Reaction of secondary and tertiary aliphatic halides with aromatic aldehydes mediated by chromium(II): a selective cross-coupling of alkyl and ketyl radicals

Article abstract of DOI:10.1016/j.tet.2007.12.039

Takai-Utimoto reactions with secondary and tertiary aliphatic halides usually failed according to previous reports. Now, significant improvements could be achieved, and especially secondary aliphatic halides can be coupled to aromatic aldehydes in yields of up to >95%. A variety of processes are competing with the desired one, and thus conditions must be adapted to the nature of the aldehyde as well as the aliphatic halide used, as the outcome of these reactions is strongly affected by the putative radical intermediates.

Full text of DOI:10.1016/j.tet.2007.12.039

Available online at www.sciencedirect.com
Tetrahedron 64 (2008) 2134e2142
www.elsevier.com/locate/tet
Reaction of secondary and tertiary aliphatic halides with aromatic
aldehydes mediated by chromium(II): a selective
cross-coupling of alkyl and ketyl radicals
a
a,b,
Ludger A. Wessjohann a, , Gisela Schmidt , Henri S. Schrekker
*
*
a Leibniz Institute of Plant Biochemistry, Weinberg 3, D-06120 Halle (Saale), Germany
b Laboratory of Technological Processes and Catalysis, Institute of Chemistry, Universidade Federal do Rio Grande do Sul,
Bento Gonc¸alves 9500, Porto Alegre-RS, CEP: 91501-970, PO Box 15003, Brazil
Received 24 September 2007; received in revised form 8 December 2007; accepted 13 December 2007
Available online 23 December 2007
Dedicated in memoriam to Martin W.G. de Bolster
Abstract
TakaieUtimoto reactions with secondary and tertiary aliphatic halides usually failed according to previous reports. Now, significant improve-
ments could be achieved, and especially secondary aliphatic halides can be coupled to aromatic aldehydes in yields of up to >95%. A variety of
processes are competing with the desired one, and thus conditions must be adapted to the nature of the aldehyde as well as the aliphatic halide
used, as the outcome of these reactions is strongly affected by the putative radical intermediates.
Ó 2007 Elsevier Ltd. All rights reserved.
Keywords: Organometallic catalysis; Chromium alkylsdsecondary and tertiary; Catalysisdcobalt; Radical intermediates; Salt effectdlithium iodide; Barbier
reactiondGrignard-type reaction
1. Introduction
through space complexation. As a further consequence, the car-
bonyl is prone to nucleophilic attack by the formed oxo-alkyl-
metal. In other words, a nucleophilic reagent must be generated
in the presence of an electrophile. Furthermore, the initially
Synthetic organic chemists have access to a huge variety of
methods for aldol reactions.1 A particularly attractive method is
the chromium-Reformatsky reaction, which is mild, and can be
performed with polyfunctional substrates.2,3 However, research
efforts toward the development of the homologous version, the
‘homoaldol’ reaction, are relatively scarce.4e11 This might be
explained by the following reasons (Fig. 1): the generation of
the reactive oxo-alkylmetal intermediate requires the formation
of a carbanion without the mesomeric stabilization of an elec-
tron-withdrawing carbonyl, as is present in aldol reactions, and
stabilization of the carbanion is only possible by internal or
* Corresponding authors. Tel.: þ49 345 5582 1301; fax: þ49 345 5582 1309
(L.A.W.); tel.: þ55 51 3308 6284; fax: þ55 51 3308 7304 (H.S.S.).
E-mail addresses: wessjohann@ipb-halle.de (L.A. Wessjohann),
schrekker@iq.ufrgs.br (H.S. Schrekker).
Figure 1. Aldol versus ‘Homoaldol’.
0040-4020/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2007.12.039

L.A. Wessjohann et al. / Tetrahedron 64 (2008) 2134e2142
2135
applied reagent and the formed organometallic species are not
allowed to possess a strong basic character due to the presence
of the acidic a-proton to avoid reprotonation to the thermody-
namically more stable enolate isomeric form. Finally, it must
react chemoselectively with a second carbonyl compound (usu-
ally an aldehyde) as reaction partner.
(5) reduction of Co(II) to Co(I) by Cr(II) to close the catalytic
cycle for Co(I). In principle, a different course, e.g., reduction
of Co(III) alkyl by Cr(II) to Co(II) alkyl followed by dispropor-
tion to Co(I) and alkyl radical is possible too, and has been
discussed previously.2a,3k
This method is characterized by its tolerance toward most
of the common functional groups, and an excellent chemose-
lectivity for aldehydes. The obtained ester chromium alcohol-
ates usually do not react further to the lactones,y which is not
the case for most previously mentioned ‘homoaldol’ methods.
