Thiol-promoted selective addition of ketones to aldehydes

Article abstract of DOI:10.1021/ol502937n

A simple and efficient thiol-mediated addition of ketones to aromatic and aliphatic aldehydes is reported. This thermodynamically controlled Pummerer/aldol reaction, which can tolerate both moisture and protic functional groups, provides a direct entry to syn-β-thioketones in high chemo- and regioselectivity. Mechanistic studies revealed that selective transformation of the aldehyde to an electrophilic thionium ion species concurrent with the generation of a nucleophilic vinyl sulfide coupling partner from the ketone is imposing cross-coupling over dimerization.

Full text of DOI:10.1021/ol502937n

Letter
pubs.acs.org/OrgLett
Thiol-Promoted Selective Addition of Ketones to Aldehydes
Regev Parnes,† Sachin Narute,† and Doron Pappo*
Department of Chemistry, Ben-Gurion University of the Negev, Beer-Sheva 84105, Israel
S
* Supporting Information
ABSTRACT: A simple and efficient thiol-mediated addition of ketones to
aromatic and aliphatic aldehydes is reported. This thermodynamically
controlled Pummerer/aldol reaction, which can tolerate both moisture and
protic functional groups, provides a direct entry to syn-β-thioketones in high
chemo- and regioselectivity. Mechanistic studies revealed that selective
transformation of the aldehyde to an electrophilic thionium ion species
concurrent with the generation of a nucleophilic vinyl sulfide coupling
partner from the ketone is imposing cross-coupling over dimerization.
he aldol reaction is one of the most important C−C bond
Scheme 1. (A) Pummerer/Mukaiyama Aldol Addition and
(B) Pummerer/Aldol Reaction (This Work)
T
forming reactions in synthetic organic chemistry.1 Having
an advanced to high level of sophistication, this reaction is
widely applied in the synthesis of polyoxygenated natural
products.1a Nonetheless, the required chemo-, regio-, and
stereoselectivity has been achieved only via kinetically
controlled conditions, and in most cases full enolization of
the ketone prior to the addition of the aldehyde is needed. This
step demands a stoichiometric amount of activators (base,
metal, etc.) and stringent conditions (such as low temperature,
anhydrous solvents) that render the entire process burdensome.
A search is therefore underway for methods that will enable the
selective enolization of ketones in the presence of enolizable
aldehydes and other acidic functionalities. While organo-
catalysts based on secondary amines can partially meet this
challenging goal,2 to the best of our knowledge, the exploitation
of nucleophilic vinyl sulfides as a surrogate for the enol group
has not been explored, although evidence for such reactivity can
be found in the literature.3 Herein, we report that thiols
mediate the cross-addition of aromatic and aliphatic aldehydes
to ketones. Under these conditions, all the reaction steps are
reversible, thereby enabling coupling between the nucleophilic
vinyl sulfide4 formed selectively from the ketone and the
electrophilic thionium ion species generated from the aldehyde
(Scheme 1B). Their coupling in polar solvents and in the
presence of Lewis or Brønsted acid catalysts provides a direct
entry to syn-β-thioketones5 with high chemo- and regioselec-
tivity. This elegant and sustainable C−C bond forming reaction
can tolerate both moisture and protic functional groups.
Moreover, under these conditions the process is irreversible,
and therefore the side reactions that result from attack of the
thionium ion by competitive nucleophiles affect the reaction’s
efficiency (for example, Pummerer rearrangement).
