Regioselectivity in the intramolecular heck reaction of a series of cyclic sulfonamides: An experimental and computational study

Article abstract of DOI:10.1002/chem.201201359

Regioselectivity in the intramolecular Heck reaction of a series of N-sulfonyl-2,5-dihydro-3-substituted pyrroles was studied. These substrates are unbiased in terms of the formed ring size of the new heterocycle. Results indicate that high levels of regi

Full text of DOI:10.1002/chem.201201359

FULL PAPER
DOI: 10.1002/chem.201201359
Regioselectivity in the Intramolecular Heck Reaction of a Series of Cyclic
Sulfonamides: An Experimental and Computational Study
Kimberly Geoghegan,[a] Paul Evans,*[a] Isabel Rozas,[b] and Ibon Alkorta*[c]
Abstract: Regioselectivity in the intramolecular Heck reaction of a series of N-sul-
fonyl-2,5-dihydro-3-substituted pyrroles was studied. These substrates are unbiased
in terms of the formed ring size of the new heterocycle. Results indicate that high
levels of regioselectivity are observed under a range of conditions, and that there
is an underlying propensity for carbon–carbon bond formation at the most hin-
dered end of the alkene. For two examples (3-Me and 3-tBu), DFT calculations
were performed and indicate that in both cases, the modelled transition state for
carbopalladation is energetically lower for the experimentally preferred isomer.
Keywords: catalysis · cyclization ·
density functional calculations
Heck reaction · palladium · regiose-
·
lectivity
Introduction
type alkene-insertion reaction (step A, Scheme 1) is consid-
ered to be irreversible, the formation of the substituted
alkene is dictated by this process. Evidence has indicated
that the regioselective outcome of the Heck reaction de-
The Heck (Mizoroki–Heck) reaction represents a general
method for the construction of carbon–carbon bonds and is
widely used in both its inter- and intramolecular guises.[1]
Various aspects of the reaction arguably contribute to this
popularity, including functional group tolerance, its use in
the construction of complex molecules, and its ability (par-
ticularly in its intramolecular form) to form bonds and struc-
tural motifs that would be difficult to achieve by alternative
means. An example is its capacity to form what would be
considered sterically demanding tertiary and quaternary
centres. Impressive asymmetric examples of this reaction
have also been reported.[2] Recently, one significant limita-
tion of the reaction, that the halide- or pseudo-halide-substi-
tuted carbon atom needed to be sp2 hybridised or lack a b-
hydrogen atom, has been circumvented; albeit only for a
limited range of substrates at this stage.[3]
Scheme 1. Regioselectivity in the inter- and intramolecular Heck reac-
tion. Ar=aryl; EWG=electron-withdrawing group (e.g., CO2R, Ar);
EDG=electron-donating group (e.g., OR).
The location, or regiochemistry, of the newly formed
carbon–carbon bond is of great interest, in part because this
selectivity sheds light on the mechanisms involved in the
overall alkenylation process. Because the Cossꢀe–Arlman-
pends on several critical features. The electronic features of
the alkene play a significant role and, typically, carbon–
carbon bond formation takes place exclusively or preferen-
tially at the most electron-poor carbon atom (indicated by
an arrow in Scheme 1).[1,4,5] For situations in which alkene
polarisation is less dramatic, for example, in compounds
with an alkyl substituent, steric effects are said to dominate
and the major site of carbon–carbon bond formation is at
the less-substituted or less-sterically encumbered alkenyl
carbon atom. It should be noted that this widely held con-
sideration implies that the aromatic portion of the inter-
mediate complex possesses greater steric burden than the
palladium fragment.[1a]
The nature of the reaction conditions employed are also
important because these dictate the identity, in terms of the
ancillary ligands and charge on palladium, of the entity un-
dergoing insertion. For example, if the reaction is performed
in a way that favours an electron-deficient, cationic, aryl–
palladium(II) species, as opposed to an electron-rich, neu-
[a] K. Geoghegan, Dr. P. Evans
Centre for Synthesis and Chemical Biology
School of Chemistry and Chemical Biology
University College Dublin, Dublin 4 (Ireland)
Fax : (+353)1-7162501
E-mail: paul.evans@ucd.ie
[b] Prof. I. Rozas
School of Chemistry
Trinity College Dublin, Dublin 2 (Ireland)
[c] Prof. I. Alkorta
Instituto de Quimica Medica (IQM-CSIC)
Juan de la Cierva, 3, 28006 Madrid (Spain)
E-mail: ibon@iqm.csic.es
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201201359.
Chem. Eur. J. 2012, 00, 0 – 0
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
1
&
ÞÞ
These are not the final page numbers!

tral, aryl–palladium(II) species (either with an aryl/vinyl tri-
flate, or with certain additives/solvents), then a dramatic re-
versal in the regiochemistry of the resulting product (indi-
cated by arrows in Scheme 1) can be observed.[4,6] Coordina-
tion between the metal and a Lewis basic group on the sub-
strate has also been used to provide regioselectivity during
intermolecular Heck reactions.[7]
of 5 into 6 to introduce the aromatic portion of the octahy-
droindole alkaloid mesembrane (7). Based on the fact that
these intramolecular Heck reactions were performed at ele-
vated temperatures, the high level of regioselectivity that
was detected appears to be indicative of an underlying reac-
tivity preference for this type of intramolecular process.
Although much has been written concerning the mecha-
nism of the Heck reaction, it is fair to say that the precise
mechanism, or mechanisms, by which the Heck reaction
proceeds are still not completely understood.[1,2] Based on
the accepted pathway, which involves a coordinatively unsa-
turated palladium(0) species (that may be stabilised by asso-
ciated ligands) as the entity that undergoes oxidative addi-
tion with 1, the formation of a neutral palladium(II) species
that resembles 8 can be envisaged (Scheme 3). Alkene asso-
In relation to the regiochemical outcome of intramolecu-
lar Heck reactions, the rules discussed above are not com-
pletely applicable because the presence of a tether linking
the reactive entities plays a dominant role. Numerous exam-
ples demonstrate how the size of the newly formed cyclic
fragment and conformational effects associated with the
tether can override any electronic bias the alkene may pos-
sess.[8] An explanation for this behavioural divergence is
linked to the ideal eclipsed orientation between the alkene
and the carbon–palladium(II) bond required for synchro-
nous carbopalladation, which, although there are exceptions,
most favourably occurs following alkene insertion by an exo
mode of action (Scheme 1). Thus, in terms of the intramo-
lecular Heck reaction, one may state that it is the exo mode
and the size of the newly formed ring that is of most impor-
tance when considering the regiochemistry that is observed.
