Long synthetic nanotubes from calix[4]arenes

Article abstract of DOI:10.1002/chem.200601545

We report the synthesis and encapsulation properties of long (up to 5 nm) molecular nanotubes 1-4, which are based on calix[4]arenes and can be filled with multiple nitrosonium (NO⁺) ions upon reaction with NO ₂/N₂O_4<

Full text of DOI:10.1002/chem.200601545

DOI: 10.1002/chem.200601545
Long Synthetic Nanotubes from Calix[4]arenes
Voltaire G. Organo, Valentina Sgarlata, Farhood Firouzbakht, and
Dmitry M. Rudkevich*[a]
Abstract: We report the synthesis and
encapsulation properties of long (up to
5 nm) molecular nanotubes 1–4, which
were studied by UV/Vis, FTIR, and
1H NMR spectroscopy in solution, and
also by molecular modeling. Entrap-
The FTIR andtitration data revealed
+
enhancedbinidng of NO
in longer
tubes, which may be due to cooperativ-
ity. The described nanotubes may serve
as materials for storing andconverting
NOx andalso offer a promise to further
develop supramolecular chemistry of
molecular containers. These findings
also open wider perspectives towards
applications of synthetic nanotubes as
alternatives to carbon nanotubes.
are basedon calix[4]arenes andcan be
ment of NO+ in 1·
A
N
+
filledwith multiple nitrosonium (NO
)
reversible, and addition of [18]crown-6
quickly recovers starting tubes 1–4.
ions upon reaction with NO2/N2O4
gases. These are among the largest
nanoscale molecular containers pre-
paredto date andcan entrap up to five
guests. The structure andproperties of
Keywords: calixarenes · molecular
recognition · nanotubes · nitrogen
oxides · supramolecular chemistry
tubular complexes 1·
N
N
Introduction
multiple bonding. The other problem is to achieve stable en-
capsulation complexes andcontrol the behavior of guests
inside. Most of the synthetic nanotubes and channels known
to date are formed by self-assembly and thus stable only
under specific, rather mild conditions.[1,5] There have been
several breakthroughs in filling SWNTs, but it is not trivial
to identify and study the encapsulated species.[6]
Here we present a full report on synthesis andencapsula-
tion properties of long synthetic nanotubes.[7] They are cova-
lently built androbust andtheir length is controlledprecise-
ly andeasily through modular synthesis. These nanotubes
can interact with NOx gases, convert them, andthus be
easily filled. As a consequence, they form kinetically and
thermodynamically stable but reversible encapsulation com-
plexes. These complexes can be studied by conventional
spectroscopic techniques. With lengths of up to 5 nm and
with up to five guests entrapped, these nanotubes are the
largest synthetic molecular containers known to date. Taken
together, our synthetic nanotubes can serve as alternatives
to SWNTs for filling purposes andfurther applications.
A novel type of molecular container is quickly emerging,
namely, the synthetic nanotube.[1–3] In contrast to other mo-
lecular containers such as cavitands, (hemi)carcerands, and
self-assembling capsules, much developed over the last
decade,[4] nanotubes have different topology, are open at
both ends, and therefore feature different guest dynamics
upon encapsulation.
The inspiration comes from naturally occurring ion chan-
nels[5] and, on the technological side, single-walled carbon
nanotubes (SWNTs).[6] The major feature of nanochannels
andnanotubes is the ability to align multiple guest species
in one dimension, which is important for nanowiring, molec-
ular andion transport, andinformation flow. Other poten-
tial applications include using nanotubes as reaction vessels
andmolecular cylinders for storage. While synthesis offers a
variety of sizes andshapes, synthetic nanotubes are still
rare. One reason is the significant technical difficulty associ-
ated with building defined and long nanostructures through
Results and Discussion
[a] Dr. V. G. Organo, Dr. V. Sgarlata, F. Firouzbakht,
Prof. Dr. D. M. Rudkevich
Department of Chemistry & Biochemistry
The University of Texas at Arlington
Arlington, TX 76019-0065 (USA)
Fax : (+1)817-272-3808
Design and synthesis: For filling purposes, stable encapsula-
tion complexes are needed. However, with synthetic nano-
tubes such stability is rarely achieved due to the lack of ad-
ditional binding sites within their interiors.[1,2] The major
E-mail: rudkevich@uta.edu
4014
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 4014 – 4023

