Chemoselectivity as a delineator of cuprate structure in Catalytic 1,4-Addition of Diorganozinc reagents to michael acceptors

Article abstract of DOI:10.1002/chem.200903310

Reaction of the cyclic thioacetal (RS)₂CHCHO [R = 1/2x-(CH ₂)₃] with HCCCOMe, followed by treatment with TsCl/DABCO (Ts = tosyl, DABCO= 1,4-diazabicyclo[2.2.2]octane) affords the mono-protected 1,4-benzoquinone dithioaceta

Full text of DOI:10.1002/chem.200903310

DOI: 10.1002/chem.200903310
Chemoselectivity as a Delineator of Cuprate Structure in Catalytic
1,4-Addition of Diorganozinc Reagents to Michael Acceptors
Matthias Welker,[a] Simon Woodward,*[a] Luis F. Veiros,*[b] and Maria Josꢀ Calhorda[c]
Abstract: Reaction of the cyclic thioa-
phoramidite ligands. The activation
energy for 1,4-methylation of the latter
OR-acetals with ZnMe2 (>95% ee)
has been determined for two CuX2 pre-
presence of CuI/ZnMe2 giving 1,4-
1
ꢀ
ꢀ
ꢀ
cetal (RS)2CHCHO [R= = ꢀ
(CH2)3
HOC6H4S
N
2
] with HCCCOMe, followed by treat-
ment with TsCl/DABCO (Ts=tosyl,
DABCO=1,4-diazabicyclo-
bond formation. The disparate behav-
ior of these two very closely related
substrates is in accordance with the for-
mation of closely related cuprate inter-
mediates that were optimized by DFT
calculations, supporting the synthetic
and kinetic studies and thus defining
the mechanisms of both pathways.
catalysts (X=OAc, 12.2 kcalmolꢀ1
;
[2.2.2]octane) affords the mono-pro-
X=OTf, 6.7 kcalmolꢀ1
The dithioacetal SR aromatizes in the
; Tf=triflate).
tected 1,4-benzoquinone dithioacetal.
The reactivity of this SR-protected 1,4-
benzoquinone has been compared with
the behavior of the analogous OR-pro-
tected acetal in copper-catalyzed addi-
tions of ZnMe2 by using chiral phos-
Keywords:
copper
·
density
functional calculations · P ligands ·
reaction mechanisms · zinc
Introduction
bers of examples of unligated lithiocuprates that are either
characterized by X-ray crystallography (B) or by solution
NMR spectroscopy (C) are now known. Recently, the first
CuIII intermediates (D) have been characterized by low-tem-
perature “rapid injection NMR” techniques.[2] There re-
mains a dichotomy between experiment and theory: theoret-
In 2000 Woodward briefly summarized what was then
known about the structures of lithium organocuprates and
the mechanisms by which they attack 1,4-Michael accept-
ors.[1] Through the work of Bertz et al.,[2] Nakamura et al.[3]
and others[4] the intimate behavior of lithium homocuprates
(e.g., structure A in Scheme 1 with X=Me, M=Li, L*=no
ligand) has indeed emerged from this “black box” towards
the rigorously defined picture of Scheme 1. Significant num-
[a] M. Welker, Prof. S. Woodward
School of Chemistry, The University of Nottingham
University Park, Nottingham, NG7 2RD (UK)
Fax : (+44)115-951-3564
E-mail: simon.woodward@nottingham.ac.uk
[b] Prof. L. F. Veiros
Centro de Quꢁmica Estrutural, Instituto Superior Tꢂcnico
1049-001 Lisboa (Portugal)
Fax : (+351)21-846-4455
E-mail: veiros@ist.utl.pt
[c] Prof. M. J. Calhorda
Departamento de Quꢁmica e Bioquꢁmica, CQB
Faculdade de CiÞncias
Universidade de Lisboa, 1749-016 Lisboa (Portugal)
Supporting information for this article (tables with atomic coordi-
nates for all optimized calculated species, X-ray structural data for
compounds: 1a, 1b, 3, and primary kinetic data) is available on the
WWW under http://dx.doi.org/10.1002/chem.200903310.
Scheme 1. Simplified mechanism of generalized copper-promoted 1,4-ad-
ꢀ
dition of M Me (M=any suitable counterion) to a representative Mi-
chael acceptor (cyclohexenone).
5620
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 5620 – 5629

FULL PAPER
ical approaches suggest that reductive elimination of D
should be the rate-limiting step of the reaction,[3] yet appre-
ciable concentrations of these species were not found in the
reaction mixture.
Results and Discussion
Synthesis and properties of dienones 1a and 1b: Compound
1a is available in good yield by a published short synthe-
sis.[11] In our hands, following the six-step procedure by Page
et al.[12] for 1b gave the product, although one ZnCl2-pro-
moted step was capricious (as Page had noted), which limit-
ed our overall yields to <10%. Unfortunately, 1b was not
available from reaction of 1a with 1,3-propane dithiol under
all conditions attempted. Alternative products in thiol reac-
tions of benzoquinone acetals have been noted before.[13] As
we wished to access significant quantities of 1b for reactivity
studies we designed an alternative route (Scheme 3).
In a parallel to the investigations for lithiocuprates that
are described above, ligand-accelerated[5] copper-catalyzed
1,4-addition of diorganozinc species ZnR2 (especially R=
Me, Et) to cyclohexenone, and related acceptors, has
become recognized as the premier method to attain high
enantioselectivities.[6] However, the relationship of the stoi-
chiometric lithium cuprate chemistry presented in Scheme 1
to that of the catalytic CuI/ZnMe2/L* systems (M=ZnMe,
L*=a phosphoramidite ligand) has remained somewhat ob-
scure. Kinetic studies of chiral copper(I)–phosphoramidite-
catalyzed additions of ZnR2 to enones are limited to early
observation of a negative non-linear effect[7] and a recent
study of the effect of the ligand structure on the rate by
Schrader et al.[8] Additional information on potential inter-
action modes between the CuX precatalyst and the chiral
ligand were gained from comprehensive NMR studies of
Gschwind et al.[9]
As the cuprate intermediates in CuI/ZnMe2/L*-promoted
reactions are very labile, we sought to harness the power of
DFT-based calculations[10] to gain insight into the potential
structures of p-complex C (M=MeZn, L*=phosphorami-
dite) in this zinc chemistry and their subsequent redox be-
havior. To maximize the computational robustness of our
regime we wished to directly relate our calculations to a pre-
dictable “real world” chemistry and were thus attracted to
propose differential reactivity of the closely related dienones
1a and 1b (Scheme 2).
Scheme 3. Alternative route to dieneones 1b and c (Ts=4-tolylsulfonyl;
DABCO=1,4-diazabicycloACTHUNTGRNEUNG[2.2.2]octane).
Formylation of 1,3-dithiane by using BuLi followed by
DMF and aqueous workup provided 2a in good isolated
yield (79%). Direct reaction of crude or distilled 2a with
butyn-2-one led to low yields of 3a (24–35%). Use of the
known[14a] self-dimerization product of 2a led to similar
yields. The poor performance of the reaction is probably a
consequence of the initially formed allenoate 4a that is
more successfully protonated to give an undesired trans-
enone that cannot subsequently cyclize. The formation of 3a
was confirmed by X-ray crystallography as it shows an unex-
ꢀ
Scheme 2. Literature[11,12] proposed reactivity for dienones 1a and b.
Feringa et al. have demonstrated that 1a undergoes
highly selective enantiopos addition of ZnEt2 in the presence
of their phosphoramidite.[11] In earlier work Page and co-
workers suggested that 1b reacts with LiCuMe2 by giving an
unexpected 1,2-addition product with stoichiometric cup-
rates.[12] We speculated that the p-complexes formed from
1a and 1b that deliver these two disparate products would
be highly similar. Further, this common cuprate structure,
related to C, must be able to accommodate the observed ex-
perimental behavior of both, 1a and 1b, providing a good
test for the validity of any computationally-derived inter-
mediates on the conjugate addition pathway.
pectedly weak nACTHNUTRGNEUNG(O H) stretch vibration in its IR spec-
trum.[14b] Under E1cb elimination conditions 1b was formed
smoothly. The overall three-step process for 1b is conven-
ient if not high yielding (15–22%). The related monocyclic
dienone 1c was also prepared by an equivalent route for re-
action comparison studies (see later). Key 13C NMR spectro-
scopic, X-ray structural and computational data for the Mi-
chael acceptors 1a and 1b are compared in Table 1. Elec-
tronically and structurally the substrates 1a and 1b are
closely related; the most significant differences are found in
1
ꢀ
the C S bond length in 1b (ꢁ = longer than its O-acetal an-
3
alogue) and the higher chemical shift of the acetal carbon in
1a.
