Scope and mechanism of the highly stereoselective metal-mediated domino aldol reactions of enolates with aldehydes

Article abstract of DOI:10.3762/bjoc.12.80

A one-pot transformation, which involves the reaction of ketones with aldehydes in the presence of metal halides to furnish tetrahydro-2H-pyran-2,4-diols in a highly diastereoselective manner, is investigated thoroughly by experiments and computations. Th

Full text of DOI:10.3762/bjoc.12.80

Scope and mechanism of the highly stereoselective
metal-mediated domino aldol reactions of enolates
with aldehydes
M. Emin Cinar*, Bernward Engelen, Martin Panthöfer, Hans-Jörg Deiseroth, Jens Schlirf
and Michael Schmittel*
Full Research Paper
Open Access
Address:
Beilstein J. Org. Chem. 2016, 12, 813–824.
Department Chemie-Biologie, OC1, Universität Siegen,
Adolf-Reichwein-Str., D-57068 Siegen, Germany
doi:10.3762/bjoc.12.80
Received: 17 November 2015
Accepted: 05 April 2016
Published: 27 April 2016
Email:
M. Emin Cinar* - emin.cinar@uni-siegen.de;
Michael Schmittel* - schmittel@chemie.uni-siegen.de
Associate Editor: S. Bräse
*
Corresponding author
©
2016 Cinar et al; licensee Beilstein-Institut.
Keywords:
License and terms: see end of document.
DFT; diastereoselective reaction; domino aldol; metal enolate;
tetrahydropyran
Abstract
A one-pot transformation, which involves the reaction of ketones with aldehydes in the presence of metal halides to furnish tetra-
hydro-2H-pyran-2,4-diols in a highly diastereoselective manner, is investigated thoroughly by experiments and computations. The
reaction was also successfully implemented on a flow micro reactor system.
Introduction
Since its discovery in the late nineteenth century the aldol reac- of their high potential to control the stereochemical outcome of
tion has become one of the most powerful tools in the field of the bond formation [28-30]. However, the other group III metal
carbon–carbon bond formation [1-5]. It is widely used in the enolates have been almost completely omitted over the years
formation of many natural products [6-11], stereoselective syn- [31]. We have already reported a domino aldol–aldol–hemi-
theses [12-16], and tandem reactions [17-19]. While the latter acetal process that furnishes racemic tetrahydro-2H-pyran-2,4-
processes usually comprise only one aldol reaction, tandem diols in a highly stereoselective manner (Scheme 1) [30-37].
reaction sequences containing two consecutive aldol steps are
mostly limited to the trimerization of enolates [20-22].
In this paper, the domino aldol–aldol–hemiacetal reaction
involving several metals (Al, Ga, In, Ti, Zr, Sn), and various
Metal enolates (Ti [23], Zr [24], Si [25], and Sn [26]) and boron aldehydes and ketones is studied experimentally and computa-
enolates [27] have adopted a considerable significance because tionally.
813

Beilstein J. Org. Chem. 2016, 12, 813–824.
Scheme 1: Synthesis of racemic tetrahydro-2H-pyran-2,4-diols rac-5 from enolates 2 and aldehydes 3.
Results
Metal effect
Sc(OTf)3 furnished 6a in 25, 52, 10 and 55% yield, respective-
ly. In almost all cases the mono aldolate 6a was the main prod-
The experiments were performed to screen suitable metal frag- uct at lower temperatures, e.g., at 0 °C, along with the domino
ments for their ability to promote the domino aldol reaction by aldol product 5a obtained in the range of 7–26% yield.
studying the reaction between propiophenone (1a) and
benzaldehyde (3a: Ar = Ph) (Scheme 2). The enolate was Higher yields of 5a were obtained at 0 °C in the presence of In
generated from propiophenone by deprotonation with lithium and Zr. Interestingly, in the case of gallium at 0 °C and Sn(IV)
diisopropylamide (LDA) at −40 °C in tetrahydrofuran (THF) at 25 °C a diastereomer of 5a also formed in 20 and 29% yields,
and was subsequently reacted with 0.33 equivalents of MCl3 or respectively. Generally, with nearly all metals, the yield of 5a
0
.25 equivalents of MCl4, respectively. The resulting metal increased dramatically when the temperature was raised to
enolate was then treated with a stoichiometric amount of 25 °C, but dropped at higher temperatures. The decreased yield
benzaldehyde (3a) and stirred for 2 h at 0 °C, room temperature at 67 °C may emerge from the irreversible formation of the
or 67 °C. The hemiacetal 5a was obtained in varying yields aldol condensation product 6ac, which is obtained in 73% yield
along with some amount of the monoaldol 6a [38,39] (obtained in presence of AlCl3, on expense of 5a.
as a mixture of two diastereomers; syn/anti ≈ 1:1) and conden-
sation product 6ac (Scheme 2, Table 1).
The X-ray structure of racemic 5a (from ethanol) could not be
solved due to the presence of a solid racemate in the orthorhom-
Aside of the ions mentioned in Table 1 the metal-mediated bic space group P212121 (no. 19). Three axial hydroxy groups,
domino aldol reaction was also probed with LaCl3, La(OTf)3, whose probability of allocation at C2 and C6 is 0.5, respective-
CeCl3, Sc(OTf)3, BF3 and SnCl2 resulting in failure. While ly, appear to be attached to the pyran ring. This can easily be
SnCl2 afforded 6a in 90% yield, LaCl3, La(OTf)3, CeCl3, and explained by the superposition of two enantiomers which are
Scheme 2: Synthesis of rac-5a–j and monoaldol products 6a–i and 6ac–ic as obtained from propiophenone (1a) in the presence of metal chloride
and substituted arylaldehydes.
814