The formation of a-methyl branched secondary alcohols is of
great importance for the synthesis of polyketide natural prod-
ucts. A method suitable for complex substrates with many
other functional groups appears feasible with the excellent
chemoselectivity of chromium(II)-mediated reactions, would
the use of secondary chromium alkyls become available. Until
recently, secondary and tertiary aliphatic halides could not be
coupled efficiently by this Barbier-type reaction,3j although
secondary and even tertiary chromium(III) alkyls have been
reported from metalechromium(III) exchange reactions.13
The reason for this might be the higher stability of the corre-
sponding radicals. Also, instability of secondary and tertiary
alkylchromium complexes toward b-hydride elimination was
suspected, but appears unlikely to us as was argued earlier.2a
However, according to a literature report, the one electron re-
duction of primary, and especially secondary and tertiary alkyl
radicals with chromium(II) should be fast,14 whereas others
report no reactivity of allyl halides with chromium(II) if the
oxo-compound as reaction partner is omitted.15 In any case,
only primary alkylchromium(III) was considered sufficiently
stable as organometallic species to give useful reactions other
than Wurtz-type homocouplings, whereas benzylic, secondary
and tertiary alkylchromium(III) intermediates are considered
to be in equilibrium with their homolytic cleavage products
(Scheme 2).16 For allylchromium, both preferences are dis-
cussed, and their application in NozakieHiyama reactions is
widely documented.2,15 Also, Co(I) reactions to secondary
and tertiary alkyl Co(III) complexes are well documented.17
This supported us to study the behavior of unactivated second-
ary and tertiary aliphatic halides in the TakaieUtimoto reac-
tion in more detail with the aim to extend the scope of this
reaction to secondary and tertiary alkyls.3j
Masked carbonyl substrates were originally used for the
synthesis of a ‘homoaldol’.4 Recently, the formation of oxo-
alkylmetals was reported for a variety of metals, however,
a disadvantage of most of them is that they are prepared by
the silyloxycyclopropane route, which limits the scope of the
reactions with respect to type (n¼1) and substitution pattern,5
or from methoxyallene, which suffers the same principal
disadvantages.6 The most important current class of reactive
oxo-alkylmetals is that of the respective zincalkyls, generated,
e.g., from alkenes by Knochel’s method.7 The possibility to
generate oxo-alkylmetals by the reaction of a low-valent metal
with a halogenated substrate would be generally applicable to
a broad range of substrates, many of which are commercially
available. Obviously, the high reactivity of organolithium and
Grignard reagents do not fulfill the requirements for this reac-
tion. Oxo-alkyllanthanides (La, Ce, Nd, and Sm) have been
produced by the direct reaction of the lanthanide metal with
ethyl 3-halopropionates.8 Unfortunately, this method is limited
to ketone electrophiles, and direct attack at the carbonyl is
a common side reaction. Metallic zinc reacts with functional-
ized aliphatic iodides to the oxo-alkylzinc reagents. A disad-
vantage of zinc is its need to be activated, and the resulting
oxo-alkylzinc has a low reactivity toward carbonyl electro-
philes.9 Transmetallation of these oxo-alkylzinc reagents into
the corresponding zincecopper couple promotes the coupling
to aldehydes in the presence of a Lewis acid.10 Oxo-alkylzinc
reagents are also successfully applied in the transmetallation
with titanium(IV) complexes, and the resulting oxo-alkyltita-
nium reagents couple selectively with aldehydes.11
Another attractive possibility to obtain ‘homoaldol’ prod-
ucts is the extension of the chromium(II)-Barbier or TakaieUti-
moto reaction (not to confuse with the related TakaieUtimoto
olefination) to oxo-substituted alkylhalides.2a,3k,12 The chro-
mium(III) oxo-alkyls are generated under cobalt catalysis and
the generally accepted mechanism consists of the following
steps (Scheme 1): (1) reduction of Co(III) to Co(I) by Cr(II);
(2) oxidative addition of an alkyl halide to Co(I) to form the
corresponding Co(III) alkyl; (3) homolytic cleavage of the
CeCo(III) bond to yield an alkyl radical and Co(II); (4) reduc-
tive trapping of the alkyl radical by Cr(II) to generate the alkyl-
chromium species, which then couples with an aldehyde; and
Scheme 2. Formation/homolytic cleavage equilibrium of chromium(III)organyls.
y
Exceptions were observed by us in compounds exerting an exceptionally
strong ThorpeeIngold effect to close such a ring, which often results in
five- or six-membered lactones. However, as in other instances this may
also be a work-up artifact, and depending on work-up conditions different pro-
portions of open and ring-closed esters can be found.
Scheme 1. Putative mechanism for the TakaieUtimoto reaction.

2136
L.A. Wessjohann et al. / Tetrahedron 64 (2008) 2134e2142
2. Results and discussion
The reaction of cyclododecyl iodide with benzaldehyde was
presented in the original study of Takai.12 Cyclododecane and
cyclododecene were the only products reported, while the al-
dehyde remained unchanged. This is an indication that the
Cr(II)-mediated reduction of a secondary aliphatic halide is
working in principle, but fails in the CeC coupling. Based
on our experience,3 we know that well-stabilized chromium
intermediates tend to react sluggishly, but this is strongly influ-
enced by both steric effects and reaction conditions such as
solvent or salt effects.2a,3c
Scheme 3. Primary versus secondary and tertiary aliphatic halides reacting
with CrCl2 without cobalt catalyst.