Chem. Eur. J. 2010, 16, 5620 – 5629
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5621

S. Woodward, L. F. Veiros et al.
Table 1. Properties of substrates 1a and b relevant for addition chemis-
try.
tion structure 7 to this product. We believe this is a structur-
al misassignment of compound 6 for the following reasons:
Property
1a
1b
13C NMR shifts [ppm][a]
C(1)
C(2)
C(3)
1) The 13C NMR data of the product from the reaction of
185.2
128.5
141.9
89.0
183.8
127.0
146.0
45.3
1b and LiCuMe2 shows two clear ipso aromatic environ-
ments but no quaternary signal in the region expected
3
ꢀ
for an sp C O carbon.
C(4)
atomic charge (NPA)[b]
2) The methyl signal of the product shows no long-range
C(1)
C(3)
+0.500
ꢀ0.173, ꢀ0.222
+0.490
ꢀ0.197, ꢀ0.210
ꢀ
cross peak to any potential quaternary C O signal in the
HMQC spectrum but instead shows a valid cross peak to
structural data [ꢄ][c]
ꢀ
one of the CH2 S signals.
ꢀ
O C(1)
1.224
1.474
1.322
1.509
1.422
1.230
1.464
1.328
1.497
1.831
ꢀ
C(1) C(2)
ꢀ
C(2) C(3)
C(3) C(4)
C(4) Y (Y=O,S)
Our proposal is confirmed by the reaction of related 1c
with LiCuMe2(CN) that leads to known 9.[16] Optimal yields
of 6 were attained with MeMgBr addition to 1b under
CuBr·SMe2/PBu3 catalysis (5 mol% each). For rapid forma-
tion of 5 the presence of cuprate reagents or copper cata-
lysts was mandatory in all reactions. In no case was any 8
formed within the limits of NMR detection in the crude
mixture. Control reactions revealed that the presence of
CuI/LA is required for ZnMe2-induced transformations of 1a
and 1b to 5 and 6.
ꢀ
ꢀ
[a] Determined at 100 MHz in CDCl3 at ambient temperature. [b] Deter-
mined by DFT calculations at the PBE1PBE/6-31G** level; the two
values at C(3) indicate the two b-carbon atoms. [c] Determined from X-
ray crystallographic studies of 1a and b; where two identical bonds were
present, an average value is given.
Reactions of dienones 1a and 1b with carbon nucleophiles:
In order to attain a direct comparison of the reactivity of 1a
and 1b both compounds were treated with ZnMe2 under
Computational screening for a ZnMe2-derived p-complex
ground state: We have used lower levels of computational
theory to provide insights into the speciation of labile cop-
per(I) species in Cu-catalyzed reactions of organoalumini-
ums by simple metal–ligand combination trials.[17] As this
approach had been quite successful for aluminium cuprates,
a similar method was applied to the problem of assessing
the viability of potential zinc cuprates and their p-complexes
related to B and C, respectively (Scheme 1). Acetate was se-
lected as a suitable “computationally small” bridging ligand
as triflate was considered to have a too great potential to in-
troduce computational (S) or coordination (O,F) complexi-
ties. Studies were carried out by using the PBE1PBE hybrid
functional and a VDZP basis set with the core electrons of
the metal atoms Cu and Zn described by pseudo-potential
functions (see Computational Details). Computational com-
identical catalytic conditions (CuACTHNUTRGNEUNG(OTf)2 (2 mol%, Tf=tri-
flate), CH2Cl2, ꢀ208C for 4 h followed by 08C for 1 h) by
using the same phosphoramidite LA (4 mol%) (Scheme 4).
bination of CuACHTNUTRGNE(UNG OAc), ZnMe2 and ligand LB (used as a sim-
plified form of phospharamidite LA) led to exothermic for-
mation of zinc cuprate B1 for which dimerization to cuprate
B2 was energetically viable with an energy balance of DG=
2.3 kcalmolꢀ1 (Scheme 5).
The validity of suitable ground-state p-complexes of zinc
cuprates B1 and B2 was tested in a wide range of docking ex-
periments with dienone 1a that lead to the mono- or dicop-
per p-complexes C1–C6 (Scheme 6) by using the same
PBE1PBE hybrid functional approach. In one case (C1) an
Scheme 4. Real and fictional products that result from organomethyl ad-
ditions to dienones 1a–c.
The two-stage temperature profile was needed to attain
complete conversion of both compounds. Dienone 1a gave
the expected[11] 1,4-addition product 5 with high enantiose-
lectivity. This behavior was identical when copper(I)thio-
additional equivalent of [CuACHTNUGTRNEUNG(OAc)(LB)2] was included to
test a recent mechanistic proposal by Gschwind.[9] In two
other cases (C2 and C3) an additional equivalent of LB or
ZnMe2 was included to test recent mechanistic suggestions
by Schrader.[8] In all but one case (C1), very significant in-
creases of the ground-state energy are encountered—these
are highly unlikely compatible with the observed efficient
catalysis at low temperatures. The closest energetic species
thene carboxylate (TC) [Cu(TC)][15] or Cu
ACTHNUGTRNE(UNG OAc)2 were used
as precatalysts. Equivalent ZnMe2/CuI/LA-mediated reaction
of 1b gave a product that was identical in every way to the
product that results from the reaction of 1b and LiCuMe2 as
1
described by Page et al.[12] Based on H NMR and IR spec-
troscopic data alone these authors had assigned the 1,2-addi-
5622
www.chemeurj.org
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 5620 – 5629

Catalytic 1,4-Addition of Diorganozinc Reagents
FULL PAPER
[LnCuACHTNUTRGNE(NUG m-Me)2ZnMe] motif. We could not attain this species.
When thus placed, the ZnMe2 unit “slipped”, becoming
weakly associated with the phosphoramidite ligand (C3), dis-
sociated completely (C4), or migrated to the carbonyl
oxygen (C5). No easily energetically accessed p-complex
could be attained from interaction of the dimer B2 with di-
enone 1a. Only the high-energy complex C6 was calculated
despite attempts to constrain the p bonds in 1a to bind both
copper centers. The presence of additional ZnMe2 did not
improve stability of B2 +1a, only a weak Lewis acid interac-
tion in the region of the carbonyl group was calculated
(structure C7, 28.6 kcalmolꢀ1 above B1, not shown).
By using the energetically most accessible C1 as a starting
point, attempts were made to profile the putative rate-deter-
mining oxidative addition/reductive elimination steps equiv-
alent to “D transition states” (Scheme 1) and to rationalize
the different chemoselectivity between 1a and 1b. Prelimi-
nary calculations revealed that the B3LYP functional pro-
vides a better description of the energy barrier associated
with the formation of 5, than the one obtained with the
PBE1PBE functional (see Computational Details). Thus, all
mechanistic studies reported below were obtained with the
B3LYP functional. Because of the large “computational
size” of these multi-metal species and the demands of transi-
tion-state calculations, the structure of the phosphoramidite
ligand was further simplified to (MeO)2PNMe2 (LC) in the
mechanistic calculations. Two isomers of the p-complex with
substrate 1a were optimized, differing on the spacial ar-
rangement around the Cu center. In one isomer the methyl
group is close to the six-membered acetal ring (C1’); this
Scheme 5. Lowest energy zinc cuprates accessible from phosphoramidite
ligated LB and CuACHTUNGTRENNUNG(OAc) in the presence of ZnMe2.
was arbitrarily set as the energy reference compound (EC ’
=
1
0.00 kcalmolꢀ1, see Scheme 7). In the other isomer, C1, the
methyl and phosphoramidite ligand have exchanged posi-
tions and, thus, the methyl ligand is on the same side of the
carbonyl group. C1 is 2.2 kcalmolꢀ1 less stable than C1’. The
geometry around the Cu center makes C1’ the starting point
for the 1,4-addition, while C1 provides the reagent for the
1,2-addition. The energy profiles calculated for both reac-
tions are represented in Scheme 7.