Beilstein J. Org. Chem. 2016, 12, 813–824.
Table 1: Effect of metals on the domino aldol reaction of 1a and 3a at different temperatures (reaction time: 2 h) on the yield of 5a and 6a.
Yield of 5a and 6a at various temperatures
Entry
Metal
at 0 °C
a (%)
at 25 °C
at 67 °C
5
6a (%)
5a (%)
6a (%)
5a (%)
6a (%)
1
2
3
4
5
6
Al
18
19
22
3
64a
8
8
−
−
2
5
−
5
Ga
In
17 (+20)b
70
19
13
14
23
22
67
60
50a
21
17
76
19
61
7
85a
Tib
Zr
26
10
20
38
76c
Sn(IV)
36 (+29)b,c
aRef. [34]. bA diastereomer of 5a. cRef. [37].
statistically and isoconformationally incorporated in the crystal
lattice. Separation of two enantiomers was achieved by using
chiral column chromatography (Chiralpak AD, Daicel) fol-
lowed by recrystallization from H2O/MeOH (1:4) providing the
appropriate crystal for X-ray analysis. Accordingly, the enan-
tiopure crystal of 5a was unambiguously assigned to a tetra-
hydro-2H-pyran-2,4-diol structure with all phenyl and methyl
groups occupying equatorial positions while the hydroxy groups
are placed in axial positions with an allocation probability of
1
.0 each (Figure 1). Within the crystal lattice the molecules
arrange in a chain along the a-axis, so that each molecule is
twisted by 180° against each other. Additionally, there are alter-
nating inter- and intramolecular hydrogen bonds between the
hydroxy groups.
Mechanistic aspects
To shed light on the mechanism, the metal to enolate ratio was
varied (Table 2), while keeping the optimum temperature for
each metal as determined in the previous experiments. The reac-
Figure 1: Crystal structure of enantiopure 5a [40].
tion was already successful with two equivalents of enolate per enolate had different effects on the reactions depending on the
metal fragment. However, higher yields were obtained at higher metals. While the yield was decreased with zirconium, it in-
loadings. For example, zirconium worked best with three creased both with aluminum and indium. In the latter case,
enolate units and tin with four. Surprisingly, an excess of an >99% yield was obtained.
Table 2: Dependence on the stoichiometric amount of propiophenone (1a) enolate with regard to the metal (reaction time: 2 h, reaction temperature:
25 °C, 5a/6a in %).
Amount of propiophenone (1a)
2 equiv 3 equiv
Metal
chloride
Entry
4 equiv
6 equiv
5a
6a
5a
6a
5a
6a
5a
6a
1
2
3
4
ZrCl4
SnCl4
AlCl3
InCl3
33
20
–
48
45
–
76
14
66
85
23
35
8
68
36
–
22
22
–
–
–
–
–
66
>99
34
–
–
–
13
–
–
815

