Concave 1,10-phenanthrolines as tailored ligands for copper(I)-catalyzed diastereoselective cyclopropanations

Article abstract of DOI:10.1002/ejoc.200500650

Alkenyloxy-substituted aryl bridgeheads have been connected to 2,9-dichloro-1,10-phenanthroline (11) either by Suzuki coupling or by nucleophilic aromatic substitution. The resulting di- or tetraalkenes 12 or 17 have been cyclized by ring-closing metathesis to give concave 1,10-phenanthrolines 14 and 19, differing from other concave 1,10-phenanthrolines 1 in their geometries. The influence of the shapes of these ligands and those of the related concave 1,10-phenanthrolines 1a-f and 2 on the stereochemistry of the copper(I)-catalyzed cyclopropanation of indene (3) and styrene (4) has been investigated, identifying factors that favour the formation of thermodynamically less stable syn-cyclopropanes 6 and 7. Wiley-VCH Verlag GmbH & Co. KGaA, 2006.

Full text of DOI:10.1002/ejoc.200500650

FULL PAPER
DOI: 10.1002/ejoc.200500650
Concave 1,10-Phenanthrolines as Tailored Ligands for Copper(
I
)-Catalyzed
Diastereoselective Cyclopropanations[‡]
Ulrich Lüning*[a] and Frank Fahrenkrug[a]
Keywords: Cyclopropanation / Homogeneous catalysis / Macrocycles / Metathesis / Supramolecular chemistry
Alkenyloxy-substituted aryl bridgeheads have been con-
nected to 2,9-dichloro-1,10-phenanthroline (11) either by Su-
zuki coupling or by nucleophilic aromatic substitution. The
resulting di- or tetraalkenes 12 or 17 have been cyclized by
ring-closing metathesis to give concave 1,10-phenan-
throlines 14 and 19, differing from other concave 1,10-phen-
anthrolines 1 in their geometries. The influence of the shapes
of these ligands and those of the related concave 1,10-phen-
anthrolines 1a–f and 2 on the stereochemistry of the cop-
per( )-catalyzed cyclopropanation of indene (3) and styrene
(4) has been investigated, identifying factors that favour the
formation of thermodynamically less stable syn-cyclopro-
panes 6 and 7.
I
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2006)
Introduction
The surprising properties of enzymes have been a stimu-
lation for many enterprises in supramolecular chemistry.[1–5]
Because the high selectivity of an enzyme is largely a prod-
uct of the concave environment of the active site,[6] geomet-
rically related molecules such as macrocycles, clefts or pin-
cers have attracted wide interest. Concave reagents have
been developed with the goal of increasing the selectivities
of reactions or catalyses through the corresponding shield-
ing of the reaction site. In particular, concave 1,10-phenan-
throlines 1 and 2 (Figure 1 and Figure 2) have shown good
selectivities in a number of reactions.[7–11]
The active and selective catalysts in many of these reac-
tions are transition metal ion complexes of the ligands 1
and 2: examples include copper() complexes[12] in the ster-
eoselective cyclopropanation of alkenes such as indene (3)
or styrene (4) by diazoacetates 5.[13] The cyclopropanation
of alkenes by diazoacetates has been investigated thor-
oughly, especially for the development of enantioselective
catalysts. Styrene (4) is usually used as substrate, and only
rarely can the cis/trans selectivity by steered in both direc-
tions in good yields.[14,15]
The concave 1,10-phenanthrolines 1 and 2 differ in their
selectivities in the cyclopropanation of both alkenes 3 and
4. While the stiff bimacrocyclic 1,10-phenanthrolines 1 with
aryl bridgeheads in position 2 and 9 give anti products (e.g.,
exo-configured cyclopropanes of indene 6 or trans-config-
Figure 1. Concave 1,10-phenanthrolines 1a–1g with direct CC-con-
nection between the aryl bridgeheads and the 1,10-phenanthroline
system.
ured cyclopropanes of styrene 7; for a definition, see foot-
note [a] in Table 1), the 1,10-phenanthroline-bridged ca-
lix[6]arene 2 predominantly gives the complementary syn
products (cis or endo, respectively).[11] This finding has also
been investigated theoretically,[16] and the ability of 2 to
adopt a bent structure seems to be the reason for the syn
selectivity. While the complexes of the stiff concave 1,10-
[‡] Concave Reagents, 47. Part 46: S. Lüthje, C. Bornholdt, U.
Lüning, Eur. J. Org. Chem., 2006, 909–915, preceding article.
[a] Otto Diels-Institut für Organische Chemie, Christian-Al-
brechts-Universität zu Kiel,
Olshausenstr. 40, 24098 Kiel, Germany
Fax: +49-431-880-1558
E-mail: luening@oc.uni-kiel.de
916
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2006, 916–923

Tailored Ligands for Diastereoselective Cyclopropanations
FULL PAPER
etry resembling a shallow bowl. In both cases, the large
1,10-phenanthroline ligand bound to the copper() ion de-
termines the stereoselectivity of the cyclopropanation. Ac-
cording to calculations,[16] a copper carbenoid is the cyclo-
propanating species, but the orientations of the carbene
moiety and the approaching alkene are different in the two
ligands 1 and 2.
With the stiff and symmetrical ligand 1, the carbene sits
in the cavity, presenting a hydrogen atom and an ester func-
tion towards the approaching alkene (see Figure 3). It is
obvious that the alkene will approach in such a way that its
substituent R will face away from the ester group, resulting
in the formation of anti products. In the opposite orienta-
tion, R and the ester group experience steric repulsion by
one another, which is increased if their mobility is reduced
by the encumbering macrocycle, so the anti selectivity is en-
hanced relative to that of ligand-free catalysis. The corre-
lation between the size of the ester group and the pro-
Figure 2. Calixarene-based ligand 2, synthesized by A,D-bridging
of calix[6]arene with a 2,9-bismethylene-1,10-phenanthroline unit,
in two representations.
Figure 3. a) A copper carbenoid, Cu+=CH–COOR, is the key in-
termediate in the copper()-catalyzed cyclopropanation of alkenes
by diazoacetates. Concave 1,10-phenanthrolines 1 or 2 determine
the syn/anti selectivity by complexing the copper() ion. The coordi-
nation to the nitrogen atoms and the concave shielding is indicated
by the dashed bonds and the half circle. b) View onto the complex
1·Cu+=CH–COOR along the axis formed by the carbene carbon
atom and the copper() ion. The macrocyclic rim consists of the
two aryl bridgeheads and the two chains Y1 and Y2, while the bot-
tom of the cavity is formed by the 1,10-phenanthroline, indicated
by the dashed lines. The copper() ion is not visible: it is hidden by
the carbene carbon atom connecting the ester group and the hydro-
gen atom. The concave shielding only allows attack from the front,
but the alkene can be oriented in two fashions, providing either anti
or syn products. The latter attack is more hindered because the
ester group is larger than the hydrogen atom. c) In complex
2·Cu+=CH–COOR (for detailed structure of the ligand 2 see Fig-
ure 2), the large calixarene moiety only allows the ester group to
point away from the cavity. For the attacking alkene, two sterically
demanding pathways compete: either the alkene approaches from
the side of the ester group, giving the syn product, or the alkene
phenanthrolines 1 possess a lamp-like geometry, with the
copper() ion being the light bulb and the bimacrocycle be-
ing the lamp shade, the bridged calixarene 2 adopts a geom- has to squeeze alongside the calixarene, producing the anti product.
