Enantioselective C-H carbene insertions with homogeneous and immobilized copper complexes

Article abstract of DOI:10.1039/c1ob05499f

The efficiency of chiral bis(oxazoline)- and azabis(oxazoline)-copper complexes in the enantioselective carbene insertion into C-H bonds of cyclic ethers in homogeneous phase strongly depends on the structure of the substrate. The immobilization on laponi

Full text of DOI:10.1039/c1ob05499f

View Article Online / Journal Homepage / Table of Contents for this issue
Organic &
Biomolecular
Chemistry
Dynamic Article Links
Cite this: Org. Biomol. Chem., 2011, 9, 6075
www.rsc.org/obc
PAPER
Enantioselective C–H carbene insertions with homogeneous and immobilized
copper complexes†
Jose´ M. Fraile,* Pilar Lo´pez-Ram-de-Viu, Jose´ A. Mayoral, Marta Rolda´n and Jorge Santafe´-Valero
Received 31st March 2011, Accepted 27th May 2011
DOI: 10.1039/c1ob05499f
The efficiency of chiral bis(oxazoline)- and azabis(oxazoline)-copper complexes in the enantioselective
carbene insertion into C–H bonds of cyclic ethers in homogeneous phase strongly depends on the
structure of the substrate. The immobilization on laponite clay by electrostatic interactions not only
allows the recovery and reuse of the heterogeneous catalysts, but in some cases also improves
enantioselectivity and overall chemoselectivity, making possible reactions that do not take place or lead
to low yields in solution, even with the commonly used Rh2[S-DOSP]4 catalyst.
bis(imine) complex was used in the synthesis of methyl phenidate
Introduction
with low enantioselectivity (18% e.e.).9
The great potential of carbene insertion reactions for the carbon-
In this context, our group described the first use of chiral
carbon bond formation from the poorly reactive carbon-hydrogen
bis(oxazoline)- and azabis(oxazoline)-copper complexes as cat-
bonds has attracted considerable interest towards this type of
alysts for the enantioselective intermolecular carbene insertion
reaction.1 Catalysis is very important to control the high reactivity
of the free carbenes in order to obtain chemoselective processes.
Dirhodium(II) compounds have been the most used and versatile
catalysts for carbene insertions, mostly the enantioselective version
of both intra- and intermolecular reactions.2 Very recently the
use of a salen-Ir complex as catalyst for the enantioselective
insertion into THF and 1,4-cyclohexadiene has been described
with very high enantioselectivity (83–99% ee).3 Coinage metals
(copper, silver, and gold) are emerging as alternatives to rhodium,4
although mostly for non-enantioselective reactions using non-
chiral trispyrazolylborate and analogous ligands.5 The use of these
metals in enantioselective carbene insertions is rather scarce and
poor results are commonly reported. Copper complexes with
chiral bis(oxazolines) and related ligands have been described
as catalysts for intramolecular insertions of carbenoids from
phenyliodonium ylides6 or diazocompounds.7 Those reactions
take place with low yields or chemoselectivities, and moderate to
low enantiomeric excess, 15–60% e.e. and 6–74% e.e. respectively.
Only very recently Maguire and coworkers have described the
efficient intramolecular insertion of a-diazosulfones with high
to excellent enantioselectivities.8 Even less examples of enantios-
elective intermolecular carbene insertions catalyzed by metals
different from rhodium can be found in the literature. A copper-
into C–H bonds of THF.10 The immobilization by electrostatic
interactions on laponite clay, forming a catalyst-support ion pair,
led to improved catalysts, not only in activity and recoverability,
but also in enantioselectivity due to a surface effect of the
support.11 In this paper we extend the study to other ethers
to analyze several factors on the different selectivities (chemo-,
regio-, diastereo-, and enantioselectivity) of the reactions. It
is important to note that, to the best of our knowledge, the
enantioselective version of these reactions had not been previously
described.
Results and discussion
Reactions of methyl phenyldiazoacetate with tetrahydropyran
(THP), 1,4-dioxane, and 1,3-dioxolane (Scheme 1) were studied
using several copper complexes with chiral bis(oxazoline)12 (1) and
azabis(oxazoline)13 (2) ligands (Chart 1). For the sake of compar-
ison, the previously described results of the reaction with THF,10
as well as those obtained with the commercially available Rh2[S-
DOSP]4 catalyst, are also included (Table 1). Two diastereomers
are obtained with THF, THP, and 1,4-dioxane. The unlike and
like terms14 have been used instead of the syn/anti nomenclature
also used in the literature. In all cases homogeneous catalysts are
compared, under the same conditions, with the analogous catalysts
immobilized by cation exchange on laponite clay (Scheme 2),15
forming an ion pair between the clay and the cationic complex,
and precluding metal leaching unless copper neutral species are
generated. The reactions were truly heterogeneous, as shown by
the lack of activity of the filtrate, and elemental analyses showed
the absence of metal leaching.