Beilstein J. Org. Chem. 2016, 12, 813–824.
In the same manner the influence of the amount of aldehyde the molecular preorganization, which is necessary for the reac-
was examined. As can be seen from the data in Table 3 the ratio tion. The reactions carried out with 1.0, 0.75 and 0.5 equiva-
of 5a to 6a decreases dramatically by increasing the amount lents of zirconium provided 5a in 76, 42 and 14% yield, respec-
of aldehyde two-fold, owing to the formation of higher amounts tively. Reducing the amount to 0.25 equivalents of zirconium
of the monoaldol product in presence of excess aldehyde furnished only 3% yield.
(
Scheme 2).
Even the concentration influences the yield of 5a. The
In order to check whether this outcome is the result of thermo- optimum concentration is 375 mM, in which the reaction
dynamic control, a second aldehyde was added to the reaction afforded 76% yield of 5a. In case of 750 mM, the yield
mixture after 2 h. The larger the amount of aldehyde in the reac- decreases drastically to 40%. However, lower concentrations
tion, the higher is the yield of monoaldol product 6a, which such as 250 mM and 187 mM do not have such an obvious in-
supports a thermodynamically controlled equilibrium as further fluence and furnish the expected product in 61 and 59% yields,
confirmed by the following observations: (1) with gallium(III) respectively.
two diastereomeric tetrahydro-2H-pyran-2,4-diols are formed at
0
°C and only one (i.e., 5a) at elevated temperature; (2) at A time dependency study clearly showed that the best yields
higher temperature, the product yield of 5a is gradually reduced were achieved after 2 hours and longer reaction times did not
on account of new 6ac (only in the case of AlCl3), which is lead to improved yields (Table 4).
expected to be formed irreversibly from the metal-bound
monoaldolate; (3) when excess benzaldehyde (3a) was added to Variation of enolate and aromatic aldehyde
the reaction with AlCl3, the yield of domino product 5a gradu- Subsequently, various aldehydes were tested in the domino
ally decreased from, for example, 42% to 30% with 1.5 and aldol reaction with propiophenone enolate in combination with
2
.0 equivalents of benzaldehyde, respectively.
different metals (Table 5). All aromatic aldehydes, even those
containing strongly coordinating substituents such as the
Likewise, the amount of metal chloride influences the yield of dimethylamino group, are accepted in this transformation. With
5
a. The higher the amount of metal chloride the less likely is anthracene-9-carbaldehyde (3f), however, the yields drastically
Table 3: Stoichiometry dependence on the amount of aldehyde (reaction time: 2 h, reaction temperature: 25 °C, 5a/6a in %).
In presence of benzaldehyde (3a) in
Entry
Metal
1 equiv
1.5 equiv
2 equiv
4 equiv
5a
6a
5a
6a
5a
6a
5a
6a
1
2
3
Zr
Al
In
65
66
78
7.0
8
40
42
53
16
39
30
19
30
34
31
40
42
5
–
–
60
–
13
–
Table 4: Time dependency of domino aldol reactions in the presence of various metal chlorides (reaction temperature: 25 °C, 5a/6a in %).
AlCl3
GaCl3
InCl3
ZrCl4
SnCl4
Entry
Time
5a
6a
5a
6a
5a
6a
5a
6a
5a
6a
1
2
3
4
5
6
10 min
30 min
60 min
34
39
48
14
11
10
8
15
22
42
70
−
70
61
40
19
–
56
60
68
87
73
75
9
23
21
43
68
67
64
26
20
16
22
6
15 (20)a 40
21 (19)a 18
34 (29)a 17
36 (29)a 22
10
10
-
120 min 66
1 day
–
–
–
3
–
–
–
–
5 days
–
–
–
5
5
aSecond diastereomer.
816

Beilstein J. Org. Chem. 2016, 12, 813–824.
decreased, most likely due to steric hindrance. Although 5b−j as in 5a since coupling constants, the shift of the 2-CH3
benzaldehyde (3a) was not used, product 5a appeared in the group and the coupling constant J(4-H, 5-H) agreed.
reaction of anthracene-9-carbaldehyde (3f) with Al and Zr
metals, a finding that requires an explanation (vide infra). The The facile formation of domino products from aromatic alde-
NMR investigations suggest the same relative configuration of hydes proposed to use this reaction also with aromatic dialde-
Table 5: Reactions of various aldehydes with propiophenone metal enolate (reaction time: 2 h, reaction temperature: 25 °C).
Yield of 5 [%]
Entry
Product
Aldehyde
AlCl3
InCl3
SnCl4
ZrCl4
40
1
2
3
4
5b
5c
5d
5e
14
47
34
3b
3c
3d
3e
44
17
–
94
72
50
36
–
60a
45
28a
45
5
5f
13b
55
–
29b
3f
6
7
5g
5h
–
–
62
50
–
–
66
51
3
g
h
3
8
9
5i
–c
–
53
35
–e
–c
–
–c
30
–e
3
i
5jd
5k
3j
10
–
–
3k
817

Beilstein J. Org. Chem. 2016, 12, 813–824.
Table 5: Reactions of various aldehydes with propiophenone metal enolate (reaction time: 2 h, reaction temperature: 25 °C). (continued)
11
5l
–
–f
–
–f
3l
aRef. [37]. bAdditionally 10% (Al) and 20% (Zr) of 5a are formed. cThe reaction was not carried out. d1,4-Bis-(2,4-dimethyl-3,5-diphenyl-3,5-dihydroxy-
tetrahydropyranyl)-benzene. eInseparable mixture. fNo reaction.
hydes (3j−l). As anticipated the reaction proceeded smoothly analysis. The crystals are in the space group P21/c. It forms
with terephthalaldehyde (3j) giving rise to product 5j (see "fibers" with alternating incorporation of the two enantiomers
Scheme 2 and Table 5), while isophthalaldehyde (3k) provided of 7h and they are held together by hydrogen bonds (intramo-
a mixture of isomers, which were not separable. The steric lecular: 1.90 Å, intermolecular: 1.93 Å) (Figure 2).
congestion in o-phthalaldehyde (3l) precluded the formation of
the domino-aldol product. However, it is known that o-phthal- Various aldehydes were also probed in the domino aldol reac-
aldehyde (3l) provides the corresponding aldol product in the tion with the indium enolate of butyrophenone (1f, Scheme 4).
presence of base [41].
Reactions involving 2-furfural, cinnamaldehyde, butyraldehyde,
isobutyraldehyde and 2-phenylpropanal did not provide the cor-
The variability in the ketone moiety proved to be rather responding tetrahydro-2H-pyrans, while the reaction with alde-
restricted (Scheme 3). While in the case of propiophenone (1a) hydes possessing p-NMe2 (3b), p-F (3c) and p-MeO (3d)
and butyrophenone (1f) moderate to good yields (≥50%) were substituted phenyl units worked in reasonable yields affording
obtained, the reaction with acetophenone (1e) did not furnish 8b–d.
any domino aldol product at all. The only acyclic aliphatic ke-
tone that led to the formation of the domino aldol product was The electronic influence of substituents at the aldehyde and/or
pinacolone (1d), which gave the tetrahydro-2H-pyran-2,4-diol ketone moiety was more systematically analyzed using series of
7
d in 20% yield (Scheme 3, Table 6). The reactions of the benzaldehydes (3a, 3c and 3d) and propiophenones (1a–c) both
cyclic ketones were only successful in the case of cyclo- substituted by H, OMe and F at the para-phenyl position.
hexanone (1h), while transformations with five- (1g) and seven- Yields increased with time as expected. The introduction of two
(
1i) membered rings failed most likely due to strain.
F or MeO substituents leads to a decrease of the yield com-
pared with 5a (Scheme 5, Table 7). The best yields were ob-
The structure of product 7h, which precipitated from the crude tained within a series in the case of ketone and aldehyde
mixture in crystalline form, was solved by single crystal X-ray possessing a donor–acceptor situation.
Scheme 3: Reaction of various ketones (1b−i) with benzaldehyde (3a) in the presence of InCl3 and ZrCl4.
818