Pleasingly the p-complex C1’ evolves smoothly to the ex-
pected enolate product E1 in a highly exothermic reaction.
A clear transition state D1 could be identified. In the transi-
tion state, D1, methyl migration from the Cu atom to the C6
Scheme 6. Dienone (1a) docking experiments with zinc cuprate B1. The
numbers (ꢀ0.8 to +22.5) indicate the free-energy costs, in kcalmolꢀ1, as-
sociated with the transformations: i) B1 +1a+[CuACHTUNRTGNE(GNU OAc)(LB)2], ii) B1 +
1a+LB, iii) B1 +1a+ZnMe2, iv) B1 +1a, v) B1 +1a+ZnMe2 (carbonyl
bound), and vi) dimerization of B1, followed by addition of 1a (B2 +1a).
ꢀ
ring is under way. The Cu C(Me) distance (2.136 ꢄ) is al-
ready 0.14 ꢄ longer than the corresponding one in C1’, but
ꢀ
the formation of the new C C bond is only beginning with a
separation of 2.211 ꢄ, still far from the value observed in
the product, E1 (1.540 ꢄ). These results are fully corroborat-
ed by the corresponding Wiberg indices (WI).[18] The
to B1 is p-complex C1, which is essentially identical to the
proposal of Gschwind for the catalyst rest state in 1,4-addi-
tions to cyclohexenone.[9]
On the basis of kinetic experiments on ZnEt2 additions to
cyclohexenone performed in 2004,[8] Schrader et al. had sug-
gested a structure equivalent to C2. However, this looks un-
likely to be attained in our case due to its relative instability
compared to the enone and/or ligand dissociation. Schrader
and co-workers also proposed a species closely related to C3
but with a 5-coordinate copper center that bridges in an
ꢀ
Wiberg index associated with the Cu C(Me) bond drops
ꢀ
from 0.33 in C1’, to 0.13 in D1, while for the new C C bond
the corresponding Wiberg index rises from 0.60 in the transi-
tion state D1, to 1.00 in the product E1, demonstrating that
the processes of bond forming and of bond breaking are
well advanced when the transition state is reached. Enolate
ꢀ
formation is also evident in E1, with a C O bond (d=
1.286 ꢄ, WI=1.32) that is longer and weaker than the C=O
Chem. Eur. J. 2010, 16, 5620 – 5629
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5623

S. Woodward, L. F. Veiros et al.
which leads to a “slipped” thioether ligand in a slightly dif-
ferent overall spacial arrangement (Scheme 8).
Species C8 shows a very facile rearrangement to the kinet-
ic aromatization product E8 via an identified transition state
Scheme 8. Calculated free energy profile for the aromatization of 1b
Scheme 7. Calculated free energy profiles of 1,4- and 1,2-additions with
substrate 1a.
ꢀ
through S Me bond formation from C8.
ꢀ
bond present in the C1’ (d=1.274 ꢄ, WI=1.38). Naturally,
D8. In D8, formation of the new S C(Me) bond is only in-
cipient (d=3.215 ꢄ, WI=0.15). In fact, these values indicate
ꢀ
the adjacent C C bond shows the reverse evolution shorten-
ꢀ
ing from 1.426 ꢄ (WI=1.25) in C1’ to 1.408 ꢄ (WI=1.44) in
E1.
that in D8, the S C(Me) interaction is only slightly stronger
than in C8 (d=3.282 ꢄ, WI=0.11). Correspondingly, the
ꢀ
ꢀ
Alternative reaction pathways to the 1,4-addition were in-
vestigated. The energetically most accessible was the 1,2-ad-
dition via transition state D2. This is 14.8 kcalmolꢀ1 above
the starting point for the 1,4-addition (C1’), thus, corre-
sponding to a less favorable path when compared with the
one previously discussed, which is in line with the experi-
mentally observed chemoselectivity. Despite extensive at-
tempts, no viable reaction vector leading to opening of the
process of C S and Cu C(Me) bond breaking is only start-
ꢀ
ing, once D8 is reached. The S C bond becomes only a little
longer and weaker, going from C8 (d=2.010 ꢄ, WI=0.73)
to D8 (d=2.076 ꢄ, WI=0.65), the same happens with the
Cu C(Me) bond (C8: d=2.008 ꢄ, WI=0.35; D8: d=
2.012 ꢄ, WI=0.33). These results indicate that D8 is a
rather early transition state.
To allow direct comparison with the behavior of dienone
1a and the erroneous 1,2-reactivity of 1b proposed by Page,
the barriers associated with potential 1,2- and 1,4-addition
of ZnMe2 to the sulfur analogues C9’ and C9 were calculated.
In both cases viable transition states (D9 and D10) and prod-
ucts (E9 and E10) could be identified (Scheme 9). The path-
ways leading to the kinetic 1,4- (E9) and 1,2-products (E10)
were found to be significantly higher in energy than that for
the aromatization (E8) (Scheme 8).
ꢀ
ꢀ
acetal 1a by Cu Me (equivalent to the reactivity of 1b)
could be detected. This is in
line with the experimental ob-
servations, as no 10 could be
identified within NMR detec-
tion limits in spectra of the
crude reaction mixture.
Related, but not observed, 1,4- and 1,2-addition reactivity
of sulfur dienone 1b with ZnMe2 was also studied from the
sulfur analogues of C1’ and C1 (C9’ and C9, respectively; see
later). However, in accordance with the experimental, a
much more favorable reaction pathway to the aromatization
product 6 could be calculated, starting from a closely related
intermediate C8. This species is more stable than C9’ (by
10.9 kcalmolꢀ1) and generated by CuI coordination to one
sulfur atom of the substrate, rather than to the double bond,
Kinetic and mechanistic investigation of the reactivity of di-
enones 1a and 1b: Although the computational models pro-
posed fit exactly to the observed experimental features of
the reactivity of substrates 1a and 1b we sought to further
correlate the calculated energy barriers to real kinetic data.
The experiments of Schrader et al. demonstrate, that obtain-
ing full low error bar kinetic analyses/rate laws of these air
sensitive systems is demanding.[8] Additionally, although it is
5624
www.chemeurj.org
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 5620 – 5629

Catalytic 1,4-Addition of Diorganozinc Reagents
FULL PAPER
Figure 1. Arrhenius plot of the reaction of 1a (2.41 mm) with ZnMe2
(2.89 mm) in toluene (18.4 mL), catalyzed by Cu
Tf=triflate) or Cu(OAc)2 (black diamonds) (both at 0.048 mm) and LA
(0.096 mm). The correlation coefficients for the two data sets are: R2(Cu-
(OAc)2)=0.989 and R2(Cu
(OTf)2)=0.988.
ACHTUNGRTEN(NNUG OTf)2 (grey squares;
AHCTUNGTRENNUNG
A
ACHTUNGTRENNUNG
Scheme 9. Calculated free energy profiles for the hypothetical 1,4- and
1,2-additions to 1b. Structures C9 and C9’ are analogous to C1 and C1’, re-
spectively, but with bounded 1b instead of 1a.
suggested to be the rate-determining step, is being compared
with Ea for a whole process. Nevertheless, the calculated
and experimental results are in good agreement. The fact
that a Cu
ACHTUNGTRENNUNG
generally accepted that the transformation of p-complex C
through intermediate D to the product enolate E is rate-lim-
iting, it is not known what kinetic advantage this has over
other elements of the process. The complex nature of these
reaction pathways and the possibilities of mixed copper spe-
ciation or even radical reaction pathways[19] means that de-
ceptively simple kinetic results may hide a mꢂlange of ef-
fects in this area. Thus, we were attracted to the approach of
Krause et al. in measuring a simple Arrhenius “activation
energy” for the whole of a cuprate-induced process.[20] Pre-
liminary studies indicated that suitable kinetic data could be
attained for the first 2 h of the transformation of 1a to 5,
calculations are performed with CuACTHUNGTRENNNUG
known that such copper(II) precursors are reduced rapidly
under the reaction conditions.[9] It might be expected that in-
creasing the Lewis acidity of the copper complex should en-
gender a faster reaction. This is proven by the fact that the
CuACHTUNTRGNEUNG(OTf)2/LA-catalyzed reaction shows a reduced activation
energy of 6.7ꢂ0.7 kcalmolꢀ1. By assuming that Arrhenius
plots of the 1,4-addition processes can be equated to the
Eyring–Polanyi equation[21] [Eq. (1)], estimates of the activa-
tion enthalpies and entropies can be attained.
z
z
k
T
ꢀDH
1
T
kB DS
ð1Þ
ln
¼
þ ln
þ
catalyzed by CuACHTUNGTRENNUNG(OAc)2 (2 mol%) and LA (4 mol%) under
R
h
R
the reaction conditions related to those used in the synthetic
studies (0.129m 1a in toluene, ꢀ108C, 1.2 equiv ZnMe2).