Departamento de Qu´ımica Orga´nica, Instituto de Ciencia de Materiales de
Arago´n and Instituto Universitario de Cata´lisis Homoge´nea, Facultad de
Ciencias, Universidad de Zaragoza – C.S.I.C., E-50009, Zaragoza, Spain.
E-mail: jmfraile@unizar.es; Fax: +34 976 762077; Tel: +34 976 761000
(ext. 3514)
† Electronic supplementary information (ESI) available: Experimental
procedures, characterization data of every compound, and NMR spectra.
See DOI: 10.1039/c1ob05499f
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 6075–6081 | 6075
©

View Article Online
Table 1 Results of the enantioselective reaction between methyl phenyldiazoacetate and different substrates catalyzed by copper complexes in solution
and immobilized on laponite claya
Entry
Substrate
THF
Ligand
Type of catalyst
Conv. (%)b,c
unlike/likec
% ee unliked
% ee liked
1
2
3
4
5
6
7
8
[Rh]e
1a
Homog
Homog
Homog
Homog
Homog
Immob
Immob
Immob
Immob
Homog
Homog
Homog
Homog
Homog
Immob
Immob
Immob
Homog
Homog
Homog
Homog
Homog
Immob
Immob
Immob
Homog
Homog
Immob
Immob
Immob
Immob
—
48
85
54
74
66
60
27
67 : 33f
64 : 36
64 : 36
74 : 26
56 : 44
75 : 25
74 : 26
51 : 49
59 : 41
63 : 37
65 : 35
58 : 42
62 : 38
51 : 49
58 : 42
63 : 37
46 : 54
39 : 61
37 : 63
32 : 68
43 : 57
29 : 71
43 : 57
37 : 63
30 : 70
—
72(90)f
59
59
0
64
nd
40
55
-12
48
2a
2b
2c
1a
84
58
4
62
39
51
16
59
2a
2b
9
2c
50
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
THP
[Rh]e
1a
37(26)g
9
52(72)g
33
28(42)g
17
2a
43
19
40
39
77
43
53
-11
51
68
58
48
38
71
41
2b
2c
1a
2a
60
2c
39
71
1,4-Dioxane
1,3-Dioxolane
[Rh]e
1a
<1(16)g
10
- (41)g
62
- (59)g
71
2a
8
3
11
7
30
38
-7
0
24
48
45
0
35
45
54
50
—
—
—
—
—
—
2b
2c
1a
2a
11
36
2c
[Rh]e
1a
21(10)g
0
49(76)g
—
—
—
—
—
1a
51
66
47
76
38
33
56
31
2a
2b
2c
—
a Reaction conditions: 2% catalyst (Cu/diazocompound ratio), substrate as solvent, slow addition (2 h) of diazo compound, reflux. b Conversion to
insertion products. As total conversion of diazocompound was observed in all cases, this value reflects the chemoselectivity to the insertion reaction.
c Determined by GC. d THF: determined by HPLC (Chiralcel OD-H); (aS,2R) is the major unlike product. THP: determined by GC (Cyclodex-b); (aS,2R)
was tentatively assigned as the major unlike product. 1,4-Dioxane and 1,3-dioxolane: determined by GC (Cyclodex-g). e Results with Rh2[S-DOSP]4.
f Ref. 16 In parenthesis% ee at -50 ◦C (increased up to 97% ee in hexane). g In parenthesis results at rt.
Scheme 2 Cationic exchange on laponite.
Chart 1 Oxazoline-containing ligands used along this work.
Catalytic activity in C–H insertion (chemoselectivity)
As in the case of Rh catalysts16 the conversion of diazocompound
to C–H insertion products is limited by side reactions, mainly
the dimerization of the diazocompound to give both cis and
trans alkenes, and then the values of conversion are representative
Scheme 1 Insertion reactions of methyl phenyldiazoacetate with different
cyclic ethers.
for the chemoselectivity of the reaction. Tests of insertion into
6076 | Org. Biomol. Chem., 2011, 9, 6075–6081
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
THF with 1a/Cu(OTf)2 ratios in the range of 0.8–2.0 showed
constant enantioselectivities for ratios higher than 1.1, whereas
catalytic activity was reduced, demonstrating that the formation
of 2 : 1 complexes has no positive effect on the insertion reaction.
A ligand/Cu ratio = 1.1 was used with the rest of ligands and
substrates in homogeneous phase. The performance of triflate was
shown to be better than that of other anions such as acetate,
-
bromide, chloride, and SbF6 .
Scheme 3 Proposed mechanism for polymerization of 1,3-dioxolane.