To date, what is lacking in relation to the intramolecular
Heck reaction are examples in which carbon–carbon bond
formation at either end of an unsymmetrically substituted
alkene would lead to identically sized, newly formed rings.[9]
If substrates of this type could be designed, then insights
into the alkene-insertion event could be uncovered. For sev-
eral years,[10] we have been interested in the intramolecular
Heck reaction of cyclic N-sulfonyl compounds and we have
recently uncovered a highly regioselective process.[11] Treat-
ment of 1 with a palladium(0) pre-catalyst gave the adduct
2, resulting from selective bond formation at position a, in
excellent isolated yield (Scheme 2). No regioisomeric mate-
Scheme 3. Plausible reaction pathways for the regioselective carbopalla-
dation of 1.
ciation, probably accompanied by the loss of one 2-electron
ligand, leads to a square-planar, or distorted square-planar,
alkene–palladium(II) complex of the type 9. Subsequent car-
bopalladation may occur via conformer 9a or 9b to generate
product 10 or 11, respectively. Because this carbopalladation
event is irreversible,[1e,g] selectivity at this stage governs the
formation of product 2 following the b-hydride elimination
of 10. Alternatives to this neutral pathway have also been
well-discussed. If a leaving group (X) with weaker ligand
donor properties, such as the triflate group, is employed,
then a cationic reaction intermediate becomes possible; for
example, intermediate 12 is of particular significance when
bidentate asymmetric ligands are used. Jutland and Amator-
e[1g,12] have proposed that under Pd
the palladium(0) source that participates in the initial oxida-
tive addition is [Pd
(PPh3)2OAc]À, which results in intermedi-
ACHTUNGTREN(NGNU OAc)2/PPh3 conditions,
AHCTUNGTRENNUNG
Scheme 2. Regioselectivity in the intramolecular Heck reaction of sulfo-
namides.
ates such as 13 undergoing carbopalladation. In addition to
the homogeneous reaction intermediates, investigations into
the impressive turnover numbers achieved by some pallada-
cycle catalysts (e.g., 32) led to the suggestion that these spe-
cies serve as a reservoir of the real catalyst, nanoparticulate
colloidal palladium(0), and that this ligand-free material
serves as an effective catalyst.[13]
rial attributable to the isomer 3 or its alkene isomer 4,
which would both result from carbon–carbon bond forma-
tion at position b, was observed. This is notable because the
selectivity observed in this case is not governed by the size
of the ring that is formed. Subsequently, this same type of
regioselectivity was successfully employed in the conversion
Thus, current evidence suggests that it is very likely that
more than one distinct reaction pathway exists for the Heck
&
2
&
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Regioselectivity in Intramolecular Heck Reactions of Sulfonamides
FULL PAPER
reaction, depending on the type of reaction conditions em-
ployed and the type of reaction substrate. Based on this, we
felt that selection of the type of intermediate undergoing
carbopalladation during the regiochemistry-establishing
event, that is, 9, 12 or 13 (Scheme 3), might enable us to
challenge the regiochemical preference uncovered in the
previous work.[11]
Herein, we report studies aimed at understanding and
probing this regioselective reaction. Specifically, we aimed
to uncover any effect that the choice of reaction conditions
has on this process, and to establish whether the identities
of the palladium catalyst or pre-catalyst, the ligand (biden-
tate, s-donor capacity, etc.), the solvent, the base or the
halide/pseudo-halide have any role. We also sought to study
any effect that the variation of the alkenyl substituent has
on the reaction, and designed alternative Heck-reaction sub-
strates that offered different polarisation of the alkenyl
group and possessed different steric considerations.
was used directly, no significant difference in the yield or re-
gioselectivity to that previously reported (Scheme 2)[11] was
encountered. The use of the palladacycle 32[15] also showed
no dramatic difference in the regiochemical outcome
(Table 1, entry 2). A feature of catalyst 32 is that the use of
extremely low loadings are often effective at temperatures
of 1308C and above, and that in fact these lower catalyst
loadings result in faster reactions, a finding that can be at-
tributed to diminished rates of active-catalyst decomposition
pathways.[13] However, we found that unsatisfactory conver-
sions were encountered at lower loadings of 32 (1 to 0.01
mol% at 1308C) and the recovery of starting material 1 was
the major outcome. Replacement of K2CO3 with homogene-
ous amine bases proved detrimental to the isolated yields of
2 (Table 1, Entries 3 and 4), but as before, no material re-
sulting from a regioisomeric intramolecular Heck reaction
was detected. As the inclusion of thallium and silver salts[16]
has been advocated for their ability to alter the outcome of
Heck-type processes (possibly due to the removal of the
halide from the coordination sphere of the palladium(II) in-
termediate; that is, cationic intermediates of the type 12),
the use of Ag2CO3 was investigated (Table 1, entry 5). In
this case an inefficient Heck process occurred from which 2
was isolated in low yield (24%), and the remaining mass
balance was attributed to 1 (35%) and traces of N-sulfonyl-
pyrrole 31 (6%), which is presumably present as a result of
redox chemistry caused by the silver species.
Similarly, the inclusion of the ammonium salt nBu4NHSO4
with a dimethylformamide (DMF)/H2O mixture[17] used as
the solvent has been proposed to facilitate the formation of
cationic organopalladium(II) intermediates, and in the past
we have used these reaction conditions to improve the re-
gioselectivity of a different intramolecular Heck reaction.[18]
However, as indicated in Table 1, entry 6, no change in re-
gioselectivity was observed under these reaction conditions
and the adduct 2 was isolated in 83% yield.
Heck-type processes are typically performed in polar sol-
vents and, under some reaction conditions, this may be
linked to the formation and stabilisation of the catalytically
active palladium(0) species.[1,19] Therefore, because we felt
that the identity of the catalytically active Pd species that
participates in the initial oxidative addition and subsequent
carbopalladation event would be key in terms of governing
the regiochemical outcome of the process, we investigated
toluene because of its low dielectric constant (Table 1,
entry 7). In this instance, a moderate yield of 2 was obtained
with a small amount of an impurity that proved chromato-
graphically inseparable with 2. Analysis of proton and
carbon NMR spectra suggested that this impurity was com-
pound 4, which was formed from exocyclic b-hydride elimi-
nation of the reaction intermediate, and that the ratio be-
tween 2 and 4 was 85:15. Compound 4 was separated from 2
by recrystallisation. In the next experiment (Table 1,
entry 8) the chelating ligand 1,3-bis(diphenylphosphino)pro-
pane (dppp) was investigated. In this case, 88% of 2 was iso-
lated and no isomeric material could be detected.
