Through-shell alkyllithium additions and borane reductions

Article abstract of DOI:10.1021/jo026649z

The through-shell borane reduction and methyllithium addition to benzaldehyde (1), benzocyclobutenone (2), and benzocyclobutenedione (3) incarcerated inside a hemicarcerand (4) with four tetramethylenedioxy bridges are reported. All guests could be reduced and methylated. Selective monoreduction and monomethylation were observed for 3. In the methyllithium addition to 4>=3, the initially formed lithium alcoholate underwent a Moore rearrangement. The reactivity of the incarcerated guests toward methyllithium increased in the order 1 < 2 ? 3 and toward borane in the order 1 ? 2 ≈ 3. Guest reactivity was correlated with the inner-phase location of the reacting carbonyl group in the preferred guest inner-phase orientation. The latter was determined from the X-ray structures of 4>=1, 4>=2, and 4>=3, from molecular mechanical calculations, and from the hemicarcerand-induced upfield shift of the guest proton resonances. In the methyllithium and n-butyllithium addition to 4>=1 and 4>=3 at elevated temperatures, selective cleavage of a host's spanner or tetramethylenedioxy bridge, respectively, was observed. The cleavage of one spanner also took place in the methyllithium addition to the 1-methyl-2-pyrrolidinone hemicarceplex. These scission reactions are initiated by the initially formed lithium alcoholates, which show enhanced basicity and nucleophilicity in the inner phase as compared to the bulk phase. Mechanisms for the host scission reactions are discussed.

Full text of DOI:10.1021/jo026649z

Th r ou gh -Sh ell Alk yllith iu m Ad d ition s a n d Bor a n e Red u ction s
Ralf Warmuth,*,†,‡ Emily F. Maverick‡, Carolyn B. Knobler‡, and Donald J . Cram‡,§
Department of Chemistry, Kansas State University, Manhattan, Kansas 66506-3701, and Department of
Chemistry and Biochemistry, University of California at Los Angeles, Los Angeles, California 90095-1569
warmuth@ksu.edu
Received November 3, 2002
The through-shell borane reduction and methyllithium addition to benzaldehyde (1), benzocy-
clobutenone (2), and benzocyclobutenedione (3) incarcerated inside a hemicarcerand (4) with four
tetramethylenedioxy bridges are reported. All guests could be reduced and methylated. Selective
monoreduction and monomethylation were observed for 3. In the methyllithium addition to 4.3,
the initially formed lithium alcoholate underwent a Moore rearrangement. The reactivity of the
incarcerated guests toward methyllithium increased in the order 1 < 2 , 3 and toward borane in
the order 1 , 2 ≈ 3. Guest reactivity was correlated with the inner-phase location of the reacting
carbonyl group in the preferred guest inner-phase orientation. The latter was determined from
the X-ray structures of 4.1, 4.2, and 4.3, from molecular mechanical calculations, and from the
hemicarcerand-induced upfield shift of the guest proton resonances. In the methyllithium and
n-butyllithium addition to 4.1 and 4.3 at elevated temperatures, selective cleavage of a host’s
spanner or tetramethylenedioxy bridge, respectively, was observed. The cleavage of one spanner
also took place in the methyllithium addition to the 1-methyl-2-pyrrolidinone hemicarceplex. These
scission reactions are initiated by the initially formed lithium alcoholates, which show enhanced
basicity and nucleophilicity in the inner phase as compared to the bulk phase. Mechanisms for the
host scission reactions are discussed.
In tr od u ction
whereas others probably take place in one of the equato-
rial entryways. The latter are better described as through-
shell reactions and are particularly interesting since the
size, shape, and polarity of the host’s portals is expected
to have a strong influence on the stability of such
through-shell transition states. The concepts of inner-
phase chemistry and through-shell reactions give rise to
many interesting questions. For example, how would an
inner-phase reaction differ from its counterpart carried
Carcerands and hemicarcerands are spherical hosts
with a well-defined, enforced inner cavitysthe inner
phaseswhich provides enough space for the incarceration
of one or more guest molecules.1,2 The guest is held within
the host by constrictive binding energy, which is a new
binding phenomenon.3 Constrictive binding energy is the
activation energy required for a guest to enter the inner
cavity of a hemicarcerand through one of the size-
restricted equatorial portals. In many hemicarcerands,
these equatorial portals are large enough to allow bulk-
phase reactants to enter and exit the inner phases and
to undergo inner-phase reactions with the incarcerated
guest.4-14 Some of these reactions are truly inner-phase
reactions, in which all reactants are fully incarcerated,
(6) Cram, D. J .; Tanner, M. E.; Thomas, R. Angew. Chem. 1991, 103,
1048-1051; Angew. Chem., Int. Ed. Engl. 1991, 30, 1024-1027.
(7) Warmuth, R.; Marvel, M. A. Chem.sEur. J . 2001, 7, 1209-1220.
(8) Warmuth, R. Angew. Chem. 1997, 109, 1406-9; Angew. Chem.
Int. Ed. Engl. 1997, 37, 1347-1350.
(9) Kurdistani, S. K.; Helgeson R.; Cram, D. J . J . Am. Chem. Soc.
1995, 117, 1659-1660.
(10) Cram, D. J .; Tanner, M. E.; Knobler, C. B. J . Am. Chem. Soc.
1991, 113, 7717-7726.
(11) (a) Farran, A.; Deshayes, K. D.; Matthews, C.; Balanescu, I. J .
Am. Chem. Soc. 1995, 117, 9614-9615. (b) Farran, A.; Deshayes, K.
D. J . Phys. Chem. 1996, 100, 3305-3307. (c) Pina, F.; Parola, A. J .;
Ferreira, E.; Maestri, M.; Armaroli, N.; Ballardini, R.; Balzani, V. J .
Phys. Chem. 1995, 99, 12701-12703. (d) Parola, A. J .; Pina, F.;
Ferreira, E.; Maestri, M.; Balzani, V. J . Am. Chem. Soc. 1996, 118,
11610-11616. (e) Place, I.; Farra´n, A.; Deshayes, K.; Piotrowiak, P.
J . Am. Chem. Soc. 1998, 120, 12626-12633. (f) Romanova, Z. S.;
Deshayes, K.; Piotrowiak, P. J . Am. Chem. Soc. 2001, 123, 11029-
11036. (g) Mendoza, S.; Davidov, P. D.; Kaifer, A. E. Chem.sEur. J .
1998, 4, 864-870.
† Kansas State University.
‡ University of California at Los Angeles.
§ Deceased J une 17, 2001.
(1) (a) Cram, D. J .; Karbach, S.; Kim, Y. H.; Baczynskyj, L.;
Kalleymeyn, G. W. J . Am. Chem. Soc. 1985, 107, 2575-6. (b) Cram,
D. J . Nature 1992, 356, 29-36. (c) Cram, D. J .; Cram, J . M. In
Container Molecules and Their Guests; Monographs in Supramolecular
Chemistry, No. 4; Stoddart, J . F., Series Ed.; The Royal Society of
Chemistry: Cambridge, U.K., 1994. (d) J asat, A.; Sherman, J . C. Chem.
Rev. 1999, 99, 931-967. (e) Warmuth, R.; Yoon, J . Acc. Chem. Res.
2001, 34, 95-105.
(2) (a) Mungaroo, R.; Sherman, J . C. Chem. Commun. 2002, 1672-
1673. (b) Chopra, N.; Sherman, J . C. Angew. Chem., Int. Ed. 1999, 38,
1955-1957. (c) Naumann, C.; Place, S.; Sherman, J . C. J . Am. Chem.
Soc. 2002, 124, 16-17.
(12) Yoon, J .; Nakamura, K. Bull. Korean Chem. Soc. 1999, 20, 765-
767.
(13) Cram, D. J .; Blanda, M. T.; Paek, K.; Knobler, C. B. J . Am.
Chem. Soc. 1992, 114, 7765-7773.
(3) Cram, D. J . Angew. Chem., Int. Ed. Engl. 1988, 27, 1009-1020.
(4) Warmuth, R. J . Inclusion Phenom. 2000, 37, 1-38.
(5) Robbins, T. A.; Cram, D. J . J . Am. Chem. Soc. 1993, 115, 12199-
12200.
(14) (a) Kang, J .; Rebek, J . J r. Nature 1996, 385, 50-52. (b) Kang,
J .; Santamar´ıa, J .; Hilmersson, G.; Rebek, J ., J r. J . Am. Chem. Soc.
1998, 120, 7389-7390. (c) Kang, J .; Hilmersson, G.; Santamar´ıa, J .;
Rebek, J ., J r. J . Am. Chem. Soc. 1998, 120, 3650-3656.
10.1021/jo026649z CCC: $25.00 © 2003 American Chemical Society
Published on Web 02/14/2003
J . Org. Chem. 2003, 68, 2077-2088
2077