Beyond 2 h (ca. 25 turnovers) the kinetic analyses were in-
validated by the generation of minor secondary by-products.
Although these did not seem to greatly affect the rate of
catalysis, extraction of the concentrations of 1a and 5 was
complicated by GC co-elution and the data were not used.
Values of DH° =+11.7 and +6.3 kcalmolꢀ1 were deter-
mined for the Cu(OAc)2- and Cu(OTf)2-catalyzed processes,
G
ACHTUNGTRENNUNG
respectively. For the activation entropies the determined
values of DS° for the acetate and triflate were ꢀ32.5 e.u.[22]
and ꢀ43.3 e.u., respectively. The observation of significant
negative values of DS° is normally associated with the pres-
ence of significant ordering in a transition state. This is also
in line with the presumption two coppers, three chiral li-
gands, one zinc, and the Michael acceptor must all be assem-
bled in the transition state presented here.
A similar procedure was applied by using Cu
copper source, which provided a quicker reaction but lead
to similar minor by-products after 2 h. From Cu(OAc)2-cata-
(OTf)2-
ACHTUNGRTEN(NUNG OTf)2 as a
AHCTUNGTRENNUNG
lyzed reactions, conducted at ꢀ20 to +58C, and Cu
ACHTUNGTRENNUNG
catalyzed reactions, conducted between ꢀ35 and ꢀ108C,
two Arrhenius plots could be constructed (Figure 1). In
every kinetic run that was investigated the enantioselectivity
remained invariant and very high (>95% ee) throughout.
The slope of Figure 1 gives an activation energy (Ea) of
It was not possible to compare the kinetic behavior of 1a
with the related dithioacetal 1b. Despite many attempts we
could not attain kinetic data of sufficient quality. The major
problem is the lower solubility of the kinetic product 6 of
the reaction, which quickly leads to coprecipitation of 6 and
1b, which is itself highly crystalline. In view of these prob-
lems and the predicted, very low, activation energy (1.3 kcal
molꢀ1, see Scheme 8) associated with this transformation an
12.2ꢂ1.2 kcalmolꢀ1 for the Cu
ACHTUGNRTEN(NUGN OAc)2/LA-catalyzed addition
of ZnMe2 to 1a. Comparison with the calculation presented
in Scheme 7 should be made with a degree of caution: a cal-
culated energy barrier for a single step (C1’!(D1)°!E1)
ꢀ
alternative approach was sought. Cleavage of the C S acetal
Chem. Eur. J. 2010, 16, 5620 – 5629
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5625

S. Woodward, L. F. Veiros et al.
bond in 1b could in principle be driven by two factors:
1) the additional energy generated through aromatization on
route to product 6; or 2) the intrinsically weaker nature of
substituted Michael acceptor binds through a p-contact to
the cooper, the dithioacetal binds through a CuI···thioether
contact. The origin of the switch in chemoselectivity be-
tween 1a and 1b can probably be best ascribed to a stronger
Cu···Y interaction in the p-complex of the thioacetal (1b,
Y=S), when compared to the equivalent species for the
acetal (1a, Y=O). The enhanced donor capability of the S
atom, when compared to its O counterpart, is illustrated by
the larger coefficients on the sulfur atoms in comparison to
the oxygen atoms on the relevant orbitals of substrates 1b
and 1a, respectively (Figure 2).
ꢀ
the C S bond. We conceived that reaction of the mono-
enone dithioacetal 11 with ZnMe2 might resolve this issue
(Scheme 10). However, in practice 11 (also available
through literature procedures[12]) showed insufficient reactiv-
ity in copper(I)/LA- or copper(I)/LD-catalyzed addition of
Figure 2. Calculated HOMOꢀ1 orbital for substrate 1a (O,O-acetal, left)
and HOMO orbital for substrate 1b (S,S-dithioacetal, right).
Experimental Section
Scheme 10. Copper thiophenecarboxylate/LD-catalyzed additions to 1a
and 11. Conditions: AlMe3 (1.5 equiv), [Cu(TC)] (2 mol%; TC=thio-
thene carboxylate), LD (4 mol%), CH2Cl2, ꢀ78 to ꢀ508C over 4 h.
General: Infrared spectra were recorded by using Perkin–Elmer 983 G
infrared and Perkin–Elmer 882 infrared spectrophotometers, or a Bruker
IFS 66 spectrometer. 1H and 13C NMR spectra were recorded on Bruker
(AM400, AV400, or DRX 400) or JEOL (JNM-GX270) spectrometers
by using CHCl3 (d=7.27 ppm) and tetramethylsilane (d=0.00 ppm) as
standard; J values are given in Hz. All spectra were recorded at ambient
temperature unless otherwise noted. Mass spectra were obtained on Fin-
nigan-MAT 1020 or Autospec VG (electron impact ionization, EI), Finni-
gan-QMS (electrospray ionization, ESI), VG-ZAB, or Finnigan MAT
8200 insturments (EI, 70 eV). GC analyses were attained by using an au-
tosampler-equipped Varian 430GC (He carried gas, flow rate
1.5 mLminꢀ1) under the conditions and columns described for each com-
pound Optical rotations were measured on Bellingham Stanley ADP 440
in units of 10ꢀ1 degcm2 gꢀ1; concentration (c) is given as g100 cmꢀ3. All
reactions involving air sensitive materials were carried out under an
argon atmosphere by using standard Schlenk techniques. Reaction sol-
vents were distilled under argon from appropriate agent immediately
prior to use. Petroleum ether refers to the fraction with b.p. 40–608C.
The following compounds were attained by literature procedures: 1a,[11]
1b,[12] LA,[17] 2a,[14] 11,[12] and LD.[23]
ZnMe2 under all conditions tried. However, use of the more
Lewis acidic AlMe3 led to sole formation of 12 in >95% ee
(optimal 60% isolated yield by using LD), whereas under
identical conditions 1b still afforded 6 as the sole product in
62% isolated yield. The viability of 1,4-addition to 11 indi-
cates that aromatization is the most likely driving force for
the formation of 6. For now we reserve comment on the
structure of the p-complexes and transition states involved
in the formation of 12. It is likely that their structures are
closely related to the zinc species presented here and this is
under current investigation in our laboratories.
Conclusion
Alternative preparation of 1b–c
2-Formyl-1,3-dithiane (2a):[14] To a flame-dried flask containing 1,3-di-
thiane (2.50 g, 20.8 mmol) in dry THF (50 mL) at ꢀ308C nBuLi
(14.0 mL, 1.6m in hexane, 22.4 mmol) was added and the reaction mix-
ture was stirred at this temperature for 1 h. The mixture was allowed to
come to ꢀ108C and dry N,N-dimethylformamide (560 mL, 7.21 mmol)
was added. The mixture was stirred for 2 h at ꢀ108C before warming to
08C and overnight stirring at room temperature. The resulting suspension
was poured into ice water (50 mL) and the mixture was extracted with
pentane (20 mL) several times. The aqueous layer was neutralized with
hydrochloric acid (1m) and then extracted several times with diethyl
ether (20 mL). The total organic fractions were combined, dried
(MgSO4), filtered, and concentrated in vacuo. The crude product was
used without further purification or distilled under reduced pressure (80–
908C at 1 mmHg) to give the desired compound as an unstable colorless
oil (1.85 g, 60%) with literature properties.[14] 1H NMR (270 MHz,
CDCl3, 258C): d=9.47 (s, 1H; CH=O), 4.09 (s, 1H; CHCH=O), 3.05–
2.83 (m, 2H; SCH2), 2.57–2.46 (m, 2H; SCH2), 2.09–1.88 ppm (m, 2H;
The divergent reactivity of the O,O-acetaldienone 1a and its
dithio analogue 1b has been resolved by a combination of
experimental and DFT procedures. Whereas the former re-
sults in highly enantioselective 1,4-addition with CuI–phos-
ꢀ
phoramidite catalysis, the latter leads to S Me bond forma-
tion and aromatization to 6 with exactly the same catalyst
system. A confusion in the literature where 6 had been mis-
assigned as the 1,2-addition product 7 has been corrected.