Results and Discussion
Commercially available 2-bromobenzene sulfonyl chloride
(14), 2-chlorobenzene sulfonyl chloride (15), 2-nitrobenzene
sulfonyl chloride (16) and then 2-bromo-3,4-dimethoxyben-
zene sulfonyl chloride (17; prepared from 2-bromo-3,4-dime-
thoxybenzene)[10a] were converted into their respective allyl-
amine adducts 18–21. Subsequent alkylation with 3-chloro-
2-methylpropene, followed by ring-closing metathesis
(RCM) in the presence of 29,[14] efficiently generated the N-
sulfonyldihydropyrrole derivatives 1 and 26–28 (Scheme 4).
The nitro derivative 27 was prepared to provide access to
the corresponding triflate (see Scheme 5). The ring-closing
metathesis took place at room temperature, but optimal
yields were obtained when the solution was heated at reflux.
With the materials 1, 26 and 28 in hand, a systematic
survey of reaction parameters was undertaken (Table 1). As
indicated in Table 1, entry 1, when a palladium(0) source
Scheme 4. Preparation of dihydropyrroles 1 and 26–28. Conditions: i) al-
lylamine, Et3N, CH2Cl2, 08C to RT; ii) a) NaH, DMF, 08C; b) 3-chloro-2-
methylpropene, 08C to RT; iii) Hoveyda–Grubbs 2nd-generation catalyst
29 (1–2 mol%), CH2Cl2, 408C.
Chem. Eur. J. 2012, 00, 0 – 0
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
3
&
ÞÞ
These are not the final page numbers!

P. Evans, I. Alkorta et al.
Table 1. Intramolecular Heck reactions of 1, 26 and 28.
Substrate
Conditions
(PPh3)4] (10 mol%), K2CO3, DMF, 1108C
Products[a]
(yield [%]; ratio)
AHCTUNGTRENNUNG
1
2
3
4
5
6
7
8
9
10
11
12
13
14
1
1
1
1
1
1
1
1
1
1
1
26
26
28
[Pd
Herrmann–Beller cat. 32 (10 mol%), K2CO3, DMF, 1108C
Pd(OAc)2 (10 mol%), PPh3 (20 mol%), Et3N, DMF, 1108C
Pd(OAc)2 (10 mol%), PPh3 (20 mol%), proton sponge 33, DMF, 1108C
Pd(OAc)2 (10 mol%), PPh3 (20 mol%), Ag2CO3, DMF, 1108C
Pd(OAc)2 (10 mol%), PPh3 (20 mol%), K2CO3, nBu4NHSO4, DMF/H2O (9:1), 1108C
Pd(OAc)2 (10 mol%), PPh3 (20 mol%), K2CO3, PhMe, 1108C
Pd(OAc)2 (10 mol%), dppp (20 mol%), K2CO3, DMF, 1108C
Pd(OAc)2 (10 mol%), K2CO3, DMF, 1108C
Pd(OAc)2 (10 mol%), R-BINAP 34 (20 mol%), K2CO3, DMF, 1108C
[Pd
(dba)2] (10 mol%), R-BINAP 34 (23 mol%), PMP 35, DMF, 1108C[c]
Pd(OAc)2 (10 mol%), PPh3 (20 mol%), K2CO3, DMF, 1108C
Pd(OAc)2 (10 mol%), tBuBrettPhos 36 (31 mol%), K2CO3, DMF, 1108C
Pd(OAc)2 (10 mol%), PPh3 (20 mol%), K2CO3, DMF, 1108C
G
2 (72)
2 (69)
2 (69)
2 (34)
R
E
T
2 (24), 1 (35), 31 (6)
2 (83)
ACHTUNGTRENNUNG
E
2/4 (59; 85:15)[b]
2 (88)
ACHTUNGTRENNUNG
T
2/4 (72; 88:12)[b]
2/4 (91; 88:12),[b] 0% ee
2 (41), 16% ee
2 (18), 26 (66)
2 (77)
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
N
30 (53)
1
[a] Isolated yields after purification by flash column chromatography. [b] Ratio determined by H NMR spectroscopy. [c] Reaction was conducted follow-
ing three freeze–pump–thaw cycles. dppp=1,3-bis(diphenylphosphino)propane.
This survey of reaction conditions suggested that either
there is an overriding propensity for compound 1 to react in
a process that forms the most substituted carbon–carbon
bond (i.e., via a transition state resembling 9a rather than
9b), or that irrespective of the different conditions, the same
type of palladium species was responsible for the formation
of the adduct 2 in each case. In relation to this latter possi-
bility, it has been argued[1g,13] that in certain situations the
actual catalyst for many Heck reactions could be nanoparti-
culate palladium(0) clusters as opposed to a discrete, homo-
geneous phosphine-bound monomeric palladium complex.
Possible support for this situation was uncovered when
ligand-free conditions were studied (Table 1, entry 9). In this
instance, a mixture of 2 and 4 was isolated in 72% combined
yield in a ratio of 88:12.
To further investigate the reaction mechanism, an asym-
metric phosphine ligand was used in the reaction because
any product enantioselectivity would be indicative of a step
in which the ligand was directly involved to induce asymme-
try. Based on literature precedent, two alternative reactions
were performed in which BINAP (34; 2,2’-bis(diphenylphos-
phino)-1,1’-binaphthyl) was used as a source of potential
asymmetry. As indicated in Table 1, entry 10, when 34 was
used in place of PPh3, the efficient formation of 2 occurred
but the resultant product was racemic. However, the use of
conditions developed by Overman et al.[20] provided 2 in
moderate yield (41%) and approximately 16% enantiomer-
ic excess (Table 1, entry 11). No regioisomeric material (3 or
4) was isolated. This result suggests that an arylpalladiu-
m(II) intermediate, in which the phosphine is associated, is
involved, at least in part, in a process that converts 1 into 2.
Table 1, Entries 12 and 13 concern the use of the chloride
26, with a view to investigating how altering the rate of oxi-
dative addition might impact on the overall process. Under
identical conditions to those that furnished 2 from bromide
1 in 90% yield (Scheme 2), the use of chloride 26 gave only
an 18% yield of 2 (Table 1, entry 12). However, use of the
electron-rich phosphine 36,[21] which is known to enhance
oxidative addition, gave 2 in 77% yield (Table 1, entry 13).
These two examples provide evidence that 2, with its quater-
nary stereogenic centre, was formed following homogeneous
catalysis with a ligand-bound palladium species. Next the di-
methoxy-substituted aryl bromide 28 was studied, which was
found to give 30 in a moderate yield of 53% (Table 1,
entry 14). In the past,[10a,d] we observed lower yields when
this more electron-rich bromide was employed in this type
of reaction.
The reactions involving halides under the range of condi-
tions discussed above all led exclusively, or in high selectivi-
ty, to the quaternary regioisomer, even in the presence of
conditions that are thought to involve a cationic palladi-
um(II) species of the type 12 (Table 1, Entries 6 and 7).