All the copper triflate complexes were active for the insertion
reaction, with yields in the range 40–85% for THF (Table 1,
entries 2–5). A significant effect of ligand on catalytic activity
(chemoselectivity to insertion) is observed, with azabox ligands
(2) being clearly superior, a fact that may be related to their
higher coordinating ability17 and that has been also observed in
other reactions.18 The substituent in position 4 of the oxazoline
ring shows also an important effect on activity (chemoselectivity),
of the oxygen to the carbene may occur with all the substrates in a
reversible way, but in this case it leads to a productive reaction. The
reason for this difference is not clear but it must be related to the
presence of the second oxygen, which increases the electrophilic
character of the carbon in position 2, favouring the attack of a
second molecule of 1,3-dioxolane.
A similar behaviour would be expected in other cases in which
the breakage of the C–O was irreversible for example in the
case of elimination of an alkene. Diisopropylether was chosen
as model for this test, because the break of the O–C bond and the
transfer of a proton would produce the elimination of propylene
(Scheme 4), in an irreversible step. Confirming this hypothesis the
product of C–H insertion was not detected. On the contrary the
product coming from the insertion of the carbene in the O–H bond
of isopropanol was always obtained with both homogeneous and
heterogeneous catalysts (Scheme 4).
i
according to results in other reactions,18 in the order Ph > Pr
t
> Bu. This methodology was then extended to the insertion in
tetrahydropyran (THP) due to the structural similarity to THF
and to the interest of the insertion product as an analogue
of methyl phenidate for dopamine and serotonin transporter
inhibition.19 As can be seen (Table 1, entries 11–14) THP is less
reactive than THF and leads to moderate to low conversions (9–
43%), in the same range of result obtained with commercially
available Rh2[S-DOSP]4 (Table 1, entry 10). This effect of ring size
is analogous to that previously observed for N-Boc-piperidine
in comparison with N-Boc-pyrrolidine with Rh catalysts.20 The
donor character of the ligand17 seems to play again a role, with the
same general trend observed for THF: 2a ≥ 2c ꢀ 2b ≥ 1a.
The next substrate tested was 1,4-dioxane. In spite of having
four active positions for insertion, very low yields (3–11%, up to
16% with Rh2[S-DOSP]4 at rt) were obtained in solution (Table
1, entries 18–22). Similar detrimental effects of the presence of
an oxygen in b-position had been observed in the enantioselective
insertion of methyl phenyldiazoacetate into both crown ethers
and 1,2-dimethoxypropane with Rh catalysts,21 and in the non-
enantioselective insertion of ethyl diazoacetate into 1,4-dioxane
with homogeneous copper5b and silver scorpionates.22 The pres-
ence of a second oxygen in a-position was tested in the case of
1,3-dioxolane. In principle position 2 in this substrate is more
activated, due to the presence of the two adjacent oxygen atoms.
Analogue Rh-catalyzed insertion reactions had been car-
ried out on 2-aryl-1,3-dioxolanes, as equivalent to a Claisen
condensation,23 but the insertion on the unsubstituted 1,3-
dioxolane, which had not been described in the literature, leads
to low yields (up to 21% under reflux, entry 26). Surprisingly,
in the case of the homogeneous reactions with copper catalysts,
the first drop of diazocompound converted the reaction solution
Scheme
diisopropylether.
4 Reaction between methyl phenyldiazoacetate and
The same reasoning would justify the lower chemoselectivity
observed with 1,4-dioxane, with a breakdown process leading
to methyl 2-hydroxy-2-phenylacetate, detected in the reaction
medium, and volatile by-products such as acetaldehyde and
ethylene.
Whereas immobilization is only moderately positive for chemos-
electivity of insertion in THF with 1a (66% vs. 48 in solution,
entries 6 and 2) and detrimental with azabis(oxazolines) (27–
60% conversion vs. 54–85% in solution, entries 7–9 vs. 3–5),
it has a significant positive effect with the rest of less reactive
substrates (Table 1). In the case of THP immobilization is highly
beneficial for phenyl substituted ligands 1a (39% vs. 9% in solution,
entries 15 and 11) and 2a (77% vs. 43% in solution, entries 16
and 12), although this effect is not observed for the isopropyl
substituted azabox 2c (entry 17). The immobilized catalysts also
led to moderate to excellent conversions to the expected insertion
product in 1,3-dioxolane (47–76%, entries 28–31), in contrast with
the homogeneous catalysts. Given that polymerization does not
take place with the immobilized catalyst, it may be speculated
the need for a coordination of both oxygens, making the carbon
atom much more reactive, impossible with the solids due to the
site isolation effect. This is not the case with diisopropyl ether,
1
into a gel-like mass. The H-NMR spectrum of this gel showed
broad signals at 4.74 (2H) and 3.71 ppm (4H), compatible with a
polymer formed by dioxolane ring opening, and confirmed by the
detection of species in the m/z range of 300–1600 with differences
of 74 (supplementary material†). Even under hexane dilution
conditions, the only detectable product of the homogeneous
reaction was this polymer. This polymerization reaction may take
place through an attack of the oxygen to the electrophilic carbon
of the carbenoid (Scheme 3), in the same fashion proposed for the
attack of THF to carbene-iron-porphyrin complexes, in which one
O–Ca bond is broken leading to a C–C double bond.24 The attack
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 6075–6081 | 6077
©

View Article Online
because the elimination of propylene is an intramolecular reaction.