Eur. J. Org. Chem. 2006, 916–923
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
917

U. Lüning, F. Fahrenkrug
FULL PAPER
portion of the anti products is additional evidence of this the bridgeheads to the macrocycle chains Y1 and Y2 would
mechanism.
have to be made differently (not bis-ortho).
In contrast to the symmetrical shielding by 1, the shallow
bowl of the calixarene derivative 2 only allows the ester
group of the corresponding copper carbenoid to point in
the direction of the approaching alkene. The substituents R
on the alkene and the ester group therefore end up syn to
each other in the final cyclopropane products. This sugges-
tion is also supported by the correlation between the size
Results and Discussion
Synthesis of New Concave 1,10-Phenanthrolines
Two approaches to less symmetric and more open macro-
of the ester group and the selectivity. With the substituents cyclic 1,10-phenanthrolines have been pursued, both trying
close to each other, the highest selectivity (cis or endo) is to change the geometry of the bimacrocyclic 1,10-phenan-
obtained with the smallest ester group. The final orienta- throlines 1 from a lamp-like cavity to a shallow pocket-like
tions of the substituents R depend on their interactions structure. The first concept leaves out one chain of the bi-
either with the ester group or with the large calixarene frag- macrocycle, but simultaneously moves the remaining chain
ment.
away from the 1,10-phenanthroline. The result is a mono-
In absolute numbers, better selectivity can be achieved macrocycle 14 (see below), with meta-substituted aryl rings
for anti products (anti/syn value 140:1) than for syn prod- in the 2- and 9-positions of the 1,10-phenanthroline. The
ucts (best syn/anti value 86:14). In order to fine-tune and construction of this macrocycle might use the well estab-
to optimize the syn selectivity, the selectivity-determining lished sequence[13,24] of Suzuki coupling, ring-closing me-
ligand 2 must be optimized. However, the calix[6]arene tathesis and hydrogenation.
macrocycle in 2 cannot easily be altered[17–19] because the
Therefore, commercially available 3-bromophenol (8) was
diameter of the calixarene is defined by the (easy) synthesis used as starting material, being alkenylated with 5-bro-
of this class of compounds.[20] Only calix[4]-, calix[6]- and mopent-1-ene in 95% yield in a Williamson ether synthesis.
calix[8]arenes are easily accessible,[21,22] and of these, only Lithiation with n-butyllithium, followed by treatment with
calix[6]arene[20] and calix[8]arene[23] have so far been A,D- trimethyl borate and hydrolysis, gave the boronic acid 10 in
bridged by a 1,10-phenanthroline unit.
good yield.
This presents the question of whether the concave 1,10-
However, boronic acid 10 slowly decomposes in solution;
phenanthrolines of type 1 might be altered in such a way therefore, crude 10 was coupled with 2,9-dichloro-1,10-
that they would not be stiff anymore but might adopt bent phenanthroline (11) under standard[25] Suzuki conditions
structures like those of the bridged calixarenes 2. In order [i.e., tetrakis(triphenylphosphane)palladium(0) as catalyst
to achieve a bowl-like geometry, either a joint (–CH2–O– in and barium hydroxide octahydrate as base in a mixture of
2) would have to be incorporated between the 1,10-phenan- dimethoxyethane (DME) and water]. The double-coupling
throline bridge and the bridgeheads, or the connection of product 12 was obtained in 62% yield, calculated from 2,9-
918
www.eurjoc.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2006, 916–923

Tailored Ligands for Diastereoselective Cyclopropanations
FULL PAPER
dichloro-1,10-phenanthroline (11) and starting from three
equivalents of the bromo ether 9.
The acyclic precursor 12 could be cyclized by ring-clos-
ing metathesis in 93% yield in the presence of the Grubbs I
catalyst. The 1,2-disubstituted double bond in 13 exists as
an almost 1:1 mixture of cis- and trans-configured alkene.
Catalytic hydrogenation with palladium on charcoal gave
the macrocyclic 1,10-phenanthroline 14, with a saturated
chain Y, in 49% yield.
In a Pd(0)-catalyzed double nucleophilic heteroaromatic
substitution, this phenol 16 reacted twice with 2,9-dichloro-
1,10-phenanthroline (11) in 96% yield.
The resulting tetraene 17 was then cyclized to give the
bimacrocyclic[26] 1,10-phenanthroline 18 in excellent yield
(95%). Like all other (bi)macrocyclic 1,10-phenanthrolines
obtained by ring-closing metathesis, 18 exists as a mixture
with cis and trans double bonds, but hydrogenation again
turns the stereoisomer into a single product and gives the
bimacrocycle 19, with saturated chains, in 98% yield.
Use of Concave 1,10-Phenanthrolines as Ligands in
Copper(I)-Catalyzed Cyclopropanations
We next investigated the influence of the geometries of
the two new concave 1,10-phenanthrolines 14 and 19 as li-
gands L on the stereochemistry of the copper()-catalyzed
The second approach towards a concave 1,10-phenan- cyclopropanation of alkenes with diazoacetates 5, in rela-
throline with a pocket-like structure goes back to the idea tion to other concave 1,10-phenanthrolines. The reaction
of a bimacrocyclic compound but allows the 1,10-phenan- conditions were chosen as in other cyclopropanations.[11,16]
throline bridge to flip as the bridge in the 1,10-phenan- Indene (3) and styrene (4) were used as alkenes, with
throline-bridged calix[6]arene 2 does. The idea was to intro- methyl, ethyl and tert-butyl diazoacetates (5a–c) as carbene
duce two joints between the bridgeheads and the bridge. If precursors. Besides 14 and 19, other concave 1,10-phenan-
2,9-dichloro-1,10-phenanthroline (11) were again chosen as throlines 1 and 2 were also used in this reaction. Table 1
starting material, phenols instead of boronic acids would and Table 2 compare the data for these new ligands.
have to be coupled with 11.
The concave 1,10-phenanthrolines 1, 2, 14 and 19 were
The synthesis of an appropriate phenol is straightforward first tested in the cyclopropanation of indene (3) and sty-
because boron derivatives can easily be cleaved oxidatively. rene (4) with ethyl diazoacetate (5b). Selected concave 1,10-
A precursor for the Suzuki coupling, the bis-pentenylated phenanthrolines (1a, 1f, 2, 14 and 19) were then also investi-
2-bromoresorcinol 15, was thus lithiated and treated with gated in cyclopropanations with the smaller methyl ester 5a
trimethyl borate. The resulting dimethylboronate was then or the larger tert-butyl ester 5c. For better comparison, all
oxidized with hydrogen peroxide in sodium hydroxide to af- reactions were carried out as batch reactions, still resulting
ford phenol 16 in 81% yield.
in acceptable to good yields of cyclopropanes (up to 80%).