Therefore, we next considered whether we could specifically
perform the same type of intramolecular Heck reaction with
&
4
&
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Regioselectivity in Intramolecular Heck Reactions of Sulfonamides
FULL PAPER
substrates in which the entity released following palladi-
um(0) oxidative addition would not coordinate to the inter-
mediate arylpalladium(II) species.
By using the nitro-substituted sulfonamide 27, the synthe-
sis of diazonium salt 38 and triflate 40 was addressed
(Scheme 5).[22] Chemoselective reduction of 27 with iron/
step conversion of 24. Although heating 42 in water also
gave phenol 43 in low yields, the 20–25% observed was
higher than for the corresponding conversion of 38 into 39
and was reproducible and thus enabled sufficient material to
be accumulated. Phenol 43 was converted into triflate 44
prior to ring-closing metathesis, which then smoothly gave
40 in good yield over two steps (Scheme 5).
The treatment of 40 with our favoured intramolecular
Heck-reaction conditions (PdACTHUNRGTNEGN(U OAc)2, PPh3, K2CO3, DMF,
1108C) provided 2; however, the main product proved to be
phenol 39 (Table 2, entry 1). We reasoned that 39 was
formed due to the presence of water under the basic reac-
tion conditions, a process presumably facilitated by the
ortho electron-withdrawing sulfonyl moiety. Therefore, a lit-
erature method for the intramolecular Heck reaction of an
enoltriflate was investigated.[27] Although this solved the
issue of triflate hydrolysis, conversion proved to be low and
2 was only isolated in 14% yield (Table 2, entry 2). In both
cases, the formation of 3 or 4 was not detected. In contrast,
the use of the palladium(0)–BINAP conditions developed
by Overman et al.[20,28] gave 2 in excellent yield, albeit with
only low enantiomeric excess (Table 2, entry 3). Interesting-
ly, the outcome of this process closely mirrored, in terms of
both yield and enantiomeric excess, the behaviour of bro-
Scheme 5. Preparation of diazonium salt 38 and triflate 40. Conditions:
i) Fe0, AcOH/EtOH/H2O (2:2:1), RT; ii) NaNO2, HBF4, H2O, 0 8C;
iii) H2O, 1008C; iv) Tf2O, pyridine, CH2Cl2, 08C; v) Hoveyda–Grubbs
2nd-generation catalyst 29 (2 mol%), CH2Cl2, 408C. Tf=triflyl.
acetic acid[23]gave the corresponding aniline 37 in high yield,
and subsequent diazonium-ion formation was achieved
under typical conditions.[24,25] This enabled the isolation of
38 in moderate to good yield in two steps following recrys-
tallisation of the initially formed precipitate from acetone/
diethyl ether. Heating 38 in water (at either pH 7 or 4)[25]
generated the desired phenol 39; however, yields were rou-
tinely poor. N-Sulfonylpyrrole 45 was always formed as a
side product, as was additional coloured material that could
not be characterised. The use of CuII salts[26] led to the ex-
clusive formation of 45.
mide
1 under identical reaction conditions (Table 1,
entry 11).
Based on a recent report that describes the first intramo-
lecular Matsuda–Heck reaction,[24] diazonium ion 38 was
considered. Several often-reported protocols for the reaction
of diazonium salts with alkenes were investigated
(Table 2).[16] The use of [Pd
ACTHNUTRGENUN(G dba)2] to catalyse a reaction
mixture that included pyrrole 45 (49%) and dihydropyrrole
46 (18%) resulted in the isolation of a low yield of 2
(18%); Table 2, entry 4). In the presence of a phosphine,
the formation of a slightly higher yield of 2 (27%) was ob-
served; however, this was again accompanied by the forma-
tion of 45 and 46 (Table 2, entry 5). The use of NaOAc in di-
chloromethane[29] reduced the quantity of the pyrrole by-
product obtained, but overall conversion remained poor
Because competitive oxidation of the dihydropyrrole was
clearly an issue, we aimed to perform the diazonium-ion–
phenol conversion on the acyclic diazonium salt 42. Com-
pound 42 was readily prepared in good yield from a two-
Table 2. Applications of triflate 40 and diazonium salt 38 in the intramolecular Heck reaction.
Substrate
Conditions
Products[a]
(yield [%]; ratio)
AHCTUNGTRENNUNG
1
2
3
4
5
6
7
40
40
40
38
38
38
38
Pd
[Pd
[Pd(dba)2] (5 mol%), R-BINAP 34 (11 mol%), PMP 35, DMF, 1108C
[Pd(dba)2] (10 mol%), K2CO3, MeCN, RT
[Pd(PPh3)4] (10 mol%), K2CO3, MeCN, RT
[Pd(dba)2] (4 mol%), NaOAc, CH2Cl2, RT
Pd(OAc)2 (10 mol%), MeOH, 508C
N
2 (15), 39 (55)
2 (14), 40 (50)
2 (93), 17% ee
E
R
E
2 (18), 45 (49), 46 (18)
2 (27), 45 (27), 46 (3)
2 (83)
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
2/4 (26; 20:80),[b] 45 (37)
[a] Isolated yields after purification by flash column chromatography. [b] A mixture of 2 and 4 was isolated and the ratio was determined by 1H NMR
spectroscopy. dba=dibenzylideneacetone.
Chem. Eur. J. 2012, 00, 0 – 0
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
5
&
ÞÞ
These are not the final page numbers!

P. Evans, I. Alkorta et al.
(Table 2, entry 6). These reactions were conducted at room
temperature, demonstrating the facile oxidative addition of
diazonium salts in comparison with their halide/pseudo-
halide counterparts. In all cases (Table 2, Entries 4–6), the
only intramolecular Heck reaction product, 2, contained the
quaternary carbon centre. However, by using a ligand-free,
RCM with catalyst 29 provided the corresponding dihydro-
pyrrole derivatives 57–59 in reasonable to good yields, even
for the sterically hindered tert-butyl-substituted example 53
and the electron-poor and Lewis basic a,b-unsaturated ester
54. The methoxy-substituted analogues 60 and 61 were effi-
ciently prepared in a similar fashion.
base-free
colloidal
palladium
procedure
(Table 2,
Each Heck-precursor compound was then submitted to
entry 7),[24,30] an interesting observation was made: although
conversion was poor and the generation of pyrrole 45 was
again a complication, a 26% yield of the intramo-
lecular Heck reaction products was isolated follow-
the reaction conditions involving Pd
ACHTUNGTREN(UNNG OAc)2/PPh3 with
K2CO3 in DMF (Table 3). In the case of styrene 57, com-
Table 3. Study of alternative alkenyl substituents in the intramolecular Heck reactio-
n.[a]
ing column chromatography. Inspection of the
proton NMR spectrum indicated that for the first
time, formation of the regioisomeric Heck adduct 4,
which does not contain a quaternary centre, was
favoured (4/2, 80:20; Table 2, entry 7). Recrystalli-
sation of the mixture of regioisomers from cyclo-
hexane provided predominantly 4, which enabled
unambiguous assignment and confirmation of the
minor byproduct encountered in Table 1, Entries 7
and 9.