Isopropyl iodide is the sterically least strained secondary
halide and thus was chosen as starting point in a reaction with
benzaldehyde. All reactions were performed under similar
conditions (DMF, 55 ꢀC, 16 h), using 2.5 equiv of chromium(II)
chloride and 10 mol % lithium iodide. The beneficial effect of
lithium iodide on many CrCl2 reactions was discussed previ-
ously, and enabled, e.g., efficient chromium-Reformatsky proc-
esses.2a,3c The reaction temperature, cobalt catalyst, lithium
iodide amount, and solvent were varied. The results are summa-
rized in Table 1. The reduction potential of chromium(II) chlo-
ride is not sufficient to reduce primary aliphatic halides, and
a substitution reaction with chloride takes place instead,
thereby producing unreactive alkyl chloride (Scheme 3).12 In
contrast, secondary and tertiary aliphatic halides are readily
reduced to form the intermediate radical, which is rapidly re-
duced in a second step by chromium(II) chloride to reversibly
form the alkylchromium(III) species.18 Performing the reaction
in the absence of a cobalt catalyst and lithium iodide resulted in
a moderate yield of alcohol 6a (entry 1). The presence of cata-
lytic or stoichiometric amounts of lithium iodide improved the
yield, independent of the amount (entries 2 and 3). In the pres-
ence of lithium iodide an increased reactivity for chromium(II)-
mediated reactions is often observed, especially in THF.2a,3c
This effect has been explained by the putative formation of bet-
ter soluble, anionic chromate(II) species, e.g., Li2[CrX4]. Inter-
estingly, this effect could neither be achieved to the same extent
with iodide (NBu4I) or lithium (LiBF4) ions alone.19 In DMF,
CrCl2 is quite well soluble, and the simple Lewis acid and/or
to a minor extent iodide-exchange catalysis might be important
too. However, as was shown in the context of our chromium-
Reformatsky work, LiI in DMF is a mediocre catalyst at most
to exchange chlorides or bromides into alkyliodides, and thus,
if such an exchange is required, the addition of excess NaI or
KI is significantly better to convert the less reactive halides
(back) to iodides.19
Substitution of DMF by THF as solvent had a dramatic
effect on the outcome of this reaction, as no product was ob-
tained under these conditions (entry 4), most likely caused by
the decreased reduction potential of chromium(II) chloride in
the presence of THF.20 Although a cobalt catalyst appears to
be unnecessary in principle, we tested its possible influence,
also in order to compare it to the original TakaieUtimoto pro-
cedure. The result was quite surprising. No additional positive
effect was observed with cyanocobalamine (vitamin B12) at
standard conditions (55 ꢀC, entry 5). However, alcohol 6a
was obtained in an excellent yield in the presence of cobalt
tetramethoxyphenylporphyrin as a cheaper vitamin B12 alter-
native (entry 6).3j Surprisingly, at lower temperature both
catalysts exerted an opposite effect (entries 7 and 8). Now,
the reaction with cyanocobalamine (vitamin B12) resulted in
an excellent yield, and a strong decrease was observed for
the cobalt tetramethoxyphenylporphyrin. At least two conclu-
sions can be drawn from this: (1) the cobalt catalyst is at least
partially involved in the reaction of secondary aliphatic halides
and can have a limited beneficial effect; and (2) the cobalt
dependent transformations (Scheme 1) of cyanocobalamine
(vitamin B12) and cobalt tetramethoxyphenylporphyrin are
differently influenced by the reaction temperature. The reason
may be found in differences in nucleophilicity, and in the re-
duction potentials of the various cobalt oxidation states, which
are known to be important,3i,12 obviously resulting in different
kinetic temperature profiles, and maybe also in solubility.
The effect of the halide on the coupling of the isopropyl
unit to benzaldehyde needed to be studied next. One might
argue that the presence of lithium iodide in DMF will always,
Table 1
Reactivity of the secondary model substrate isopropyl iodide with
benzaldehydea,b
Entry
Cobalt catalyst
Remark
Yieldc (%)
1
2
3
4
5
6
7
8
a
None
None
None
None
No LiI
52
72
LiI (10 mol %)
LiI (250 mol %)
THF as solvent
55 ꢀC
70
0
70
d
Vitamin B12
Co(porphin)e
55 ꢀC
>95
>95
52
d
Vitamin B12
Co(porphin)e
20 ꢀC
20 ꢀC
CrCl2 (2.50 equiv), cobalt catalyst (5 mol %), LiI (10 mol %), benzalde-
hyde (1.00 equiv), and isopropyl iodide (1.10 equiv).
b
Reaction without isopropyl iodide resulted in 68% pinacol coupling prod-
uct 6aa (presented in Table 2).
c
Isolated yield based on benzaldehyde.
Cyanocobalamine.
Cobalt tetramethoxyphenylporphyrin.
d
e

L.A. Wessjohann et al. / Tetrahedron 64 (2008) 2134e2142
2137
Table 2
The influence of the halide on the reaction with benzaldehydea
electron transfer to the aldehyde under formation of ketyl rad-
ical 7, which transfers its unpaired electron to the coordinated
allyl halide in intermediate 8. A second single electron transfer
by chromium(II) to the aldehyde affords diradical 10, which
couples to the desired product 11. Questionable is the formation
of ketyl radical 7, as this should give pinacol coupling byprod-
uct, which is not commonly observed in NozakieHiyama reac-
tions to any noteworthy extend. Also does it contradict the very
low reactivity of ketones in NozakieHiyama reactions. Even-
more, allyl intermediate 9 is also formed in the absence of al-
dehyde, and this approach has been used in Wurtz-type
dimerization reactions,22 whereas Mulzer reports instances
where the omission of aldehydes prevents reaction.15 Unfortu-
nately, there is no conclusive experimental proof available for
any of these suggested mechanisms to date, and indeed several
mechanisms may apply dependent on the nature of the alkyl ha-
lide, the chromium salt and its ligands (incl. solvent), and maybe
also of the oxo-compound. The principle difference is the nature
of the intermediate organochromium(III) species, which may be
anionic or radical, depending on the type of organic halide used.
The radical character of alkylchromium(III) complexes is sup-
ported by the addition reactions to stabilized alkenes,14,23
whereas anion stabilization as in Reformatsky reactions favors
anionic behavior.2a,3 Furthermore, the likelihood of a radical
based mechanism for benzyl, secondary and tertiary alkyl ha-
lides is higher than for primary alkyl and allyl halides as allyl
radicals are more effectively trapped by chromium(II)
chloride.18
With these mechanistic implications in mind, we started to
study the role of the aldehyde in the reaction with isopropyl
iodide, and the results are summarized in Table 3. Substituents
on the aromatic ring of benzaldehyde exerted a dramatic effect
on the yield of alcohol 6. Good yields of alcohols 6b and 6c were
obtained with p-methoxy benzaldehyde and m-methoxy benzal-
dehyde, respectively (entries 2 and 3). The moderate yield with
o-methoxy benzaldehyde 5d implies that the reaction is highly
sensitive toward steric interactions, a common observation in
Cr(II)-mediated reactions (entry 4).2 Surprisingly, a substituent
that is expected to affect the electrophilicity of the aldehyde
only to a minor extent, like a methyl, resulted in a strongly
reduced yield (entry 5). Electron-withdrawing substituents
like a bromo or a fluoro or a trifluoromethyl favor, in different
degrees, the formation of pinacol coupling products (entries
Entry
X
Yieldb (%)
1
2
3
a
I (4a)
Br (4b)
Cl (4c)
>95
44
0c
CrCl2 (2.50 equiv), cyanocobalamine (vitamin B12)¼CN-B12 (5 mol %),
LiI (10 mol %), benzaldehyde (1.00 equiv), and isopropyl halide (1.10 equiv).
b
Isolated yield based on benzaldehyde.