Theoretical simulation has shown that very closely related
zinc cuprates can account for the reactivity change. This
strongly supports the involvement of a common [Zn
ACHTUNGERTN(NUNG m-
X)Cu(m-X)CuR] motif in all these catalytic reactions (X=a
ACHTUNGTRENNUNG
suitable bridging ligand, e.g., OAc, OTf, thiophene carboxyl-
ate; Tf=triflate). In the present case, whereas the acetal-
5626
www.chemeurj.org
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 5620 – 5629

Catalytic 1,4-Addition of Diorganozinc Reagents
FULL PAPER
CH2); 13C NMR (68 MHz, CDCl3, 258C): d=188.2, 47.5, 25.3, 24.7 ppm
(2C); IR (CHCl3 solution): n˜ =2909, 2821, 1716, 1672, 1386, 1280, 1093,
1001, 907 cmꢀ1; MS (EI+): m/z (%): 148.00 (15) [M]+, 119.00 (100),
75.03 (12).
phy (petroleum ether/Et2O 2:1) to afford 1b as a beige solid (117 mg,
80%), which is stable for weeks when kept under argon in freezer. Rf =
0.30 (petroleum ether/Et2O 1:1); m.p. 58–608C; 1H NMR (400 MHz,
CDCl3, 258C): d=7.19 (d, J=10.0 Hz, 2H; CH=CHCO), 6.23 (d, J=
10.0 Hz, 2H; CH=CHCO), 3.01 (m, 4H; SCH2CH2CH2S), 2.09 ppm (m,
2H; SCH2CH2 CH2S); 13C NMR (100 MHz, CDCl3, 258C): d=183.8,
146.0 (2C), 127.0 (2C), 45.3, 26.2 (2C), 23.2 ppm; IR (CHCl3 solution):
n˜ =2912, 2360, 1662, 1620, 871 cmꢀ1; MS (EI): m/z (%): 313.91 (47),
198.02 [M]+ (100), 156.96 (30), 124.00 (14), 74.02 (24). This material had
identical properties to that prepared by the six-step synthesis described
by Page et al.[12]
2,2-Bis(ethylthio)acetaldehyde (2b): To a solution of 40% glyoxal in
water (387 mL, 33.75 mmol) and ethanethiol (5 mL, 67 mmol; STENCH!)
a solution of HCl (2m, 0.4 mL) was added at room temperature. The het-
erogeneous mixture turned to white after 15 min and was then stirred
vigorously overnight. The organic phase was separated and the aqueous
phase was extracted with CH2Cl2 (2ꢀ10 mL) and the combined organic
phases were washed once with dilute aqueous solution of Na2CO3. The
organic layer was filtered through a cotton plug and the bulk of the sol-
vent was removed by distillation at atmospheric pressure (408C) Light
petrol (b.p. 40–608C, 5 mL) was added and distillation was resumed to
ensure the residual oil was free from noxious ethanethiol. The residual
oil was then distilled under reduced pressure (70–758C at 1 mmHg) to
afford the desired product as a colorless oil (2.21 g, 40%). 1H NMR
(270 MHz, CDCl3, 258C): d=9.17 (d, J=4.2 Hz, 1H; CH=O), 4.23 (d,
4,4-Bis(ethylthio)cyclohexa-2,5-dienone (1c): A solution of alcohol 3b
(145 mg, 0.62 mmol) and DABCO (140 mg, 1.25 mmol) in CH2Cl2
(1.5 mL) was stirred at room temperature for 5 min. Solid TsCl (179 mg,
0.94 mmol) was added in a single portion. The resulting orange mixture
was stirred at room temperature for 4 h. The reaction conversion was
monitored by TLC (petroleum ether/Et2O 2:1). An aqueous solution of
NaHCO3 (5 mL) was added and the product was extracted with CH2Cl2
(4ꢀ10 mL). The combined organic phases were dried (MgSO4) and con-
centrated in vacuo. The crude oil was purified by column chromatogra-
phy (petroleum ether/Et2O 2:1) to afford 1c as a dark oil (103 mg, 78%).
ꢀ
J=4.2 Hz, 1H; CH S), 2.60 (dq, J=7.4, 3.9 Hz, 4H; CH2), 1.26 ppm (t,
J=7.4 Hz, 6H; CH3); 13C NMR (75 MHz, CDCl3, 258C): d=189.3, 55.8,
24.8 (2C), 14.5 ppm (2C); IR (CHCl3 solution): n˜ =2976, 2932, 2875,
2828, 2709, 1712, 1453, 1424, 1379, 1267, 1053, 1020, 978, 827 cmꢀ1
;
Rf =0.48 (petroleum ether/Et2O 2:1); H NMR (400 MHz, CDCl3, 258C):
1
HRMS (EI): m/z: calcd for C6H12OS2: 164.032959 [M]+; found:
d = 6.88 (d, 2H, J = 10.0 Hz, CH=CHCO), 6.26 (d, 2H, J = 10.0 Hz,
164.033660.
CH=CHO), 2.47 (q, 4H, J
= 7.6 Hz, CH2-S), 1.20 ppm (t, 6H, J =
7.6 Hz, CH3); 13C NMR (100 MHz, CDCl3, 258C): d = 184.6, 149.8 (2C),
11-Hydroxy-1,5-dithiaspiro
formyl-1,3-dithiane (2a) (500 mg, 3.37 mmol) in acetonitrile (6 mL) was
added in flame-dried Schlenk tube under argon. But-3-yn-2-one
ACHTUNGTREN[NGNU 5.5]undec-7-en-9-one (3a): Freshly distilled 2-
128.1 (2C), 53.6, 24.6 (2C), 14.2 ppm (2C); IR (CHCl3, solution): n˜
=
3018, 2985, 2923, 23180, 1656, 1612, 869 cmꢀ1; MS (EI): m/z (%): 91.1
(15), 105.0 (15), 121.1 (100), 154.0 (22), 214.0 (9); HRMS (EI): m/z: calcd
for C10H14OS2: 214.0481; found: 214.0483.
a
(275 mg, 4.05 mmol) and K2CO3 (1.09 g, 3.37 mmol) were successively
added to this solution at room temperature. The resulting yellow solution
was cooled to 08C, stirred, and allowed to come to room temperature
overnight. The mixture turned slowly red–orange in color. The reaction
was quenched with saturated aqueous solution of NH4Cl, extracted with
CH2Cl2, dried (MgSO4), and concentrated in vacuo. The crude mixture
was purified by flash chromatography (petroleum ether/Et2O 2:1) to give
the desired product as an orange solid (0.263 g, 36%). Rf =0.15 (petro-
leum ether/Et2O 1:1); m.p. 97–988C; 1H NMR (270 MHz, CDCl3, 258C):
d=6.62 (dd, J=10.8 Hz, 2.0 Hz, 1H; =CHCO), 6.00 (d, J=10.8 Hz, 1H;
CH=CHCO), 4.55 (t, J=2.4 Hz, 1H; CHOH), 3.14–2.70 (m, 6H), 2.20–
1.80 ppm (m, 2H); 13C NMR (100 MHz, CDCl3, 258C): d=195.8, 144.7,
129.4, 69.5, 53.4, 41.1, 27.0, 25.1, 23.2 ppm; IR (CHCl3 solution): n˜ =3484,
2913, 1689, 1602, 1383, 1082, 1056, 875 cmꢀ1; MS (EI+):: m/z (%): 216.03
(37.1) [M]+, 172.00 (100), 115.97 (22.0), 97.98 (14.9).
Generalized additions of ZnMe2 to 1a–c
(S)-11-Methyl-1,5-dioxaspiro
Schlenk tube was charged with CuAHCNUTGTRENNUNG
A
A flame-dried
(15.6 mg, 4 mol%), and CH2Cl2 (1 mL), under argon. The resulting sus-
pension was stirred for 15 min and the mixture was cooled to ꢀ208C. A
solution of dimethylzinc (1.2m in toluene, 0.90 mL, 1.08 mmol, 1.5 equiv)
was added dropwise and the mixture was stirred for 5 min at ꢀ208C.