In summary, the use of either molecular nitrogen
or triflate as a leaving group in the Heck process
proved less efficient than the corresponding bro-
mide due to poor conversion and the formation of
side products. Results indicate that from reactions
in which the carbopalladation is likely to occur via
Substrate
R
R’
Products[b]
(yield [%])
AHCTUNGTRENNUNG
1
2
3
4
5
57
58
59
60
61
H
Ph
62 (45)
H
H
OMe
OMe
tBu
CO2Me
Ph
63 (41), 68 (12), 72 (34)
73 (42)
65 (54)
tBu
66 (39), 71 (18)
[a] Conditions: PdACHTNUGTRNE(UNG OAc)2 (10 mol%), PPh3 (20 mol%), K2CO3, DMF, 1108C. [b] Iso-
a cationic organopalladium(II) intermediate species lated yields after purification by flash column chromatography.
(e.g., 12), regiosiomer 2 was routinely observed.
The only exception to this was found in when a pre-
formed, colloidal palladium(0) mixture preferentially
formed the regioisomer 4 (Table 2, entry 7).
pound 62 was isolated in 45% yield and its structure was de-
termined by X-ray crystallography (Table 3, entry 1).[32]
Only traces of starting material 57 were isolated following
this reaction, and no material consistent with a regioisomer-
ic Heck adduct 67 was detected. Next, the reaction of 58
was investigated, and in this case, isolable and characterisa-
ble amounts of the regiosiomeric adduct 68 were obtained
alongside the major product 63 (Table 3, entry 2). After sep-
aration of the isomers by column chromatography, 63 was
isolated in 41% yield and 68 was isolated in 12% yield.
Proton NMR spectroscopy of the crude reaction mixture in-
dicated that 63 and 68 were formed in a ratio of 4:1, and
their structures were confirmed by X-ray crystallography
(Figure 1).[32] In this process, the remaining mass balance
was provided by pyrrole 72 (34%).
At this point, after screening a range of conditions and
different aryl leaving groups there was a clear preference
for the formation of the more sterically congested adduct 2
over 3 (or 4). Therefore, it was felt that an investigation of
alternative alkenyl substituents might be instructive. 3-
Phenyl-, 3-tert-butyl- and 3-methyl-ester-substituted unsatu-
rated bromides 49–51 were targeted based on their different
electronic and steric properties. As indicated in Scheme 6,
compounds 49 and 50 were prepared by allylic bromination
of 47 and 48 using N-bromosuccinimide (NBS),[31] whereas
51 is commercially available. Alkylation of these compounds
with 18 gave the RCM precursors 52–54, and subsequent
As indicated in Table 3, entry 3, for the case of methyl
ester 59, no products resulting from an intramolecular Heck
reaction (e.g., 64) were detected; instead, pyrrole 73 was
formed as the sole isolable product (42%). It seems reason-
able to speculate that this product resulted from a base-
mediated sulfinate-elimination process, which was then fol-
lowed by aromatisation of the product.[10b] Based on the suc-
cess of the styrene and tert-butyl substituents, the corre-
sponding dimethoxy-substituted Heck precursors 60 and 61
were considered. As indicated in Table 3, Entries 4 and 5,
the corresponding quaternary regioisomer was either the
sole (65; 54%) or the major product (66; 39%) in each
Scheme 6. Ring-closing metathesis for the synthesis of substituted sulpho-
namides 57–61. Conditions: i) NBS, CHCl3, sealed tube, 658C; ii) 18 (or
21), NaH, DMF, 08C to RT; iii) Hoveyda–Grubbs 2nd generation catalyst
29 (2 mol%), CH2Cl2, 408C. NBS = N-Bromosuccinimide.
&
6
&
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Regioselectivity in Intramolecular Heck Reactions of Sulfonamides
FULL PAPER
Figure 1. X-ray crystallographic structures of 63 and 68 (ellipsoids are
shown at the 50% probability level).
case. Although successful Heck reactions were only ob-
served for the phenyl- and tert-butyl-substituted systems, the
findings clearly demonstrate that in both instances bond for-
mation at the more congested carbon atom is favoured, and
leads to the quaternary benzo-fused bicyclic product.
Mechanistic details of Heck reactions have been success-
fully investigated computationally by using density function-
al theory (DFT) methods.[5,33] Based on these precedents, we
felt that a similar study might provide a basis for the selec-
tive carbon–carbon bond formation that was observed ex-
perimentally in the processes described above. By following
a stepwise procedure, from simple computational methods
to the highest level that can be readily considered, key inter-
mediates in the Heck reactions of 1 and 58 (R=Me and
tBu) were studied. Results were obtained by using the B97D
functional,[34] which includes dispersion forces, with the 6-
31G* basis set[35] for all the atoms except palladium and bro-
mine, results for which were calculated by using the pseudo-
potential LANL2DZ(d) and LANL2DZ functionals, respec-
tively.[36] In addition, a solvent continuum model (PCM)
using the DMF parameter was used to model the reaction
solvent.[37] All of the energetic values reported herein corre-
spond to the free energy plus the solvation energy. This
model enabled the characterisation of the possible com-
plexes involved in the regiochemical-establishing sequence
(Scheme 7).
Firstly, the final possible reaction products (without palla-
dium) were compared for both Me and tBu substituents. As
expected in both cases, the tri-substituted alkenyl products 3
and 68, formed by following the experimentally observed,
minor pathway, were more stable by 19.0 and 48.2 kJmolÀ1
for R=Me and tBu, respectively. Next, the possible inter-
mediates 10 and 11 and the transition states leading to their
formation (TSmaj and TSmin, respectively) were considered
for both the major pathway (based on experimental results)
and the corresponding minor pathway. Both conformations
of the square-planar alkene complex, in which R=Me or
tBu, were investigated. As shown, the relative energies for
conformer 9a versus 9b depend on the identity of the R
substituent.
Scheme 7. DFT calculations performed for the regiochemistry-establish-
ing sequence in which R=Me or tBu.
state was found to be associated with the major pathway.
However, the difference in relative energies between the
major and minor transition states, 13.1 and 19.0 kJmolÀ1 for
R=Me and tBu, respectively, was less dramatic than antici-
pated on the basis of the experimental results obtained (for
R=Me in particular). As was the case for the initial alkene
complexes 9a and 9b, the relative energies of the carbopal-
ladated intermediates 10 and 11 were dependant on the
identity of R.