Warmuth et al.
CHART 1
out in bulk phases? Could one completely reverse the
course of a reaction by disfavoring low-energy solution
pathways and favoring high-energy pathways through
incarceration? Would it be possible to change the regio-
and stereoselectivity of reactions?9 How would the sur-
rounding host alter the reactivity of the incarcerated
guest?15-18
A main objective of our research efforts is to explore
the scope of inner-phase reactions and through-shell re-
actions and to investigate the effect of cavity constraint,
portal size, and portal shape on guest reactivity. Such
research can lead to the development of artificial en-
zymes: tailored hemicarcerands as unimolecular reaction
flasks that mimic the binding sites of enzymes and allow
one to perform highly stereo- and regioselective trans-
formations of molecules incarcerated in their inner
phases.
have carbonyl groups with different reactivities, which
increase in bulk-phase reactions in the order 3 ≈ 2 < 1.
Our investigation shows that the order of reactivity is
strongly altered by incarceration. Furthermore, in the
course of our investigations, we discovered several in-
teresting reactions between the intermediary, incarcer-
ated lithium alcoholates and the surrounding hemicar-
cerand leading to the controlled scission of the host and
an enlargement of one of its portals. In particular, we
report the first hemicarcerand with a missing spanner
group. Previous work has shown that such hemicar-
cerands with one extended portal are very useful for
the incarceration of large and/or thermally sensitive
guests.7,9,20
Resu lts a n d Discu ssion
See Chart 1 for the structures of most of the compounds
discussed in this paper.
Here we report a detailed investigation of through-shell
reactions involving bulk-phase organometallic reactants
and also the highly selective borane reduction of benzal-
dehyde (1), benzocyclobutenone (2), and benzocyclobutene-
dione (3) inside the known hemicarcerand 4.19 We have
chosen guests 1-3 for two reasons: (1) The three guests
have different shapes, leading to different orientations
of their carbonyl groups with respect to the equatorially
located portals of hemicarcerand 4. (2) The three guests
Syn th esis of Hem ica r cep lexes. We prepared the
1-methyl-2-pyrrolidinone (NMP) hemicarceplex 4.NMP
as described for the corresponding N,N-dimethylaceta-
mide (DMA) hemicarceplex 4.DMA by replacing the
solvent DMA by NMP.19,20 Refluxing 4.NMP in diphenyl
ether for 2 days gave hemicarcerand 4.19 We prepared
hemicarceplex 4.3 as described earlier.8 Heating empty
4 in neat 1 or 2 at 150-170 °C yields 4.1 (95% yield)
and 4.2 (80% yield), respectively.
Th r ou gh -Sh ell Bor a n e Ad d ition s. In CPK models,
diborane is small enough to pass through one of the host’s
equatorial portals and hence is suitable for inner-phase
reductions. Indeed, refluxing 4.1 for 16 h in a BH3‚THF
(15) Warmuth, R. J . Chem. Soc., Chem. Commun. 1998, 59-60.
(16) Ko¨rner, S. K.; Tucci, F. C.; Rudkevich, D. M.; Heinz, T.; Rebek,
J ., J r. Chem.sEur. J . 2000, 6, 188-195.
(17) Warmuth, R.; Kerdelhue´, J .-L.; Sa´nchez Carrera, S.; Langen-
walter, K. J .; Brown, N. Angew. Chem., Int. Ed. 2002, 41, 96-99.
(18) van Wageningen, A. M. A.; Timmerman, P.; van Duynhoven,
J . P. M.; Verboom, W.; van Veggel, F. C. J . M.; Reinhoudt, D. N.
Chem.sEur. J . 1997, 3, 639.
(19) Robbins, T.; Knobler, C. B.; Bellew, D.; Cram, D. J . J . Am.
Chem. Soc. 1994, 116, 111-122.
(20) Yoon, J .; Sheu, C.; Houk, K. N.; Knobler, C. B.; Cram, D. J . J .
Org. Chem. 1996, 61, 9323-9339.
2078 J . Org. Chem., Vol. 68, No. 6, 2003

Alkyllithium Additions and Borane Reductions
F IGURE 2. 1H NMR spectrum (500 MHz in (CD2)4O at 25
°C) of hemicarceplex 4.9. Signals assigned to the guest
protons are marked with arrows.
F IGURE 1. Proposed guest orientation of 8 inside the inner
phase of hemicarcerand 4.
induced upfield shift ∆δ of the guest protons. The 1H
NMR spectrum of hemicarceplex 4.9 is shown in Figure
2. From ∆δ(H1) 0.91 and ∆δ(H2) 3.43 of the guest
protons,21,22 we predict that the favored orientation of the
elliptical guest is such that the borolane hydrogen is
located in the shielded region of one of the two cavitands
of 4, which explains the low reactivity of 9. The broaden-
ing of the signal assigned to the eight inward-pointing
Hi atoms of the methylene spanners of 4 (δ 4-4.7) is most
likely due to the restricted rotation of 9 around the
equatorial C2 axis of 4 and is consistent with the shape,
the predicted orientation, and the reactivity of 9 in the
inner phase.
in THF solution (1 M) did give incarcerated benzyl alcohol
4.5, but in only 13% yield, together with 65% unreacted
4.1 and 15% tetrahydrofuran hemicarceplex 4.THF.11
The latter hemicarceplex results most likely from dis-
sociation of 4.1 and/or 4.5 followed by incarceration of
a solvent molecule into empty 4 during the reaction.
Comparison with an authentic sample prepared by
heating empty 4 in benzyl alcohol (170 °C, 70 h, 95%
yield) confirmed the formation of 4.5. In contrast to their
bulk-phase reactivity, the through-shell reduction of
incarcerated 2 and 3 is faster than the reduction of
incarcerated 1. Refluxing 4.2 or 4.3 in excess BH3‚THF
in THF (1 M) (4.2, 40 h; 4.3, 49 h) gave 85% 4.6 and
5% 4.THF for the reduction of 4.2 (90% conversion),
and 86-90% 4.7 and ca. 5% 4.THF for the reduction
of 4.3 (91-95% conversion). The tetrahydrofuran hemi-
carceplex most likely was formed via a dissociation/
complexation sequence analogous to that described above.
Interestingly, despite the large excess of the reducing
agent, the through-shell reduction of 4.3 led to selective
monoreduction in contrast to bulk-phase reductions of
R-diketones.
11B NMR studies confirmed the formation of 9. Due to
1
quadrupole broadening, the H-coupled 11B NMR spec-
trum of 4.9 shows a broad 11B signal (line width 12.6
ppm) at δ 28.3 relative to that of BF3‚O(C2H5)2 (δ 0) with
no fine structure. The measured chemical shift is con-
sistent with the postulated structure of 9; e.g., δ 28.1 for
1,3,2-dioxaborolane (11).23
The preferred monoreduction must be a consequence
of the inaccessibility of the carbonyl group of 8 to the
bulk-phase BH3‚THF complex. We assume a guest ori-
entation inside the inner phase such that the borane is
located close to an equatorial host portal and the carbonyl
is in the inner space of a host’s cavitand (Figure 1). In
this orientation, Lewis complex formation between the
borane and an inward-pointing oxygen electron lone pair
of a tetramethylenedioxy bridge or a solvent molecule is
possible. The energetic burden associated with guest
reorientation in addition to steric factors might prevent
the through-shell interaction of the second carbonyl
oxygen with a bulk-phase borane molecule, which is
required for further reduction.
However, after aqueous workup and chromatographic
purification of 4.7, refluxing 4.7 in excess 1 M
BH3‚THF in THF for 90 h reduced guest 7 to the
dioxyborolane 9 in 85% yield. The shell of 4 protects the
guest, which survived the aqueous workup. Hydrolysis
in water-saturated CHCl3 at room temperature afforded
the corresponding cis-benzocyclobutene-7,8-diol hemicar-
ceplex 4.10 and required several weeks. The insensitiv-
ity of 9 toward hydrolysis can be rationalized by its inner-
phase orientation, suggested by the hemicarcerand-
Room -Tem p er a tu r e Meth yllith iu m Ad d ition to
In ca r cer a ted Ben za ld eh yd e. Addition of excess meth-
yllithium (MeLi) to 4.1 below 0 °C resulted in no re-
action. After 9 h at room temperature, 11% expected
product 4.12 was formed and was isolated together with
65% unreacted hemicarceplex 4.1. The addition of
N,N,N′,N′-tetramethylethylenediamine (TMEDA) to this
reaction mixture did not alter the outcome, suggesting
that the low reactivity of the incarcerated guest is not a
result of a higher aggregation state of MeLi. Comparison
1
of the product H NMR with that of an authentic sample
prepared by heating empty 4 in 1-phenylethanol (170 °C,
70 h, 95% yield) confirms the formation of 4.12.
(21) We used the proton chemical shifts of free cis-benzocyclobutene-
7,8-diol (10) in CDCl3 to estimate those of free 9. cis-Benzocyclobutene-
7,8-diol was prepared via NaBH4 reduction of 1 and showed spectral
data identical to those reported by Boyd et al. (ref 22).
(22) Boyd, D. R.; Sharma, N. D.; Stevenson, P. J .; Chima, J .; Gray,
D. J .; Dalto, H. Tetrahedron Lett. 1991, 32, 3887-3890.
(23) (a) McAchran, G. E.; Shore, S. G. Inorg. Chem. 1966, 5, 2044-
2046. (b) No¨th, H.; Wrackmeyer, B. In NMR. Basic Principles and
Progress; Diehl, P., Fluck, E., Kosfeld, R., Eds.; Springer-Verlag: New
York, 1978; Vol. 14, p 5.
J . Org. Chem, Vol. 68, No. 6, 2003 2079