Probably the fact that breakdown in the case of 1,4-dioxolane does
not require the participation of two molecules of catalyst makes
the increase in chemoselectivity not so noticeable, with the highest
values not exceeding the 30–38% obtained using azabox ligands
(entries 24–25).
It is worthy to note that insertions of methyl phenyldiazoacetate
into THP, 1,4-dioxane, and 1,3-dioxolane had not been previously
described. Thus, taking into account the reactivity problems
detected in homogeneous phase, even in the case of Rh2[S-
DOSP]4 and in agreement with those described for closely related
substrates, the values of conversion to insertion products (77%
with THP, 38% with 1,4-dioxane, and 76% with 1,3-dioxolane)
obtained with the immobilized catalysts are remarkable.
moderate, up to 53% ee in the unlike isomers (entry 12) and up
to 71% ee in the like isomers (entry 14). The substituents in box
and azabox follow the same trend as that observed with THF.
The results of enantioselectivity unlike with THP are similar to
those obtained with Rh2(S-DOSP)4 under the same conditions
(52% ee, entry 10), although in this case again enantioselectivity
can be improved by lowering the temperature. On the contrary the
results of enantioselectivity like are in general better with copper
complexes.
One interesting consequence of the immobilization was the
modification of the stereoselectivities, highly dependent on the
nature of the ligand. Whereas the immobilization of azabox-Cu
complexes produces only marginal variations of enantioselectivity
(entries 7–9 and 16–17), the immobilization of 1a-Cu produces
a significant improvement in enantioselectivity of unlike isomers,
84% ee vs. 59% ee in solution with THF (entry 6), 68% ee vs. 33% ee
in solution with THP (entry 15). Perhaps less important, but also
significant, is the simultaneous variation of diastereoselectivity.
In Fig. 1 it can be seen how the correlation between enantio-
and diastereoselectivity completely changes after immobilization,
mainly with THF. Interestingly, with the immobilized catalysts
the general trend is to obtain at the same time high diastereo- and
enantioselectivity. It is important to note that the immobilized Cu
catalysts allow reaching enantioselectivities in the same range as
those obtained with Rh2(S-DOSP)4, but in general with higher
yields of insertion products.
It might be speculated whether the changes in yield and
stereoselectivity are due to a surface effect, as proposed for cyclo-
propanation reactions,25 or to a change in the reaction mechanism
with the immobilized catalyst. The mechanism was checked by
kinetic isotopic effect, determined by competitive reaction in
the presence of equimolar amounts of THF and THF-d8. An
isotopic effect kH/kD = 2 was obtained for both homogeneous
and heterogeneous reactions, showing that the changes observed
in stereoselectivities cannot be ascribed to deep changes in the
reaction mechanism, but to changes in the relative energies of the
transition states leading to the four products. This value is lower
Stereoselectivity
The insertions into THF in homogeneous phase lead to the unlike
product as major diastereoisomer with moderate to low selectivity
(56 : 44–74 : 26, entries 1–5). Enantioselectivities in the unlike iso-
mer are also moderate, up to 64% ee (entry 5). Phenyl or isopropyl
substituents in box and azabox lead to similar enantioselectivity,
better than tert-butyl, a situation in contrast with the behaviour of
the same ligands in the related cyclopropanation reaction. Another
important feature is the relationship between diastereoselectivity
and enantioselectivity, as can be seen in Fig. 1. In general lower
diastereoselectivity is associated with higher enantioselectivity.
The results with THF are slightly worse than those obtained with
Rh2(S-DOSP)4 under the same conditions (67 : 33 diastereomeric
ratio, 72% ee syn, entry 1).16b With Rh catalysts enantioselectivity
was improved by lowering the temperature or using hexane as
◦
solvent (up to 97% ee at -50 C).16b However none of those
methods improved enantioselectivity in the case of 2a-Cu(OTf)2.