The addition of thionium ions to enol ethers is an important
variation of the Mukaiyama aldol reaction, which has been
studied by several groups.1c,4b,8 The Mukaiyama group
investigated the coupling of the dithioacetal of benzaldehyde
3a to Sn(II) enolate or the TMS-enol ether of cyclohexanone
with a stoichiometric amount of triphenylcarbenium tetra-
fluoroborate (TrBF4) as the activator (Scheme 1A).9 The
stereochemistry of the products was not determined at the
time, but the current study confirmed that the addition of
dithioacetal 3a to tin enolates is anti-selective (23:77), while
that to TMS-enol ethers is highly syn-selective (94:6). In the
current study, we explored the possibility of developing a
The reactivity of thionium ions toward nucleophiles has been
extensively studied and applied in numerous C−C bond
forming reactions for over a century.6,7 Usually, the formation
of thionium ions from sulfoxides (the Pummerer reaction)7a−d,g
or dithioacetal7h−j or directly from aldehydes and thiols
(Connective Pummerer)7k−p requires stoichiometric amounts
of an activator [such as Ac2O, Tf2O, TMSCl, dimethylmethyl-
thiosulfonium fluoroborate (DMTSF)7h,i] and/or strong Lewis
acids [TiCl4, SnCl2, Hg(OAc)2], thereby limiting the reaction
to nucleophiles that are unreactive toward the activators.
Received: October 4, 2014
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ol502937n | Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
a
thermodynamically controlled addition reaction of aldehydes
and ketones mediated by thiols (Scheme 1B).
Table 1. Optimization of Reaction Conditions
RSH (3 equiv)
catalyst (5 mol %)
First, we screened for suitable reaction conditions. The
experiments included mixing of benzaldehyde 1a (1 equiv),
cyclohexanone 2a (3 equiv), and ethanethiol (3 equiv) under
different sets of conditions. Examination of the different
solvents across the polarity spectrum [copper(II) trifluoro-
methanesulfonate (5 mol %),10 rt] showed that the
thioacetalation steps proceeded smoothly in most organic
solvents, with the exception of DMF and DMSO (Figure 1),
1a + 2a ⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯→ 5a (R = Et)
solvent, conditions
syn/
c
b
entry
catalyst
R
solvent
conditions
5a [%]
anti
1
2
3
4
5
Cu(OTf)2
Cu(OTf)2
Cu(OTf)2
Cu(OTf)2
Cu(OTf)2
TfOH
Et
Et
Et
Et
Et
Et
Et
Et
iPr
Bn
Ph
−
MeNO2
MeNO2
MeNO2
TFE
rt, 6 h
76
4.3:1
−
1:1
d
0 °C, 24 h
80 °C, 4 h
rt, 2 h
NR
70
94
92
85
80
75
5.5:1
5.5:1
11:1
7.1:1
7.1:1
2.5:1
7.1:1
1:1
TFE
10 °C, 6 h
−40 °C, 2 h
rt, 6 h
e
6
TFE
7
In(OTf)3
Sc(OTf)3
Cu(OTf)2
Cu(OTf)2
Cu(OTf)2
Cu(OTf)2
TFE
8
TFE
rt, 6 h
9
TFE
rt, 6 h
5b, 84
5c, 64
5d, 51
10
11
12
TFE
rt, 18 h
TFE
rt, 6 h
d
MeNO2
60 °C, 24 h
NR
−
a
Conditions: benzaldehyde 1a (1 equiv), cyclohexanone 2a (3 equiv),
thiol (3 equiv), and Cu(OTf)2 (5 mol %) in nitromethane or TFE (0.3
b
c
M). Isolated yield of pure product. The diastereomeric ratio was
d
e
determined by 1H NMR. NR = no reaction. TfOH (200 mol %) was
used.
The next set of experiments were performed to probe the
reaction mechanism and to clarify the stereochemical aspects of
this multicomponent transformation. In the absence of
ethanethiol both benzaldehyde 1a and cyclohexanone 2a
were recovered (Table 1, entry 12). Thus, the possibility that
the reaction proceeded via an aldol condensation/Michael
addition pathway was ruled out.11 4-tert-Butylcyclohexanone 2b
was chosen for modeling the attack direction of the thionium
ion species (Scheme 2A). In principle, four diastereoisomers 6a
(syn/equatorial attack), 6b (syn/axial attack), 6c (anti/
Figure 1. Reaction solvent screening. Conditions: 1a (0.2 mmol), 2a
(0.6 mmol), EtSH (0.6 mmol), catalyst (5 mol %), solvent (0.3 M), rt,
6 h; the products’ molar ratios were determined by HPLC analysis
using mesitylene as the internal standard. DCE = 1,2-dichloroethane,
THF = tetrahydrofuran, TFT = α,α,α-trifluorotoluene, DMF = N,N-
dimethylformamide, DMSO = dimethyl sulfoxide, TFE = 2,2,2-
trifluoroethanol.