Eur. J. Org. Chem. 2006, 916–923
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
919

U. Lüning, F. Fahrenkrug
FULL PAPER
Table 1. syn/anti selectivities[a] (endo-6/exo-6 or cis-7/trans-7) in the copper()-catalyzed cyclopropanation of indene (3) and styrene (4)
with diazoacetates 5 in the presence of different concave 1,10-phenanthrolines 1, 2,[11,19] 14 or 19 as ligands L.
Alkene
Diazoacetate
No ligand
1a
1b
1c
1d
1e
1f
2
14
19
3
3
3
4
4
4
5a
5b
5c
5a
5b
5c
33:67
43:57
29:71
43:57
38:62
33:67
5:95
4:96
3:97
24:76
23:77
12:88
15:58
29:71
2:98
62:38
33:67
12:88
86:14
76:24
70:30
74:26
67:33
54:46
51:49
56:44
62:38
60:40
54:46
52:48
61:39
58:42
63:37
48:52
42:58
67:33
1:99
1:99
2:98
2:98
16:84
16:84
18:82
15:85
[a] With acyclic alkenes, cis- and trans-cyclopropanes are formed, while cyclic alkenes give exo- and endo-configured bicyclic cyclopro-
panes. Therefore, syn summarizes cis and endo, anti summarizes trans and exo.
Table 2. Yields of the cyclopropanes 6 and 7 obtained in the copper()-catalyzed cyclopropanation of indene (3) and styrene (4) with
diazoacetates 5 in the presence of different concave 1,10-phenanthrolines 1, 2, 14 or 19 as ligands L, together with yields of the side
products dialkyl maleate and dialkyl fumarate when measurable (n.m. = not measurable). The yields were determined by GC with use
of an internal standard.
Alkene
Diazoacetate No ligand
14
19
1a
1b
1c
1d
1e
1f
2
Yield of 6 or 7 / Yield of maleate and fumarate
3
3
3
4
4
4
5a
5b
5c
5a
5b
5c
42 / n.m.
65 / 13
51 / 23
65 / 13
69 / 19
64 / 12
41 / n.m.
34 / 16
34 / 9
59 / 9
59 / 14
47 / 11
59 / n.m.
56 / 16
46 / 15
69 / 5
80 / 12
76 / 9
70 / n.m.
67 / 19
55 / 26
79 / 11
80 / 15
72 / 15
66 / n.m.
51 / 23
44 / 39
63 / 25
53 / 29
45 / 41
32 / –
39 / 23
35 / 14
54 / –
54 / 19
45 / 16
66 / 23 68 / 22 66 / 22 71 / 22
63 / 27 63 / 29 73 / 24 69 / 30
Since it is known that the competing formation of fumarate cyclic concave 1,10-phenanthrolines 1 produce the anti
and maleate can be reduced by slow addition of the diazo product trans-7 in a fourfold excess (cis-7/trans-7 Ϸ 20:80).
compound, these yields should be improvable.[27] In order
The second ligand with a structure fundamentally dif-
to allow comparison of the stereochemistry of the cyclopro- ferent from that of the bimacrocycles 1 is 19, in which two
panation of cyclic and acyclic alkenes, the prefixes syn and oxygen atoms have been placed between the bridgeheads
anti are used. For a definition, see footnote [a] in Table 1.
and the 1,10-phenanthroline moiety. This allows movement
As is already known from the cases of some concave of the plains of the 1,10-phenanthroline and the macrocy-
1,10-phenanthrolines, such as 1g, with direct aryl-bridge- cle, consisting of the bridgeheads and the octamethylene
head 1,10-phenanthroline connections,[11] stiff concave 1,10- chains, with respect to one another. With regard to the syn/
phenanthrolines 1 show good syn/anti selectivity, which can anti selectivity, the preference of the concave 1,10-phenan-
be enhanced by use of bulky diazoacetates (see Table 1; e.g., throlines 1 for the formation of the anti products exo-6 and
endo/exo ratio of Ͻ2 to Ͼ98 for addition to indene 3).
trans-7 is also lost with 19. In contrast, in many experiments
In contrast to this large preference for the formation of the thermodynamically less stable syn products are the main
the anti compounds, the monomacrocycle 14 is no longer compounds.
anti-selective. With indene (3) and any diazoacetate 5a–c,
In summary, both strategies to alter the stiff bimacro-
the syn product endo-6 is the major isomer formed. Indeed, cycles 1 have been successful: the less stable cyclopropan-
14 is syn-selective, although the best selectivity is only ation products endo-6 and cis-7 are formed in (slight) excess
62:38. However, the exact syn/anti ratios should not be with the new ligands, while the stiff ligands 1 showed a
overinterpreted. At 50:50, 10% more or less are caused by strong preference for the formation of the anti compounds.
only 1 kJmol–1 energy difference for the activation barriers. Compounds 14 and 19, however, cannot yet compete with
With styrene (4), 14 exhibits the same behaviour. Slightly the calixarene derivative 2, but the route to a syn-selective
more syn than anti product 7 is formed, while the bimacro- ligand (endo- or cis-selective) for the cyclopropanation of
920
www.eurjoc.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2006, 916–923

Tailored Ligands for Diastereoselective Cyclopropanations
FULL PAPER
70 eV): m/z (%) = 242, 240 (13, 12) [M]+, 174, 172 (38, 36) [M –
C5H8]+, 69 (100) [C5H9]+. C11H13BrO (241.12): calcd. C 54.79,
H 5.43; found C 54.80, H 5.72.
alkenes is cleared and the geometries of the new ligands 14
and 19, unlike that of the bridged calixarene 2, can be tailored
with modified building blocks to give further concave
1,10-phenanthrolines with yet other bridgeheads and chains.
3-(Pent-4-enyloxy)benzeneboronic Acid (10): 1-Bromo-3-(pent-4-en-
yloxy)benzene (9, 1.68 g, 6.97 mmol) was dissolved in tetra-
hydrofuran (25 mL). At –78 °C, n-butyllithium (2.5  in hexane,
3.1 mL, 7.75 mmol) was added and the mixture was stirred for 1 h
at –78 °C. After addition of trimethyl borate (2.60 mL, 23.3 mmol),
stirring was continued for 2 h while the mixture was allowed to warm
to room temperature. After hydrolysis with water (20 mL), the layers
were separated, and the aqueous layer was extracted with diethyl
ether (3×20–30 mL). The combined organic layer was washed with
saturated aqueous sodium chloride solution (20–30 mL) and dried
with magnesium sulfate, and the solvent was evaporated in vacuo.
Crude yield: 1.40 g. Because of decomposition of 10 in solution, the
crude product was used directly in the next synthetic step.