From these results, it can be concluded that the key step
in the formation of the major-pathway product depends
mostly on the stability of the TS, despite the stability of the
final products (which favours the formation of products
through the minor pathway) and the stability of the inter-
mediate complexes (which are substituent dependent).
In summary, this DFT study provides corroborative evi-
dence for the experimental study, suggesting that the regio-
chemical outcome is kinetically favoured and that product
selection occurs because a lower-energy transition state is
associated with the formation of the quaternary centre.
However, based on the relatively low difference in energies
between the competing transition states, it would appear
that for a reaction performed at elevated temperatures,
these DFT studies can only partly address the origin of the
selectivity.
Conclusion
The preference for the formation of a quaternary centre
over a tertiary centre following an intramolecular Heck re-
action was studied for a series of differently substituted N-
sulfonyldihydropyrroles. Results indicated that for the
methyl-substituted substrate 1, irrespective of a range of
typical experimental protocols, a high selectivity for the for-
The corresponding transition states in the pathway to the
carbopalladated intermediates 10 and 11 were also comput-
ed and the associated energies were considered. For both
substituents (R=Me and tBu), the most stable transition
Chem. Eur. J. 2012, 00, 0 – 0
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
7
&
ÞÞ
These are not the final page numbers!

P. Evans, I. Alkorta et al.
1165 cmÀ1; HRMS (EI): m/z calcd for C14H17NO2S+: 264.1058 [M+];
found: 264.1058.
mation of the quaternary regioisomeric adduct 2 was ob-
served (see Table 1). Attempts to employ triflate as a leav-
ing group (40) in the Heck reaction and thereby invoke a
cationic reaction pathway did not alter the regiochemical
outcome observed in the reactions of bromide 1 or chloride
26. The use of the diazonium salt 38 proved to not be
straightforward based on the propensity of 38 to form by-
products that result from dihydropyrrole oxidation and
proton–nitrogen exchange. However, by using an in situ
formed, colloidal palladium(0) catalyst, the alternate regio-
chemistry was observed preferentially, albeit in low overall
yield. More work is required to understand and build upon
this initial finding. Alternative substituents were investigat-
ed and it was found that the phenyl-substituted compounds
57 and 60 gave only the product of quaternary carbon–
carbon bond formation. In the case of the tert-butyl-substi-
tuted precursors 58 and 61, mixtures of products were de-
tected. Separation of these mixtures led to the isolation and
characterisation of products resulting from both regiochemi-
cal outcomes, in which the major product (ca. 3.5:1) was the
quaternary isomer. The presence of different substituents
(R=Me, Ph, tBu), which have different steric and electronic
effects, resulted in a preference for the formation of the
more hindered, quaternary carbon–carbon bond.
DFT studies were performed to probe the experimentally
observed results, and were calculated by using the classic
neutral Heck reaction pathway for the methyl and tert-
butyl-substituted compounds. Calculations were performed
for the chemical species likely to be involved in the regio-
chemistry-establishing event. Although differences were
noted for the methyl substituent versus the bulkier and
more electron-donating tert-butyl substituent, the energy of
the transition state leading to carbopalladation is lower in
both cases for the experimentally preferred intermediate
that leads to formation of the quaternary carbon centre.
11-tert-Butyl-8-thia-9-azatricyclo[7.2.1.02,7]dodeca-2(7),3,5,10-tetraene-8,8-
dioxide (68): Rf =0.3 (c-Hex/EtOAc, 3:1); m.p. 72–748C; 1H NMR
(400 MHz, CDCl3): d=7.76–7.71 (m, 1H), 7.48–7.41 (m, 1H), 7.39–7.33
(m, 1H), 7.20 (d, J=8.0 Hz, 1H), 6.10 (s, 1H), 4.42 (d, J=12.0 Hz, 1H),
4.16–4.07 (m, 1H), 3.28 (d, J=4.0 Hz, 1H), 1.07 ppm (s, 9H); 13C NMR
(100 MHz, CDCl3): d=160.6, 140.9, 135.6, 131.2, 129.7, 127.6, 127.1,
125.6, 64.8, 43.6, 33.2, 29.1 ppm; IR (NaCl, in CH2Cl2): n˜max =3086, 2967,
1642, 1336, 1156 cmÀ1; HRMS (EI): m/z calcd for C14H17NO2S+: 263.0980
[M+]; found: 263.0988.
3-tert-Butyl-N-benzenesulfonyl pyrrole (72): light-brown waxy solid; Rf =
0.5 (c-Hex/EtOAc, 3:1); 1H NMR (400 MHz, CDCl3): d=7.83 (d, J=
7.5 Hz, 2H), 7.59 (t, J=7.5 Hz, 1H), 7.50 (t, J=8.0 Hz, 2H), 7.07 (s, 1H),
6.89 (s, 1H), 6.25 (s, 1H), 1.18 ppm (s, 9H); 13C NMR (100 MHz,
CDCl3): d=140.8, 139.5, 133.7, 129.4, 126.8, 121.1, 115.2, 112.9, 31.0,
27.0 ppm; IR (NaCl, in CH2Cl2): n˜max =3139, 3066, 1584, 1368, 1161 cmÀ1
;
HRMS (EI): m/z calcd for C14H17NO2S+: 263.0980 [M+]; found:
263.0976.
Acknowledgements
The authors would like to thank University College Dublin for financial
support and the National University of Ireland for an NUI Travelling
Studentship (KG). We also thank Dr. Helge Mꢃller-Bunz for X-ray crys-
tallography and Mr. Johannes E. M. N. Klein for constructive comments.
Membership (PE) of COST CM0804 is acknowledged. Computational
work was carried out with financial support from the Ministerio de Edu-
caciꢄn y Ciencia (Project No. CTQ2009–13129-C02–02) and Comunidad
Autꢄnoma de Madrid (Project MADRISOLAR2, ref. S2009/PPQ-1533).
[1] For general reviews concerning the Heck reaction, see: a) R. F.
Heck, Org. React. 1982, 27, 345–390; b) S. E. Gibson, R. Middleton,
Contemp. Org. Synth. 1996, 3, 447–471; c) J. T. Link, Org. React.
2002, 60, 157–534; d) G. Zeni, R. C. Larock, Chem. Rev. 2006, 106,
4644–4680; e) J. P. Knowles, A. Whiting, Org. Biomol. Chem. 2007,
5, 31–44; f) N. T. S. Phan, M. Van Der Sluys, C. W. Jones, Adv.
Synth. Catal. 2006, 348, 609–679; g) for a monograph, see: The Miz-
oroki–Heck Reaction (Ed.: M. Oestreich), Wiley, New York, 2009.
[2] For reviews in the area of asymmetric Heck reactions, see: a) D.