Warmuth et al.
F IGURE 4. Proposed mechanism for the formation of 4.13
and 4.14 upon addition of methyllithium to 4.3.
F IGURE 3. Schematic representation of the proposed orien-
tation of lithium alcoholate 16 in the inner phase of hemicar-
cerand 4. This orientation provides sufficient coordination of
the lithium cation by electron lone pairs of the alcoholate and
a tetramethylenedioxy bridge of 4.
of 4, the inner-phase quench of this equilibrium mixture
yields guests 13 and 14 (Figure 4). We conclude that the
rearrangement of 16 takes place only upon raising the
temperature from -78 °C to room temperature and not
earlier, at -78 °C, since (1) 13 but not 14 is isolated upon
addition of 1 equiv of MeLi to free benzocyclobutenedione
at -78 °C followed by water quench at -78 °C25 and (2)
Swenton et al. previously observed a similar rearrange-
ment for a noncomplexed benzocyclobutenedione/vinyl-
lithium adduct in solution only after warming this adduct
to room temperature with the exclusion of water.26
Meth yllith iu m Ad d ition a t 0 °C to In ca r cer a ted
Ben zocyclobu ten ed ion e. To our surprise, the same
reaction, carried out at -10 or 0 °C, gave only trace
amounts (<5%) of the previously observed products as
determined from the integration of the characteristic
upfield-shifted methyl protons of 13 and 14 in the 1H
NMR spectrum of the crude reaction mixture. Instead,
we isolated diol host 19 (X ) H)9,20 as the main product
in 70% yield. We explain the cleavage of both ether C-O
bonds of a tetramethylenedioxy bridge of 4 via two
subsequent base-induced â-eliminations. The following
observation supports the action of incarcerated lithium
alcoholate 16 or 17 and not bulk-phase methyllithium
to initiate the cleavage of the first C-O bond: Methyl-
lithium does not cleave diol host 19 (X ) H) or the
Meth yllith iu m Ad d ition a t -78 °C to In ca r cer a ted
Ben zocyclobu ten ed ion e. The addition of excess meth-
yllithium to a solution of 4.3 in THF/Et2O at -78 °C
followed by H2O quench at -78 °C after 30 h resulted in
the formation of two new hemicarceplexes, 4.13 (65%)
and 4.14 (25%). Although a large excess of MeLi was
used, no double addition product 15 was formed in this
through-shell addition reaction.
We assume that the initially formed lithium alcoholate
16 has an inner-phase orientation which allows the
coordination of the lithium cation to the alkoxide and
carbonyl oxygen of 16 and to an inward-pointing electron
lone pair of each of the two oxygens of a tetramethyl-
enedioxy bridge as shown in Figure 3. In this orientation,
the addition of a second equivalent of MeLi will be
difficult.
While the product 4.13 was expected, the formation
of 4.14 is more surprising. Moore showed that R-hy-
droxycyclobutenones thermally rearrange to furanones.24,25
An inner-phase Moore rearrangement of 8-hydroxy-8-
methylbenzocyclobuten-7-one (13) could explain the for-
mation of 7-methylphthalide (14). Since hemicarceplex
4.13 is thermally stable at room temperature, it is
unlikely that this rearrangement took place at room
temperature during the workup. We explain the forma-
tion of 4.14 as follows (Figure 4). The through-shell MeLi
addition leads to lithium alcoholate 16. Due to the
shielding nature of the surrounding hemicarcerand,
addition of water at -78 °C does not quench 16 at this
temperature. When the temperature is raised, 16 rear-
ranges to 17 via the quinodimethane 18a ,b, similar to
the Moore rearrangement mentioned above.24,25 At a
temperature sufficiently high for water to enter the cavity
27
benzene hemicarceplex 4.C6H6 under the same condi-
tions (excess MeLi, 0 °C, 8 h, THF/ether). We conclude
that the first ether bond cleavage must be initiated via
hemicarcerand deprotonation by the incarcerated lithium
alcoholate, yielding 20 (X ) Li) (Figure 5). Subsequent
cleavage of the second ether linkage most likely proceeds
via methyllithium, a stronger base than the alcoholate.
Interestingly, alcoholates in general are too weak as
bases to induce cleavage of ethers via â-elimination. In
control experiments, lithium alcoholate 21 (0.4 M) does
not cleave the toluene hemicarceplex 4.C6H5CH328 even
after 14 days at 0 °C or after 24 h in refluxing THF/
hexane. This corresponds to a rate acceleration for the
innermolecular â-elimination by a factor of .104 com-
pared to the intermolecular â-elimination. The inner-
phase lithium alcoholate concentration is about 8 mol/L,
(26) Swenton, J . S.; J ackson, D. K.; Manning, M. J .; Raynolds, P.
W. J . Am. Chem. Soc. 1978, 100, 6182-6188.
(24) (a) Perri, S. T.; Foland, L. D..; Decker, O. H. W.; Moore, H. W.
J . Org. Chem. 1986, 51, 3068-3070. (b) Foland, L. D.; Karlsson, J . O.;
Perri, S. T.; Schwabe, R.; Xu, S. J .; Patil, S.; Moore, H. W. J . Am. Chem.
Soc. 1989, 111, 975-989.
(25) (a) Liebeskind, L. S.; Iyer, S.; J ewell, C. F., J r. J . Org. Chem.
1986, 51, 3067-3068. (b) Liebeskind, L. S. Tetrahedron 189, 45, 3053-
3060.
(27) Hemicarceplex 4.C6H6 was prepared as described in ref 19 and
gave a 1H NMR spectrum that is consistent with the spectrum reported
in ref 19.
(28) (a) Kerdelhue´, J .-L.; Langenwalter, K. J .; Warmuth, R. J . Am.
Chem. Soc. 2003, 125, 973-986. (b) Makeiff, D. A.; Pope, D. J .;
Sherman, J . C. J . Am. Chem. Soc. 2000, 122, 1337-1342.
2080 J . Org. Chem., Vol. 68, No. 6, 2003

Alkyllithium Additions and Borane Reductions
in the presence of TMEDA, no products could be isolated
corresponding to 13 or 14; these would, according to CPK
models, emerge partially from the inner phase through
one equatorial portal. However, small amounts of the diol
host 19 (X ) H) (10%) and host 20 (X ) H) (10%) were
isolated together with recovered starting material (62%).
Both products are probably formed via the mechanism
outlined above. Control experiments show the stability
of 4 toward n-BuLi under the chosen conditions. This,
and in particular the isolation of 20 (X ) H), indicates
that n-BuLi slowly adds to incarcerated 3 and that 19 is
formed via the mechanism described above. Steric hin-
drance due to the larger butyl group probably prevents
a faster reaction and hence a higher yield during the
reaction time of 12 h. Such a bulk-phase reactant size
selectivity was previously observed by Kurdistani et al.
in their studies of the alkylation and isotopic exchanges
of various incarcerated phenols.9 Better spatial fixation
of the alcoholate in the vicinity of an equatorial portal,
due to the butyl substituent, may lead to â-elimination
at -78 °C, which is faster than the addition of n-BuLi to
incarcerated 3.
F IGURE 5. Mechanism for base-induced cleavage of 4.3 to
yield the diol host 19 (X ) Li).
Meth yllith iu m Ad d ition to In ca r cer a ted Ben zo-
cyclobu ten on e. A completely different reaction behavior
was observed for 4.2. Addition of MeLi to 4.2 in THF/
ether below -6 °C resulted in no reaction. After 5 h at
room temperature a complex product mixture formed,
composed of the 7-methylbenzocyclobuten-7-ol hemicar-
ceplex 4.22 (25%), the 2-methylacetophenone hemicar-
ceplex 4.23 (ca. 1%, as determined from the integration
of the characteristic guest signal at δ -1.40 in the 1H
NMR spectrum of the crude product mixture), and the
hemicarcerand 24 (25%) and its 2-methylacetophenone
hemicarceplex 24.23 (40%). We supported our assign-
F IGURE 6. Proposed transition state for the lithium alkoxide-
induced cleavage of hemicarcerand 4, yielding hemicarcerand
20 (X ) Li).
on the basis of an inner-phase volume of approximately
200 Å3, which we estimated from the X-ray structures of
4.1, 4.2, and 4.3 (see below). Even if we take the 20-
fold higher local concentration into account, the inner
molecular â-elimination is still more than 100-fold faster
than the intermolecular â-elimination. Three factors
might contribute to this rate acceleration: (1) the absence
of aggregation of the alcoholate in the inner phase,29 (2)
the poor ability of 4 to “solvate” the incarcerated alco-
holate, leading to an increased basicity,10 (3) lithium
coordination by the alkoxide lone pair and an oxygen lone
pair of the cleaved C-O bond. Coordination fixes the
alkoxide oxygen in close proximity to the â-hydrogen of
the bridge. Via a six-membered transition state the
lithium cation can efficiently compensate for the buildup
of negative charge at the phenoxide during the concerted
syn-elimination (Figure 6).30
ment by comparison with authentic samples prepared as
follows: Formation of 4.23 by heating empty 4 in neat
2-methylacetophenone failed. However, heating empty
24, which has one enlarged portal due to the missing
spanner, in neat 23 gave 24.23. The latter can be
converted into 4.23 with BrCH2Cl/K2CO3 in DMF (80%
yield).31 Irradiation of incarcerated 23 at λ > 300 nm in
CDCl3 gave 4.22 via photoinduced electrocyclization.32
Our proposed mechanism to explain the formation of
24 is related to the SN2cA mechanism of acetal cleavage
Bu tyllith iu m Ad d ition a t -78 °C to In ca r cer a ted
Ben zocyclobu ten ed ion e. Upon addition of n-butyl-
lithium (n-BuLi) at -78 °C to 4.3 in hexane/THF (1:5)
(29) Exner, J . H.; Steiner, E. C. J . Am. Chem. Soc. 1974, 96, 1782-
1787.
(30) (a) Bartsch, R. A. Acc. Chem. Res. 1975, 8, 239-245. (b) Sicher,
J . Angew. Chem. 1972, 84, 177-191; Angew. Chem., Int. Ed. Engl.
1972, 11, 200-214.
(31) Using the same strategy, we prepared authentic 4.14 via the
sequence 24 + 13 f 24.14; 24.14 + BrCH2Cl + K2CO3 f 4.14 to
prove the formation of 4.14 in the MeLi addition to 4.3 at -78 °C
(see also the Experimental Section).
J . Org. Chem, Vol. 68, No. 6, 2003 2081