Beilstein J. Org. Chem. 2016, 12, 813–824.
Table 6: Variation of the ketone in the domino aldol reaction with benzaldehyde (3a) in the presence of InCl3 and ZrCl4 (reaction time: 2 h, reaction
temperature: 25 °C).
Yield of 7 [%]
Yield of 7 [%]
Entry
Product Ketone
Product Ketone
InCl3
ZrCl4
InCl3
ZrCl4
1
7b
–a
40
7fb
70
50
1b
1f
2
3
4
7c
7d
7e
–a
20
–
45
–
7g
7h
7i
–
5
–
–
5
–
1
g
1
c
d
1
1h
–
1e
1i
aThe reaction was not carried out. bRef. [42].
Figure 2: (a) Crystal structure of 7h and (b) its arrangement in the crystal [43].
Scheme 4: Reaction of n-butyrophenone (1f) with various aldehydes (3b−d) in presence of InCl3 (reaction time: 2 h, reaction temperature: 25 °C).
Domino aldol reaction by using a CYTOS™
microreactor
rate of 1 mL min−1 caused precipitation of the compounds and
consequently blocking the reactor, flow rates from 2 mL min−1
Using the CYTOS™ microreactor, a continuous reactor, the onwards were used to run the reaction (Table 8). The break-
following results were obtained, which are well in agreement down of the yields with increasing flow rates is easily ex-
with the outcome of the batch experiments. Since a slow flow plained with the short reaction time.
819

Beilstein J. Org. Chem. 2016, 12, 813–824.
Scheme 5: Domino aldol reactions of different aldehydes and ketones possessing p-H, p-F and p-MeO substituents at the phenyl units with ZrCl4.
Products see Table 7.
Table 7: Time dependent domino aldol reactions of different aldehydes and ketones having H, F and MeO units at para-position of phenyl units using
ZrCl4. Yields are in % and numbers in brackets are for mono aldol products (reaction time: 2 h, reaction temperature: 25 °C).
1
a
1b
F (3c)
9a
1c
F (3c)
9c
H (3a)
F (3c)
MeO (3d) H (3a)
MeO (3d) H (3a)
MeO (3d)
3a–c
Entry Time [min]
5a
5c
5d
7a
9b
7b
9d
5
9
(+9)
7
5
5
8
25
(+7)
26
(+11)
15
(+19)
1
2
3
4
5
6
7
5
(+10)
(+12)
(+15)
(+18)
(+14)
8
23
(+10)
10
(+15)
6
8
10
(+15)
28
(+8)
26
(+11)
17
(+15)
10
20
30
45
60
120
(+16)
(+20)
(+19)
19
27
(+8)
15
(+17)
8
12
(+15)
14
(+17)
29
(+7)
28
(+20)
20
(+9)
(+16)
(+16)
27
35
(+10)
16
(+16)
12
(+16)
15
(+19)
20
(+16)
32
(+12)
32
(+12)
25
(+11)
(+18)
44
51
(+9)
20
(+13)
19
(+19)
20
(+17)
30
(+17)
37
(+6)
36
(+17)
29
(+10)
(+19)
60
52
(+10)
35
(+12)
24
(+18)
21
(+18)
37
(+15)
40
(+6)
44
(+11)
30
(+9)
(+20)
76
62
(+7)
45
(+5)
40
(+20)
29
(+17)
54
(+14)
45
(+4)
58
(+14)
32
(+14)
(+8)
Table 8: Reaction of benzaldehyde (3a) with indium propiophenone enolate in the CYTOSTM Labsystem (5a/6a in %).
2
5
−
2 °C
34 °C
47 °C
Flow rate
mL min−1]
Average rxn.
time [min]
Entry
[
a
6a
5a
6a
5a
6a
1
2
3
4
5
1
2
3
5
9
−
−
−
−
−
60
30
20
12
6
62
63
21
20
12
14
12
8
59
58
20
22
10
14
9
50
48
12
25
14
12
6
6
5
820