Enantioselectivity of the like isomers follows a similar trend, up
to 55% ee.
than that was obtained for the Rh-catalyzed reaction (kH/kD
=
3),16b probably due to a slightly different geometry of the transition
state, whereas the same value of 2 was found for the Rh-catalyzed
insertion into cyclohexene. This value is in agreement with the
insertion as the rate limiting step, a situation different from that of
the cyclopropanation of styrene with ethyl diazoacetate promoted
by the same catalysts, in which the formation of the copper-carbene
intermediate is the rate limiting step.26
In order to get some deeper insight in the stereochemical course
of the reactions, the reaction products were separated by semi-
preparative chiral HPLC27 (see ESI†). The four insertion products
in THF were obtained in enantiopure form. In this way the
absolute configuration of the major unlike product was confirmed
to be (2R,aS) with [a]2D5 -88.1◦ (c 0.159, CHCl3).‡ The major like
product showed also negative value of the specific rotation, [a]D25
-19.3◦ (c 0.7, CHCl3).
Fig. 1 Relationship between diastereo- (% unlike) and enantioselectivity
(% ee unlike) of the reactions of methyl phenyldiazoacetate with THF (᭜)
and THP (᭡) in homogeneous (filled symbols, tendency in full line) and
heterogeneous (open symbols, tendency in dashed line) phase.
In order to check the absolute configuration, an epimerization of
the a-carbon of this product under basic conditions was tested to
The insertions into THP in homogeneous phase also lead
to the unlike product as major diastereoisomer with similar
selectivity (51 : 49–65 : 35, entries 11–14), again in the same range
as that obtained with Rh (entry 10). Enantioselectivities are also
‡ The value reported in ref. 16 is [a]2D5 -74.0◦ (c 0.2, CHCl3). In our case
the value obtained for the minor unlike isomer, absolute configuration
(2S,aR), is [a]2D5 +78.1◦ (c 0.167, CHCl3).
6078 | Org. Biomol. Chem., 2011, 9, 6075–6081
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
obtain the corresponding mixture of unlike/like products with the
same absolute configuration in position 2 of the furan (Scheme 5).
Unexpectedly the reaction with sodium methoxide led to a 25 : 75
mixture of unlike:like in racemic form. To discard the possible
abstraction of the proton in position 2 of THF, the reaction
was carried out in d4-methanol obtaining in this way the same
mixture of products but deuterated only in a-position and in
the methyl ester (Scheme 5). The ring opening must then take
place through a mechanism outlined in Scheme 3, similar to that
proposed to explain the ring opening of bicyclic N,O-acetal serine
derivatives.28 Unfortunately this result did not allowed the absolute
configuration of the like products to be determined.
Fig. 2 Newman projections of like and unlike compounds of insertion
into 1,4-dioxane.
ee, entries 24–25) and reduction for box 1a (45% vs. 71% ee in
solution, entry 23).
In the case of 1,3-dioxolane, only products of insertion in
position 2 of the dioxolane ring were detected in all cases, showing
a very high regioselectivity. Products of insertion in position 4
had been obtained with 2-aryl-1,3-dioxolanes as substrates, in
◦
proportions ranging from 29% at 50 ◦C to 18% at -30 C.23
Enantioselectivity was moderate with all the ligands, with values in
the range of 31–56% ee (entries 28–31). The best enantioselectivity
was obtained with the ligand bearing tert-butyl groups (56% ee),
in contrast with results obtained with the rest of substrates. This
result is analogous to that obtained with Rh2[S-DOSP]4 under the
same conditions (49% ee, entry 26), and slightly worse than the
result at r.t. (76% ee), although this higher ee is accompanied by
an important decrease in yield up to only 10%.
Scheme 5 Epimerization reaction of the major like product of insertion
into THF.
Recoverability
With respect to the absolute configuration of the major isomer
of insertion into THP, the separation by semi-preparative HPLC
was performed in a similar way to that described for the products
of THF insertion, but in this case the separation was more
difficult, due to the lower solubility of the mixture (ESI†). Both
unlike enantiomers were obtained in enantiopure form, but like
enantiomers were obtained only as a 5/95 mixture. The major
unlike compound showed [a]2D5 -76.4◦ (c 0.55, CCl4). Taking
into account the similar nature of THP and THF, the induction
mechanism should be the same and (2R,aS) configuration can be
tentatively assigned to the major unlike product, also confirmed
by the same sign of the specific rotation. Additionally the sign is
also the same as the product with 2 chlorine atoms in 3,4 positions
of the aromatic ring described in the literature.19a
An additional advantage of immobilized catalysts is the possibility
of recovery and reuse (Table 2). In the case of THF recovery was
very efficient with box 1a in three reuses, leading to a total TON of
134 with 75 : 25 diastereoselectivity and 83% ee, much better than
TON 24 with 64 : 36 and 59% ee of the homogeneous catalyst.
With the rest of catalysts only one or two reuses were tried, with
similar results to those obtained in the first run, showing the high
stability of the catalysts under the reaction conditions.