Mc Cartney, P. J. Guiry, Chem. Soc. Rev. 2011, 40, 5122–5150;
b) A. B. Dounay, L. E. Overman, Chem. Rev. 2003, 103, 2945–2963.
[3] a) L. Firmansjah, G. C. Fu, J. Am. Chem. Soc. 2007, 129, 11340–
11341; b) K. S. Bloome, R. L. McMahen, E. J. Alexanian, J. Am.
Chem. Soc. 2011, 133, 20146–20148.
Experimental Section
Representative procedure for the Heck reaction: A degassed solution of
compound 58 (170 mg, 0.49 mmol, 1 equiv) in DMF (4 mL) was treated
with PdACHTUNGTRENNUNG(OAc)2 (11 mg, 0.049 mmol, 10 mol%), PPh3 (26 mg, 0.098 mmol,
[4] W. Cabri, I. Candiani, Acc. Chem. Res. 1995, 28, 2–7.
[5] a) H. von Schenck, B. Akermark, M. Svensson, J. Am. Chem. Soc.
2003, 125, 3503–3508; b) R. J. Deeth, A. Smith, J. M. Brown, J. Am.
Chem. Soc. 2004, 126, 7144–7151.
[6] a) J. Ruan, J. Xiao, Acc. Chem. Res. 2011, 44, 614–626; b) P.
Wucher, L. Caporaso, P. Roesle, F. Ragone, L. Cavallo, S. Mecking,
I. Gçttker-Schnetmann, Proc. Natl. Acad. Sci. USA 2011, 108, 8955–
8959; c) G. K. Datta, H. von Schenk, A. Hallberg, M. Larhad, J.
Org. Chem. 2006, 71, 3896–3903.
20 mol%) and K2CO3 (135 mg, 0.98 mmol, 2 equiv) under a N2 atmos-
phere, and the solution was heated at 1108C for 18 h. The reaction mix-
ture was cooled and EtOAc (10 mL) and H2O (10 mL) were added. The
resultant aqueous layer was separated and further extracted with EtOAc
(2ꢂ10 mL), and the combined organic extracts were dried over MgSO4.
Filtration was followed by solvent removal under reduced pressure to
give the crude product. Purification by flash column chromatography (c-
Hex/EtOAc, 15:1) provided product 63 (51 mg, 41%) as a colourless
solid, as well as co-products 68 (16 mg, 12%) and pyrrole 72 (45 mg,
34%).
1-tert-Butyl-8-thia-9-azatricyclo[7.2.1.02,7]dodeca-2(7),3,5,10-tetraene-8,8-
dioxide (63): Rf =0.4 (c-Hex/EtOAc, 3:1); m.p. 164–1688C; 1H NMR
(400 MHz, CDCl3): d=7.75 (dd, J=8.0, 1.5 Hz, 1H), 7.68 (dd, J=8.0,
1.0 Hz, 1H), 7.45 (td, J=8.0, 1.0 Hz, 1H), 7.38 (td, J=8.0, 1.5 Hz, 1H),
6.43 (d, J=4.0 Hz, 1H), 6.36 (d, J=4.0 Hz, 1H), 4.36 (d, J=12.0 Hz,
1H), 4.21 (d, J=12.0 Hz, 1H), 1.28 ppm (s, 9H); 13C NMR (100 MHz,
CDCl3): d=142.1, 137.3, 134.9, 133.5, 130.8, 129.4, 128.6, 125.9, 63.6, 56.7,
32.8, 28.9 ppm; IR (NaCl, in CH2Cl2): n˜max = 3112, 2965, 2935, 1330,
[7] For an example, see: G. K. Datta, M. Larhed, Org. Biomol. Chem.
2008, 6, 674–676.
[8] For a selection of intramolecular examples in which the bias of elec-
tron-poor alkenes has been overcome, see: a) O. Kwon, D. S. Su, D.
Meng, W. Deng, D. C. DꢅAmico, S. J. Danishefsky, Angew. Chem.
1998, 110, 1978–1981; Angew. Chem. Int. Ed. 1998, 37, 1877–1880;
b) B. M. Trost, H. C. Tsui, F. D. Toste, J. Am. Chem. Soc. 2000, 122,
3534–3535; c) P. Diaz, F. Gendre, L. Stella, B. Charpentier, Tetrahe-
dron 1998, 54, 4579–4590; d) J. W. Dankwardt, L. A. Flippin, J. Org.
Chem. 1995, 60, 2312–2313; e) A. Ashimori, B. Bachand, L. E.
Overman, D. J. Poon, J. Am. Chem. Soc. 1998, 120, 6477–6487.
&
8
&
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Regioselectivity in Intramolecular Heck Reactions of Sulfonamides
FULL PAPER
[9] D. J. Kucera, S. J. OꢅConnor, L. E. Overman, J. Org. Chem. 1993, 58,
5304–5306.
[24] F. A. Siqueira, J. G. Taylor, C. R. D. Correia, Tetrahedron Lett. 2010,
51, 2102–2105.
[25] P. Hanson, J. R. Jones, A. B. Taylor, P. H. Walton, A. W. Timms, J.
Chem. Soc. Perkin Trans. 2 2002, 1135–1150.
[26] T. Cohen, A. G. Dietz, Jr., J. R. Miser, J. Org. Chem. 1977, 42,
2053–2058.
[27] B. F. Sun, C. L. Wang, R. Ding, J. Y. Xu, G. Q. Lin, Tetrahedron Lett.
2011, 52, 2155–2158.
[28] T. Takemoto, M. Sodeoka, H. Sasai, M. Shibasaki, J. Am. Chem.
Soc. 1993, 115, 8477–8478.
[29] K. Kikukawa, K. Nagira, F. Wada, T. Matsuda, Tetrahedron 1981, 37,
31–36.
[30] a) G. Mehta, S. Sengupta, Tetrahedron Lett. 1996, 37, 8625–8626;
b) S. Cacchi, G. Fabrizi, A. Goggiamani, A. Sferrazza, Synlett 2009,
973–977.
[10] a) P. Evans, T. McCabe, B. S. Morgan, S. Reau, Org. Lett. 2005, 7,
43–46; b) P. Evans, J. Org. Chem. 2007, 72, 1830–1833; c) S. Kelleh-
er, P. Y. Quesne, P. Evans, Beilstein J. Org. Chem. 2009, 5, No. 69;
d) K. Geoghegan, S. Kelleher, P. Evans, J. Org. Chem. 2011, 76,
2187–2194.
[11] J. E. M. N. Klein, K. Geoghegan, N. Mꢀral, P. Evans, Chem.
Commun. 2010, 46, 937–939.
[12] a) C. Amatore, A. Jutand, Acc. Chem. Res. 2000, 33, 314–321; b) A.