FULL PAPER
drawback of encapsulation complexes with gases, including
those formedby SWNTs, also lies in their relatively low
thermodynamic stability.[8] An alternative approach is based
on reversible chemical transformation of gases upon encap-
sulation. In this case, they produce reactive intermediates
with much higher affinities for the receptor molecules.
Higher stabilities of such host–guest complexes result in
better sensors andalso offer attractive opportunities for the
design of chemical reagents from gases, as well as conceptu-
ally novel materials for gases andfrom gases. This approach
has been successful for NOx. Kochi, Rathore, andco-work-
ers showedthat, when convertedto the cation raidcals,
simple calix[4]arenes can strongly bindNO gas with forma-
Two pairs of phenolic oxygen atoms, orientedin opposite di-
rections, provide diverse routes to enhance the tube length
modularly. For the connection, diethylene glycol bridges
were chosen, which not only provide relatively high confor-
mational rigidity of the tubular structure, but also seal the
walls by minimizing the gaps between the calixarene mod-
ules.
The synthesis of nanotubes 2–4 is basedon a straightfor-
ward, modular strategy which incorporates reliable William-
son-type alkylations and provides high yields (Schemes 1
tion of cationic calixarene–nitrosonium species.[9] In these,
+
the NO molecule is transformedinto nitrosonium (NO
cation, which is tightly encapsulatedinside the calixarene
)
cavity. We recently discovered that calix[4]arenes reversibly
+
interact with NO2/N2O4 andentrap highly reactive NO
cation within their p-electron-rich interiors.[10] NO+ is gener-
atedfrom N O4, which is known to disproportionate to NO+
2
À
NO3 . Only one NO+ ion was foundper cavity; very high
association constants (Kassoc @106 mÀ1) were determined. We
took advantage of this unique chemistry between calixar-
enes andNO gases for the design and filling of nanotubes
x
1–4.
Scheme 1. Synthesis of nanotubes 2–4.
and2). Trimeric tube 2 was synthesizedin 70% yieldby
coupling of tetratosylate 5[11] with two equivalents of diol
6[7b] in boiling THF with NaH as base. Tube 3, which con-
tains four calixarenes, was preparedby reaction of bis-calix-
arene diol 7 with ditosylate 8 in 64% yield(NaH, K CO3,
2
THF). Finally, reaction of two equivalents of diol 7 with tet-
ratosylate 5 under the same conditions afforded pentameric
nanotube 4 in a remarkable 82% yield. The successful use
of K2CO3 in addition to NaH in these last cases can possibly
be explainedby template effects. Some calixarene–crown
derivatives are known to strongly complex K+ ion.[3a]
Syntheses of precursors for tubes 3 and 4 are basedon
conventional calixarene transformations. 25,27-Bis[2-benzy-
loxy)ethyloxy]-26,28-dihydroxycalix[4]arene (9) was ob-
tainedby alkylation of the parent, commercially available
calix[4]arene with 2-(benzyloxy)ethanol p-toluenesulfonate
In the design of nanotubes 1–4, several calix[4]arenes are
rigidly connected from both sides of their rims with two
symmetrical bridges. This is possible for a 1,3-alternate con-
formation. Calix[4]arenes in a 1,3-alternate conformation
are more rigidthan other conformers andhave a cylindrical
inner tunnel, formedby two cofacial pairs of aromatic rings
orientedorthogonally along the cavity axis. According to a
number of X-ray studies, this tunnel is 6 Š in diameter.[7]
andK CO3 in hot MeCN in 71% yield. It was further alky-
2
latedwith 2-( tert-butyldimethylsiloxy)ethanol p-toluenesul-
fonate andCs CO3 in DMF with the formation of 1,3-alter-
2
nate calixarene 10 in 67% yield. Derivative 10 was then de-
silylatedwith acetyl chloride in MeOH, andthe resulting
Chem. Eur. J. 2007, 13, 4014 – 4023
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4015

D. M. Rudkevich et al.
SWNTs are still a challenge be-
cause of their poor solubility.
Monitoring trappedguests by
conventional spectroscopy in
synthetic nanotubes has also
been difficult. For example, the
first calixarene-derived nano-
tubes, designed as channels for
small metal ions, showedonly
weak complexation abilities, as
the calixarene tunnel did not
bindmetal cations (Ag +, K+,
Cs+).[2] The situation is differ-
ent when calixarenes andNO
x
gases are employed. As men-
tionedearlier, simple calix[4]ar-
enes reversibly interact with
NO2/N2O4 andentrap the reac-
tive NO+ cation within their p-
electron-rich interiors.[10] Stable
nitrosonium complexes, for ex-
ample, 14, were quantitatively
Scheme 2. Synthesis of calix[4]arene precursors 7–13.
diol 11 was coupledto ditosylate 12[12] in THF in the pres-
ence of NaH as base. This afforded tubular bis-calixarene 13
in 32% yield, which was subsequently debenzylated (H2,
Pd/C, AcOH, THF) to give 7 in 82% yield. This was
smoothly (86%) convertedto ditosylate 8 by using p-tolue-
nesulfonyl chloride, Et3N, and4-imd ethylaminopyrinde
(DMAP) in CH2Cl2.
isolated upon addition of a Lewis acid (SnCl4).
Our findings with nanotubes 1–4 are as follows:
1) Addition of an excess of NO2/N2O4 to nanotubes 1–4 in
(CHCl2)2 in the presence of SnCl4 or BF3·Et2O resulted
in quantitative formation of nitrosonium complexes
1·
U
N
Nanotubes 1–4 have an inner tunnel of 6 Š in diameter
andare 17, 26, 35, and45 Š long, respectively (Figure 1).
Tubes 3 and 4 have molecular weights of approximately 2.3
and2.8 kDa, which definitely qualifies them as nanostruc-
tures.
formedwhen nanotubes 1–4 were mixedwith commer-
cially available nitrosonium salt NO+SbF6 in (CHCl2)2.
À
Complexes 1·
A
G
1
Vis, FTIR, and H NMR spectroscopy. They possess typi-
Figure 1. MacroModel 7.1 (Amber* Force Field) representation of nano-
tubular structures 3 and 4, a side view. Hydrogen atoms removed for
clarity. The X-ray structures of shorter tubes 1 and 2 have been reporte-
d.[7a]
Encapsulation studies: For monitoring the entrapment pro-
cesses in SWNTs, a combination of different spectroscopic
[6]
andmicroscopic techniques has been applied.
Especially
Scheme 3. Nitrosonium complexes of calixarene-basednanotubes. Coun-
important are TEM andFTIR spectroscopy, as they allow
the internal location of molecules andtheir molecular vibra-
tions, respectively, to be studied. Solution studies with
+
terions omittedfor clarity. When NO 2/N2O4 is usedas a source of NO
,
À
multiple NO3 counterions are formed, which are situated outside the
tubes.
4016
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 4014 – 4023