In the case of THP most catalysts were efficiently recovered, even
with improved catalytic activity as it happens with 2c. Whereas 1a-
Cu loses activity and selectivity upon recovery probably due to a
partial loss of ligand, the higher stability of 2a-Cu and 2c-Cu
complexes17 makes them more suitable for this reaction, keeping
the activity and enantioselectivity for 3 cycles.
In the case of 1,4-dioxane the presence of the oxygen in the
ring produces a change in the priority of the substituents of
the stereogenic centre of the ring, and hence in its absolute
configuration. This means that the product coming from the
attack by the same face will have different relative configuration
unlike/like with respect to that of THP. This relative configuration
was assigned from the shielding effect of the phenyl group on
protons in C3 in the most stable antiperiplanar conformation
(Fig. 2).29 Due to this change, the like preference obtained with
1,4-dioxane does not represent any change in reaction mechanism.
Enantioselectivities were moderate in solution with azabox 2a
and 2c, whereas the highest values were obtained with box 1a
(71% ee, entry 19). This result is clearly better than that obtained
with Rh2[S-DOSP]4. at r.t. (59% ee, entry 18). The use of the
immobilized catalysts produced variations in enantioselectivity,
but without any pattern, with improvements for azabox (50–54%
The low chemoselectivity of the insertion into 1,4-dioxane, with
the generation of the corresponding by-products, that poison the
catalyst or favor leaching of the chiral ligand, may be the origin for
the poor recyclability of the immobilized catalysts in this reaction.
Only azabox 2c allows efficient recovery regarding both activity
and selectivity in one recycle.
The high chemoselectivity obtained with immobilized catalysts
in the insertion into 1,3-dioxolane, probably due to site isolation
as pointed above, allows efficiently recovering the catalysts with
similar enantioselectivity and even improved activity, as in the
case of 2c. Unfortunately the lower stability of 2b-Cu makes the
recovery inefficient, with loss of enantioselectivity.
It is the clear that there is a close relationship between
chemoselectivity and catalyst recovery, as observed in the closely
related cyclopropanation,30 because of the detrimental effect of
by-products.
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 6075–6081 | 6079
©

View Article Online
Table 2 Results of recovery of the copper complexes immobilized on laponite clay in the enantioselective reaction between methyl phenyldiazoacetate
and different substratesa
Substrate
THF
Ligand
Run
Conv. (%)b,c
unlike/likec
% ee unliked
% ee liked
1a
1
2
3
4
5
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
66
71
72
65
37
60
55
61
50
56
62
39
25
31
77
54
59
43
53
57
7
10
12
30
34
26
38
37
8
51
54
49
66
58
44
47
38
37
76
86
85
75 : 25
77 : 23
75 : 25
72 : 28
54 : 46
74 : 26
74 : 26
73 : 27
59 : 41
58 : 42
56 : 44
58 : 42
55 : 45
55 : 45
63 : 37
60 : 40
59 : 41
46 : 54
46 : 54
47 : 53
43 : 57
45 : 55
47 : 53
37 : 63
38 : 62
36 : 64
30 : 70
28 : 72
35 : 65
—
84
83
82
81
28
58
57
56
62
61
58
68
33
31
58
62
58
39
42
36
48
18
7
39
40
39
40
24
51
51
54
59
59
56
41
16
13
60
65
59
71
70
67
45
41
25
54
33
20
50
46
37
—
—
—
—
—
—
—
—
—
—
—
—
2a
2c
1a
2a
2c
1a
2a
2c
1a
2a
2b
2c
THP
1,4-Dioxane
11
9
12
36
30
16
38
27
26
33
34
30
56
42
36
31
28
27
1,3-Dioxolane
—
—
—
—
—
—
—
—
—
—
—
a Reaction conditions: 2% catalyst (Cu/diazocompound ratio), substrate as solvent, slow addition (2 h) of diazo compound, reflux. b Conversion to
insertion products. As total conversion of diazocompound was observed in all cases, this value reflects the chemoselectivity to the insertion reaction.
c Determined by GC. d THF: determined by HPLC (Chiralcel OD-H); (aS,2R) is the major unlike product. THP: determined by GC (Cyclodex-b);
(aS,2R) was tentatively assigned as the major unlike product. 1,4-Dioxane and 1,3-dioxolane: determined by GC (Cyclodex-g).
reactions. An additional advantage of the immobilized catalysts is
Conclusions
the improvement of the stereoselectivity, up to 84% ee with THF
Homogeneous box- and azabox-copper complexes are able to
and 68% ee with THP combined with high diastereoselectivity.
catalyze the enantioselective insertion of methyl phenyldiazoac-
In summary the immobilized copper complexes are promising
etate into C–H bonds of cyclic ethers. Chemo-, diastereo-, and
catalysts for carbene insertion reactions, given that in the most de-
enantioselectivities strongly depend on the nature of the substrate.
