossi, Tetrahedron 1993, 49,
4495.
´
[11] C. Costentin, P. Hapiot, M. Medebielle, J.-M. Saveant, J. Am. Chem. Soc.
2000, 122, 5623.
[12] R. G. Scamehorn, J. F. Bunnett, J. Org. Chem. 1977, 42, 1449.
[13] D. R. Carver, J. S. Hubbard, J. F. Wolfe, J. Org. Chem. 1982, 47, 1036.
[14] N. Kornblum, Angew. Chem. Int. Engl. 1975, 14, 734.
[15] G. A. Russell, R. K. Norris, E. J. Panek, J. Am. Chem. Soc. 1971, 93, 5839.
´
[16] J.-M. Saveant, Adv. Phys. Org. Chem. (Ed. T. T. Tidwell), Academic Press,
2000, 35, 117.
On the basis of our findings, further studies are currently in
progress in order to develop this photoinduced reductive process
of dithiane derivatives as an environmentally friendly alternative
to the current reduction methodologies for the de-oxygenation
of carbonyl groups.
[17] N. Kornblum, Aldrichim. Acta 1990, 23, 71.
˜ ´ ˜
[18] A. B. Penenory, J. E. Argu¨ello, in Handbook of Synthetic Photochem-
istry, Chapter 10 (Eds: A. Albini, M. Fagnoni), WILEY-VCH Verlag GmbH
& Co. KGaA, Weinheim, 2010, p. 319.
[19] T. W. Greene, P. G. M. Wuts, Protective Groups in Organic Synthesis, 3rd
edn, John Wiley & Sons, New York, 1999.
[20] P. J. Kocienski, Protecting Groups, George Thieme, Sttugart, 1994.
[21] E. J. Corey, D. Seebach, Angew. Chem. Int. Ed. Engl. 1965, 4, 1075.
[22] D. Seebach, E. J. Corey, J. Org. Chem. 1975, 40, 231.
[23] B. T. Grobel, D. Seebach, Synthesis 1977, 357.
[24] P. C. Bulman Page, M. B. van Niel, J. C. Prodger, Tetrahedron 1989, 45,
7643.
Acknowledgements
This work was partly supported by Secretar´ıa de Ciencia y
´
Tecnolog´ıa de la Universidad Nacional de Cordoba (SECYT-UNC),
Consejo Nacional de Investigaciones Cient´ıficas y Tecnicas (CON-
´
´
´
[25] E. Juaristi, H. A. Jimenez-Vazquez, J. Org. Chem. 1991, 56, 1623.
[26] A. B. Pierini, A. B. Penenory, R. A. Rossi, J. Org. Chem. 1985, 50, 2739.
´
´
ICET) and Agencia de Promocion Cient´ıfica y Tecnica (ANPCYT-
FONCYT), Argentina. G.O.-M. gratefully acknowledges the receipt
of a fellowship from CONICET.
˜ ´ ˜
˜ ´ ˜
[27] E. R. N. Bornancini, S. M. Palacios, A. B. Penenory, R. A. Rossi, J. Phys.
Org. Chem. 1989, 2, 255.
˜ ´ ˜
[28] A. B. Penenory, R. A. Rossi, Gazz. Chim. Ital. 1995, 125, 605.
[29] A. L. J. Beckwith, S. M. Palacios, J. Phys. Org. Chem. 1991, 4, 404.
˜ ´ ˜
[30] J. E. Argu¨ello, A. B. Penenory, R. A. Rossi, J. Org. Chem. 2000, 65, 7175.
REFERENCES
[31] E. C. Ashby, J. N. Argyropoulos, J. Org. Chem. 1985, 50, 3274.
[32] E. C. Ashby, W. S. Park, A. B. Goel, W. Y. Su, J. Org. Chem. 1985, 50, 5184.
[33] H. Firouzabadi, N. Iranpoor, H. Hazarkhani, J. Org. Chem. 2001, 66,
7527.