Side product: 52% pinacol coupling product 6aa.
c
via nucleophilic substitution catalysis, deliver isopropyl io-
dides (and/or chlorides from CrCl2). But this process, although
fast with higher alkalihalides (NaI, KI), is very slow with LiI.
Only minimal exchange is observed if secondary halides are
reacted with lithium iodide in DMF, in contrast to NaI. The re-
actions were performed under identical conditions (DMF,
55 ꢀC, 16 h), using 2.5 equiv chromium(II) chloride, 5 mol %
cyanocobalamine (vitamin B12), and 10 mol % lithium iodide.
The results are summarized in Table 2. Importantly, the
TakaieUtimoto reaction is highly sensitive to the rate at which
the alkylchromium species is generated.12 The yield of alcohol
6a was strongly reduced in the case of the bromide (entry 2),
and no product 6a was obtained when isopropyl chloride was
applied (entry 3). The reaction with chloride 4c exclusively re-
sulted in the formation of pinacol coupling product 6aa.
To place these results in a broader perspective, it is useful to
discuss the mechanistic background of chromium(II)-mediated
reactions. Most researchers relate their results to the principle
of the mechanisms presented in Schemes 1 and 2. The initially
formed radical 1 is reduced in a second step by chromium(II)
chloride to generate the organochromium species 2, which con-
tains an alkyl group of carbanionic nature. A consecutive nucle-
ophilic attack on the aldehyde affords chromium alcoholate 3.21
An alternative mechanism (Scheme 4) was proposed by Mulzer
for the NozakieHiyama reaction, the addition of allyl halides
to aldehydes, using rather special, oxygen-rich substrates,
which may exert a special behavior.15 Mulzer proposed a single
Scheme 4. Alternative mechanism for CrCl2-mediated reactions.

2138
L.A. Wessjohann et al. / Tetrahedron 64 (2008) 2134e2142
Table 3
The reaction of isopropyl iodide with aromatic aldehydesa,b
Entry
R1
Product
Yieldc (%)
1
H (5a)
6a
6b
6c
6d
6e
6f
>95
70d
2
4-OMe (5b)
3-OMe (5c)
2-OMe (5d)
4-Me (5e)
3
63
37
4
5
55
17e
6
4-Br (5f)
4-F (5g)
4-CF3 (5h)
7
6g
6h
6i
51f
8
0g
9
Furan-2-yl (5i)
2-Naphthaldehyde (5j)
2-Naphthaldehyde (5j)
39
37þ11h
10
11i
a
6j
6j
29þ36j
CrCl2 (2.50 equiv), cyanocobalamine (vitamin B12)¼CN-B12 (5 mol %), LiI (10 mol %), aldehyde (1.00 equiv), and isopropyl iodide (1.10 equiv).
Reactions of other substrate combinations than isopropyl iodide/benzaldehyde were not optimized for yield.
Isolated yield based on aldehyde.
b
c
d
e
f
Isolation of 5% 6bb, dehydrated 6b.
Isolation of 31% pinacol coupling product 6ff.
Pinacol coupling product detected by GCeMS of the crude product, but not isolated.
Isolation of 61% pinacol coupling product 6hh.
Isolation of 11% 6jj (presented in Scheme 5).
g
h
i
CrCl2(TMEDA) instead of CrCl2.
Isolation of 36% 6jj (presented in Scheme 5).
j
6e8). Unfortunately, aliphatic aldehydes were completely un-
might favor a competitive homo-Wurtz-coupling reaction of
the halide alone, and a fast generation favors the homo-pinacol
coupling reaction of the aldehyde only. Probably, benzalde-
hyde 5a provides the optimum kinetics for the hetero-cross-
coupling under the standard conditions used for this study
(entry 1).
reactive under these reaction conditions. Furan-2-yl 5i was cou-
pled in moderate yield (entry 9). Obviously, the aldehyde
partner needs to have the proper redox or radical stabilizing
potential in order to allow good yield, e.g., through conjugation
to an aromatic system.
This extreme sensitivity of the reaction toward small
changes in the aldehydes’ electron density and/or radical sta-
bilization properties implies the involvement of a radical inter-
mediate. The outcome of the reaction appears to depend on the
rate at which the ketyl radical is formed. A slow generation
Disappointing was the yield with 2-naphthaldehyde. How-
ever, this is caused by the competing generation of the mech-
anistically highly interesting side product 6jj (Scheme 5).
TMEDA and other electron-donating amine ligands are known
to increase the reduction potential of chromium(II) chloride.20
Scheme 5. Proposed mechanism for the formation of aldehyde 6jj.

L.A. Wessjohann et al. / Tetrahedron 64 (2008) 2134e2142
2139
Table 4
The reaction of secondary aliphatic iodides with benzaldehydea,b
Entry
R1
R2
Product
Yieldc (%)
1
2
3
4
a
Me
Et
Me (4a)
Me (4d)
6a
>95
44
6k
6l
6me
e(CH2)5e (4e)
58
10d
(CH2)2CO2Me
Me (4feBr)d
CrCl2 (2.50 equiv), cyanocobalamine-vitamin B12¼CN-B12 (5 mol %), LiI (10 mol %), benzaldehyde (1.00 equiv), and aliphatic iodide (1.10 equiv).