After this time, benzoquinone monoacetal 1a (120 mg, 0.72 mmol) dis-
solved in CH2Cl2 (1 mL) was added dropwise. The resulting yellow solu-
tion was stirred at ꢀ208C for 4 h. The reaction was quenched with
NH4Cl, extracted with Et2O (20 mL), dried (MgSO4), and concentrated
in vacuo to afford a dark oil (113 mg, 87%). Analysis by 1H NMR and
GC indicated a conversion of 89% to (S)-5 with 99% ee. 1H NMR
(400 MHz, CDCl3, 258C): d=7.35 (d, J=10.4 Hz, 1H; CH=CHCO), 6.04
(d, J=10.4 Hz, 1H; =CHCO), 4.00–3.90 (m, 4H; OCH2CH2CH2O), 2.48–
2.42 (m, 3H; CH-CH2), 2.09–2.03 (m, 1H; OCH2CH2CH2O), 1.59–1.55
(m, 1H; OCH2CH2CH2O), 1.08 ppm (d, J=6.0 Hz, 3H; CH3); 13C NMR
(100 MHz, CDCl3, 258C): d=199.3, 143.8, 129.9, 95.5, 60.7 (2C), 41,9,
38.9, 25.3, 13.9 ppm; IR (CHCl3 solution): n˜ =2934, 2875, 1682, 1388,
1114, 986 cmꢀ1; MS (EI+): m/z (%): 182.10 [M]+ (37), 154.10 (34),
140.05 (49), 124.05 (100), 112.05 (54), 82.00 (41); HRMS (ES+): m/z:
calcd for C10H14O3: 182.0943 [M]+; found: 182.09433; GC (assay on Lipo-
dex A column, 1108C isothermal): retention time: 15.0 min ((R)-5) and
15.2 min ((S)-5).
4,4-Bis(ethylthio)-5-hydroxycyclohex-2-enone (3b): Preparation of 3b
was performed in an equivalent manner to 3a from aldehyde 2b (2.00 g,
12.17 mmol) dissolved in acetonitrile (15 mL) and butyn-2-one (994 mg,
14.60 mmol) followed by addition of a single portion of K2CO3 (1.68 g,
12.17 mmol). Crude 3b, an oil, was purified by column chromatography
(petroleum ether/Et2O 2:1) to afford the desired compound as a yellow
liquid (678 mg, 24%). Rf =0.20 (petroleum ether/Et2O 1:1); 1H NMR
(400 MHz, CDCl3, 258C): d=6.70 (dd, J=10.1, 0.7 Hz, 1H; CH=
CHCO), 5.95 (d, J=10.1 Hz, 1H; CH=CH-CO), 4.12 (t, J=6.3 Hz, 1H;
CH-OH), 3.22 (s, 1H; OH), 2.84–2.82 (m, 2H; CH2), 2.78–2.57 (m, 4H;
CH2S), 1.21 ppm (dt, J=7.5, 1.6 Hz, 6H; CH3); 13C NMR (100 MHz,
CDCl3, 258C, TMS): d=196.4, 148.4, 128.6, 71.1, 62.8, 42.4, 24.1, 23.3,
14.2 (2C) ppm; IR (CHCl3 solution): n˜ =3534, 3009, 2976, 3932, 2874,
1684, 1611, 1449, 1379, 1341, 1300, 1265, 1240, 1163, 1088, 1057, 975, 909,
859, 822 cmꢀ1; MS (ES+): m/z (%): 487.11 (10), 419.08 (22), 256.05 (10),
255.05 [M+Na]+ (100), 233.07 [M+H]+ (3), 171.05 (68); HRMS (ES+):
m/z: calcd for C10H16O2S2: 255.04839 [M+Na]+; found: 255.0474.
Methyl-(3-(p-tolylthio)propyl)sulfane (6): A procedure as described for
compound
5
was followed by using copper(II)triflate (4.8 mg,
0.013 mmol) and benzoquinone dithioacetal 1b (130 mg, 0.66 mmol). The
reaction was performed in CH2Cl2 (2 mL). Upon addition of 1b the reac-
tion mixture became light brown. After 24 h at ꢀ208C, no starting mate-
rial could be detected by TLC. The mixture was quenched with NH4Cl,
filtered through celite to remove some insoluble material, extracted with
Et2O, dried (MgSO4), and concentrated in vacuo to afford 6 as a brown
oil (90 mg, 65%). Proton NMR indicated total conversion towards the ar-
omatized product. 1H NMR (400 MHz, CDCl3, 258C:) d=7.30 (d, J=
8.4 Hz, 2H; CHar), 6.77 (d, J=8.4 Hz, 2H; =CHar), 5.10 (brs, 1H; OH),
2.91 (t, J=7.2 Hz, 2H; CH2), 2.59 (t, J=7.2 Hz, 2H; CH2), 2.06 (s, 3H;
CH3), 1.85 ppm (quintuplet, J=7.2 Hz, 2H; CH2); 13C NMR (100 MHz,
CDCl3, 258C): d=154.9, 133.5 (2C), 126.3, 116.0 (2C), 345, 32.7, 28.3,
15.4 ppm; IR (CHCl3 solution): n˜ =3593, 2919, 1599, 1584, 1493, 1320,
1,5-DithiaspiroACHTUNGTRENNUNG[5.5]undeca-7,10-dien-9-one (1b): A solution of alcohol 3a
(160 mg, 0.74 mmol) and DABCO (166 mg, 1.48 mmol) in CH2Cl2 (3 mL)
was stirred at room temperature for 5 min. Solid TsCl (212 mg,
1.12 mmol) was added in a single portion. The resulting orange mixture
was stirred at room temperature for 4 h. The reaction conversion was
monitored by TLC (petroleum ether/Et2O 2:1). An aqueous solution of
NaHCO3 (10 mL) was added and the product was extracted with CH2Cl2
(4ꢀ10 mL). The combined organic phases were dried (MgSO4) and con-
centrated in vacuo. The crude oil was purified by column chromatogra-
Chem. Eur. J. 2010, 16, 5620 – 5629
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5627

S. Woodward, L. F. Veiros et al.
1167, 1094, 960 cmꢀ1; HRMS (ESꢀ): m/z: calcd for C10H14OS2: 213.0408
[MꢀH]ꢀ; found: 213.0417. These data are not consistent with the struc-
ture 7 proposed in the literature,[12] which is a miss assignment.
Computational Details
The calculations were performed by using the Gaussian 03 software pack-
age.[25] The PBE1PBE functional was used for the screening of p-com-
plexes (Scheme 6) and for preliminary mechanistic studies. That function-
al uses a hybrid-generalized gradient approximation (GGA), including
25% mixture of Hartree–Fock[26] exchange with DFT[10] exchange-corre-
lation, given by Perdew, Burke, and Ernzerhof functional (PBE).[27] How-
ever, we verified that the energy barrier for 1,4-addition to acetal 1a cal-
culated with the B3LYP functional (13.0 kcalmolꢀ1) was considerably
closer to the experimental activation energy (Ea =12.2 kcalmolꢀ1), than
the corresponding value obtained with PBE1PBE (7.2 kcalmolꢀ1). Thus,
all mechanistic calculations were repeated by using the B3LYP function-
al. This is also an hybrid functional, including a mixture of Hartree–Fock
exchange (20%) with DFT exchange correlation, given by Beckeꢅs three-
parameter functional with the Lee, Yang, and Parr correlation functional,
which includes both local and non-local terms.[28] All calculations were
performed without symmetry constraints. The optimized geometries were
obtained with the LanL2DZ basis set[29] augmented with an f-polarization
4-(Ethylthio)phenol (9): A similar procedure as described for compound
4 was followed, starting from benzoquinone dithioacetal 1c (195 mg,
0.91 mmol). After 3.5 h, no starting material could be detected by TLC.