Experimental Section
General Remarks: The following chemicals were obtained commer-
cially and were used without further purification: N,N-dimethyl-
formamide (Fluka, Ն99.8%), dimethoxyethane (Aldrich), 3-bro-
mophenol (Aldrich), 5-bromopent-1-ene (Fluka), trimethyl borate
(Fluka), n-butyllithium 2.5  in hexanes (Aldrich), tetrakis(triphen-
ylphosphane)palladium(0) (Aldrich), barium hydroxide octa-
hydrate (Merck), benzylidenebis(tricyclohexylphosphane)dichlo-
roruthenium (Fluka). Pd/C (10%) (Merck). 2,9-Dichloro-1,10-
phenanthroline (11)[28,29] and 2-bromo-1,3-bis(pent-4-enyloxy)ben-
zene (15)[24] were prepared by literature procedures. Dry solvents
were obtained with suitable desiccants: tetrahydrofuran was dis-
tilled from lithium aluminium hydride and ethyl acetate was dis-
tilled from calcium chloride. Column chromatography was carried
out on basic alumina (Fluka, activity I) or silica gel (Macherey–
Nagel, activity I). The preparative, centrifugally accelerated, thin-
layer chromatograph (Chromatotron) was a model 7924T from
Harrison Research. 1H NMR and 13C NMR spectra were recorded
on Bruker AC 200 (200 MHz), AM 300 (300 MHz or 75 MHz) or
DRX 500 (500 MHz or 125 MHz) instruments, with tetramethylsil-
ane as internal standard. IR spectra were measured on a Perkin–
Elmer 1600 Series device. MS spectra were recorded on a Finnigan
MAT 8230 machine. Elemental analyses were carried out on a Vari-
oEl instrument (Elementaranalysensysteme GmbH). Gas
chromatography was performed with a Varian 3400 Gas Chroma-
tograph fitted with an Optima 1/25 m column from Macherey–Na-
gel. Comment: all (concave) 1,10-phenanthrolines tend to contain
or absorb water.[25] The water content was calculated from the ele-
mental analyses. As further structural corroboration, high-resolu-
tion mass spectra were recorded for all 1,10-phenanthrolines.
2,9-Bis[3-(pent-4-enyloxy)phenyl]-1,10-phenanthroline (12): Crude 3-
(pent-4-enyloxy)benzeneboronic acid (10, 1.40 g) and 2,9-dichloro-
1,10-phenanthroline (11, 580 mg, 2.33 mmol) were dissolved in di-
methoxyethane (100 mL) and water (25 mL). After addition of bar-
ium hydroxide octahydrate (3.20 g, 10.1 mmol) and tetrakis(tri-
phenylphosphane)palladium(0) (270 mg, 233 µmol), the mixture
was heated to reflux for 16 h. Water (30 mL) and dichloromethane
(30 mL) were then added, and the layers were separated. The aque-
ous layer was extracted with dichloromethane (3×30 mL), and the
combined organic layer was extracted with saturated aqueous so-
dium chloride solution (20 mL) and dried with magnesium sulfate.
The solvent was evaporated in vacuo, and the residue was purified
by chromatography (silica gel, dichloromethane) and recrystallized
from dichloromethane/n-hexane, giving 615 mg (62%) of 12, m.p.
92–93 °C. 1H NMR (300 MHz, CDCl3): δ = 1.79 (mc, 4 H,
OCH2CH2), 2.30 (mc, 4 H, CH2CH=), 4.20 (t, J = 6.4 Hz, 4 H,
OCH2), 4.99–5.13 (m, 4 H, =CH2), 5.90 (tdd, Jt = 6.7, Jd = 10.2,
Jd = 17.1 Hz, 2 H, CH=), 7.04 (ddd, J Ϸ 1, J = 2.6, J Ϸ 8 Hz, 4 H,
4Ј,4ЈЈ-H), 7.45 (t, J Ϸ 8 Hz, 2 H, 5Ј,5ЈЈ-H), 7.76 (s, 2 H, 5,6-H),
8.03 (ddd, J Ϸ 1, J Ϸ 2, J Ϸ 8 Hz, 2 H, 6Ј,6ЈЈ-H), 8.09 (dd, J Ϸ 2,
J = 2.6 Hz, 2 H, 2Ј,2ЈЈ-H), 8.12 (d, J = 8.4 Hz, 2 H, 3,8-H), 8.28
(d, J = 8.4 Hz, 2 H, 4,7-H) ppm. 13C NMR (75 MHz, CDCl3): δ =
28.6 (t, OCH2CH2), 30.2 (t, CH2CH=), 67.2 (t, OCH2), 113.4 (d,
2Ј,2ЈЈ-C), 115.2 (t, =CH2), 116.0 (d, 4Ј,4ЈЈ-C), 119.9, 120.0, (2×d,
3,6Ј,6ЈЈ,8-C), 126.0 (d, 5,6-C), 127.9 (s, 4a,6a-C), 129.6 (d, 5Ј,5ЈЈ-
C), 136.8 (d, 4,7-C), 137.9 (d, CH=), 140.8 (s, 1Ј,1ЈЈ-C), 145.9 (s,
1-Bromo-3-(pent-4-enyloxy)benzene (9): 3-Bromophenol (8, 3.67 g,
21.2 mmol) was dissolved in dry N,N-dimethylformamide (40 mL),
and potassium carbonate (9.00 g, 65.2 mmol), potassium iodide
(500 mg, 3.01 mmol) and 5-bromopent-1-ene (2.50 mL, 16.8 mmol)
were added. After the mixture had been stirred for 16 h at 60 °C,
the solvent was evaporated in vacuo and the residue was dissolved
in a mixture of sodium hydroxide (2 , 30 mL) and diethyl ether
(30 mL). The layers were separated and the aqueous layer was ex-
tracted with diethyl ether (3×30 mL). The combined organic layer
was washed with aqueous sodium hydroxide (2 , 3×30 mL) and
saturated aqueous sodium chloride solution (30 mL) and was dried
with magnesium sulfate. The solvent was evaporated in vacuo and
the residue was purified by filtration through silica gel with cyclo-
hexane/ethyl acetate (10:1). Evaporation of the solvents gave 4.87 g
(95%) of 9. GC (Optima 1/25 m, temperature program: 5 min at
10a,10b-C), 156.4 (s, 2,9-C), 159.5 (s, 3Ј,3ЈЈ-C) ppm. IR (KBr): ν =
˜
3073, 3026 (arom. C–H), 2932, 2854 (aliph. C–H), 1640 (aliph.
C=C), 1596, 1475 (arom. C=C), 1100 (C–O–C) cm–1. MS (EI,
70 eV): m/z (%) = 500 (100) [M]+, 459 (25) [M – C3H5]+, 445 (38)
[M – C4H7]+, 431 (12) [M – C5H9]+. HR-MS: C34H32N2O2, found
500.24640, calcd. 500.24637 (–0.1 ppm), C3313CH32N2O2, found
501.24960, calcd. 501.24973 (0.3 ppm). C34H32N2O2 (500.63):
calcd. C 81.57, H 6.44, N 5.60, C34H32N2O2 ×0.2 H2O: calcd.