Scheme 2. Mechanistic Study of the Addition of Aromatic
Aldehydes and Cyclohexanone Derivatives
but β-thioketone 5a was formed only in acetonitrile, nitro-
methane, and 2,2,2-trifluoroethanol (TFE), all polar solvents
with high dielectric constants that are capable of stabilizing
charge separation. Experimentation with different catalysts (see
Figure S1B in the Supporting Information (SI)) showed that
hard Lewis acids, such as CuCl2, Sc(OTf)3, Fe(OTf)3,
In(OTf)3, BF3·OEt2, and Cu(OTf)2, promoted the reaction,
while softer Lewis acids [CuBr, Zn(OTf)2, Ni(OTf)3, and
Fe(OTf)2] and Brønsted acids, such as p-toluenesulfonic acid
(PTSA) and acetic acid, were not effective. On the other hand
the reaction with trifluoromethanesulfonic acid (TfOH, 5 mol
%) reached full conversion. For reasons of ease of handling
Cu(OTf)2 was chosen as the catalyst for further investigation.
The initial screening identified conditions that produced β-
thioketone 5a in 76% yield, but with moderate diastereose-
lectivity (Table 1, entry 1). Further studies revealed that both
the stereoselectivity and the efficiency of the process were
temperature dependent (Table 1, entries 1−3). A change of the
solvent from nitromethane to TFE improved both the yield
(92%) and the diastereomeric product ratio (syn/anti = 5.5:1;
Table 1, entry 5) of the reaction. When the reaction was carried
out at −40 °C with 2 equiv of TfOH, high selectivity (syn/anti
= 11:1; Table 1, entry 6) was obtained. Other thiols may also
be used (Table 1, entries 9−11), while ethanethiol was found to
be the most effective. Notably, compounds 5a−d were obtained
as single products, and other possible side products, such as bis-
aldol addition or condensation products, were not observed.3
a
a
Conditions: (a) EtSH (3 equiv), Cu(OTf)2 (5 mol %), TFE. (b)
NaOH, EtOH−H2O (3:1), rt, 1.5 h, quantitative. (c) Similar
conditions to a, except that MeNO2 was used as the solvent.
B
dx.doi.org/10.1021/ol502937n | Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
equatorial attack), and 6d (anti/axial attack) could have been
formed in this reaction. In practice, a moderate degree of net
syn-selectivity [(6a and 6b)/(6c and 6d), ca. 4.9:1) and a high
preference for attack from the equatorial direction [(6a and
6c)/(6b and 6d), ca. 9.4:1)] were observed. Overall, β-
thioketone 6a was isolated in 71% yield after column
chromatography. The possibility that this selectivity derives
from the difference in stability of the isomers was ruled out,
since the thia-Michael addition of ethanethiol to 2-benzylidene-
4-(tert-butyl)cyclohexanone (7) under similar reaction con-
ditions [Cu(OTf)2 (5 mol %), rt] was found to be nonselective
in MeNO2, affording the four diastereoisomers 6a−d in equal
ratios, and inefficient in TFE.
Scheme 3. Addition of Aliphatic Aldehydes to Ketones
Based on steric arguments and previous findings, it may be
assumed that the ethyl group of the thionium ion is cis to the
H-atom (Scheme 2B).8f The fact that similar stereoratios are
observed for different Lewis and Brønsted acids suggests that
the catalysts are not intimately involved with the thionium ion
during the transition states.1c Therefore, an open transition
state model8f,12 is more likely. It is suggested that nonbonded
interactions between R4 of the thionium ion and the
cyclohexane ring in TS A (giving the anti-product) cause a
destabilizing effect, which favors the syn-product through TS S.