The best results are obtained with THF (up to 64% ee), whereas
manding reactions they allow reaching higher chemoselectivities,
with similar stereoselectivities, than the commonly used Rh2(S-
1,4-dioxane is much less reactive (up to 11% conversion) and 1,3-
DOSP)4 catalyst, an advantage probably due to the site isolation
dioxolane leads only to polymerization products.
effect.
When the same complexes are immobilized by electrostatic
interactions with laponite clay, the catalytic performance consider-
Experimental
ably improves. In all cases the productivity increases, either in one
single reaction, up to 38% conversion with 1,4-dioxane vs. 11%
in solution, or by efficient recovery in at least three consecutive
runs. Moreover, the immobilized catalysts are able to efficiently
promote the insertion in position 2 of 1,3-dioxolane, with up
to 56% ee preventing the insertion in position 4 and other side
General procedure for carbene insertion with heterogeneous
catalysts
A suspension of dried heterogeneous catalyst (100 mg) in an-
hydrous substrate (THF, THP, 1,4-dioxane, or 1,3-dioxolane,
6080 | Org. Biomol. Chem., 2011, 9, 6075–6081
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
10 mL) with n-decane (100 mg) as internal standard was heated
under reflux under an inert atmosphere. A solution of methyl
phenyldiazoacetate (1 mmol) in anhydrous substrate (10 mL)
was slowly added during 2 h with a syringe pump. Once the
addition had finished, the reaction mixture was stirred and heated
under reflux for 30 min. The catalyst was filtered off and washed
with substrate (5 mL). The yield and diastereoselectivity were
determined by GC and enantioselectivities were determined by
HPLC or GC. The catalyst was dried under vacuum and reused
under the same conditions.
9 J. M. Axten, R. Ivy, L. Krim and J. D. Winkler, J. Am. Chem. Soc.,
1999, 121, 6511.
10 J. M. Fraile, J. I. Garc´ıa, J. A. Mayoral and M. Rolda´n, Org. Lett., 2007,
9, 731.
11 J. M. Fraile, J. I. Garc´ıa, C. I. Herrer´ıas, J. A. Mayoral and E. Pires,
Chem. Soc. Rev., 2009, 38, 695.
12 G. Desimoni, G. Faita and K. A. Jørgensen, Chem. Rev., 2006, 106,
3561.
13 M. Glos and O. Reiser, Org. Lett., 2000, 2, 2045; H. Werner, R. Vicha,
A. Gissibl and O. Reiser, J. Org. Chem., 2003, 68, 10166.
14 IUPAC. Compendium of Chemical Terminology, (the “Gold Book”)
2nd ed. (compilation A. D. McNaught, A. Wilkinson), Blackwell
Scientific Publications, Oxford, 1997. XML on-line corrected version:
http://goldbook.iupac.org (2006-), (compilation M. Nic, J. Jirat, B.
Kosata; updates A. Jenkins). 10.1351/goldbook.L03423.
15 J. M. Fraile, J. I. Garc´ıa, C. I. Herrer´ıas, J. A. Mayoral and M. A.
Harmer, J. Catal., 2004, 221, 532.
Acknowledgements
16 H. M. L. Davies and T. Hansen, J. Am. Chem. Soc., 1997, 119, 9075;
H. M. L. Davies, T. Hansen and M. R. Churchill, J. Am. Chem. Soc.,
2000, 122, 3063.
Financial support from the Spanish Ministerio de Ciencia e
Innovacio´n (projects CTQ2008-05138, CTQ2008-00187, and Con-
solider Ingenio 2010 CSD2006-0003), and the Diputacio´n General
de Arago´n (E11 Group co-financed by the European Regional
Development Funds) is gratefully acknowledged. M. R. and J.
S.-V. are indebted to DGA and CSIC respectively for their grants.
17 J. M. Fraile, J. I. Garc´ıa, C. I. Herrer´ıas, J. A. Mayoral, O. Reiser, A.
Socue´llamos and H. Werner, Chem.–Eur. J., 2004, 10, 2997.
18 C. Geiger, P. Kreitmeier and O. Reiser, Adv. Synth. Catal., 2005,
347, 249; L. Aldea, J. M. Fraile, H. Garc´ıa-Mar´ın, J. I. Garc´ıa, C.
I. Herrer´ıas, J. A. Mayoral and I. Pe´rez, Green Chem., 2010, 12, 435.
19 P. C. Meltzer, P. Wang, P. Blundell and B. K. Madras, J. Med. Chem.,
2003, 46, 1538; B. K. Madras, M. A. Fahey, G. M. Miller, R. de la Garza,
M. Goulet, R. D. Spealman, P. C. Meltzer, S. R. George, B. F. O’Dowd,
A. A. Bonab, E. Livnid and A. J. Fischman, Eur. J. Pharmacol., 2003,
479, 41.