C 80.99, H 6.48, N 5.56; found C 81.03, H 6.79, N 5.48.
100 °C, 10 °Cmin–1, 20 min at 250 °C): tRet = 12.1 min, purity: 4,13-Dioxa-1,3(1,3)-dibenzena-2(2,9)-1,10-phenanthrolinacyclotride-
99%. 1H NMR (300 MHz, CDCl3): δ = 1.87 (mc, 2 H, OCH2CH2), caphan-8-ene (13): 2,9-Bis[3-(pent-4-enyloxy)phenyl]-1,10-phenan-
2.23 (mc, 2 H, CH2CH=), 3.94 (t, J = 6.4 Hz, 2 H, OCH2), 4.99– throline (12, 600 mg, 1.20 mmol) was dissolved in dichloromethane
5.10 (m, 2 H, =CH2), 5.84 (tdd, Jt = 6.7, Jd = 10.2, Jd = 17.1 Hz,
1 H, CH=), 6.80–6.84 (m, 1 H, 4-H), 7.04–7.14 (m, 3 H, 2,5,6- thenium (49.3 mg, 59.9 mmol, 5 mol-%) was added. After stirring
H) ppm. 13C NMR (75 MHz, CDCl3): δ = 28.3 (t, OCH2CH2), 30.0
for 16 h at room temperature, the mixture was filtered three times
(t, CH2CH=), 67.3 (t, OCH2), 113.5 (d, 4-C), 115.4 (t, =CH2), through basic aluminium oxide with dichloromethane. Evaporation
117.7 (d, 2-C), 122.8 (s, 1-C), 123.6 (d, 6-C), 130.5 (d, 5-C), 137.6
of the solvent in vacuo gave 527 mg (93%) of 13. 1H NMR
(d, =CH), 159.8 (s, 3-C) ppm. IR (film): ν = 3076 (arom. C–H), (300 MHz, CDCl3): δ = 1.86, 2.34 (2×mc, 8 H, CH2), 4.12, 4.18
(120 mL) and benzylidenebis(tricyclohexylphosphane)dichlororu-
˜
2975, 2941, 2874 (aliph. C–H), 1640 (aliph. C=C), 1590, 1573, 1468
(2×t, J Ϸ 6, J Ϸ 6 Hz, 4 H, OCH2), 5.47 (mc, 1.15 H, trans-
(arom. C=C), 1064 (arom. C–Br), 1019 (C–O–C) cm–1. MS (EI, CH=CH)*, 5.55 (2×mc, 0.85 H, cis-CH=CH)*, 7.04 (mc, 2 H,
Eur. J. Org. Chem. 2006, 916–923
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
921

U. Lüning, F. Fahrenkrug
4Ј,4ЈЈ-H), 7.40 (t, J Ϸ 8 Hz, 2 H, 5Ј,5ЈЈ-H), 7.53 (mc, 2 H, 6Ј,6ЈЈ- dium chloride solution (30 mL). After drying with magnesium sul-
FULL PAPER
H), 7.76, 7.77 (2×s, 2 H, 5,6-H), 8.09, 8.11 (2×d, J Ϸ 8, J Ϸ 8 Hz,
2 H, 3,8-H), 8.28, 8.28 (2×d, J Ϸ 8, J Ϸ 8 Hz, 2 H, 4,7-H), 8.25,
fate, the solvent was evaporated in vacuo and the residue was puri-
fied by chromatography (cyclohexane/ethyl acetate, 10:1) yielding
8.44 (2×mc, 2 H, 2Ј,2ЈЈ-H) ppm. * The stereoisomer ratio was 200 mg (81%) of 16, which is partly molten at room temperature.
1:1.3. Because of the small difference in chemical shift, the assign-
GC (Optima 1/25 m, program: 5 min at 100 °C, 10 °Cmin–1, 20 min
ment may be reversed. 13C NMR (75 MHz, CDCl3): δ = 23.8, 27.9, at 250 °C): tRet = 17.2 min, purity: 93%. 1H NMR (500 MHz,
28.0, 29.3 (4×t, CH2), 65.0, 66.9 (2×t, OCH2), 111–121 (8×d,
2Ј,2ЈЈ,3,4Ј,4ЈЈ,6Ј,6ЈЈ,8-C), 126.1, 126.1 (2×d, 5,6-C), 128.0 (s, 4a,6a-
C), 129.3, 129.3, 130.0, 130.5 (4×d, 5Ј,5ЈЈ-C, CH=CH), 136.9 (d,
CDCl3): δ = 1.93 (mc, 4 H, OCH2CH2), 2.25 (mc, 4 H, CH2CH=),
4.05 (t, J = 6.6 Hz, 4 H, OCH2), 4.99–5.09 (m, 4 H, =CH2), 5.56
(s, 1 H, OH), 5.86 (tdd, Jt = 6.7, Jd = 10.3, Jd = 17.1 Hz, 4 H,
4,7-C), 141.2, 141.4 (2×s, 1Ј,1ЈЈ-C), 146.1 (s, 10a,10b-C), 157.0, CH=), 6.56 (d, J = 8.3 Hz, 2 H, 3,5-H), 6.75 (dd, J = 7.9, J =
157.4 (2×s, 2,9-C), 159.8, 160.1 (2×s, 3Ј,3ЈЈ-C) ppm. IR (KBr): ν
8.7 Hz, 1 H, 4-H) ppm. 13C NMR (125 MHz, CDCl3): δ = 28.3 (t,
OCH2CH2), 30.1 (t, CH2CH=), 68.6 (t, OCH2), 106.2 (d, 3,5-C),
˜
= 2926, 2862 (aliph. C–H), 1598, 1488 (arom. C=C), 1052 (C–O–
C) cm–1. MS (EI, 70 eV): m/z (%) = 472 (100) [M]+, 364 (59) 115.1 (t, =CH2), 118.9 (d, 4-C), 135.5 (s, 1-C), 137.8 (d, CH=),
[C24H16O2N2]+. HR-MS: C32H28N2O2, found 472.21510, calcd. 146.6 (s, 2,6-C) ppm. IR (KBr): ν = 3535 (O–H), 3075 (arom. C–
˜
472.21509 (–0.0 ppm), C3113CH28N2O2, found 473.21820, calcd.
473.21844 (0.5 ppm). C32H28N2O2 (472.58): calcd. C 81.33, H 5.97,
N 5.93, C32H28N2O2 ×0·3H2O: calcd. C 80.41, H 6.03, N 5.86;
found C 80.63, H 6.27, N 5.58.