This C−C bond forming step is highly selective and leads to
high diastereoselectivity under kinetic conditions (low temper-
ature for short reaction time, as in Table 1, entry 6). Yet, under
thermodynamic conditions (a prolonged reaction time at a
temperature >0 °C), the stereochemical outcome is signifi-
cantly affected due to several reversible processes. Elimination
of the thiol group from the product followed by nonselective
thia-Michael addition is a major concern.13 Another destructive
process is the retro-addition cleavage to return starting
materials. When the reaction’s product β-thioketone 5a (initial
dr = 5.8:1) was resubmitted to the reaction conditions,
dithioacetal 3a was detected by HPLC analysis in 70% yield
together with a 30% yield of 5a (dr = 1:1, Scheme 2C).
Next, the Pummerer/aldol addition of aliphatic aldehydes to
ketones was examined. It would be quite advantageous to
couple ketones with aliphatic aldehydes, since the aldol addition
of ketones to aliphatic aldehydes is not favorable from both
thermodynamic and kinetic points of view, and in most cases
this reaction resulted exclusively in self-coupling of the
aldehyde. Indeed, this result was obtained when isovaleralde-
hyde 1c was reacted with 4′-hydroxyacetophenone 2c in the
absence of thiol [Cu(OTf)2, MeNO2, rt]. However, the
addition of ethanethiol changed the mechanistic scheme, and
the cross-addition product 8 was isolated in 50% yield
[Cu(OTf)2 (10 mol %), TFE, 90 °C, sealed tube, Scheme
3A]. The foundations for this significant chemoselective
process are based on prior observations by the Cohen group,
who reported that the elimination of thiophenol from
dithioacetals of ketones to form vinyl sulfides takes place
several orders of magnitude faster than its elimination from
dithioacetals of aldehydes.14 The rate of this E1-type
elimination depends on the stability of the carbonium ion
that remains after the loss of one thiol group.15 To further
probe this unique chemoselectivity, aldehyde 1c was mixed
with ethanethiol under similar reaction conditions (rt to 60 °C,
24 h). GC-MS and NMR analyses of the crude reaction
products (see SI) revealed that dithioacetal 3b was formed
almost exclusively with only a trace amount of homodimeriza-
tion products (Scheme 3B). In contrast, when 2-isopropyl-5-
methyl-2-hexenal (9), which is the self-aldol condensation
product of 1c, was reacted with ethanethiol [Cu(OTf)2 (10 mol
%), TFE, 60 °C, 5 h], dithioacetals 10 and 3b were obtained in
a 4.6:1 ratio. The formation of 3b from 10 may take place via
thionium ion intermediate I. These experiments suggest that
the homocoupling of dithioacetal 3b is both kinetically
unfavorable and reversible under the reaction conditions,
emphasizing the mechanistic distinction between this three-
component addition reaction and the classic aldol reaction.
Overall, we propose that the formation of an electrophilic
thionium ion II (Scheme 4, eq 1) from the aldehyde and the
Scheme 4. Proposed Mechanism
concurrent selective generation of a nucleophilic vinyl sulfide
III from the ketone (Scheme 4, eq 2) are the key factors that
dictate the unusual chemoselectivity observed in this reaction.
Finally, we set out to investigate the scope of this unique
thiol-mediated reaction (Figure 2). It is clear that both
electron-deficient and -rich aromatic aldehydes are suitable
coupling partners. The addition of benzaldehyde 1a to 3-
methylcyclohexanone at low temperature afforded product 15
(77% yield) in high regioselectivity, and the coupling of 1a with
2-methylcyclohexanone under kinetic conditions (−40 °C)
took place at C-6 affording compound 16a in 70% yield (dr =
1:1:5:12), whereas, under thermodynamically controlled
conditions (rt) a mixture of two regioisomers 16a (34%
yield, dr = 0:1:2:31) and the C-2 substitution product, 16b
(24% yield, single isomer), were obtained. In general, the
addition of aliphatic aldehydes required harsher conditions, as
the reactions failed to reach full conversion (compounds 14−
15 and 26−27). Importantly, protic functional groups, such as
−OH (products 21a, 25−26) and CO2H (23), did not require
any protection.
C
dx.doi.org/10.1021/ol502937n | Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
(2) (a) Wende, R. C.; Schreiner, P. R. Green Chem. 2012, 14, 1821−
1849. (b) Dondoni, A.; Massi, A. Angew. Chem., Int. Ed. 2008, 47,
4638−4660. (c) MacMillan, D. W. Nature 2008, 455, 304−8.