Beilstein J. Org. Chem. 2016, 12, 813–824.
Discussion
on hexa-coordinate indium [51] as implemented in Gaussian 09
The experimental results indicated that two diastereomeric tetra- owing to the predicted good results in our earlier work [30]. The
hydro-2H-pyran-2,4-diols were formed at 0 °C in the presence remaining coordination sites of indium were occupied by two
of gallium and at 25 °C with Sn(IV), but only one (i.e., 5a) at THF molecules. The minima and transition states of the calcu-
elevated temperature. Such finding is indicative of thermo- lated structures were verified by analyzing the harmonic vibra-
dynamic control in the reaction. In the first step, presumably a tional frequencies, using analytical second derivatives. To
metal di-, tri- or tetraenolate is formed based on the ratio of predict the energies plausibly, as recommended for organome-
enolate to metal chloride. Because in principle a metal dieno- tallic compounds, single point calculations with M06 func-
late is sufficient as the nucleophilic component, additional tional [52] were performed using the same basis sets
enolate ligands may simply act as "innocent bystander ligands" (Scheme 6).
in the reaction cascade. Since we have been able to obtain 5a
with GaCl3 in a diastereomerically pure form at 25 °C – Complexation of the metal enolate E with benzaldehyde
although at low temperature a sizeable amount of 20% of a (PhCHO) is followed by the exergonic first aldol addition
second diastereomer was formed – it is reasonable to assume a showing a small activation barrier of 1.82 kcal mol−1 via a
reversible formation of the metal-bound tetrahydro-2H-pyran- half−chair like transition state (TS-C-A1), which is in accord
2
,4-diol. Under thermodynamic control all large substituents R with the anti-selective aldol addition of titanium enolates
methyl, phenyl) are placed in the equatorial position which [53,54]. TS-C-A1 leads to the formation of anti-aldolate A1,
leads for all metal ions excluding Ga to only 1 out of 16 possessing ΔGrel of −6.60 kcal mol−1. In the next step, A1 is
(
possible diastereoisomers.
attacked by a second enolate at higher temperature via the
bicyclic transition state TS-A1-A2 (ΔGrel = 4.18 kcal mol−1)
To shed more light on the mechanism, DFT calculations were with a chair–chair conformation. In the last step, intramolecular
carried out using the Gaussian 09 program [44]. Gas-phase opti- cyclization with a relative TS energy of 1.22 kcal mol−1 (TS-
mization of geometries was performed by using the B3LYP A2-A3) takes place, which furnishes the metal-bound hemi-
[
45-47] method with Pople’s split-valence 6-31G(d) basis set on acetal in a boat conformation (A3) with a relative free energy of
C, H, O atoms and double-ζ quality basis set (LANL2DZ) [48- −7.49 kcal mol−1. Hydrolysis of A3 provides tetrahydro-2H-
5
0] containing Hay and Wadt’s effective core potential (ECP) pyran-2,4-diol 5a.
Scheme 6: DFT calculations on the formation of A3, hydrolysis of which provides 5a, at M06/6-31G(d)/LANL2DZ//B3LYP/6-31G(d)/LANL2DZ level
(ΔGrel with unscaled zpe are in kcal mol−1).
821