The mixture was quenched with NH4Cl, extracted with Et2O, dried
(MgSO4), and concentrated in vacuo to afford the product as a yellow
solid (99 mg, 59%). m.p. 38–408C; 1H NMR (400 MHz, CDCl3, 258C):
d=7.29 (d, J=8.4 Hz, 2H; CHar), 6.78 (d, J=8.4 Hz, 2H; CHar), 2.90 (q,
J=7.2 Hz, 2H; CH2), 1.24 ppm (t, J=7.2 Hz, 2H; CH3); 13C NMR
(100 MHz, CDCl3, 258C): d=154.8, 133.4 (2C), 126.3, 115.9 (2C), 29.8,
14.6 ppm; IR (CHCl3 solution): n˜ =3595, 2929, 1600, 1584, 1494, 1257,
1170 cmꢀ1; MS (EI): m/z (%): 97.0 (20), 125.0 (23), 126.0 (26), 139.0 (32),
154.0 (100); HRMS (EI): m/z: calcd for C8H10OS: 154.0452; found:
154.0443. This material was identical to an example described in the liter-
ature.[16]
7-Methyl-1,5-dithiaspiroACHTUNGTRENNUNG[5.5]undecan-9-one (12): A flame-dried Schlenk
[23]
tube was charged with [Cu(TC)][15] (2.3 mg, 2 mol%), ligand LD
ACTHNUTRGNEUNG
function,[30] for Cu and for Zn, and a standard 6-31G(d,p)[31] for the re-
(11.9 mg, 4 mol%), and CH2Cl2 (1.0 mL) under argon. The suspension
was stirred at room temperature for 15 min and then cooled to ꢀ788C.
Trimethylaluminium (2m in hexane, 0.45 mL, 0.90 mmol, 1.5 equiv) was
added dropwise. The mixture was stirred for 2 min at ꢀ788C after which
time enone 11 (120 mg, 0.60 mmol) dissolved in CH2Cl2 (2.0 mL) was
added dropwise. The solution turned yellow and was stirred while it was
allowed to warm slowly from ꢀ78 to ꢀ108C over 5 h. Complete disap-
pearance of the starting material (followed by TLC, petroleum ether/
Et2O 2:1) was observed. The reaction was quenched with aqueous
NH4Cl, extracted with CH2Cl2 (20 mL), dried (MgSO4), and concentrated
in vacuo. The crude material was purified by flash chromatography (pe-
troleum ether/Et2O 2:1) to afford the product as a yellow oil (78 mg,
60% yield, >95% ee). The enantiomeric excess was measured by NMR
after derivatization with (1R,2R)-(+)-1,2-dipheneylethylene diamine.[24]
[a]D24 =+10.6 (c=1.00, CHCl3); 1H NMR (400 MHz, CDCl3, 258C): d=
3.06 (ddd, J = 14.4, 11.2, 3.1 Hz, 1H; CH2aCH2S), 2.91 (ddd, J = 14.4,
11.2, 3.1 Hz, 1H; CH2aS), 2.83–2.69 (m, 3H; CH2b-CH2aS, CH2a), 2.62–
2.51 (m, 1H; CH2b), 2.55 (m, 1H; CHCO), 2.49–2.41 (m, 1H; CH2a), 2.38
(m, 1H; CHCO), 2.35 (m, 1H; CHCH3), 2.21–2.15 (m, 1H; CH2b), 2.15–
2.05 (m, 1H; CH2bS), 1.97–1.86 (m, 1H; CH2bS), 1.20 ppm (d, J=6.6 Hz,
3H; CH3); 13C NMR (100 MHz, CDCl3, 258C): d=209.5, 53.8, 45.0, 41.6,
38.3, 35.2, 26.3, 24.5, 17.0 ppm; IR (CHCl3 solution): n˜ =2966, 2912, 1712,
1453, 1417, 1381, 1340, 1278, 1239, 1143, 909 cmꢀ1; MS (ES+): m/z (%):
455.12 (21), 271.08 (12), 240.06 (11), 239.05 [M+Na]+ (100), 217.07
[M+H]+ (14); HRMS (ES+): m/z: calcd for C10H16OS2: 239.05348
[M+Na]+; found: 239.0538.
maining elements. Transition-state optimizations were performed with
the synchronous transit-guided quasi-Newton method (STQN) developed
by Schlegel et al.[32] Frequency calculations were performed to confirm
the nature of the stationary points, yielding one imaginary frequency for
the transition states and none for the minima. Each transition state was
further confirmed by following its vibrational mode downhill on both
sides, and obtaining the minima presented on the energy profiles. The
free energy values presented in Schemes 6–9 and discussed along the text
were obtained at 298.15 K and 1 atm by conversion of the zero-point-cor-
rected electronic energies with the thermal energy corrections based on
the calculated structural and vibrational frequency data. A natural popu-
lation analysis (NPA)[33] and the resulting Wiberg indices[18] were used for
a detailed study of the electronic structure and bonding of the optimized
species. The orbital drawings were obtained by using the program MO-
LEKEL 4.0.[34]
Acknowledgements
M.W. is grateful to the FP6 Programme for providing a Marie Curie
Early Stage Fellowship under Contract MEST-CT-2005-019780 (INDAC-
CHEM). Helpful input from the COST D40 Programme on “Innovative
Catalysis” and from the UK government (EPSRC grant EP/F033478/1) is
acknowledged by S.W. We thank Prof. A. J. Blake and Dr. W. Lewis and
Dr. R. Fryatt for their input into the crystallographic and preliminary
synthetic studies. M.J.C. thanks FCT and FEDER for financial support
(PPCDT/QUI/58925/2004).
Kinetic studies of the addition of ZnMe2 to 1a: The kinetic measure-
ments were performed under an argon atmosphere in a flame-dried
round bottom flask. The temperature was controlled by a cold thermom-
eter from Brannan and was maintained constant within ꢂ0.5 K by using
a Thermo Haake DC50K75 cryostat. A solution of ligand LA (65 mg,
4 mol%) and CuACHTUNGTRENNUNG(OAc)2 (11.7 mg, 2 mol%) in freshly distilled toluene
(16.0 mL) was prepared and stirred at room temperature for 30 min.
After that time the solution was cooled to the appropriate temperature
and diethylzinc (1.2m in toluene, 3.26 mL, 3.91 mmol) was added to the
reaction mixture. At the timepoint t=0, a previously cooled solution of
dienone 1a (500 mg, 3.00 mmol) in toluene (4.0 mL) was added to the re-
action mixture (end volume: 23.26 mL, starting concentration of 1a:
0.129m). Aliquots were taken under an argon counterflow at regular in-
tervals (by using a Pasteur pipette that had been previously cooled in
liquid nitrogen) and immediately hydrolyzed in a pre-cooled mixture of
methanol already present in the reaction cooling bath. The organic com-
ponents were analyzed by gas chromatography (Lipodex A, isothermal
1108C): retention times: 15.0 min ((R)-5), 15.2 min ((S)-5), and 16.0 min
(1a).
[1] S. Woodward, Chem. Soc. Rev. 2000, 29, 393–401.
[2] a) Detection of copper
S. Cope, M. Murphy, C. A. Ogle, B. J. Taylor, J. Am. Chem. Soc.
2007, 129, 7208–7209; b) copper(III) species with coordinated li-
ACHTUNGTRENUN(NG III) intermediate “D” by NMR: S. H. Bertz,
AHCTUNGTRENNUNG
gands: E. R. Bartholomew, S. H. Bertz, S. Cope, D. C. Dorton, M.
Murphy, C. A. Ogle, Chem. Commun. 2008, 1176–1177.
[3] a) Recent key overviews of theoretical approaches: S. Mori, E. Na-
kamura in Modern Organocopper Chemistry (Ed.: N. Krause),
Wiley-VCH, Weinheim, 2002, pp. 315–346. b) Mechanism of
LiCuR2 addition to p-extended Michael acceptors: N. Yoshikai, T.
Yamashita, E. Nakamura, Chem. Asian J. 2006, 1, 322–330.
c) “Dummy” ligand effects in 1,4-addition: M. Yamanaka, E. Naka-
mura, J. Am. Chem. Soc. 2005, 127, 4697–4706.
[4] a) Diastereoselectivity in p-complex formation: W. Henze, T. Gꢆrt-
ner, R. M. Gschwind, J. Am. Chem. Soc. 2008, 130, 13718–13726.
Recent key reviews of cuprate solution structure: b) R. M.
Gschwind, Chem. Rev. 2008, 108, 3029–3053; c) T. Gartner, R. M.
5628
www.chemeurj.org
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 5620 – 5629

Catalytic 1,4-Addition of Diorganozinc Reagents
FULL PAPER
Gschwind in The Chemistry of Organocopper Compounds (Eds.: Z.
Rappoport, I. Marek), Wiley, New York, 2009.
[25] Gaussian 03, Revision C.02, M. J. Frisch, G. W. Trucks, H. B. Schle-
gel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, J. A. Montgom-
ery, Jr., T. Vreven, K. N. Kudin, J. C. Burant, J. M. Millam, S. S.