(d) Seayad, J.; List, B. Org. Biomol. Chem. 2005, 3, 719−724.
(e) Dalko, P. I.; Moisan, L. Angew. Chem., Int. Ed. 2004, 43, 5138−
5175. (f) Notz, W.; Tanaka, F.; Barbas, C. F., III. Acc. Chem. Res. 2004,
37, 580−591. (g) Dalko, P. I.; Moisan, L. Angew. Chem., Int. Ed. 2001,
40, 3726−3748.
(3) (a) Kumar, A.; Akanksha. Tetrahedron Lett. 2007, 48, 8730−8734.
(b) Mukaiyama, T.; Narasaka, K.; Hokonoki, H. J. Am. Chem. Soc.
1969, 91, 4315−4317.
(4) (a) Zyk, N. V.; Beloglazkina, E. K.; Belova, M. A.; Dubinina, N. y.
S. Russ. Chem. Rev. 2003, 72, 769−786. (b) Takeda, T.; Kaneko, Y.;
Fujiwara, T. Tetrahedron Lett. 1986, 27, 3029−3032. (c) Rodriguez, A.
D.; Nickon, A. Tetrahedron 1985, 41, 4443−4448.
(5) Kondo, T.; Mitsudo, T.-a. Chem. Rev. 2000, 100, 3205−3220.
(6) (a) Smith, L. H.; Coote, S. C.; Sneddon, H. F.; Procter, D. J.
Angew. Chem., Int. Ed. 2010, 49, 5832−5844. (b) Feldman, K. S.
Tetrahedron 2006, 62, 5003−5034. (c) Bur, S. K.; Padwa, A. Chem.
Rev. 2004, 104, 2401−2432.
(7) (a) Peng, B.; Huang, X.; Xie, L.-G.; Maulide, N. Angew. Chem.,
Int. Ed. 2014, 53, 8718−8721. (b) Huang, X.; Maulide, N. J. Am. Chem.
Soc. 2011, 133, 8510−8513. (c) Feldman, K. S.; Fodor, M. D. J. Am.
Chem. Soc. 2008, 130, 14964−14965. (d) Feldman, K. S.; Karatjas, A.
G. Org. Lett. 2006, 8, 4137−4140. (e) Padwa, A.; Bur, S. K.
Tetrahedron 2007, 63, 5341−5378. (f) Padwa, A.; Nara, S.; Wang, Q.
Tetrahedron Lett. 2006, 47, 595−597. (g) Padwa, A.; Waterson, A. G. J.
Org. Chem. 2000, 65, 235−244. (h) Trost, B. M.; Sato, T. J. Am. Chem.
Soc. 1985, 107, 719−21. (i) Trost, B. M.; Murayama, E. J. Am. Chem.
Soc. 1981, 103, 6529−30. (j) Mukaiyama, T.; Narasaka, K.; Hokonoki,
H. J. Am. Chem. Soc. 1969, 91, 4315−4317. (k) Smith, L. H.; Nguyen,
T. T.; Sneddon, H. F.; Procter, D. J. Chem. Commun. 2011, 47,
10821−10823. (l) Miller, M.; Vogel, J. C.; Tsang, W.; Merrit, A.;
Procter, D. J. Org. Biomol. Chem. 2009, 7, 589−597. (m) Ovens, C.;
Vogel, J. C.; Martin, N. G.; Procter, D. J. Chem. Commun. 2009, 3101−
3103. (n) Ovens, C.; Martin, N. G.; Procter, D. J. Org. Lett. 2008, 10,
1441−1444. (o) Miller, M.; Tsang, W.; Merritt, A.; Procter, D. J.
Chem. Commun. 2007, 498−500. (p) McAllister, L. A.; McCormick, R.
A.; James, K. M.; Brand, S.; Willetts, N.; Procter, D. J. Chem.Eur. J.
2007, 13, 1032−1046.
(8) (a) Trost, B. M.; Burns, A. C.; Bartlett, M. J.; Tautz, T.; Weiss, A.