H), 2974, 2941, 2874 (aliph. C–H), 1640, 1619 (aliph. C=C), 1506,
1469 (arom. C=C), 1094 (C–OH) cm–1. MS (EI, 70 eV): m/z (%) =
262 (32) [M]+, 126 (80) [C6H6O3]+, 69 (100) [C5H9]+. C16H22O3
(262.35): calcd. C 73.25, H 8.45, C16H22O3 ×0.1H2O: calcd.
C 72.75, H 8.47; found C 72.59, H 8.99.
4,13-Dioxa-1,3(1,3)-dibenzena-2(2,9)-1,10-phenanthrolinacyclotride-
caphane (14): Palladium on charcoal (10%, 20.0 mg) was mixed 2,9-Bis[2,6-bis(pent-4-enyloxy)phenyloxy]-1,10-phenanthroline (17):
with ethyl acetate (5.00 mL, filtered through basic aluminium oxide
prior to use, in order to remove acid traces) and hydrogen was
bubbled through the mixture for 30 min. This activated mixture
was then mixed with a solution of 4,13-dioxa-1,3(1,3)-dibenzena-
2(2,9)-1,10-phenanthrolinacyclotridecaphan-8-ene (13, 200 mg,
423 mmol) in ethyl acetate (12 mL, quality as above) and hydrogen
was bubbled through the solution for 2 h while stirring at room
temperature was continued. Stirring of the solution in an atmo-
sphere of hydrogen was continued for 16 h at room temperature.
The solvent was evaporated in vacuo and the residue was dissolved
in dichloromethane, filtered through basic aluminium oxide, puri-
fied by chromatography with a chromatotron (neutral aluminium
oxide, dichloromethane/n-pentane, 1:1) and recrystallized from
dichloromethane/n-hexane, yielding 97 mg (48%) of 14, m.p. 156–
2,9-Dichloro-1,10-phenanthroline (11, 43.0 mg, 173 mmol) and 2,6-
bis(pent-4-enyloxy)phenol (16, 135 mg, 515 mmol) were dissolved
in dry N,N-dimethylformamide (10 mL). After addition of caesium
carbonate (300 mg, 921 µmol), the mixture was stirred for 18 h at
100 °C. After evaporation of the solvent in vacuo, the residue was
dissolved in aqueous sodium hydroxide solution (2 , 30 mL) and
dichloromethane (30 mL). After separation of the layers, the aque-
ous layer was extracted with dichloromethane (3×40 mL). The
combined organic layer was washed with aqueous sodium hydrox-
ide solution (2 , 3×40 mL) and concentrated aqueous sodium
chloride solution (30 mL). After drying with magnesium sulfate,
the solvent was evaporated in vacuo, and the residue was dissolved
in dichloromethane and filtered through basic aluminium oxide.
After evaporation of the solvent, the residue was recrystallized
158 °C. 1H NMR (300 MHz, CDCl3): δ = 0.94, 1.23, 1.50, 1.74, from dichloromethane/n-hexane, giving 116 mg (96%) of 17, m.p.
1.84 (5 mc, 12 H, CH2), 4.12 (t, J = 5.6 Hz, 4 H, OCH2), 7.01 (mc, 34 °C. 1H NMR (300 MHz, CDCl3):
1.48 (mc, 8 H,
2 H, 4Ј,4ЈЈ-H), 7.40 (t, J Ϸ 8 Hz, 2 H, 5Ј,5ЈЈ-H), 7.45–7.50 (m, 2 H, OCH2CH2), 1.76 (mc, 8 H, CH2CH=), 3.77 (t, J = 6.4 Hz, 8 H,
δ
=
6Ј,6ЈЈ-H), 7.70 (s, 2 H, 5,6-H), 8.08 (d, J = 8.4 Hz, 2 H, 3,8-H),
8.25–8.30 (m with d at 8.28, J = 8.4 Hz, 4 H, 2Ј,2ЈЈ,4,7-H) ppm.
13C NMR (75 MHz, CDCl3): δ = 23.8, 26.3, 28.1 (3×t, CH2), 67.5
(t, OCH2), 113.9, 115.5, 120.0, 121.3 (4×d, 2Ј,2ЈЈ,3,4Ј,4ЈЈ,6Ј,6ЈЈ,8-
C), 126.2 (d, 5,6-C), 128.1 (s, 4a,6a-C), 129.3 (d, 5Ј,5ЈЈ-C), 136.9
OCH2), 4.6–4.8 (m, 8 H, =CH2), 5.52 (tdd, Jt = 6.6, Jd = 10.6, Jd
= 16.7 Hz, 4 H, CH=), 6.56 (d, J = 8.4 Hz, 4 H, 3Ј,3ЈЈ,5Ј,5ЈЈ-H),
7.10 (t, J = 8.4 Hz, 2 H, 4Ј,4ЈЈ-H), 7.26 (d, J = 8.6 Hz, 2 H, 3,8-
H), 7.61 (s, 2 H, 5,6-H), 8.12 (d, J = 8.6 Hz, 2 H, 4,7-H) ppm.
13C NMR (75 MHz, CDCl3): δ = 28.5 (t, OCH2CH2), 29.7 (t,
(d, 4,7-C), 141.5 (s, 1Ј,1ЈЈ-C), 146.3 (s, 10a,10b-C), 157.7 (s, 2,9-C), CH2CH=), 68.9 (t, OCH2), 108.3 (d, 3Ј,3ЈЈ,5Ј,5ЈЈ-C), 112.8 (d, 3,8-
160.0 (s, 3Ј,3ЈЈ-C) ppm. IR (KBr): ν = 3032 (arom. C–H), 2924, C), 114.4 (t, =CH2), 123.6 (d, 5,6-C), 124.9 (d, 4Ј,4ЈЈ-C), 125.9 (s,
˜
2860 (aliph. C–H), 1602, 1486 (arom. C=C), 1098 (C–O–C) cm–1.
MS (EI, 70 eV): m/z (%) = 474 (100) [M]+, 364 (88) [C24H16- (s, 10a,10b), 152.6 (s, 2Ј,2ЈЈ,6Ј,6ЈЈ-C), 161.5 (s, 2,9-C) ppm. IR
1Ј,1ЈЈ-C), 133.7 (s, 4a,6a-C), 138.0 (d, CH=), 139.0 (d, 4,7-C), 143.7
N2O2]+. HR-MS: C32H30N2O2, found 474.23060, calcd. 474.23074
(0.3 ppm), C3113CH30N2O2, found 475.23370, calcd. 475.23407
(0.8 ppm). C32H30N2O2 (474.59): calcd. C 80.98, H 6.37, N 5.90,
C32H30N2O2 ×0.2H2O: calcd. C 80.37, H 6.41, N 5.86; found
C 80.51, H 6.57, N 5.75.
(KBr): ν = 3074 (arom. C–H), 2935, 2861 (aliph. C–H), 1640 (aliph.
˜
C=C), 1603, 1458 (arom. C=C), 1100 (C–O–C) cm–1. MS (EI,
70 eV): m/z (%) = 700 (100) [M]+, 659 (94) [M – C3H5]+, 645 (26)
[M – C4H7]+. HR-MS: C44H48N2O6, found 700.35130, calcd.