20 H. M. L. Davies, C. Venkataramani, T. Hansen and D. W. Hopper, J.
Am. Chem. Soc., 2003, 125, 6462.
21 H. M. L. Davies and J. Yang, Adv. Synth. Catal., 2003, 345, 1133.
22 H. V. R. Dias, R. G. Browning, S. A. Richey and C. J. Lovely,
Organometallics, 2004, 23, 1200.
Notes and references
1 M. P. Doyle, R. Duffy, M. Ratnikov and L. Zhou, Chem. Rev., 2010,
110, 704.
2 H. M. L. Davies and R. E. J. Beckwith, Chem. Rev., 2003, 103, 2861.
3 H. Suematsu and T. Katsuki, J. Am. Chem. Soc., 2009, 131, 14218.
4 M. M. D´ıaz-Requejo and P. J. Pe´rez, Chem. Rev., 2008, 108, 3379.
5 M. M. D´ıaz-Requejo, T. R. Belderra´ın, M. C. Nicasio, S. Trofimenko
and P. J. Pe´rez, J. Am. Chem. Soc., 2002, 124, 896; A. Caballero, M. M.
D´ıaz-Requejo, T. R. Belderra´ın, M. C. Nicasio, S. Trofimenko and P.
J. Pe´rez, Organometallics, 2003, 22, 4145; A. Caballero, M. M. D´ıaz-
Requejo, T. R. Belderra´ın, M. C. Nicasio, S. Trofimenko and P. J. Pe´rez,
J. Am. Chem. Soc., 2003, 125, 1446; H. V. R. Dias, R. G. Browning, S.
A. Richey and C. J. Lovely, Organometallics, 2004, 23, 1200; J. Urbano,
T. R. Belderra´ın, M. C. Nicasio, S. Trofimenko, M. M. D´ıaz-Requejo
and P. J. Pe´rez, Organometallics, 2005, 24, 1528; G. P. A. Yap, F. Jove,
J. Urbano, E. Alvarez, S. Trofimenko, M. M. D´ıaz-Requejo and P. J.
Pe´rez, Inorg. Chem., 2007, 46, 780; J. Pe´rez, D. Morales, L. A. Garc´ıa-
Escudero, H. Mart´ınez-Garc´ıa, D. Miguel and P. Bernad, Dalton Trans.,
2009, 375.
6 P. Mu¨ller and C. Bole´a, Helv. Chim. Acta, 2002, 85, 483.
7 H.-J. Lim and G. A. Sulikowski, J. Org. Chem., 1995, 60, 2326; M. P.
Doyle and W. Hu, J. Org. Chem., 2000, 65, 8839; M. P. Doyle and I. M.
Phillips, Tetrahedron Lett., 2001, 42, 3155.
8 C. J. Flynn, C. J. Elcoate, S. E. Lawrence and A. R. Maguire, J. Am.
Chem. Soc., 2010, 132, 1184; C. R. Slattery and A. R. Maguire, Org.
Biomol. Chem., 2011, 9, 667.
23 H. M. L. Davies, J. Yang and J. Nikolai, J. Organomet. Chem., 2005,
690, 6111.
24 H. M. Mbuvi and L. K. Woo, Organometallics, 2008, 27, 637.
25 A. I. Ferna´ndez, J. M. Fraile, J. I. Garc´ıa, C. I. Herrer´ıas, J. A. Mayoral
and L. Salvatella, Catal. Commun., 2001, 2, 165; M. R. Castillo, L.
Fousse, J. M. Fraile, J. I. Garc´ıa and J. A. Mayoral, Chem.–Eur. J.,
2007, 13, 287; J. I. Garc´ıa, B. Lo´pez-Sa´nchez, J. A. Mayoral, E. Pires
and I. Villalba, J. Catal., 2008, 258, 378; J. M. Fraile, J. I. Garc´ıa, G.
Jime´nez-Ose´s, J. A. Mayoral and M. Rolda´n, Organometallics, 2008,
27, 2246.
26 J. M. Fraile, J. I. Garc´ıa, V. Mart´ınez-Merino, J. A. Mayoral and L.
Salvatella, J. Am. Chem. Soc., 2001, 123, 7616.
27 Y. Okamoto and T. Ikai, Chem. Soc. Rev., 2008, 37, 2593.
28 C. Aydillo, G. Jime´nez-Ose´s, J. H. Busto, J. M. Peregrina, M. M.
Zurbano and A. Avenoza, Chem.–Eur. J., 2007, 13, 4840.
29 H. M. L. Davies and P. Ren, Tetrahedron Lett., 2001, 42, 3149.
30 H. Werner, C. I. Herrer´ıas, M. Glos, A. Gissibl, J. M. Fraile, I. Pe´rez, J.
A. Mayoral and O. Reiser, Adv. Synth. Catal., 2006, 348, 125.
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 6075–6081 | 6081
©