Reactions of other substrate combinations than isopropyl iodide/benzaldehyde were not optimized for yield.
Isolated yield based on benzaldehyde.
b
c
d
e
Bromide used instead of iodide.
Alcohol 6m is easily transformed to lactone 6mm with p-TsOH in benzene.
Using CrCl2(TMEDA) instead of CrCl2 resulted in an im-
a wide variety of functional groups, including esters (homoeno-
lates). Thus, bromide ester substrate 4feBr was tested, but
resulted in a poor yield of hydroxy ester 6m (entry 4), which
was even lower when the iodide of 4f was applied. Hydroxy
ester 6m could be isolated, but spontaneously reacted further
to lactone 6mm if kept at room temperature or if subjected to
acidic (work-up) conditions.
The reaction of tert-butyl halides with benzaldehyde was
studied as well (Scheme 6). The reaction with Cr(II) should
generate quite stable, sterically demanding, and thus unreactive
radicals. This should be seen in the outcome of these reactions.
Indeed, low yields were obtained with iodide 14a (7%) and bro-
mide 14b (19%), and application of chloride 14c did not result in
the formation of alcohol 15. Instead competing pinacol coupling
of benzaldehyde to product 6aa occurred in high yield (up to
73% yield). Obviously, the side reactions dominate when the
organochromium species is unable to react further because the
radical is too stabilized or sterically hindered to participate in
cross-coupling or Wurtz-homocoupling.
The typical chemoselectivity of TakaieUtimoto reactions
toward aldehydes in the presence of a ketone was also observed
for the reaction with secondary aliphatic halides. Only alcohol
6a was obtained from the reaction of isopropyl iodide with
benzaldehyde in a competition experiment with acetophenone
(Scheme 7). However, the yield of alcohol 6a is lower when
compared to an experiment without acetophenone (Table 2,
entry 1). A possible explanation might be found in the compet-
itive formation of the acetophenone ketyl and benzaldehyde ke-
tyl radicals. The isopropylchromium reagent cannot react fast
with the bulky acetophenone or the derived radical, and this
might decrease the amount of either reduction equivalents or
proved yield of 36% of aldehyde 6jj (entries 10 and 11). A
direct nucleophilic attack to the naphthyl core rather than to
the aldehyde position can be considered highly unlikely for
any isopropylanion based species, although an ‘ene-type’ reac-
tion to 13 is feasable too. Considering a radical mechanism
similar to those of Mulzer15 and Takai,18 or as proposed for
Sm(II) reactions,24 one might speculate that product 6jj is
formed through diradical intermediate 12. This generalized
mechanism can also explain the formation of homocoupling
products and why the outcome of these radical reactions is
extremely sensitive to the structure of both substrates. In addi-
tion to the mechanistic insight, this reaction also has synthetic
implications, as all other organometallic reagents will react
with the aldehyde exclusively, and not by breaking up the
aromatic systems.
The results with sterically more demanding secondary ali-
phatic iodides are summarized in Table 4. Application of sec-
butyl iodide or cyclohexyl iodide both resulted in moderate
yields of alcohols 6k and 6l, respectively (entries 2 and 3). It
is known that the stability of alkylchromium species decreases
with an increasing chain-length of the alkyl group.13b Thus the
persistence of aldehyde after incomplete reaction with bulkier
iodides instead of pinacol coupling might be explained by
a faster generation or greater stability of the alkyl radical com-
pared to the ketyl radical, and as a consequence this undergoes
radical homocoupling reactions like the Wurtz-type dimeriza-
tion. The effect of a sterically more demanding secondary ali-
phatic iodide is the same as that obtained by rendering the
aldehyde less reactive (cf. Table 3, entry 4). The biggest syn-
thetic value of the TakaieUtimoto reaction is its tolerance to
Scheme 6. TakaieUtimoto reactions with tertiary aliphatic halides.

2140
L.A. Wessjohann et al. / Tetrahedron 64 (2008) 2134e2142
6hh,32 6i,33 6j,34 6k,35 6l,36 6mm,37 and 1538 were in accor-
dance with the literature data. Thin-layer chromatography
was carried out on Merck silica 60/F-254 aluminum-backed
plates. Flash chromatography was performed using Merck
silica gel 60 (40e60 mm). NMR spectra were recorded in
CDCl3. Chemical shifts d are quoted in parts per million
(ppm), and coupling constants J are given in hertz (Hz).
4.2. Syntheses
Scheme 7. Competition experiment between benzaldehyde (1.00 equiv) and
acetophenone (1.00 equiv) in the Cr(II)-mediated reaction with isopropyl
iodide (1.10 equiv).
4.2.1. General procedure for the reaction of secondary
and tertiary aliphatic halides with chromium(II) chloride
and aldehydes/ketones
benzaldehyde ketyl radical, respectively. This might enhance
side reactions of the isopropyl radical.
To chromium(II) chloride (200 mg, 1.63 mmol, 2.50 equiv),
cyanocobalamine (vitamin B12) (44 mg, 33 mmol, 0.05 equiv),
and LiI (8.7 mg, 65 mmol, 0.10 equiv) was added 2.5 ml DMF
under vigorous stirring. After a few minutes, the aldehyde/
ketone (0.65 mmol, 1.00 equiv) and aliphatic halide
(0.72 mmol, 1.10 equiv) were added in this order. The resulting
mixture was stirred for 16 h at 20 ꢀC, and the reaction was
quenched with 2.5 ml water. The water layer was extracted
with diethyl ether, and the combined organic fractions were
washed with H2O and aqueous NaCl. The organic layer was
dried over MgSO4 and concentrated in vacuo. The crude prod-
uct was purified by flash chromatography, usually with a
hexane/diethyl ether mixture as eluent.