Beilstein J. Org. Chem. 2016, 12, 813–824.
Computationally predicted A2 has a lower free energy than the tive, and (c) display a sufficiently large ion radius so that the
hemiacetal A3, which is responsible for the formation of prod- reaction cascade can take place in the periphery of the metal.
uct 5a. This finding suggests that the hydrolysis occurs on The latter requirement is apparently prohibitive for a boron(III)-
the stage of A2 furnishing A2OH. Afterwards, intramolecular mediated reaction because the ion radius of the B3+ ion is very
ring closure of A2OH with a relative activation barrier of small (25 pm). In the case of tin(II), a high yield of 6a (up to
2
3.8 kcal mol−1 leads to 5a in an exergonic process (Scheme 7). 90%) was observed but formation of 5a was not detected which
At higher temperature, the condensation product 6ac emerges indicates a hindrance for the second aldol addition. Presumably
from the dehydration of 6a, taking place via an irreversible due to the low charge density of the Sn2+ ions the second car-
reaction, which is accountable for the decrease of the yield at bonyl function is not sufficiently activated for the last
higher temperatures.
ketone–ketone–aldol step. For lanthanum and cerium and
maybe even for tin(II) the size may cause problems since these
ions are too big. The distance between the reactants is probably
too large for a bond formation (Table 9).
In the case of 9-anthracenylaldehyde (3f) employing Zr or Al,
formation of 5a, which was not observed in the presence of
indium, is detected. Most likely the size of the anthracenyl
moiety decreases the rate of the second aldol reaction or the
hemiacetal formation, so that deprotonation and subsequently a
retro-aldol reaction takes place (Scheme 8). The formation of a
domino aldol product with two anthracenyl residues was not ob-
served most probably due to the steric demand of the anthra-
cenyl unit.
As illustrated in Scheme 8, the conversion of the aldehyde to
the ketone moiety was also witnessed with benzaldehyde (3a)
and 4-methoxybenzaldehyde (3d) in the reaction with 4'-fluoro-
propiophenone (1b) enolate and when reacting 4-methoxybenz-
aldehyde (3d) with propiophenone (1a) enolate. Here, there is
also no interaction between the metal center and the aryl ring of
Scheme 7: The follow-up reactions of A2OH and 6a at M06/6-31G(d)//
B3LYP/6-31G(d) level (ΔGrel with unscaled zpe are in kcal mol−1).
The mechanism depicted in Scheme 6 is in agreement with the the aldehyde possible. So, a competition between the second
following set of requirements with regard to the metal center: it aldol or hemiacetal formation and the deprotonation should be
should (a) exhibit Lewis acidity, (b) be sufficiently electroposi- considered.
Table 9: Effect of the properties of metals on the yields of 5a (reaction time: 2 h).
Metal
χa
Ionic radiusb [pm]
Charge density Z2/r [e2Å−1]
Yield 5a [%]
SnCl2
ZrCl4
SnCl4
TiCl4
1.72
1.22
1.72
1.32
2.01
1.47
1.82
1.49
1.08
1.08
1.20
102 (CN 2)
80 (CN 5)
76 (CN 5)
56 (CN 5)
25 (CN 4)
53 (CN 4)
61 (CN 4)
76 (CN 4)
117 (CN 6)
115 (CN 6)
89 (CN 6)
3.92
–
20.00
21.05
28.57
36.00
16.98
14.80
11.84
7.69
76
34
56
–
BCl3
AlCl3
52
93
85
–
GaCl3
InCl3
LaCl3
CeCl3
Sc(OTf)3
7.82
–
10.11
–
a χ: Electronegativity (According to Allred and Rochow). b CN = Coordination number [55].
822

Beilstein J. Org. Chem. 2016, 12, 813–824.
stage of this manuscript. We are indebted to the High-Perfor-
mance-Computing (HPC) Linux Cluster HorUS of University
of Siegen for computational support.
References
1.
Wurtz, C. A. Bull. Soc. Chim. Fr. 1872, 17, 436–442.
2
.
Nielsen, A. T.; Houlihan, W. J. The Aldol Condensation. In Organic
Reactions; Cope, A. C., Ed.; Wiley: New York, 1968; Vol. 16, pp 1 ff.
Mekelburger, H. B.; Wilcox, C. S. Formation of Enolates. In
Comprehensive Organic Synthesis. Selectivity, Strategy, and Efficiency
in Modern Organic Chemistry; Trost, B. M.; Fleming, I., Eds.;
Pergamon Press: Oxford, 1991; Vol. 2, pp 99–131.
3.
4.
5.
Scheme 8: Proposed mechanism for the formation of benzaldehyde in