˜ ´ ˜
[1] A. B. Pierini, A. B. Penenory, M. T. Baumgartner, in Electron Transfer
Reactions in Organic Synthesis (Ed.: P. Vanelle), Research Signpost,
Trivandrum, 2002, p. 63.
[34] S. K. De, Synthesis 2004, 2837.
wileyonlinelibrary.com/journal/poc
Copyright ß 2011 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2011, 24 1136–1143

SRN1 REACTIONS OF 2-ARYL-1,3-DITHIANE ANIONS
[35] D. R. Arnold, X. Du, J. Chen, Can J. Chem. 1995, 73, 307.
[36] F. G. Bordwell, Acc. Chem. Res. 1988, 21, 456.
[45] F. G. Bordwell, X. Zhang, Acc. Chem. Res. 1993, 26, 510.
[46] F. G. Bordwell, X. Zhang, R. Filler, J. Org. Chem. 1993, 58, 6067.
[47] F. G. Bordwell, A. H. Clemens, D. E. Smith, J. Begemann, J. Org. Chem.
1985, 50, 1151.
[48] A plot of the pK values for para-substituted acetophenones vs.
Hammett s constant gave a linear correlation. F. G. Bordwell, F. J.
Cornforth, J. Org. Chem. 1978, 43, 1763. For further examples, see
Bordwell pKa Table (acidity in DMSO). http://www.chem.wisc.edu/
areas/reich/pkatable and references cited therein.
[37] After addition of iodide 4 to a solution of anions 3a–d in DMSO,a red
color was immediately developed, probably due to a charge transfer
complex formation between these species. R. Foster, Organic Char-
ge-Transfer Complexes, Academic Press, London and New York, 1969.
[38] Oxidation of 1,3-dithiane anion by stable nitroxides such as TEMPO
has been previously reported. A. J. Herrera, A. Studer, Synthesis 2005,
1389.
´
[49] H. G. Viehe, Z. Janousek, R. Merenyi, L. Stella, Acc. Chem. Res. 1985, 18,
917.
[50] L. Stella, J. N. Harvey, Synthetic utility of the captodative effect, in
Radicals in Organic Synthesis, Vol. I, Ch 3.3 (Eds: P. Renaud, M. P. Sibi),
Wiley-VCH, Weinheim, 2001, p. 360.
[51] A very low reactivity of 1,3-dithiane anion towards BrPh in liquid
ammnonia (4% of the substitution product detected by GC)
was previously reported, but no additional information was
provided. M. F. Semmelhack, T. Bargar, J. Am. Chem. Soc. 1980,
102, 7765.
[52] Thiophenol was formed by the SRN1 reaction between the alkyl
thiolate dianion 10 PhI, followed by fragmentation of the radical
anion of the substitution product. R. A. Rossi, S. M. Palacios, J. Org.
Chem. 1981, 46, 5300.
[39] The poor mass balance of the reactions performed in the presence of
tert-BuOK in excess or by the addition of DTBN can probably be
ascribed to an increasing amount of neopentane, from hydrogen
abstraction of the neopentyl radical intermediate. This highly volatile
hydrocarbon could not be detected under the reactions conditions.
[40] March’s Advances in Organic Chemistry, 6th edn. (Eds: M. B. Smith,
J. March), John Wiley & Sons, New Jersey, 2007, p 1574.
[41] E. C. Ashby, J. Bowers, R. De Priest, Tetrahedron Lett. 1980, 21, 3541.
[42] M. Weber, H. Fisher, J. Am. Chem. Soc. 1999, 121, 7381.
[43] A. L. J. Beckwith, C. J. Easton, T. Lawrence, A. K. Serelis, Aust. J. Chem.
1983, 36, 545.
[44] Cyclised derivatives from the reaction of radical 15 with
1,3-propanedithiolate anion and hydrocarbons 9a and 10a were
also detected as traces by GC-MS.
J. Phys. Org. Chem. 2011, 24 1136–1143
Copyright ß 2011 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/poc