Warmuth et al.
F IGURE 7. Proposed mechanism for the formation of hemicarcerand 24 and guest inside-out rotation of intermediate
hemicarcerand 26.
(Figure 7).33 After the addition of methyllithium to
incarcerated 2, the alcoholate 25 undergoes an inner
molecular nucleophilic attack at the acetal carbon of a
host spanner. Spaciotemporal effects are probably re-
sponsible for the strongly increased nucleophilicity of
incarcerated 25.34 In the newly formed product 26, host
and guest are linked together via a dioxymethylene
linkage. Once 26 is formed, the covalently bound guest
can easily rotate from the inner phase through the
equatorial host portal to the periphery of the hemicar-
cerand shell, where the acetal bond is cleaved during the
workup to give 24 (Figure 7). Such inside-out rotation of
a guest that is covalently bound to the portal should be
possible at temperatures low enough for constrictive
binding to prevent the dissociation of the corresponding
hemicarceplex with a nonlinked guest.7 Alternatively, 26
with the guest inside is cleaved during the workup,
followed by dissociation or base-catalyzed opening of the
cyclobutenol ring to give 24.23.
Consistent with our interpretation, we observed a
higher (4.22 + 4.23)/(24 + 24.23) ratio of approxi-
mately 1:1 (by 1H NMR) when the reaction was quenched
after 30 min at room temperature. Although 24.23 is
always formed as a major product, we never observed
formation of 24.22. Either 24.22 dissociates during the
workup to give empty 24, preventing its spectroscopic
detection, or the base-catalyzed ring-opening of 22 is
much faster inside 24 than in the inner phase of 4. The
latter interpretation is reasonable, since the presence of
two phenoxides in the molecular shell of deprotonated
24 should allow for a very efficient catalysis of the
F IGURE 8. Partial 1H NMR spectra (CDCl3, 360 MHz, 22
°C) of the hemicarceplex 4.NMP after 0 min (a), 85 min (b),
reaction without the need of a strongly solvated highly
polar hydroxide ion to enter the hydrophobic inner phase
of 24. Studies designed to elucidate these mechanisms
are currently under way in our laboratories. Interest-
13 h (c), and 44 h (d) of exposure to MeLi in THF/ether at 0°C
followed by aqueous workup. (e) Partial 1H NMR spectrum
(CDCl3, 360 MHz, 22 °C) of hemicarcerand 24 showing the four
ingly, in this methyllithium addition reaction, we did not
observe the formation of diol host 19 (X ) H) or host 20
(X ) H). On the other hand, 24 was not observed in the
alkyllithium additions to 4.3. Thus, small structural
changes of the incarcerated lithium alcoholate have a
very strong effect on the mode of its innermolecular
reactivity.
doublets (ratio 1:3:2:1) which are assigned to the outward-
pointing spanner protons. Signals assigned to the outward-
pointing spanner protons of 4.27 and 4.NMP are marked
with filled squares and filled triangles, respectively. Signals
assigned to the protons of incarcerated 27 and NMP are
marked with arrows and filled circles, respectively.
Meth yllith iu m Ad d ition to 4.NMP . Our results in
the alkyllithium additions to 4.2 and 4.3 led us to see
if 19 and/or 24 might also be formed from other incarcer-
ated alkoxides. This was the basis for investigating the
reaction between 4.NMP and methyllithium. Indeed,
addition of MeLi to 4.NMP in THF/ether at 0 °C led to
(32) Wagner, P. J .; Subrahmanyam, D.; Park, B.-S. J . Am. Chem.
Soc. 1991, 113, 709-710.
(33) (a) Kresge, A. J .; Weeks, D. P. J . Am. Chem. Soc. 1984, 106,
7140-7143. (b) Amyes, T. L.; J encks, W. P. J . Am. Chem. Soc. 1989,
111, 7900-7909.
(34) Menger, F. M. Acc. Chem. Res. 1993, 26, 206-212.
2082 J . Org. Chem., Vol. 68, No. 6, 2003

Alkyllithium Additions and Borane Reductions
the complete consumption of 4.NMP after 70 h, and 24
was isolated in 70% yield. We followed the progress of
this reaction by 1H NMR spectroscopy (Figure 8). Im-
mediately after the MeLi addition, the intensity of the
NMP guest signals decreased, while new guest signals
at δ -0.21 (s), -0.45 (m), -0.87 (s), and -0.9 (m)
appeared. We assign these new signals to guest protons
of the 1,2-dimethylpyrrolinol hemicarceplex 4.27 on the
basis of the following observations: (1) The FAB-MS
spectrum showed an intensive signal at m/ z 2365 (58%),
which is the mass expected for the methyllithium addi-
tion product. (2) The newly formed hemicarceplex 4.27
is stable in solution at room temperature. (3) The guest
quantitatively escapes the inner phase of 4 during the
column chromatographic workup on silica gel, leaving the
empty hemicarcerand. We assume that the acidic silica
gel leads to the guest protonation, which facilitates
hemicarceplex dissociation.10 All three observations are
consistent with the formation of 4.27. After longer
reaction times, the intensity of the proton signals of
27 slowly decreased accompanied by the buildup of
new signals, which are assigned to host 24 (Figure 8).
As discussed earlier for incarcerated lithium alcohol-
ate 21 (Figure 7), we assume that the lithium alcohol-
ate 28 underwent a nucleophilic transacetalization, lead-
ing to 29, which is hydrolyzed to give 24 during the
workup.
F IGURE 9. Stereoview of the hemicarceplex 4.benzaldehyde‚
2nitrobenzene. All H atoms are omitted; two of the four
disordered benzaldehyde guest positions are shown (the other
two are generated by the center of symmetry).
F IGURE 10. Stereoview of the hemicarceplex 4.benzocyclo-
butenone‚2nitrobenzene. All H atoms are omitted; one of the
two disordered benzocyclobutenone guest positions is shown
(the other is generated by the center of symmetry).
The efficient synthesis of 4.NMP, which can be
prepared in high yield (40-45%) from cavitand 30,10 and
the absence of other cleavage products such as 19 and
20 make the MeLi addition to 4.NMP a fast method for
the preparation of empty hemicarcerand 4 or host 24.
Hemicarcerand 24 should be a very useful intermediate
for the incarceration of guests that are too large to pass
through an equatorial portal of 4 and for the incorpora-
tion of catalytic functional groups into the host shell.
Cor r ela tion betw een Rea ctivity a n d Gu est Con -
for m a tion . The observed inner-phase guest reactivity
decreased in the order 3 . 2 > 1 in the through-shell
methyllithium additions and in the order 3 ≈ 2 > 1 in
the through-shell borane reductions. Why are 2 and 3
F IGURE 11. Stereoview of the hemicarceplex 4.benzocyclo-
butenedione‚2nitrobenzene. All H atoms are omitted; one of
the two disordered benzocyclobutenedione guest positions is
shown (the other is generated by the center of symmetry).
more reactive than benzaldehyde 1 in the inner phase
as compared to the liquid bulk phase? The selectivity in
the MeLi addition to incarcerated 3 and also the fact that
the addition takes place at -78 °C suggest that the
reacting carbonyl group is located close to an equatorially
located entryway of 4. To determine the orientations of
the guests 1-3 in the inner phase of host 4, we undertook
X-ray crystal studies of the three hemicarceplexes. All
three complexes were crystallized by slow evaporation
J . Org. Chem, Vol. 68, No. 6, 2003 2083

Warmuth et al.
F IGURE 12. Top views of 4.1, 4.2, and 4.3, left to right. “Feet”, nitrobenzene, and all H atoms are removed for clarity. Disordered
-O(CH2)4O- bridging linkers are shown at the bottom left and top right. The entire guest model is shown (site occupancy 1.0).
CHART 2. Hem ica r cer a n d -In d u ced Up field Sh ifts of Gu est P r oton s in 4.1, 4.2, a n d 4.3
from chloroform/nitrobenzene, and all three are ap-
proximately isomorphous with other complexes of the
same host crystallized from this solvent mixture.19 In the
crystals, 4 lies on a center of symmetry, requiring
disorder in the noncentrosymmetric guests. One molecule
of nitrobenzene is located in each “foot” region; these
regions are formed by four -CH2CH2C6H5 groups that
help to make the host soluble in nonpolar solvents. In
4.1 (Figure 9), the final refinement model includes two
guest molecules at 0.25 occupancy, roughly orthogonal
to each other. In 4.2 and 4.3 (Figures 10 and 11), one
molecule of guest was refined at 0.5 occupancy. The
additional guest moieties generated by the center of
symmetry bring the site occupancies to 1.0 in each
complex. Though the guests and bridges are disordered,
it is clear that the carbonyl O atom lies in a polar cap of
the host in each complex, while the second carbonyl O
F IGURE 13. Energy-minimized structure of hemicarceplex
in 4.3 is in a portal, between bridges (Figures 9-11).
Host 4 is able to adjust to the dimensions and chemical
nature of its guests,19 and 4.1, 4.2, and 4.3 demon-
strate this adaptability in two ways. First, the two
partial-occupancy, planar molecules of 1 in the complex
are aligned between bridges, with the shortest guest-to-
host distances being H(guest)‚‚‚bridgehead O(host), 2.3-
2.6 Å. In 4.2 the guest is also aligned between bridges,
and the shortest guest-to-host distances are H(guest)‚‚‚
bridgehead O(host), 1.9-2.4 Å, and O(guest)‚‚‚spanner
H(host), 2.8 Å. In contrast, 3 in its complex is aligned
between portals, with the shortest guest-to-host distance
being O(guest)‚‚‚spanner H(host), 2.2-2.3 Å. Second,
though the two hemispheres of host 4 may twist with
respect to each other to accommodate guests,19 in 4.1,
4.2, and 4.3, as in all other examples of the isomor-
phous nitrobenzene solvate series, they are eclipsed.
However, the distances between the planes of the bridge-
head O atoms of 4 in connected hemispheres differ (3.77
Å in 4.1, 3.87 Å in 4.2, and 3.99 Å in 4.3), showing
the flexibility of the -O(CH2)4O- linkers. Figure 12
4.2 (MM2 force field, Macromodel).35,36 Phenethyl feet and
host hydrogen atoms have been omitted. Atom coloring is as
follows: O, red; host C, gray; guest C, black; H, white.
shows the alignment of the guest molecules in the host
cavity.
Since the crystal study of 4.2 gave ambiguous starting
positions for guest atoms, we determined the lowest-
energy orientation by molecular mechanics calculations
(Figure 13). In the calculated structure the carbonyl
group is aligned with the host’s polar D4 axis and par-
tially protrudes into a polar cap of 4. When the calculated
internal coordinates for the guest were applied to the
refinement model in the crystal structure, the orientation
of the guest was confirmed (Figures 10 and 12).
The solid-state inner-phase orientation of 1 and 3 and
the predicted orientations of 2 are consistent with the
observed host-induced upfield shift of the guest protons
of 1-3 (Chart 2). Protons that are preferentially located
in a polar cap are shielded more strongly in comparison
to protons that are located in the central part of the inner
2084 J . Org. Chem., Vol. 68, No. 6, 2003