the reaction of 9-anthracenylaldehyde (3f) with Zr and Al.
Heathcock, C. H. Modern Enolate Chemistry: Regio- and
Stereoselective Formation of Enolates and the Consequence of
Enolate Configuration on Subsequent Reactions. In Modern Synthetic
Methods; Scheffold, R., Ed.; VHCA: Basel, 1992; Vol. 6, pp 1 ff.
Mahrwald, R. Aldol Reactions; Springer Science+Business Media B.V.:
Heidelberg, 2009. doi:10.1007/978-1-4020-8701-1
Conclusion
The present results demonstrate a domino aldol reaction
working with several substrates and metals that is far superior to
the other method with TiCp2 [20,22], which could only be real-
ized with one single substrate resulting in a formal trimeriza-
tion. The variations in the metal fragment are promising with
regard to the development of an enantioselective version of the
above reaction and further variations in the substrates. DFT
calculations unveil the mechanism for the stereoselective forma-
tion of 5a.
6. Evans, D. A.; Fitch, D. M.; Smith, T. E.; Cee, V. J. J. Am. Chem. Soc.
000, 112, 10033–10046. doi:10.1021/ja002356g
2
7.
Crimmins, M. T.; Katz, J. D.; Washburn, D. G.; Allwein, S. P.;
McAtee, L. F. J. Am. Chem. Soc. 2002, 124, 5661–5663.
doi:10.1021/ja0262683
8
.
.
Tanaka, H.; Sawayama, A. M.; Wandless, T. J. J. Am. Chem. Soc.
2003, 125, 6864–6865. doi:10.1021/ja035429f
Song, H. Y.; Joo, J. M.; Kang, J. W.; Kim, D.-S.; Jung, C.-K.;
Kwak, H. S.; Park, J. H.; Lee, E.; Hong, C. Y.; Jeong, S.; Jeon, K.;
Park, J. H. J. Org. Chem. 2003, 68, 8080–8087.
doi:10.1021/jo034930n
9
Supporting Information
Experimental section, copies of 1H and 13C NMR spectra
of compounds, Cartesian coordinates and CIF files of 5a
and 7h.
10.Kalesse, M.; Cordes, M.; Symkenberg, G.; Lu, H.-H. Nat. Prod. Rep.
014, 31, 563–594. doi:10.1039/c3np70102f
2
1
1.Shiina, I. Chem. Rec. 2014, 14, 144–183. doi:10.1002/tcr.201300022
2.Mukaiyama, T.; Banno, K.; Narasaka, K. J. Am. Chem. Soc. 1974, 96,
1
7
503–7509. doi:10.1021/ja00831a019
Supporting Information File 1
1
3.Mukaiyama, T. Org. React. 1982, 28, 203–331.
Experimental section, copies of 1H and 13C NMR spectra
of compounds and Cartesian coordinates.
1
4.Braun, M. Formation of C–C- Bonds by Addition of Enolates to
Carbonyl Groups. In Stereoselective Synthesis: C-C Bond Formation
by Addition to C=O, C=N and Reactions Involving Olefinic Double
Bonds; Hoffmann, W.; Mulzer, J.; Schaumann, E.; Hoffmann, J., Eds.;
Houben-Weyl, Methods of Organic Chemistry, Vol. E 21b; Thieme
Verlag: Stuttgart, 1995; pp 1603–1666.
[http://www.beilstein-journals.org/bjoc/content/
supplementary/1860-5397-12-80-S1.pdf]
Supporting Information File 2
1
5.Machajewski, T. D.; Wong, C.-H. Angew. Chem., Int. Ed. 2000, 39,
CIF file of compound 5a.
1
352–1375.
doi:10.1002/(SICI)1521-3773(20000417)39:8<1352::AID-ANIE1352>3.
.CO;2-J
[http://www.beilstein-journals.org/bjoc/content/
supplementary/1860-5397-12-80-S2.cif]
0
1
1
1
1
6.Andrushko, V.; Andrushko, N., Eds. Stereoselective Synthesis of Drugs
and Natural Products; John Wiley & Sons: Hoboken, 2013; Vol. 1,
pp 215–249.
Supporting Information File 3
CIF file of compound 7h.
7.Wang, X.; Meng, Q.; Nation, A. J.; Leighton, J. L. J. Am. Chem. Soc.
[http://www.beilstein-journals.org/bjoc/content/
2002, 124, 10672–10673. doi:10.1021/ja027655f
supplementary/1860-5397-12-80-S3.cif]
See for aldol reactions as a first step in a domino process.
8.Bazin, S.; Feray, L.; Siri, D.; Naubron, J.-V.; Bertrand, M. P.
Chem. Commun. 2002, 2506–2507. doi:10.1039/B206695E
See for aldol reactions as a second step in a domino process.
9.Davies, H. M. L.; Lian, Y. Acc. Chem. Res. 2012, 45, 923–935.
doi:10.1021/ar300013t
Acknowledgements
Support from the DFG (Graduiertenkolleg, SFB), Degussa AG
and the Fonds der Chemischen Industrie is gratefully acknowl-
edged. We thank Dr. Thomas Koy and Wolfgang Henn for
valuable preparative work and helpful discussions in the early
See for aldol reactions as a later step in a domino process.
20.Ghorai, M. K.; Halder, S.; Das, S. J. Org. Chem. 2015, 80, 9700–9712.
doi:10.1021/acs.joc.5b01768
823