Iyengar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani,
N. Rega, G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K.
Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda,
O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian,
J. B. Cross, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann,
O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, P. Y.
Ayala, K. Morokuma, G. A. Voth, P. Salvador, J. J. Dannenberg,
V. G. Zakrzewski, S. Dapprich, A. D. Daniels, M. C. Strain, O.
Farkas, D. K. Malick, A. D. Rabuck, K. Raghavachari, J. B. Fores-
man, J. V. Ortiz, Q. Cui, A. G. Baboul, S. Clifford, J. Cioslowski,
B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. L.
Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A. Na-
nayakkara, M. Challacombe, P. M. W. Gill, B. Johnson, W. Chen,
M. W. Wong, C. Gonzalez, J. A. Pople, Gaussian, Inc., Wallingford
CT, 2004.
[26] W. J. Hehre, L. Radom, P. von R. Schleyer, J. A. Pople, Ab Initio
Molecular Orbital Theory, Wiley, New York, 1986.
[27] a) J. P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 1997, 78,
1396; b) J. P. Perdew, Phys. Rev. B 1986, 33, 8822.
[28] a) A. D. Becke, J. Chem. Phys. 1993, 98, 5648; b) B. Miehlich, A.
Savin, H. Stoll, H. Preuss, Chem. Phys. Lett. 1989, 157, 200; c) C.
Lee, W. Yang, G. Parr, Phys. Rev. B 1988, 37, 785.
[29] a) T. H. Dunning, Jr., P. J. Hay, Modern Theoretical Chemistry, Vol. 3
(Ed.: H. F. Schaefer III), Plenum Press, New York, 1976, p. 1; b) P. J.
Hay, W. R. Wadt, J. Chem. Phys. 1985, 82, 270; c) W. R. Wadt, P. J.
Hay, J. Chem. Phys. 1985, 82, 284; d) P. J. Hay, W. R. Wadt, J. Chem.
Phys. 1985, 82, 299.
[5] D. J. Berrisford, C. Bolm, K. B. Sharpless, Angew. Chem. 1995, 107,
1159–1170; Angew. Chem. Int. Ed. 1995, 42, 1059–1070.
[6] Key overviews of development of this area: a) A. Alexakis, J. E.
Bꢆckvall, N. Krause, O. Pꢇmies, M. Diꢂguez, Chem. Rev. 2008, 108,
2796–2823; b) J. Christoffers, G. Koripelly, A. Rosiak, M. Roessle,
Synthesis 2007, 1279–1300; c) T. Jerphagnon, M. G. Pizzuti, A. J.
Minnaard, B. L. Feringa, Chem. Soc. Rev. 2009, 38, 1039–1075.
[7] L. A. Arnold, R. Imbos, A. Mandoli, A. H. M. de Vries, R. Naaz,
B. L. Feringa, Tetrahedron 2000, 56, 2865–2878.
[8] T. Pfretzschner, L. Kleenmann, B. Janza, K. Harms, T. Schrader,
Chem. Eur. J. 2004, 10, 6048–6057.
[9] H. Zhang, R. M. Gschwind, Chem. Eur. J. 2007, 13, 6691–6700.
[10] R. G. Parr, W. Yang, Density Functional Theory of Atoms and Mole-
cules, Oxford University Press, New York, 1989.
[11] a) R. Imbos, M. H. G. Brilman, M. Pineschi, B. L. Feringa, Org. Lett.
1999, 1, 623–625; b) R. Imbos, M. H. G. Brilman, M. Pineschi, B. L.
Feringa, Org. Lett. 1999, 1, 1873; c) R. Imbos Ph.D. Thesis, Universi-
ty of Groningen (Germany), 2002.
[12] P. C. Bulman Page, S. A. Harkin, A. P. Marchington, M. B. Van Niel,
Tetrahedron 1989, 45, 3819–3838.
[13] a) P. de March, M. Escoda, M. Figueredo, J. Font, Tetrahedron Lett.
1995, 36, 8665–8668; b) P. G. Dormer, A. M. Kassim, J. L. Leazer, F.
Lang, F. Xu, K. A. Savary, E. G. Corley, L. DiMichele, J. O. DaSilva,
A. O. King, D. M. Tschaen, R. D. Larsen, Tetrahedron Lett. 2004, 45,
5429–5432.
[14] a) A. I. Meyers, R. C. Strickland, J. Org. Chem. 1972, 37, 2579–
2583; b) CCDC-756136 (1a), CCDC-756137 (1b), and CCDC-
756138 (3) contain the supplementary crystallographic data for this
paper. These data can be obtained free of charge from The Cam-
bridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_
request/cif.
[30] A. W. Ehlers, M. Bçhme, S. Dapprich, A. Gobbi, A. Hçllwarth, V.
Jonas, K. F. Kçhler, R. Stegmann, A. Veldkamp G. Frenking, Chem.
Phys. Lett. 1993, 208, 111.
[31] a) R. Ditchfield, W. J. Hehre, J. A. Pople, J. Chem. Phys. 1971, 54,
724; b) W. J. Hehre, R. Ditchfield, J. A. Pople, J. Chem. Phys. 1972,
56, 2257; c) P. C. Hariharan, J. A. Pople, Mol. Phys. 1974, 27, 209;
d) M. S. Gordon, Chem. Phys. Lett. 1980, 76, 163; e) P. C. Hariharan,
J. A. Pople, Theor. Chim. Acta 1973, 28, 213.
[15] G. D. Allred, L. S. Liebeskind, J. Am. Chem. Soc. 1996, 118, 2748–
2749.
[16] S. Ghersetti, A. Mangini, F. Taddei, Annali di Chimica 1965, 55,
358–364.
[17] S. Woodward, Synlett 2007, 1490–1500.
[32] a) C. Peng, P. Y. Ayala, H. B. Schlegel, M. J. Frisch, J. Comput.
Chem. 1996, 17, 49; b) C. Peng, H. B. Schlegel, Isr. J. Chem. 1993,
33, 449.
[18] a) K. B. Wiberg, Tetrahedron 1968, 24, 1083; b) Wiberg indices are
electronic parameters related to the electron density between atoms.
They can be obtained from a natural population analysis and pro-
vide an indication of the bond strength.
[19] For radical effects in 1,4-addition chemistry, see: a) K. Li, A. Alexa-
kis, Angew. Chem. 2006, 118, 7762–7765; Angew. Chem. Int. Ed.
2006, 45, 7600–7603; b) S. H. Bertz, Org. Biomol. Chem. 2005, 3,
392–394.
[33] a) J. E. Carpenter, F. Weinhold, THEOCHEM 1988, 169, 41; b) J. E.
Carpenter, Ph.D. Thesis, University of Wisconsin (Madison), 1987;
c) J. P. Foster, F. Weinhold, J. Am. Chem. Soc. 1980, 102, 7211;
d) A. E. Reed, F. Weinhold, J. Chem. Phys. 1983, 78, 4066; e) A. E.
Reed, F. Weinhold, J. Chem. Phys. 1983, 78, 1736; f) A. E. Reed,
R. B. Weinstock, F. Weinhold, J. Chem. Phys. 1985, 83, 735; g) A. E.
Reed, L. A. Curtiss, F. Weinhold, Chem. Rev. 1988, 88, 899; h) F.
Weinhold, J. E. Carpenter, The Structure of Small Molecules and
Ions, Plenum Press, New York, 1988, p. 227.
[20] J. Canisius, A. Gerold, N. Krause, Angew. Chem. 1999, 111, 1727–
1730; Angew. Chem. Int. Ed. 1999, 38, 1644–1646.
[21] J. C. Polanyi, Science 1987, 236, 680–690, and references therein.
[22] Non-SI unit of molar entropy used, e.u.=4.184 JKꢀ1 molꢀ1
.
[34] MOLEKEL 4.0, P. Flꢈkiger, H. P. Lꢈthi, S. Portmann, J. Weber,
Swiss Center for Scientific Computing, Manno, Switzerland, 2000.
[23] M. Vuagnoux-d’Augustin, A. Alexakis, Chem. Eur. J. 2007, 13,
9647–9662.
[24] A. Alexakis, J. C. Frutos, P. Mangeney, Tetrahedron: Asymmetry
1993, 4, 2431–2434.
Received: December 2, 2009
Published online: April 13, 2010
Chem. Eur. J. 2010, 16, 5620 – 5629
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5629