Alkyllithium Additions and Borane Reductions
CHART 3. Restr icted In ter n a l Rota tion of
In ca r cer a ted Ben za ld eh yd e in Its Low -En er gy a n d
High -En er gy Or ien ta tion s
of 1 in the transition-state region for the inner-phase
guest rotation leads to an unfavorable internal entropy
contribution and might contribute to the higher ∆Gq of 1
(Chart 3).
Hence, guest reactivity is not just a function of the
reactive group location with respect to the host’s equato-
rial portals, but is also very strongly correlated to the
guest mobility and the probability that the reactive group
will be exposed to an equatorial portal, which is required
for a through-shell reaction with a bulk-phase reactant.
F IGURE 14. Partial 1H NMR (400 MHz, CD2Cl2/CDCl3 (7:
1)) of 4.1 at different temperatures. Signals assigned to the
inward-pointing spanner protons of 4 are marked with filled
triangles.
phase.7,8,19,20,28 Thus, very similar guest orientations are
expected in solution and allow us to rationalize the
observed order of guest reactivity.
Con clu sion s
In 4.3, one guest carbonyl is favorably aligned for
efficient through-shell reaction without the necessity of
guest reorientation, which explains the high reactivity
of 3 even at -78 °C. On the other hand, in 4.1 and 4.2
a through-shell reaction requires 1 and 2 to tilt consider-
ably away from their favored orientation, which would
explain their low reactivity toward MeLi as compared to
that of 4.3. We suggest that the higher reactivity of 2
as compared to 1 is related to the ease of adapting a
reactive orientation in which the carbonyl group is
pointing toward a portal of 4, and that the energy is
smaller for 2. A qualitative measure for this energy might
be the barrier for guest rotation ∆Gq around the equato-
rial C2 axis of 4. We determined ∆Gq for 1 by low-
temperature 1H NMR studies. At 170.5 K, two sets of
inward-pointing hydrogen resonances are observed as a
consequence of the reduction of the host symmetry in the
limit of slow guest rotation around an equatorially located
host C2 axis (Figure 14, top). These multiplets coalesce
at Tc ) 184.4 K (Figure 14, middle), from which we
Through-shell and inner-phase chemistries clearly
differ from “conventional” chemistry in the bulk phase
with respect to their rates and their selectivity. In
accordance with earlier results,9,10 the inner-phase and
through-shell reactions reported here show the following
characteristic features:
(1) Functional groups of the guest that are located in
the inner cavity of a host’s polar cavitand are less reactive
than those exposed to an equatorial portal, which have
the potential for high through-shell reactivity.
(2) The reactivity of bulk-phase reactants is largely
influenced by their size and shape relative to the size
and shape of the host’s equatorial portals.
(3) If functional groups are protected in the guest’s
most favorable orientation, reactivity depends on the
inner-phase rotational mobility of the guest.
Furthermore, our studies on the through-shell alkyl-
lithium additions to 1-3 and NMP show that the basicity
and nucleophilicity of incarcerated lithium alcoholates
strongly exceed by several orders of magnitude those of
bulk-phase alcoholates, resulting in efficient innermo-
lecular elimination or nucleophilic transacetalization
reactions and the formation of hemicarcerands with one
extended portal in the host shell. In these innermolecular
reactions, small structural changes of the guest have a
sound effect on the reaction mode. Especially noteworthy
is the highly efficient formation of host 24 via the
addition of methyllithium to 4.NMP. This reaction
should be very useful for the attachment of a single tether
to a hemicarcerand, which is desirable for hemicarcerand
immobilization. Studies in this direction are currently
under way.
calculated ∆Gq
) 8.9 ( 0.3 kcal/mol.37
184.4K
The inner-phase rotation of 2, on the other hand, could
not be frozen out in (CD2)4O at 164 K, and therefore,
we predict a lower barrier as compared to that of 1,
consistent with our interpretation of the through-shell
reactivity of 1 and 2. The higher rotational barrier ∆Gq
of 1 is surprising in light of the larger size of 2. The
restriction of rotational mobility of the C(1)-Ccarbonyl bond
(35) (a) Goodman, J . M.; Still, W. C. J . Comput. Chem. 1991, 12,
1110-1117. (b) Allinger, N. L. J . Am. Chem. Soc. 1977, 99, 8127-
8134. (c) Still, W. C.; Tempczyk, A.; Hawley, R. C.; Hendrickson, T. J .
Am. Chem. Soc. 1990, 112, 6127-6129.
(36) (a) Houk, K. N.; Nakamura, K.; Sheu, C.; Keating, A. E. Science
1996, 273, 627-629. (b) Nakamura, K.; Houk, K. N. J . Am. Chem.
Soc. 1995, 117, 1853-1854. (c) Sheu, C.; Houk, K. N. J . Am. Chem.
Soc. 1996, 118, 8056-8070.
(37) Gu¨nther, H., Ed. NMR-Spektroskopie, 3rd ed.; Georg Thieme
Verlag: Stuttgart, Germany, 1992; pp 310-312.
Exp er im en ta l Section
All reactions were conducted under argon unless otherwise
stated. Tetrahydrofuran was freshly distilled from benzophe-
J . Org. Chem, Vol. 68, No. 6, 2003 2085