Beilstein J. Org. Chem. 2016, 12, 813–824.
2
2
2
1.Barba, F.; de la Fuente, J. L. J. Org. Chem. 1996, 61, 8662–8663.
43.CCDC 1470045 contains the supplementary crystallographic data for
this paper. The data can be obtained free of charge from The
Cambridge Crystallographic Data Centre via
doi:10.1021/jo952158l
2.Yun, S.-S.; Suh, I.-H.; Choi, S.-S.; Lee, S. Chem. Lett. 1998, 27,
985–986. doi:10.1246/cl.1998.985
http://www.ccdc.cam.ac.uk/data_request/cif
3.Ghosh, A. K.; Shevlin, M. The Development of Titanium Enolate-based
Aldol Reactions. In Modern Aldol Reactions; Mahrwald, R., Ed.;
Wiley-VCH: Weinheim, 2008; pp 63–125.
44.Gaussian 09, Revision D.01; Gaussian, Inc.: Wallingford CT, 2010.
45.Becke, A. D. J. Chem. Phys. 1993, 98, 1372–1377.
doi:10.1063/1.464304
2
2
2
4.Kanno, K.-I.; Takahashi, T. Zr(IV) and Hf(IV) Lewis Acids. In Acid
Catalysis in Modern Organic Synthesis; Yamamoto, H.; Ishihara, K.,
Eds.; Wiley-VCH: Weinheim, 2008; Vol. 2, pp 825–858.
5.Mukaiyama, T.; Matsuo, J.-I. Boron and silicon enolates in crossed
aldol reaction. In Modern Aldol Reactions; Mahrwald, R., Ed.;
Wiley-VCH: Weinheim, 2008; pp 127–160.
46.Becke, A. D. J. Chem. Phys. 1993, 98, 5648–5652.
doi:10.1063/1.464913
47.Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785–789.
doi:10.1103/PhysRevB.37.785
48.Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 270–283.
doi:10.1063/1.448799
6.Renzetti, A.; Marrone, A.; Gérard, S.; Sapi, J.; Nakazawa, H.; Re, N.;
Fontana, A. Phys. Chem. Chem. Phys. 2015, 17, 8964–8972.
doi:10.1039/C4CP05412A
49.Wadt, W. R.; Hay, P. J. J. Chem. Phys. 1985, 82, 284–298.
doi:10.1063/1.448800
50.Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 299–310.
doi:10.1063/1.448975
2
2
2
7.Dias, L. C.; Aguilar, A. M. Chem. Soc. Rev. 2008, 37, 451–469.
doi:10.1039/B701081H
51.Wells, R. L.; Kher, S. S.; Baldwin, R. A.; White, P. S. Polyhedron 1994,
13, 2731–2735. doi:10.1016/S0277-5387(00)83430-0
See for: Indium trichloride crystallizes from THF as InCl3(THF)3.
52.Zhao, Y.; Truhlar, D. G. Theor. Chem. Acc. 2008, 120, 215–241.
doi:10.1007/s00214-007-0310-x
8.Donohoe, T. J. Contemp. Org. Synth. 1996, 3, 1–18.
doi:10.1039/CO9960300001
9.Evans, D. A.; Janey, J. M.; Magomedov, N.; Tedrow, J. S.
Angew. Chem., Int. Ed. 2001, 40, 1884–1888.
doi:10.1002/1521-3773(20010518)40:10<1884::AID-ANIE1884>3.0.CO
53.Annunziata, R.; Cinquini, M.; Cozzi, F.; Borgia, A. L. J. Org. Chem.
1992, 57, 6339–6342. doi:10.1021/jo00049a053
54.Kanemasa, S.; Mori, T.; Tatsukawa, A. Tetrahedron Lett. 1993, 51,
8293–8296. doi:10.1016/S0040-4039(00)61414-4
55.Huheey, J. E.; Keiter, E. A.; Keiter, R. L. Inorganic Chemistry:
Principles of Structure and Reactivity, 4th ed.; Harper Collins: New
York, 1993.
;2-9
3
0.Cinar, M. E.; Schmittel, M. J. Org. Chem. 2015, 80, 8175–8182.
doi:10.1021/acs.joc.5b01256
And references therein.
3
1.Dénès, F.; Pérez-Luna, A.; Chemla, F. Chem. Rev. 2010, 110,
2366–2447. doi:10.1021/cr800420x
And references therein.
32.Schmittel, M.; Burghart, A.; Malisch, W.; Reising, J.; Söllner, R.
J. Org. Chem. 1998, 63, 396–400. doi:10.1021/jo971650x
33.Schmittel, M.; Burghart, A.; Werner, H.; Laubender, M.; Söllner, R.
J. Org. Chem. 1999, 64, 3077–3085. doi:10.1021/jo981793z
34.Schmittel, M.; Ghorai, M. K.; Haeuseler, A.; Henn, W.; Koy, T.;
Söllner, R. Eur. J. Org. Chem. 1999, 2007–2010.
License and Terms
This is an Open Access article under the terms of the
Creative Commons Attribution License
doi:10.1002/(SICI)1099-0690(199909)1999:9<2007::AID-EJOC2007>3
(http://creativecommons.org/licenses/by/2.0), which
permits unrestricted use, distribution, and reproduction in
any medium, provided the original work is properly cited.
.0.CO;2-2
3
3
3
3
3
4
5.Schmittel, M.; Ghorai, M. K. Synlett 2001, 1992–1994.
doi:10.1055/s-2001-18766
6.Schmittel, M.; Haeuseler, A.; Nilges, T.; Pfitzner, A. Chem. Commun.
The license is subject to the Beilstein Journal of Organic
Chemistry terms and conditions:
2003, 34–35. doi:10.1039/B209536J
7.Haeuseler, A.; Henn, W.; Schmittel, M. Synthesis 2003, 2576–2589.
doi:10.1055/s-2003-42450
(http://www.beilstein-journals.org/bjoc)
8.Kobayashi, S.; Hachiya, I. J. Org. Chem. 1994, 59, 3590–3596.
doi:10.1021/jo00092a017
The definitive version of this article is the electronic one
which can be found at:
9.Cuperly, D.; Petrignet, J.; Crévisy, C.; Grée, R. Chem. – Eur. J. 2006,
12, 3261–3274. doi:10.1002/chem.200501555
0.CCDC 1470044 contains the supplementary crystallographic data for
this paper. The data can be obtained free of charge from The
Cambridge Crystallographic Data Centre via
doi:10.3762/bjoc.12.80
http://www.ccdc.cam.ac.uk/data_request/cif
4
1.Giorgi, G.; Arroyo, F. J.; López-Alvarado, P.; Menéndez, J. C.
Tetrahedron 2011, 67, 5582–5589. doi:10.1016/j.tet.2011.05.115
2.Schmittel, M.; Söllner, R.; Drauz, K.; Günther, K.
Tetrahydropyran-2,4-diole, Verfahren zu deren Herstellung und
Verwendung. Deutsches Patent- und Markenamt DE 19911 198 A1,
May 11, 2000.
4
824