H. J. Am. Chem. Soc. 2012, 134, 1474−1477. (b) Mori, I.; Ishihara, K.;
Flippin, L. A.; Nozaki, K.; Yamamoto, H.; Bartlett, P. A.; Heathcock,
C. H. J. Org. Chem. 1990, 55, 6107−6115. (c) Mori, I.; Bartlett, P. A.;
Heathcock, C. H. J. Am. Chem. Soc. 1987, 109, 7199−7200. (d) Trost,
B. M.; Sato, T. J. Am. Chem. Soc. 1985, 107, 719−721. (e) Ohshima,
M.; Murakami, M.; Mukaiyama, T. Chem. Lett. 1985, 1871−1874.
(f) Mori, I.; Bartlett, P. A.; Heathcock, C. H. J. Org. Chem. 1990, 55,
5966−5977.
Figure 2. Substrates scope. For the exact reaction conditions, see SI.
In summary, a simple and efficient thiol-mediated addition of
ketones to aromatic and aliphatic aldehydes was developed.
There are several reasons why this Pummerer/aldol reaction is
unique. Namely, both aldehyde and ketone coupling partners
are selectively activated by the same thiol. The incorporation of
vinyl sulfide as a surrogate for enolate eliminates the need to
activate the ketone in advance. The reaction exhibits a distinctly
mechanistic scheme compared to the classic aldol reaction, with
all steps being reversible. This method provides a single-step
process for the synthesis of important β-thioketone building
blocks.16 Finally, this multicomponent process can be carried
out in air in the presence of water and protic functional groups
(protective groups are not needed).
ASSOCIATED CONTENT
* Supporting Information
■
S
(9) Ohshima, M.; Murakami, M.; Mukaiyama, T. Chem. Lett. 1985,
14, 1871−1874.
(10) Copper(II) trifluoromethanesulfonate from Strem [catalog no.
29-5000] was used in this study.
Full experimental procedures, characterization data, and NMR
spectra. This material is available free of charge via the Internet
at http://pubs.acs.org.
(11) (a) Perin, G.; Mesquita, K.; Calheiro, T. P.; Silva, M. S.;
Lenardao, E. J.; Alves, D.; Jacob, R. G. Synth. Commun. 2013, 44, 49−
̃
AUTHOR INFORMATION
Corresponding Author
■
58. (b) Abaee, M. S.; Cheraghi, S.; Navidipoor, S.; Mojtahedi, M. M.;
Forghani, S. Tetrahedron Lett. 2012, 53, 4405−4408.
(12) Heathcock, C. H.; Hug, K. T.; Flippin, L. A. Tetrahedron Lett.
1984, 25, 5973−5976.
*E-mail: pappod@bgu.ac.il.
Author Contributions
†R.P. and S.N. contributed equally.
Notes
(13) Alt, I.; Rohse, P.; Plietker, B. ACS Catal. 2013, 3, 3002−3005.
(14) (a) Oida, T.; Tanimoto, S.; Ikehira, H.; Okano, M. Bull. Chem.
Soc. Jpn. 1983, 56, 959−960. (b) Cohen, T.; Herman, G.; Falck, J. R.;
Mura, A. J. J. Org. Chem. 1975, 40, 812−813.
(15) Cohen, T.; Mura, A. J.; Shull, D. W.; Fogel, E. R.; Ruffner, R. J.;
Falck, J. R. J. Org. Chem. 1976, 41, 3218−3219.
(16) Guha, C.; Mondal, R.; Pal, R.; Mallik, A. Chem. Sci. 2013, 125,
1463−1470.
The authors declare no competing financial interest.
REFERENCES
■
(1) (a) Kan, S. B. J.; Ng, K. K. H.; Paterson, I. Angew. Chem., Int. Ed.
2013, 52, 9097−9108. (b) Machajewski, T. D.; Wong, C.-H. Angew.
Chem., Int. Ed. 2000, 39, 1352−1375. (c) Mahrwald, R. Chem. Rev.
1999, 99, 1095−1120.
D
dx.doi.org/10.1021/ol502937n | Org. Lett. XXXX, XXX, XXX−XXX