Warmuth et al.
none ketyl just prior to use. 1H NMR spectra and 13C NMR
spectra were recorded at the indicated frequency. FAB mass
spectra were determined with 3-nitrobenzyl alcohol matrix.
FT-IR spectra were obtained in CHCl3 solution. CHN elemen-
tary analyses were obtained from Desert Analytics, Tucson,
AZ. Gravity chromatography was performed on silica gel
(70-230 mesh). Thin-layer chromatography involved polyeth-
ylene-backed plates (silica gel, 0.25 mm). Preparative thin-
layer chromatography involved glass-backed plates (silica gel
60, 0.5 or 1 mm). Compounds 2,38 3,38,39 14,40 and 3010 were
synthesized according to literature procedures.
Gen er a l P r oced u r e for th e Syn th esis of Hem ica r ce-
p lexes 4.Gu est a n d 24.Gu est (P r oced u r e A). A 1 mol
equiv sample of hemicarcerand 4 (or 24) and 50 mol equiv of
the guest were sealed under argon in a glass ampule and were
heated for the specified time at the specified temperature.
After the reaction mixture had cooled to room temperature,
the ampule was opened. The reaction mixture was poured into
a 10-fold volume of methanol. If the free guest was a solid,
the content of the ampule was dissolved in a minimum volume
of CHCl3, which was poured into the 10-fold volume of
methanol. The precipitated crude hemicarceplex was collected
on a glass sinter (pore size 10-15 µm), washed with methanol,
and dried at high vacuum. Purification by column chromatog-
raphy on SiO2 with CH2Cl2 as the mobile phase or by
preparative TLC on SiO2 with CHCl3 as the mobile phase gave
the pure hemicarceplex 4.guest (24.guest).
(38). Anal. Calcd for C151H144O25: C, 76.89; H, 6.15. Found:
C, 76.91; H, 6.05.
1-P h en yleth a n ol Hem ica r cep lex 4.12. Application of
procedure A (70 h, 170 °C, preparative TLC on SiO2 (0.5 mm))
gave 4.12 (95% yield) as a white powder. 1H NMR (360 MHz,
CDCl3, 25 °C): δ 7.16-7.26 (m, 40H), 6.91 (br s, 8H), 6.17 (d,
2H, guest-H), 5.66 (d, 8H), 5.18 (t, 2H, guest-H), 4.84 (t, 8H),
4.6-3.9 (3 v br s, 16 H), 3.85 (t, 8H), 3.51 (t, 1H, guest-H),
3.05 (q, 1H, guest-CHCH2OH), 2.47-2.75 (m, 32H), 2.1-1.8
(m, 16H), -2.42 (d, 3H, guest-CH3). FAB-MS: m/z [M + H]+,
2371 (25); [M - 12 + H]+, 2251 (85). Anal. Calcd for
C
152H146O25: C, 76.94; H, 6.20. Found: C, 76.88; H, 6.20.
7-Meth ylp h th a lid e Hem ica r cep lex 24.14. Application
of procedure A (36 h, 185 °C, preparative TLC on SiO2 (0.5
mm)) gave 24.14 (70% yield) as a white powder. 1H NMR (400
MHz, CDCl3, 60 °C): δ 7.83 (br s, 1H, OH), 7.43 (br s, 1H,
OH), 7.30-7.10 (m, 40H, C6H5), 7.07 (s, 1H, ArH on bowl), 7.00
(s, 1H, ArH on bowl), 6.96 (s, 1H, ArH on bowl), 6.94 (s, 1H,
ArH on bowl), 6.93 (s, 1H, ArH on bowl), 6.92 (s, 1H, ArH on
bowl), 6.89 (s, 1H, ArH on bowl), 6.86 (s, 1H, ArH on bowl),
6.57 (d, 1H, guest), 6.26 (t, 1H, guest), 5.89 (d, 1H, OCH2O
outer), 5.86 (d, 1H, OCH2O outer), 5.85 (d, 1H, OCH2O outer),
5.78 (d, 1H, OCH2O outer), 5.54 (d, 1H, OCH2O outer), 5.47
(d, 1H, guest), 5.42 (d, 1H, OCH2O outer), 5.42 (d, 1H, OCH2O
inner), 5.27 (d, 1H, OCH2O outer), 5.07 (t, 1H, CH methine),
5.02 (t, 1H, CH methine), 4.96 (t, 1H, CH methine), 4.81 (t,
1H, CH methine), 4.79 (t, 1H, CH methine), 4.66 (t, 1H, CH
methine), 4.63 (t, 1H, CH methine), 4.62 (t, 1H, CH methine,),
4.62 (d, 1H, OCH2O inner), 4.56 (d, 1H, OCH2O inner), 3.87
(d, 1H, OCH2O inner), 3.84 (d, 1H, OCH2O inner), 4.22-3.76
(m, 15H, OCH2CH2), 3.76 (t, 1H, guest), 3.63 (m, 1H, OCH2-
CH2), 3.62 (q, 1H, CHCH3, guest), 3.25 (br s, 1H, OCH2O
inner), 3.12 (br s, 1H, OCH2O inner), 2.78-2.35 (m, 32H,
CH2CH2Ph), 2.1-1.6 (m, 16H, OCH2CH2), -2.01 (d, 3H, CH3,
guest). FAB-MS: m/z [M + H]+, 2384.6 (50); [M - 23 + H]+,
2238.1 (100). Anal. Calcd for C152H144O26: C, 76.49; H, 6.08.
Found: C, 76.70; H, 6.13.
Ben za ld eh yd e Hem ica r cep lex 4.1. Application of pro-
cedure A (5 days, 170 °C, preparative TLC on SiO2 (0.5 mm))
1
gave 4.1 (95% yield) as a white powder. H NMR (360 MHz,
CDCl3, 25 °C): δ 7.16-7.26 (m, 40H), 6.98 (s, 8H), 6.58 (s, 1H,
guest-CHO), 6.55 (d, 2H, guest-H), 5.75 (t, 2H, guest-H), 5.66
(d, 8H), 4.85 (t, 8H), 4.05 (d, 8H), 3.88 (br s, 16H), 3.64 (t, 1H,
guest-H), 2.47-2.75 (m, 32H), 1.90 (br s, 16H). FAB-MS: m/z
[M + H]+, 2356 (100); [M - 1 + H]+, 2250 (43). Anal. Calcd
for C151H142O25: C, 76.96; H, 6.07. Found: C, 76.89; H, 6.10.
Ben zocyclobu ten on e Hem ica r cep lex 4.2. Application
of procedure A (8 days, 150 °C, column chromatography on
SiO2) gave 4.2 (80% yield) as a white powder. H NMR (400
2-Meth yla cetop h en on e Hem ica r cep lex 24.23. Applica-
tion of procedure A (70 h, 185 °C, preparative TLC on SiO2
(0.5 mm)) gave 24.23 (65% yield) as a white powder. 1
1
MHz, CDCl3, 25 °C): δ 7.16-7.26 (m, 40H), 6.96 (s, 8H), 6.17
(d, 1H, guest-H), 5.73 (t, 1H, guest-H), 5.62 (d, 8H), 4.85 (t,
8H), 4.67 (d, 1H, guest-H), 4.08 (d, 8H), 3.92 (br s, 16H), 3.62
(t, 1H, guest-H), (s, 2H, guest-CH2), 2.47-2.75 (m, 32H), 1.88
(br s, 16H). FAB-MS: m/z [M + H]+, 2368.4 (100); [M - 2 +
H]+, 2250 (43). Anal. Calcd for C152H142O25: C, 77.04; H, 6.04.
Found: C, 77.27; H, 6.01.
H NMR
(400 MHz, CDCl
3, 25 °C): δ 7.10-7.30 (m, 40H, C6H5), 7.01
(s, 2H, ArH on bowl), 6.92 (s, 2H, ArH on bowl), 6.91 (s, 2H,
ArH on bowl), 6.86 (s, 2H, ArH on bowl), 6.77 (s, 2H, OH,
disappears upon D2O addition), 6.24 (d, 1H, guest-H), 5.93 (d,
1H, OCH
2O outer), 5.89 (d, 1H, OCH2O outer), 5.82 (d, 1H,
OCH
2O outer), 5.73 (tb, 1H, guest-H), 5.49 (d, 1H, OCH2O
Ben zocyclobu ten ed ion e Hem ica r cep lex 4.3.8 Applica-
outer), 5.42 (d, 1H, OCH2O outer), 5.00 (t, 1H, CH methine),
4.96 (t, 2 H, CH methine), 4.92 (d, 2 H, OCH2O inner), 4.73 (t,
2 H, CH methine), 4.61 (t, 1H, CH methine), 4.57-4.65 (m, 5
H, OCH2O inner, CH methine), 4.55 (br d, 1H, guest-H), 3.8-
4.24 (m, 14 H, OCH2CH2), 3.82 (d, 2 H, OCH2O inner), 3.7-
3.78 (m, 2H, OCH2CH2), 3.66 (t, 1H, guest-H), 3.44 (d, 2 H,
OCH2O inner), 2.4-2.8 (m, 32H, CH2CH2Ph), 1.75-2.0 (m,
16H, OCH2CH2), 1.64 (s, 3H, guest-CH3), -1.29 (s, 3H, guest-
CH3). FAB-MS: m/z [M + H]+, 2371 (50); [M - 23 + H]+, 2238
(100). Anal. Calcd for C152H146O25: C, 76.94; H, 6.20. Found:
C, 76.97; H, 5.97.
tion of procedure A (4 days, 145 °C, column chromatography
1
on SiO2) gave 4.3 (30% yield) as a slightly yellow powder. H
NMR (500 MHz, CDCl3, 25 °C): δ 7.16-7.27 (m, 40H), 6.96
(s, 8H), 5.76 (dd, 2H, guest-H), 5.62 (d, 8H), 5.25 (dd, 2H, guest-
H), 4.86 (t, 8H), 4.10 (d, 8H), 3.96 (br s, 16H), 2.47-2.75 (m,
32H), 1.92 (br s, 16H). FT-IR (CHCl3, 25 °C, cm-1): ν(CO) 1782.
FAB-MS: m/z [M + H]+, 2382.9 (100); [M - 3 + H]+, 2250.8
(39). Anal. Calcd for C152H140O26: C, 76.62; H, 5.92. Found:
C, 76.87; H, 5.77.
Ben zyl Alcoh ol Hem ica r cep lex 4.5. Application of pro-
2-Meth yla cetop h en on e Hem ica r cep lex 4.23 (P r oce-
d u r e B). Hemicarceplex 24.23 (18 mg, 7.6 µmol), anhydrous
potassium carbonate (500 mg), excess BrCH2Cl (1 mL), and
degassed DMF (4 mL) were stirred for 48 h at 60 °C under
argon. The cooled reaction mixture was poured into methanol/
water (1:1, 50 mL). The precipitate was filtered off, washed
with water and methanol, and dried at high vacuum. Prepara-
tive TLC (SiO2, mobile phase CH2Cl2) provided 12 mg of
hemicarceplex 24.23 as a white powder (67% yield). 1H NMR
(360 MHz, CDCl3, 25 °C): δ 7.1-7.3 (m, 40H), 6.92 (s, 8H),
6.63 (t, 1H, guest-H), 6.49 (d, 1H, guest-H), 5.66 (d, 8H), 4.85
(t, 8H), 4.45 (d, 1H, guest-H), 4.24 (br s, 8H), 3.97 (br s, 16H),
3.63 (t, 1H, guest-H), 2.45-2.75 (m, 32H), 1.84 (br s, 19H),
-1.40 (s, 3H, guest-CH3). FT-IR (CHCl3, 25 °C, cm-1): ν(CO)
1696. FAB-MS: m/z [M + H]+, 2383 (63); [M - 23 + H]+, 2250
cedure A (70 h, 170 °C, preparative TLC on SiO2 (0.5 mm))
gave 4.5 (95% yield) as a white powder. H NMR (360 MHz,
1
CDCl3, 25 °C): δ 7.1-7.4 (m, 40H), 6.94 (s, 8H), 6.09 (d, 2H,
guest-H), 5.66 (d, 8H), 5.37 (t, 2H, guest-H), 4.86 (t, 8H), 4.71
(br s, 1H, guest-H), 4.20 (d, 8H), 3.89 (br s, 16H), 3.24 (t, 1H,
guest-H), 2.47-2.75 (m, 32H), 1.91 (br s, 16H), 1.75 (d, 2H,
guest-CH2), -2.59 (t, 1H, guest-H). FAB-MS: m/z [M + H]+,
2358 (37); [M - CH2O + H]+, 2328 (100); [M - 5 + H]+, 2250
(38) Du¨rr, H.; Nickels, H.; Pacala, L. A.; J ones, M., J r. J . Org. Chem.
1980, 45, 973-980.
(39) Cava, M. P.; Mangold, D.; Muth, K. J . Org. Chem. 1964, 29,
2947-2948.
(40) Canonne, P.; Plamondon, J .; Akssira, M. Tetrahedron 1988, 44,
2903-2912.
2086 J . Org. Chem., Vol. 68, No. 6, 2003