3. Conclusion
In conclusion, the TakaieUtimoto reaction can be used to
couple secondary and to some extent tertiary aliphatic halides
with aromatic aldehydes. Ketones and aliphatic aldehydes are
not reactive. The outcome of the reactions are, however, very
sensitive toward small changes in the alkyl halide as well as
in the aldehyde and the conditions, which need to be optimized
for each system. This behavior can be explained by a mecha-
nism in which an alkyl radical and a ketyl radical selectively
cross-couple to afford the product. A better stabilized radical
at either end results in a shift from cross-coupling to homocou-
pling, i.e., with electron-rich aldehydes’ reactivity leans toward
reactions of the alkyl radical, like the Wurtz coupling, whereas
with either an alkyl halide that generates a less reactive alkyl
radical or an electron poor aldehyde the pinacol coupling reac-
tion is favored. This version of the TakaieUtimoto reaction
principally allows the synthesis of secondary ‘homoaldols’,
and is chemoselective toward aldehydes, important features
for polyketide synthesis applications, but conditions need to
be optimized to allow useful yields. These observations might
result in a better understanding of the chromium(II)-mediated
reactions of radical stabilizing alkylchromium species like
secondary, tertiary, allylic, or benzylic ones, in general.
4.2.2. 1-(3-Methoxy-phenyl)-2-methyl-propan-1-ol (6c)
Flash chromatography afforded 6c as a colorless oil in 63%
yield. Rf¼0.24 (hexane/diethyl ether¼80:20); 1H NMR
(400 MHz) d 7.25 (dd, J¼8.2, 8.2, 1H), 6.90e6.87 (m, 2H),
6.81 (m, 1H), 4.33 (d, J¼7.0, 1H), 3.81 (s, 3H), 1.95 (m, 1H),
1.85 (br s, 1H), 1.00 (d, J¼7.0, 3H), 0.81 (d, J¼6.6, 3H); 13C
NMR (100 MHz) d 159.5 (C), 145.4 (C), 129.1 (CH), 118.9
(CH), 112.8 (CH), 112.0 (CH), 79.9 (CH), 55.2 (CH3), 35.2
(CH), 19.1 (CH3), 18.2 (CH3); IR (neat): 3430 (w), 2958 (w),
789 (w), 702 (w) cmꢁ1; HRMS-EI (70 eV) m/z calcd for
C11H16O2 (Mþ) 180.1150, found 180.1158. Anal. Calcd for
C11H16O2: C, 73.30; H, 8.95. Found: C, 73.10; H, 8.54.
4.2.3. 1-(2-Methoxy-phenyl)-2-methyl-propan-1-ol (6d)
Flash chromatography afforded 6d as a colorless oil in 37%
yield. Rf¼0.20 (hexane/diethyl ether¼85:15); 1H NMR
(400 MHz) d 7.26e7.21 (m, 2H), 6.94 (dd, J¼7.4, J¼7.4,
1H), 6.88 (d, J¼8.2, 1H), 4.49 (d, J¼7.4, 1H), 3.84 (s, 3H),
2.38 (br s, 1H), 2.06 (m, 1H), 1.03 (d, J¼6.6, 3H), 0.79 (d,
J¼7.0, 3H); 13C NMR (100 MHz) d 156.6 (C), 131.5 (C),
128.2 (CH), 128.1 (CH), 120.5 (CH), 110.5 (CH), 77.2
(CH), 55.2 (CH3), 34.1 (CH), 19.6 (CH3), 18.5 (CH3); IR
(neat): 3433 (w), 2958 (m), 1237 (s), 754 (s) cmꢁ1; HRMS-
EI (70 eV) m/z calcd for C11H16O2 (Mþ) 180.1150, found
180.1159.
4. Experimental
4.1. General remarks
All reactions were carried out under an argon atmosphere
in flame-dried glassware using standard syringe and septa
techniques. The commercial reagents 4aee, 5bej, 14aec,
16, cyanocobalamine (vitamin B12), cobalt tetramethoxyphe-
nylporphyrin, lithium iodide, chromium(II) fluoride, TMEDA,
and chromium(II) chloride (99.9% from Strem Chemicals)
were used as purchased. Substrate 4f was synthesized accord-
ing to a literature procedure, and its purity was determined by
1H NMR.25 THF was distilled from potassium/benzophenone.
Absolute DMF was purchased from Fluka. Benzaldehyde 5a
was distilled from potassium hydride. Spectral data of the
known compounds 6a,26 6aa,27 6b,28 6bb,29 6e,30 6ff,31
4.2.4. 1-(4-Bromo-phenyl)-2-methyl-propan-1-ol (6f)
Flash chromatography afforded 6f as a colorless oil in 17%
yield. Rf¼0.13 (hexane/diethyl ether¼85:15); 1H NMR
(400 MHz) d 7.46 (d, J¼8.2, 2H), 7.19 (d, J¼8.2, 2H), 4.35

L.A. Wessjohann et al. / Tetrahedron 64 (2008) 2134e2142
2141
R. V. A.; Wessjohann, L. A. J. Org. Chem. 2002, 67, 1975; (g) Schrekker,
H. S.; Micskei, K.; Hajdu, C.; Patonay, T.; de Bolster, M. W. G.; Wessjo-
hann, L. A. Adv. Synth. Catal. 2004, 346, 731; (h) Wessjohann, L. A.;
Wild, H.; Schrekker, H. S. Tetrahedron Lett. 2004, 45, 9073; (i)
Wessjohann, L. A.; Schrekker, H. S. Tetrahedron Lett. 2007, 48, 4323;