700.35126 (–0.1 ppm), C4313CH48N2O6, found 701.35400, calcd.
701.35461 (0.9 ppm). C44H48N2O6 (700.86): calcd. C 75.40, H 6.90,
N 4.00; found C 75.23, H 7.01, N 3.94.
2,6-Bis(pent-4-enyloxy)phenol (16): 2-Bromo-1,3-bis(pent-4-en-
yloxy)benzene (15, 4.93 g, 15.2 mmol) was dissolved in tetra-
hydrofuran (50 mL). At –78 °C, n-butyllithium (2.5  in hexane,
6.80 mL, 17.0 mmol) was added and the mixture was stirred for 1 h
at –78 °C. Trimethyl borate (4.90 mL, 43.9 mmol) was added and
stirring was continued for 2 h, while the mixture was warmed to
room temperature. After the mixture had been cooled to –78 °C,
aqueous sodium hydroxide (3 , 30 mL) and aqueous hydrogen
peroxide (30%, 30 mL) were added simultaneously, and the mixture
was stirred for 18 h at room temperature, after which it was ex-
tracted with diethyl ether (4×30 mL). The combined organic layer
was washed with water (3×30 mL) and concentrated aqueous so-
2,11,13,22,23,25-Hexaoxa-1,12(1,3,2)-dibenzena-24(2,9)-1,10-phen-
anthrolinabicyclo[10.10.3]pentacosaphan-6,17-diene
(18):
2,9-
Bis[2,6-bis(pent-4-enyloxy)phenyloxy]-1,10-phenanthroline
(17,
410 mg, 585 mmol) was dissolved in dichloromethane (58.5 mL, c
= 0.01 mmolL–1). After addition of benzylidenebis(tricyclohexyl-
phosphane)dichlororuthenium (24.1 mg, 29.3 µmol, 5 mol-%), the
mixture was stirred for 16 h at room temperature and filtered three
times through basic aluminium oxide with dichloromethane. Evap-
oration of the solvent in vacuo gave 365 mg (95%) of 18. 1H NMR
(300 MHz, CDCl3): δ = 1.0–1.9 (m with 3×mc at 1.24, 1.45, 1.77,
922
www.eurjoc.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2006, 916–923

Tailored Ligands for Diastereoselective Cyclopropanations
FULL PAPER
16 H, CH2), 3.54, 3.85, 3.92 (3×mc, 8 H, OCH2), 4.60, 4.70, 4.94 After addition of ca. 25 mg of n-hexadecane (+/– 0.01 mg) as GC
(3×mc, 4 H, CH=CH), 6.45–6.55 (m, with d at 6.49, J = 8.3 Hz, standard, the products were analyzed by GC: Optima 1/25 m, 80 °C
4 H, 3Ј,3ЈЈ,5Ј,5ЈЈ-H), 7.07, 7.08, 7.12 (3×t, J = 8.3, J = 8.3, J =
for 5 min, 10 °Cmin–1 until 140 °C, 1 min, 2 °Cmin–1 until 160 °C,
8.3 Hz, 2 H, 4Ј,4ЈЈ-H), 7.37, 7.38, 7.39 (3×d, J = 8.6, J = 8.6, J = 1 min, 20 °Cmin–1 until 240 °C, 2 min.
8.6 Hz, 2 H, 3,8-H), 7.58, 7.62, 7.64 (3×s, 2 H, 5,6-H), 8.12, 8.15,
8.17 (3×d, J = 8.6, J = 8.6, J = 8.6 Hz, 2 H, 4,7-H) ppm. 13C NMR
[1] Encyclopedia of Supramolecular Chemistry (Eds.: J. L. Atwood,
(75 MHz, CDCl3): δ = 22.5, 26.4, 26.8, 28.6, 28.8, 29.1 (6×t, CH2),
65.4, 66.1, 68.6 (3×t, OCH2), 106.2, 106.8, 108.1 (3×d,
3Ј,3ЈЈ,5Ј,5ЈЈ-C), 113.0, 113.2 (2×d, 3,8-C), 123.7, 125.3, 125.3
(3×d, 4Ј,4ЈЈ,5,6-C), 125.9, 125.9 (2×s, 1Ј,1ЈЈ-C), 129.7, 130.1, 130.3
(3×d, CH=CH), 131.7 (s, 4a,6a-C), 139.1, 139.2 (2×d, 4,7-C),
143.8 (s, 10a,10b-C), 152.3, 152.3, 152.7 (3×s, 2,9-C), 161.4 (s,
J. Steed), Marcel Dekker, New York, 2004.
[2] J. W. Steed, J. L. Atwood, Supramolecular Chemistry, John
Wiley & Sons, Chichester, New York, Weinheim, Brisbane, Sin-
gapore, Toronto, 2000.
[3] J.-M. Lehn, Supramolecular Chemistry, VCH, Weinheim, New
York, Basel, Cambridge, Tokyo, 1995.
[4] Comprehensive Supramolecular Chemistry (Ed.: J. L. Atwood),
Pergamon Press, Oxford, 1996.
2Ј,2ЈЈ,6Ј,6ЈЈ-C) ppm. IR (KBr): ν = 2933, 2869 (aliph. C–H), 1602
˜
(aliph. C=C), 1458 (arom. C=C), 1103 (C–O–C) cm–1. MS (EI,
70 eV): m/z (%) = 644 (100) [M]+. HR-MS: C40H40N2O6, found
644.28850, calcd. 644.28864 (0.2 ppm), C3913CH40N2O6, found
645.29150, calcd. 645.29199 (0.8 ppm). C40H40N2O6 (644.76):
calcd. C 74.51, H 6.25, N 4.34, C40H40N2O6 ×0.5H2O: calcd.
C 73.49, H 6.32, N 4.28; found C 73.50, H 6.40, N 4.13.
[5] F. Vögtle, Supramolekulare Chemie, Teubner, Stuttgart, 1989.
[6] See biochemistry textbooks, e.g.: D. Voet, J. G. Voet, C. W.
Pratt, Lehrbuch der Biochemie, Wiley-VCH, Weinheim, 2002.
[7] U. Lüning, M. Müller, M. Gelbert, K. Peters, H. G.
von Schnering, M. Keller, Chem. Ber. 1994, 127, 2297–2306.
[8] M. Hagen, U. Lüning, Chem. Ber./Recueil 1997, 130, 231–234.
[9] M. Gelbert, U. Lüning, Supramol. Chem. 2001, 12, 435–444.
[10] B. Meynhardt, U. Lüning, C. Wolff, C. Näther, Eur. J. Org.
Chem. 1999, 2327–2335.
[11] F. Löffler, M. Hagen, U. Lüning, Synlett 1999, 1826–1828.
[12] U. Lüning, M. Müller, Chem. Ber. 1990, 123, 643–645.
[13] U. Lüning, F. Fahrenkrug, M. Hagen, Eur. J. Org. Chem. 2001,
2161–2163.