Alkyllithium Additions and Borane Reductions
(100). Anal. Calcd for C153H146O25: C, 76.99; H, 6.25. Found:
C, 77.36; H, 6.22.
(br s, 8H, OCH2CH2), 3.61 (t, 1H, guest-H), 2.46-2.78 (m, 32H,
CH2CH2Ph), 1.94 (br s, 16H, OCH2CH2), 1.32 (v br s, 1H, guest-
H). 13C NMR (125.6 MHz, CDCl3, 25 °C): δ 186.8 (guest), 155.7
(guest), 148.6, 146.3 (guest), 143.8, 141.9, 138.7, 132.3 (guest),
130.6 (guest), 128.6, 128.5, 126.0, 122.1 (guest), 119.4 (guest),
114.3, 99.0, 85.6 (guest), 72.2, 37.2, 34.6, 32.6, 27.7. FAB-MS
(NBA-matrix): m/z [M + H]+, 2384 (100); [M - 7 + H]+, 2250
(42). Anal. Calcd for C152H142O26: C, 76.56; H, 6.00. Found:
C, 76.33; H, 6.05.
7-Meth ylp h th a lid e Hem ica r cep lex 4.14. Application of
procedure B on hemicarceplex 24.14 gave hemicarceplex 4.14
as a white powder. 1H NMR (400 MHz, CDCl3, 25 °C): δ 7.16-
7.26 (m, 40H), 6.91 (s, 8H), 6.81 (d, 1H, guest-H), 6.75 (t, 1H,
guest-H), 5.63 (br s, 8H), 5.29 (d, 1H, guest-H), 4.87 (t, 8H),
4.32 (v br s, 8 H), 3.9-4.1 (m, 16H), 3.70 (q, 1H, guest-
CHCH3O), 3.70 (t, 1H, guest-H), 2.47-2.75 (m, 32H), 1.84 (br
s, 16H), -2.15 (d, 3H, guest-CH3). FT-IR (CHCl3, 25°C, cm-1):
ν(CO) 1781. FAB-MS: m/z [M + H]+, 2398.1 (100); [M - 14 +
H]+, 2249.5 (85). Anal. Calcd for C153H144O26: C, 76.61; H, 6.05.
Found: C, 76.73; H, 5.77.
Hem ica r cep lex 4.9 via Bor a n e Red u ction of 4.7.
Application of procedure C to hemicarceplex 4.7 (reflux, 90
h, preparative TLC on SiO2 with CH2Cl2 as the mobile phase)
gave hemicarceplex 4.9 (85% yield). 1H NMR (500 MHz,
(CD2)4O/5% D2O, 25 °C): δ 7.13-7.3 (m, 40H, C6H5), 7.01 (br
s, 8H, ArH on bowl), 6.54 (dd, 2H, guest-H), 5.68 (d, 8H,
OCH2O outer), 4.92 (s, 2H, guest-H), 4.82 (t, 8H, CH methine),
3.9-4.7 (v br s, 8H, OCH2O inner), 4.00 (br s, 16H, OCH2-
CH2), 3.92 (dd, 2H, guest-H), 2.47-2.70 (m, 32H, CH2CH2Ph),
1.91 (br s, 16H, OCH2CH2), -0.6 to -0.1 (v br s, 1H, BH). 13C
NMR (125.6 MHz, CDCl3, 25 °C): δ 148.2, 145.7 (guest), 144.4,
141.8, 138.6, 128.6, 128.4, 127.7 (guest), 126.0, 122.7 (guest),
114.0, 97.5, 78.0 (guest), 71.8, 37.1, 34.5, 32.5, 29.7, 28.1. 11B
NMR (160.5 MHz, CDCl3, 25 °C): δ 28.3. FT-IR (CHCl3, 25
°C, cm-1): ν(B-H) 2655. FAB-MS (NBA-matrix): m/ z ([M +
H]+, 2395.7 (75); [M - 9 + H]+, 2249.4 (100). Anal. Calcd for
C152 H143O26B: C, 76.18; H, 6.01. Found: C, 76.23; H, 6.00.
1-Meth yl-2-p yr r olid in on e Hem ica r cep lex 4.NMP .20
A
solution of 30 (2 g, 1.97 mmol),10 butane-1,4-diol dimesylate
(2 g, 8.13 mmol), and anhydrous Cs2CO3 (10 g) in degassed
NMP (500 mL) was stirred for 24 h at room temperature under
argon. Butane-1,4-diol dimesylate (2 g, 8.13 mmol) was added,
and stirring was continued for 24 h at room temperature and
for 48 h at 60 °C. The reaction mixture was cooled to room
temperature and poured into water/brine (1:1, 4 L). The
precipitate was filtered off, washed with water (500 mL) and
methanol (100 mL), and redissolved in CHCl3 (20 mL). The
CHCl3 solution was dropwise added to vigorously stirred
methanol (200 mL). The precipitated crude product was
filtered off, washed with methanol (10 mL), and dried at high
vacuum overnight. The crude product was purified by column
chromatography on SiO2 with CHCl3 as the mobile phase to
give 4.NMP as a white solid (1.02 g, 44% yield). 1H NMR (360
MHz, CDCl3, 25 °C): δ 7.25-7.12 (m, 40H), 6.86 (s, 8H), 5.81
(d, 8H), 4.85 (t, 8H), 4.32 (d, 8H), 3.95 (br s, 16H), 2.40-2.73
(m, 32H), 1.99 (m, 18H), -0.58 (t, 2H, guest), -0.76 (m, 2H,
guest), -0.88 (s, 3H, guest).
Hyd r olysis of Hem ica r cep lex 4.9. Hemicarceplex 4.9
(10 mg, 4.2 µmol) dissolved in water-saturated CDCl3 was left
standing at room temperature in a NMR tube. The guest
slowly hydrolyzed within 3 weeks to yield 4.cis-benzocy-
clobutene-7,8-diol (4.10). Purification by preparative TLC on
SiO2 with CHCl3 as the mobile phase gave 4.10 as a white
1
solid (9 mg, 90% yield). H NMR (500 MHz, CDCl3, 25 °C): δ
Bor a n e Red u ction of 4.1 (P r oced u r e C). A solution of
BH3‚THF in THF (1 M, 5 mL) was added under argon to of
4.1 (12 mg, 5.1 µmol). The solution was refluxed for 16 h,
cooled to room temperature, and quenched by slow syringe
addition of water (5 mL). CHCl3 (10 mL) was added. Both
layers were separated, and the aqueous layer was extracted
twice with CHCl3 (5 mL). The combined organic layers were
dried over MgSO4 and concentrated to 1 mL. Methanol (10 mL)
was added. The precipitate was filtered off, washed with
methanol, and dried at the high vacuum for 10 h to yield a
hemicarceplex mixture (12 mg), whose composition was de-
termined by the intensities of selected 1H NMR signals of all
incarcerated guest molecules. The mixture contained 13% 4.5,
13% 4.THF,19 and 67% 4.1.
7.16-7.3 (m, 40H, C6H5), 6.95 (s, 8H, ArH on bowl), 5.64 (d,
8H, OCH2O outer), 5.08 (m, 2H, guest-H), 5.00 (m, 2H, guest-
H), 4.84 (t, 8H, CH methine), 4.33 (d, 8H, OCH2O inner), 3.95
(br s, 16H, OCH2CH2), 3.76 (d, 2H, guest-H), 2.47-2.73 (m,
32H, CH2CH2Ph), 1.90 (br s, 16H, OCH2CH2), 1.41 (d, 2H,
guest-H). 13C NMR (125.6 MHz, CDCl3, 25 °C) δ 148.4, 145.7
(guest), 144.0, 141.8, 138.7, 128.6, 128.4, 127.9 (guest), 126.0,
121.8 (guest), 113.9, 98.9, 83.0 (guest), 73.0 (guest), 72.3, 37.1,
34.5, 32.6, 29.7, 27.7. FAB-MS: m/z [M + H]+, 2386 (100).
Anal. Calcd for C152H144O26: C, 76.49; H, 6.08. Found: C, 76.81;
H, 6.15.
Meth yllith iu m Ad d ition to 4.1 (P r oced u r e D). Hemi-
carceplex 4.1 (55 mg, 23.3 µmol) was dissolved in dry THF (5
mL) under argon. MeLi in diethyl ether (1.4 M, 4 mL) was
added, and the mixture was stirred for 8 h, cooled to -78 °C,
and quenched with saturated NH4Cl solution (5 mL). The two
phases were separated, and the aqueous layer was extracted
with chloroform (10 mL). The combined organic layers were
dried over MgSO4 and concentrated. The crude product
mixture was purified by preparative TLC on SiO2 with CHCl3/
5% EtOAc as the mobile phase to yield 64% recovered 4.1
(35 mg) and 11% 4.1-phenylethanol (4.12) (6 mg).
Bor a n e Red u ction of 4.2. Application of procedure C to
hemicarceplex 4.2 (reflux, 40 h) gave a crude hemicarceplex
mixture containing 85% 4.benzocyclobutenol (4.6) as deter-
1
mined by H NMR. Purification by preparative TLC on SiO2
with CH2Cl2 as the mobile phase gave 4.6 as a white solid.
1H NMR (400 MHz, CDCl3, 25 °C): δ 7.15-7.3 (m, 40H), 6.95
(s, 8H), 6.07 (t, 1H, guest-H), 5.64 (d, 8H), 5.54 (d, 1H, guest-
H), 4.85 (t, 8H), 4.31 (m, 1H, guest-H), 4.26 (d, 8H), 4.00 (m,
8H), 3.86 (m, 8H), 2.47-2.75 (m, 32H), 2.65 (d, 1H, guest-H),
1.85-1.95 (m, 16H), 1.26 (s, 1H, guest-OH), 1.01 (dd, 1H,
guest-H), 0.9 (dd, 1H, guest-H). FAB-MS: m/z [M + H]+,
2370.1 (37); [M - CH2O + H]+, 2340.3 (100); (M - 6 + H]+,
2249.6 (22). Anal. Calcd for C152H144O25: C, 77.01; H, 6.15.
Found: C, 77.31; H, 5.97.
Bor a n e Red u ction of 4.3. Application of procedure C to
hemicarceplex 4.3 (reflux, 49 h) gave a crude hemicarceplex
mixture containing 90% hemicarceplex 4.8-hydroxybenzocy-
clobutene-7-one (4.7), 4% 4.THF,19 and 6% 4.3 as deter-
mined by 1H NMR spectroscopy. Purification by preparative
TLC on SiO2 with CH2Cl2 gave 4.7 as a white solid (81%
isolated yield). 1H NMR (500 MHz, CDCl3, 25 °C): δ 7.13-7.3
(m, 40H, C6H5), 7.10 (t, 1H, guest-H) 7.05 (br s, 8H, ArH of
bowl), 6.63 (d, 1H, guest-H), 5.68 (d, 8H, OCH2O outer), 5.10
(d, 1H, guest-H), 4.91 (t, 8H, CH methine), 4.24 (d, 8H, OCH2O
inner), 4.08 (br s, 8H, OCH2CH2), 4.04 (t, 1H, guest-H), 3.99
Meth yllith iu m Ad d ition to 4.3 a t -78 °C. Application
of procedure D to hemicarceplex 4.3 (THF/ether (8:5), -78
°C, 30 h) and purification of the crude products by preparative
TLC on SiO2 with CH2Cl2 gave 65% 4.8-hydroxy-8-methyl-
benzocyclobutene-7-one (4.13) and 25% 4.9-methylphthalide
(4.14). The following are the data for 4.13. 1H NMR (360
MHz, CDCl3, 25 °C): δ 7.42 (d, 1H, guest-H), 7.16-7.25 (m,
40H), 6.87 (s, 8H,), 6.8 (t, 1H, guest-H), 5.71 (d, 8H), 4.93 (d,
1H, guest-H), 4.82 (t, 8H), 4.0-4.7 (m, 24H), 4.00 (d, 1H, guest-
H), 2.86 (s, 1H, guest-OH), 2.47-2.75 (m, 32H), 1.87 (br s,
16H), -2.04 (s, 3H, guest-CH3). FT-IR (CHCl3, 25 °C, cm-1):
ν(CO) 1762. FAB-MS: m/z [M + H]+, 2397 (84); [M - 13 +
H]+, 2250 (100). Anal. Calcd for C153H144O26: C, 76.61; H, 6.05.
Found: C, 76.30; H, 6.17.
Meth yllith iu m Ad d ition to 4.3 a t 0 °C. Application of
procedure D to hemicarceplex 4.3 (0 °C, 8 h) and purification
J . Org. Chem, Vol. 68, No. 6, 2003 2087