(j) Wessjohann, L. A.; Schmidt, G.; Schrekker, H. S. Synlett 2007,
2139; (k) Wessjohann, L. A.; Gabriel, T.; Gutsche, A. S.; Schrekker,
H. S. Unpublished results; (l) Wessjohann, L. A.; Gabriel, T.; Schrekker,
H. S. Unpublished results.
(d, J¼6.6, 1H), 1.91 (m, 1H), 1.80 (br s, 1H), 0.97 (d, J¼6.6,
3H), 0.80 (d, J¼6.6, 3H); 13C NMR (100 MHz) d 142.5
(C), 131.21 (CH), 128.3 (CH), 121.1 (C), 79.2 (CH), 35.2
(CH), 18.8 (CH3), 18.0 (CH3); IR (neat): 3396 (w), 2959
(m), 908 (w), 826 (w), 782 (w), 734 (m) cmꢁ1; HRMS-EI
(70 eV) m/z calcd for C10H13BrO (Mþ) 228.0150, found
228.0145.
4. (a) Hoppe, D.; Zschage, O. Angew. Chem. 1989, 101, 67; (b) Ahlbrecht,
H.; Beyer, U. Synthesis 1999, 365.
4.2.5. 1-(4-Fluoro-phenyl)-2-methyl-propan-1-ol (6g)
Flash chromatography afforded 6g as a colorless oil in 51%
yield. Rf¼0.15 (hexane/diethyl ether¼85:15); 1H NMR
(300 MHz) d 7.27 (dd, J¼8.6, J¼5.5, 2H), 7.02 (dd, J¼8.6,
8.6, 2H), 4.35 (d, J¼6.8, 1H), 1.92 (m, 1H), 1.80 (br s, 1H),
0.99 (d, J¼6.6, 3H), 0.78 (d, J¼6.8, 3H); 13C NMR
(75.50 MHz) d 161.9 (d, J¼244, C), 139.1 (d, J¼3.0, C),
128.0 (d, J¼8.3, CH), 114.9 (d, J¼21.1, CH), 79.3 (CH), 35.4
(CH), 19.0 (CH3), 18.3 (CH3); IR (neat): 3420 (w), 2960 (w),
1604 (w), 1510 (m), 842 (w), 776 (w) cmꢁ1; MS-EI m/z:
168.1 (5%, Mþ), 125.0 (100%, MþꢁC3H7).
5. (a) Crimmins, M. T.; Nantermet, P. G. Org. Prep. Proced. Int. 1993, 25,
43; (b) Kuwajima, I.; Nakamura, E. Comprehensive Organic Synthesis;
Trost, B. M., Fleming, I., Eds.; Pergamon: Oxford, 1991; Vol. 2.
6. Hormuth, S.; Reissig, H.-O.; Dorsch, D. Angew. Chem., Int. Ed. Engl.
1993, 32, 1449.
7. (a) Langer, F.; Schwink, L.; Devasagayaraj, A.; Chavant, P.-Y.; Knochel,
P. J. Org. Chem. 1996, 61, 8229; (b) Knochel, P. J. Organomet. Chem.
2003, 680, 136.
8. Fukuzawa, S.-I.; Sumimoto, N.; Fujinami, T.; Sakai, S. J. Org. Chem.
1990, 55, 1628.
9. (a) Tamaru, Y.; Nakamura, T.; Sakaguchi, M.; Ochiai, H.; Yoshida, Z.
Chem. Commun. 1998, 610; (b) Houkawa, T.; Ueda, T.; Sakami, S.;
Asaoka, M.; Takei, H. Tetrahedron Lett. 1996, 37, 1045.
10. Yah, M. C. P.; Knochel, P.; Santa, L. E. Tetrahedron Lett. 1988, 29,
3887.
4.2.6. 1-Isopropyl-1,4-dihydro-naphthalene-2-
carbaldehyde (6jj)
11. (a) Rozema, M. J.; Sidduri, A.; Knochel, P. J. Org. Chem. 1992, 57, 1956;
(b) Ochiai, H.; Nishihara, T.; Tamaru, Y.; Yoshida, Z.-I. J. Org. Chem.
1988, 53, 1343; (c) Armstrong, J. D., III; Hartner, F. W., Jr.; DeCamp,
A. E.; Volante, R. P.; Shinkai, I. Tetrahedron Lett. 1992, 33, 6599;
(d) DeCamp, A. E.; Kawaguchi, A. T.; Volante, R. P.; Shinkai, I.
Tetrahedron Lett. 1991, 32, 1867.
Flash chromatography afforded 6jj as a colorless oil in 36%
yield. Rf¼0.28 (hexane/diethyl ether¼90:10); 1H NMR
(400 MHz) d 9.60 (s, 1H), 7.22e7.18 (m, 4H), 7.14 (dd,
J¼2.3, 5.9, 1H), 3.86 (m, 1H), 3.73 (m, 1H), 3.59 (m, 1H),
1.96 (m, 1H), 0.91 (d, J¼7.0, 3H), 0.66 (d, J¼6.6, 3H); 13C
NMR (100 MHz) d 192.6 (CH), 149.7 (CH), 144.4 (C), 136.5
(C), 134.1 (C), 129.6 (CH), 127.7 (CH), 126.2 (CH), 126.1
(CH), 43.4 (CH), 35.0 (CH), 32.6 (CH2), 20.7 (CH3), 18.7
12. Takai, K.; Nitta, K.; Fujimura, O.; Utimoto, K. J. Org. Chem. 1989, 54,
4732.
13. (a) Sneeden, R. P. A.; Zeiss, H. H. J. Organomet. Chem. 1971, 26, 101; (b)
Nishimura, K.; Kuribayashi, H.; Yamamoto, A.; Sakuji, I. J. Organomet.
Chem. 1972, 37, 317.