Jutand, Pure Appl. Chem. 2004, 76, 565–576; c) for a recent exam-
ple, see: F. Proutiere, F. Schoenebeck, Angew. Chem. 2011, 123,
8342–8345; Angew. Chem. Int. Ed. 2011, 50, 8192–8195.
[13] a) M. T. Reetz, J. G. de Vries, Chem. Commun. 2004, 1559–1563;
b) J. G. de Vries, Dalton Trans. 2006, 421–429.
[14] For an example, see: T. Vorfalt, K. J. Wannowius, H. Plenio, Angew.
Chem. 2010, 122, 5665–5668; Angew. Chem. Int. Ed. 2010, 49, 5533–
5536, and references therein.
[15] a) M. Beller, H. Fischer, W. A. Herrmann, K. ꢆfele, C. Brossmer,
Angew. Chem. 1995, 107, 1992–1993; Angew. Chem. Int. Ed. Engl.
1995, 34, 1848–1849; b) W. A. Herrmann, C. Brossmer, K. ꢆfele,
C. P. Reisinger, T. Priermeier, M. Beller, H. Fischer, Angew. Chem.
1995, 107, 1989–1992; Angew. Chem. Int. Ed. Engl. 1995, 34, 1844–
1848.
[16] A. Putey, L. Joucla, L. Picot, T. Besson, B. Joseph, Tetrahedron
2007, 63, 867–879.
[17] a) S. Lemaire-Audoire, M. Savignac, C. Dupuis, J. P. GenÞt, Tetrahe-
dron Lett. 1996, 37, 2003–2006; b) K. S. A. Vallin, M. Larhed, A.
Hallberg, J. Org. Chem. 2001, 66, 4340–4343.
[18] J. E. M. N. Klein, H. Mꢃller-Bunz, Y. Ortin, P. Evans, Tetrahedron
Lett. 2008, 49, 7187–7190.
[19] I. P. Beletskaya, A. V. Cheprakov, Chem. Rev. 2000, 100, 3009–3066.
[20] A. Ashimori, B. Bachand, M. A. Calter, S. P. Govek, L. E. Overman,
D. J. Poon, J. Am. Chem. Soc. 1998, 120, 6488–6499.
[31] T. Ohmura, K. Masuda, I. Takase, M. Suginome, J. Am. Chem. Soc.
2009, 131, 16624–16625.
[32] CCDC-876850 (62), CCDC-876851 (63) and CCDC-876852 (68) con-
tain the supplementary crystallographic data for this paper. These
data can be obtained free of charge from The Cambridge Crystallo-
graphic Data Centre via www.ccdc.cam.ac.uk/data_request/cif. Com-
pound 62: C16H13NO2S; Mr =283.33; monoclinic; space group P21;
unit cell parameters: a=8.0994(2), b=18.8070(3), c=8.8124(2) ꢇ;
a= 908; V=1336.41(5) ꢇ3; T=100(2) K; Z=4; 26989 reflections
measured, 5602 unique reflections (Rint =0.0336); R1 = 0.0270 and
wR2 =0.0726 (all data). Compound 63: C14H17NO2S; Mr =263.35;
monoclinic, P21/n; unit cell parameters: a=11.6193(2), b=9.0842(1),
c=12.0434(2) ꢇ; a= 908; V=1231.04(3) ꢇ3; T=100(2) K; Z=4;
29893 reflections collected, 4350 unique (Rint =0.0295); R1 = 0.0466
and wR2 =0.1057 (all data). Compound 68: C14H17NO2S; Mr =
263.35; monoclinic, P21/n; unit cell parameters: a=10.8789(1), b=
7.5258(1), c=16.5713(2) ꢇ; a= 908; V=1339.61(3) ꢇ3; T=
100(2) K; Z=4; 41893 reflections collected, 4756 unique (Rint
0.0266); R1 = 0.0332 and wR2 =0.0914 (all data).
=
[33] C. Bꢈcktorp, P. O. Norrby, Dalton Trans. 2011, 40, 11308–11314, and
references therein.
[21] L. M. Chapman, B. Adams, L. T. Kilman, A. Makriyannis, Tetrahe-
dron Lett. 2010, 51, 1517–1522.
[34] S. Grimme, J. Comput. Chem. 2006, 27, 1787–1799.
[35] a) P. C. Hariharan, J. A. Pople, Theor. Chim. Acta 1973, 28, 213–
222; b) P. C. Hariharan, J. A. Pople, Mol. Phys. 1974, 27, 209–214.
[36] a) P. J. Hay, W. R. Wadt, J. Chem. Phys. 1985, 82, 299–310; b) T. H.
Dunning, Jr., P. J. Hay, Modern Theoretical Chemistry, Vol 3 (Ed.:
H. F. Schaefer III), Plenum, New York, 1976.
[37] V. Barone, M. Cossi, J. Tomasi, J. Chem. Phys. 1997, 107, 3210–
3221.
[22] For reviews concerning diazonium-ion Matsuda–Heck reactions,
see: a) A. Roglans, A. Pla-Quintana, M. Moreno-MaÇas, Chem. Rev.
2006, 106, 4622–4643; b) J. G. Taylor, A. V. Moro, C. R. D. Correia,
Eur. J. Org. Chem. 2011, 1403–1428; c) F. X. Felpin, L. Nassar-
Hardy, F. Le Callonnec, E. Fouquet, Tetrahedron 2011, 67, 2815–
2831.
[23] a) L. Wang, P. Li, Z. Wu, J. Yan, M. Wang, Y. Ding, Synthesis 2003,
2001–2004; b) G. A. Heropoulos, S. Georgakopoulos, B. R. Steele,
Tetrahedron Lett. 2005, 46, 2469–2473; c) Y. Liu, Y. Lu, M. Prashad,
Received: April 20, 2012
Revised: July 6, 2012
ˇ
O. Repic, T. J. Blacklock, Adv. Synth. Catal. 2005, 347, 217–219.
Published online: && &&, 0000
Chem. Eur. J. 2012, 00, 0 – 0
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
9
&
ÞÞ
These are not the final page numbers!

P. Evans, I. Alkorta et al.
What the Heck: A series of cyclic sul-
fonamides (R=Me, Ph and tBu; X =
Br, Cl, Tf, N2BF4) were studied in the
intramolecular Heck reaction. The out-
come of this process was controlled by
the presence of the substituent rather
than the ring size of the heterocycle
formed, and in all cases, formation of
the quaternary product was observed
(see scheme).
Heck Reaction
K. Geoghegan, P. Evans,* I. Rozas,
I. Alkorta* ..................... &&&& — &&&&
Regioselectivity in the Intramolecular
Heck Reaction of a Series of Cyclic
Sulfonamides: An Experimental and
Computational Study
&
10
&
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!