Warmuth et al.
of the reaction product by preparative TLC on SiO2 with
CH2Cl2 as the mobile phase gave 65% hemicarcerand 19
(X ) H).
structure determined at 25 °C), P21/c, θmax ) 50°, Cu KR radi-
ation. The host is centrosymmetric; the noncentrosymmetric
guests are disordered about the center of symmetry in the host
cavity. Two of the four bridges (linkers) that join the host bowls
are also disordered. Each guest was modeled as a rigid group
or groups. One molecule of nitrobenzene is located in each of
the cavities formed by the four -CH2CH2C6H5 “feet”. The
present structures belong to an isomorphous series of ni-
trobenzene solvates of the same host with different guests.19
The Crystallographic Information Files (CIF) are available as
Supporting Information and have been deposited with the
Cambridge Crystallographic Data Center (CCDC 196141,
CCDC 196142, CCDC 196143).
Hemicarceplex 4.1‚2C6H5NO2: a ) 16.736(13) Å, b )
20.479(15) Å, c ) 20.138(14) Å, â ) 99.53(2)°, V ) 6809(2) Å3,
Z ) 2, 6870 unique reflections, 2461 > 2σ(I), GOF ) 1.730, R
) 0.20 (295 parameters). The guest was refined as two
independent rigid groups, nearly orthogonal to each other (89°)
at convergence.
n -Bu tyllith iu m Ad d ition to 4.3 a t -78 °C. Application
of procedure D to hemicarceplex 4.3 (hexane/THF (1:5), 0.4
M n-BuLi, 0.2 M TMEDA, -78 °C, 12 h) and purification of
the reaction product by preparative TLC on SiO2 with CH2Cl2
as the mobile phase gave 10% hemicarcerand 19 (X ) H), 10%
hemicarcerand 20 (X ) H), and 62% recovered 4.3. The
following are the data for 20 (X ) H). 1H NMR (400 MHz,
CDCl3, 25 °C): δ 7.15-7.30 (m, 40H), 6.89 (s, 1H), 6.896 (s,
2H), 6.85 (s, 1H), 6.84 (s, 2H), 6.81 (s, 2H), 6.59 (s, 1H), 6.06
(d, 1H), 5.97 (d, 2H), 5.87 (d, 2H), 5.86 (d, 2H), 5.82 (m, 1H)
5.07-5.21 (m, 2H), 4.85 (t, 2 H), 4.83 (t, 4H), 4.81 (t, 2H), 4.28
(m, 2 H), 4.27 (d, 2 H), 4.23 (d, 2 H), 4.20 (d, 2 H), 4.16 (dd, 2
H), 3.8-4.0 (m, 12 H), 2.10-2.75 (m, 34H), 2.1-1.75 (m,
14H). FAB-MS: m/z [M + H]+, 2251 (100). Anal. Calcd for
C
144H138O24: C, 76.85; H, 6.09. Found: C, 76.36; H, 6.01.
Meth yllith iu m Ad d ition to 4.2 a t 22 °C. Application of
procedure D to hemicarceplex 4.2 (22 °C, 5 h) gave 25%
hemicarceplex 4.7-methylbenzocyclobuten-7-ol (4.22), 1%
hemicarceplex 4.23, 40% hemicarceplex 24.23, and 25%
Hemicarceplex 4.2‚2C6H5NO2: a ) 16.650(11) Å, b )
20.442(14) Å, c ) 20.043(13) Å, â ) 98.998(16)°, V ) 6738(8)
Å3, Z ) 2, 6721 unique reflections, 2965 > 2σ(I), GOF ) 1.668,
R ) 0.17 (334 parameters). Internal coordinates for the guest,
taken from the energy-minimized structure (Figure 12), are
similar to those for benzocyclobutenone 1-oxime.41
Hemicarceplex 4.3‚2C6H5NO2: a ) 16.720(7) Å, b ) 20.559-
(8) Å, c ) 20.168(8) Å, â ) 98.524(12)°, V ) 6856(5) Å3, Z ) 2,
6593 unique reflections, 2795 > 2σ(I), GOF ) 1.536, R ) 0.16
(300 parameters). Approximate internal coordinates for the
guest were taken from the structure of benzocyclobutenedi-
one.42
1
hemicarcerand 24 as determined from the H NMR spectrum
of the crude products. Purification of the reaction products was
achieved by preparative TLC on SiO2 with CH2Cl2 as the
1
mobile phase. The following are the data for 4.22. H NMR
(400 MHz, CDCl3, 25 °C): δ 7.16-7.25 (m, 40H,), 6.94 (s, 8H),
5.93 (d, 1H, guest), 5.64 (d, 8H), 5.39 (d, 1H, guest), 4.98 (t,
1H, guest), 4.83 (t, 8H), 4.53-4.48 (m, 2H, guest), 4.33 (d, 8H),
3.97 (s, 16H), 4.00 (d, 1H, guest-H), 2.47-2.75 (m, 32H), 1.9-
1.8 (m, 17H), 1.67 (s, 1H, guest), 0.90 (s, 3H, CH3). FT-IR
(CHCl3, 25 °C, cm-1): ν 3016 (m), 2947 (m), 2873 (w), 1603
(w), 1496 (w), 1474 (m), 1440 (s), 1374 (w), 1316 (m) 1224 (m),
1154 (m), 1106 (m) 1064 (m), 1018 (m) 988 (s). FAB-MS: m/z
[M + H]+, 2384 (38); [M - 19 + H]+, 2251 (100). Anal. Calcd
for C153H148O25: C, 76.99; H, 6.25. Found: C, 76.81; H, 5.98.
Ack n ow led gm en t. We thank Professor Kendall N.
Houk and Professor Harold W. Moore for helpful dis-
cussions and Dr. Rachel A. Watson-Clark for her as-
sistance with recording the 11B NMR spectrum. We
thank the U.S. Public Health Service for supporting
Grant GM-12640 (D.J .C., principal investigator). The
major part of this work was conducted at UCLA. R.W.
thanks the Alexander von Humboldt Foundation (Ger-
many) for a Feodor-Lynen fellowship and Kansas State
University for startup funds to continue this work at
Kansas State University.
1
The following are the data for 24. H NMR (360 MHz, CDCl3,
25 °C): δ 7.72 (br s, 2H, OH), 7.10-7.30 (m, 40H, C6H5), 6.97
(s, 2H, ArH on bowl), 6.88 (s, 2H, ArH on bowl), 6.85 (s, 2H,
ArH on bowl), 6.84 (s, 2H, ArH on bowl), 5.98 (d, 1H, OCH2O
outer), 5.92 (d, 3H, OCH2O outer), 5.83 (d, 2H, OCH2O outer),
5.71 (d, 1H, O OCH2O outer), 4.75-4.9 (t, 7 H, CH methine),
4.61 (t, 1H, CH methine), 4.53 (m, 2 H, OCH2CH2), 4.34 (d, 2
H, OCH2O inner), 4.14-4.28 (m, 7H, OCH2O inner, OCH2CH2),
3.75-4.07 (m, 12H, OCH2O inner OCH2CH2), 2.10-2.75 (m,
32H, CH2CH2Ph), 2.1-1.75 (m, 16H, OCH2CH2). FT-IR (CHCl3,
25 °C, cm-1): ν 3350 (v br) 3011 (m), 2948 (m), 2874 (m), 1602
(w), 1578 (w), 1496 (w) 1474 (s), 1441 (s), 1374 (m), 1317 (s),
1154 (m) 1106 (m), 1065 (m), 1018 (s), 988 (s). FAB-MS: m/z
[M + H]+, 2238 (100). Anal. Calcd for C143H136O24‚CH2Cl2: C,
74.43; H, 5.99. Found: C, 74.57; H, 5.83.
Meth yllith iu m Ad d ition to 4.NMP a t 0 °C. Application
of procedure D to hemicarceplex 4.NMP (THF/ether (1:1), 0
°C, 70 h) and purification of the reaction product by prepara-
tive TLC on SiO2 with CHCl3 as the mobile phase gave 70%
hemicarcerand 24.
X-r a y Diffr a ction . Crystal data common to all three struc-
tures: 4.guest‚2C6H5NO2 (crystallized from CHCl3/C6H5NO2,
Su p p or tin g In for m a tion Ava ila ble: 11B NMR spectrum
of 4.9, COSY spectrum of 4.7, 4.9, 4.10, 24.23, and 24,
TOCSY spectrum of 24.14, H/D exchange experiments of
24.23, and Crystallographic Information File (CIF) for 4.1,
4.2, and 4.3. This material is available free of charge via
the Internet at http://pubs.acs.org.
J O026649Z
(41) Viossat, B.; Rodier, N.; Andrieux, J .; Plat, M. Acta Crystallogr.
1986, C42, 824-825.
(42) Allen, F. H.; Trotter, J . J . Chem. Soc. B 1970, 916-920.
2088 J . Org. Chem., Vol. 68, No. 6, 2003