[14] A remarkable example of both enantioselective and diastereo-
selective cyclopropanations of styrene: T. Niimi, T. Uchida, R.
Irie, T. Katsuki, Adv. Synth. Catal. 2001, 343, 79–88, and refs.
[15] Another example of a selective cis-cyclopropanation of styrene
but with moderate enantioselectivity: A. Puglisi, M. Benaglia,
R. Annunziata, A. Bologna, Tetrahedron Lett. 2003, 44, 2947–
2951.
[16] M. Bühl, F. Terstegen, F. Löffler, B. Meynhardt, S. Kierse, M.
Müller, C. Näther, U. Lüning, Eur. J. Org. Chem. 2001, 2151–
2160.
[17] While the diameter of the 1,10-phenanthroline-bridged calix[6]-
arene 2 is fixed, the OH groups may be substituted: see
refs.[18,19]
[18] S. Konrad, Ph.D. thesis, Christian-Albrechts-Universiät zu
Kiel, 2005.
[19] F. Löffler, Ph.D. thesis, Christian-Albrechts-Universiät zu Kiel,
2001.
[20] H. Ross, U. Lüning, Tetrahedron Lett. 1997, 38, 4539–4542.
[21] C. D. Gutsche, Calixarenes, The Royal Society of Chemistry,
Cambridge, 1989 and 1992.
2,11,13,22,23,25-Hexaoxa-1,12(1,3,2)-dibenzena-24(2,9)-1,10-phen-
anthrolinabicyclo[10.10.3]pentacosaphane (19): Palladium on char-
coal (10%, 40.0 mg) was mixed with ethyl acetate (5.00 mL, filtered
through basic aluminium oxide prior to use, in order to remove
acid traces) and hydrogen was bubbled through the mixture for
30 min, after which it was combined with
a solution of
2,11,13,22,23,25-hexaoxa-1,12(1,3,2)-dibenzena-24(2,9)-1,10-phen-
anthrolinabicyclo[10.10.3]pentacosaphane-6,17-diene (18, 315 mg,
449 mmol) in ethyl acetate (20 mL, quality as above). Hydrogen
was bubbled through the mixture for 2 h while stirring at room
temperature was continued. Stirring of the solution in an atmo-
sphere of hydrogen was continued for 16 h at room temperature.
Total conversion was monitored by TLC. The solvent was evapo-
rated in vacuo and the residue was dissolved in dichloromethane
and filtered through basic aluminium oxide. The solvent was evapo-
rated and the residue was recrystallized from dichloromethane/n-
hexane, yielding 285 mg (98%) of 19, m.p. 168 °C. 1H NMR
(500 MHz, CDCl3): δ = 0.70, 0.76, 0.87, 1.41 (4×mc, 24 H, CH2),
3.87 (mc, 8 H, OCH2), 6.60 (d, J = 8.4 Hz, 4 H, 3Ј,3ЈЈ,5Ј,5ЈЈ-H),
7.11 (t, J = 8.4 Hz, 2 H, 4Ј,4ЈЈ-H), 7.22 (d, J = 8.6 Hz, 2 H, 3,8-
H), 7.61 (s, 2 H, 5,6-H), 8.10 (d, J = 8.6 Hz, 2 H, 4,7-H) ppm.
13C NMR (125 MHz, CDCl3): δ = 24.0, 27.4, 28.4 (3×t, CH2), 69.0
(t, OCH2), 108.1 (d, 3Ј,3ЈЈ,5Ј,5ЈЈ-C), 112.3 (d, 3,8-C)*, 123.7 (d, 5,6-
C)*, 125.3 (d, 4Ј,4ЈЈ-C)*, 125.8 (s, 4a,6a-C), 133.3 (s, 10a,10b-C)°,
139.0 (d, 4,7-C), 143.8 (s, 1Ј,1ЈЈ-C)°, 152.6 (s, 2Ј,2ЈЈ,6Ј,6ЈЈ-C), 161.8
[22] C. D. Gutsche, Calixarenes Revisited, The Royal Society of
Chemistry, Cambridge, 1998.
(s, 2,9-C) ppm. *, ° assignment may be reversed. IR (KBr): ν =
˜
[23] S. Konrad, C. Näther, U. Lüning, Eur. J. Org. Chem. 2005,
2929, 2855 (aliph. C–H), 1602 (aliph. C=C), 1466 (arom. C=C),
1100 (C–O–C) cm–1. MS (EI, 70 eV): m/z (%) = 648 (100) [M]+, 605
(7) [M – C3H8]+. HR-MS: C40H44N2O6, found 648.31990, calcd.
648.31995 (0.1 ppm), C3913CH44N2O6, found 649.32330, calcd.
649.32330 (0.0 ppm). C40H44N2O6 (648.79): calcd. C 74.05, H 6.84,
N 4.32, C40H44N2O6 ×0.4 H2O: calcd. C 73.24, H 6.88, N 4.27;
found C 73.10, H 7.03, N 4.10.
2330–2337.
[24] U. Lüning, M. Abbass, F. Fahrenkrug, Eur. J. Org. Chem. 2002,
3294–3303.
[25] U. Lüning, F. Fahrenkrug, Eur. J. Org. Chem. 2004, 3119–3127.
[26] As for related bimacrocyclic 1,10-phenanthrolines,[12,25] it is
conceivable that the double macrocyclization can also connect
two alkenyloxy groups within one bridgehead to form a meta-
cyclophane instead of a bimacrocycle, but for all related com-
pounds with the same chain length (eight atoms between the
phenolic oxygen atoms),[12,25] spectroscopic studies and also X-
ray analysis have exclusively identified the bimacrocyclic com-
pounds.
[27] M. Hagen, Ph.D. thesis, Christian-Albrechts-Universität zu
Kiel, 1999.
[28] S. Toyota, C. R. Woods, M. Benagalia, J. S. Siegel, Tetrahedron
Lett. 1998, 39, 2697–2700.
[29] J. Lewis, T. D. O’Donoghue, J. Chem. Soc. Dalton Trans. 1980,
Cyclopropanation: Under argon, ca. 2.5 to 3.5 mg (+/– 0.01 mg) of
copper() triflate hemi-benzene complex (Aldrich) were placed in a
vial, and 440 equivalents [based on copper()] of alkene 3 or 4, and
1.2 equivalents of the ligands 1a–f, 2, 14 or 19, dissolved in 1,2-
dichloroethane (c = 0.01 molL–1), were added. After addition of
50 equivalents of the diazoacetate 5, the mixture was stirred at
room temp. for 24 h. At the beginning of the reaction, quick devel-
opment of gas was frequently noticed. After filtration of the mix-
ture through silica gel with diethyl ether as eluent, most of the
solvent mixture was evaporated in vacuo until ca. 5 mL remained.
1,2-Dichloromethane was then added to give ca. 10 mL of solution.
736–742.
Received: August 26, 2005
Published Online: December 13, 2005
Eur. J. Org. Chem. 2006, 916–923
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
923