(CH3); IR (neat): 2960 (m), 1685 (s), 791 (w), 740 (m) cmꢁ1
MS-EI m/z: 198.1 (100%, MþꢁH2).
;
14. Kochi, J. K.; Powers, J. W. J. Am. Chem. Soc. 1970, 92, 137.
15. (a) Mulzer, J.; Strecker, A. R.; Kattner, L. Tetrahedron Lett. 2004, 45,
8867; (b) Hoppe, D.; Mulzer, J. Methods of Organic Chemistry: Stereose-
lective Synthesis, 4th ed.; Helmchen, G., Ed.; Thieme (Houben-Weyl):
Stuttgart, 1995; Vol. E21b.
Acknowledgements
16. Espenson, J. H. Acc. Chem. Res. 1992, 25, 222.
Marek Smoluch from the Vrije Universiteit Amsterdam, Dr.
Jurgen Schmidt and Christine Kuhnt are acknowledged for
¨
performing mass spectrometry analyses. H.S.S. was supported
by the state of Saxony-Anhalt (HWP).
17. Brown, K. L. Chem. Rev. 2005, 105, 2075.
18. Takai, K.; Matsukawa, N.; Takahashi, A.; Fujii, T. Angew. Chem., Int. Ed.
1998, 37, 152.
19. Wessjohann, L. A. Habilitation Thesis, Ludwig Maximilians Universitat
Munchen: Munchen, 1998.
¨
¨
¨
20. (a) Kochi, J. K.; Mocadlo, P. E. J. Am. Chem. Soc. 1966, 4094; (b)
Okude, Y.; Hirano, S.; Hiyama, T.; Nozaki, H. J. Am. Chem. Soc.
1977, 3179.
Supplementary data
Supplementary data and representative spectra associated
with this article can be found in the online version, at
doi:10.1016/j.tet.2007.12.039.
21. (a) Kochi, J. K.; Singleton, D. M. J. Am. Chem. Soc. 1968, 90, 1582; (b)
Kochi, J. K.; Davis, D. D. J. Am. Chem. Soc. 1964, 86, 5264.
22. (a) Okude, Y.; Hiyama, T.; Nozaki, H. Tetrahedron Lett. 1977, 18, 3829;
(b) Sustmann, R.; Altevogt, R. Tetrahedron Lett. 1981, 22, 5167.
23. (a) Lubbers, T.; Schafer, H. J. Synlett 1992, 743; (b) Tashtoush, H. I.;
¨
¨
References and notes
Sustmann, R. Chem. Ber. 1993, 126, 1759; (c) Hackmann, C.; Schafer,
¨
´
H. J. Tetrahedron 1993, 49, 4559; (d) Auge, J.; Gil, R.; Kalsey, S. Tetra-
hedron Lett. 1999, 40, 67.
´
1. Palomo, C.; Oiarbide, M.; Garcıa, J. M. Chem. Soc. Rev. 2004, 33, 65 and
references therein.
24. Curran, D. P.; Fevig, T. L.; Jasperse, C. P.; Totleben, M. J. Synlett 1992,
943.
2. For reviews, see: (a) Wessjohann, L. A.; Scheid, G. Synthesis 1999, 1; (b)
Furstner, A. Chem. Rev. 1999, 99, 991; (c) Smith, K. M. Coord. Chem.
25. Olah, G. A.; Karpeles, R.; Narang, S. C. Synthesis 1982, 963.
26. Palmer, M. J.; Kenny, J. A.; Walsgrove, T.; Kawamoto, W.; Wills, M.
J. Chem. Soc., Perkin Trans. 1 2002, 416.
27. Choudary, B. M.; Chowdari, N. S.; Jyothi, K.; Kantam, M. L. J. Am.
Chem. Soc. 2002, 124, 5341.
¨
Rev. 2006, 250, 1023; (d) Takai, K. Org. React. 2004, 64, 253.
3. (a) Gabriel, T.; Wessjohann, L. Tetrahedron Lett. 1997, 38, 1363; (b)
Gabriel, T.; Wessjohann, L. Tetrahedron Lett. 1997, 38, 4387; (c)
Wessjohann, L.; Gabriel, T. J. Org. Chem. 1997, 62, 3772; (d) Wessjo-
hann, L.; Wild, H. Synthesis 1997, 512; (e) Wessjohann, L.; Wild, H.
Synlett 1997, 731; (f) Schrekker, H. S.; de Bolster, M. W. G.; Orru,
28. Kise, N.; Hirata, Y.; Ueda, N. J. Org. Chem. 2001, 66, 862.
29. Limmert, M. E.; Roy, A. H.; Hartwig, J. F. J. Org. Chem. 2005, 70, 9364.

2142
L.A. Wessjohann et al. / Tetrahedron 64 (2008) 2134e2142
30. Yang, W. K.; Cho, B. T. Tetrahedron: Asymmetry 2000, 11, 2947.
31. Wang, L.; Zhang, L. Tetrahedron 1998, 54, 11129.
32. Li, C.-J.; Meng, Y.; Yi, X.-H.; Ma, J.; Chan, T.-H. J. Org. Chem. 1998, 63,
7498.
34. Kulasegaram, S.; Kulawiec, R. J. J. Org. Chem. 1997, 62, 6547.
35. Nichols, M. A.; McPhail, A. T.; Arnett, A. M. J. Am. Chem. Soc. 1991,
113, 6222.
36. Salvi, N. A.; Chattopadhyay, S. Tetrahedron 2001, 57, 2833.
37. Asao, N.; Ohishi, T.; Sato, K.; Yamamoto, Y. Tetrahedron 2002, 58, 8195.
38. Clerici, A.; Porta, O. J. Org. Chem. 1985, 50, 76.
33. Kusakabe, M.; Kitano, Y.; Kobayashi, Y.; Sato, F. J. Org. Chem. 1989, 54,
2085.