Nanotubes
FULL PAPER
cal features of simpler calix[4]arene–NO+ species, de-
scribedearlier. [9,10]
2) Of particular importance is the characteristic deep
purple color. The broadcharge-transfer bands responsi-
ble for this are observedat lmax ꢀ550 nm in the absorp-
tion spectra of all these nanotubes. Charge transfer only
occurs when NO+ guests are tightly entrappedinside the
calixarene cavities.[9,10] Accordingly, the filling process
can be monitored visually. Upon stepwise addition of
À
1
NO2/N2O4 or NO+SbF6 in (CDCl2)2, the H NMR sig-
nals of empty tubes 1–4 andcomplexes 1·
(NO+)2–
4·
(NO+)5 can be seen separately andin slow exchange.
AHCTREUNG
A
A
This is typical for host–guest complexes with high ex-
change barriers (DGꢀ >15 kcalmolÀ1) and/or large asso-
ciation constants (Kassoc >106 mÀ1).[4a,b]
3) The presence of the guests andtheir location insied
nanotubes 1·
ventional NMR analysis (Figure 2). Structural groups in-
volvedin complexation were iedntifiedby
1H NMR,
COSY, andNOESY experiments. Chemical shifts of the
ArOCH2 andCH 2OCH2 protons, situatedbetween d=
A
G
AHCTREUNG
2.5 and3.75 ppm, andto a lesser extent the aromatic
protons are very sensitive to encapsulation. In addition
to charge transfer, strong cation–dipole interactions be-
tween the calixarene oxygen atoms andthe entrapped
NO+ take place. All three groups of signals for the
propyl ArOPr protons in 1·
N
N
edsignificantly downfield(
Ddꢀ1 ppm) relative to the
signals obtainedfrom empty tubes 1–4 (Figure 2). This
implies that two NO+ ions are locatedat the ends of the
nanotubes andoccupy the terminal calixarene compart-
ments. Downfieldshifts
( Dd>1 ppm) of the glycol
ArOCH2 andCH OCH2 protons, situated in the middle
2
of the tubular structures, were also observed. This indi-
cated that the middle calixarene(s) in the longer tubes
+
are filledwith NO as well.
4) According to molecular modeling, the NO+-fillednano-
Figure 2. Selectedportions of the 1H NMR spectra (500 MHz, (CDCl2)2,
295 K) of A) nanotube 1, B) fillednanotube 1·
D) fillednanotube 2·
(NO+)3, E) nanotube 3, F) fillednanotube 3·
G) nanotube 4, andH) fillednanotube 4·
(NO+)5. The residual solvent sig-
nals are markedwith filledcircles. Complexes 1·
(NO+)2–4·(NO+)5 were
(NO+)2, C) nanotube 2,
tubes 1·
N
N
A
N
structures with all-gauche conformations about the glycol
AHCTREUNG
À
À
C C bonds, whereas empty tubes 1–4 have all-anti C C
conformations, which was also evident from the X-ray
structures of shorter tubes 1 and 2.[7a] Experimental
proof came from FTIR studies in (CHCl2)2 andCCl . In
A
ACHTREUNG
preparedfrom empty tubes 1–4 andNO 2/N2O4 in the presence of SnCl4.
4 À
empty tubes 1–4 the bandcorresponding to the anti C C
conformation at n˜ =1335 cmÀ1 was observed. In com-
arene units closer.[14] The para aromatic CH protons of
plexes 1·
(NO+)2–4·
G
tubes 1·
N
N
new bandat n˜ =1355 cmÀ1 appeared, which is character-
shielded and are seen somewhat upfield, at d=6.2–
6.4 ppm (Figure 2). These signals can be usedas a char-
acteristic signature for complex formation.
À
istic for the gauche C C conformer (Figure 3). These ab-
sorptions are attributedto CH wagging.[13] One reasona-
2
ble explanation for such conformational change might be
participation of the basic glycol CH2OCH2 oxygen atoms
in NO+ complexation. To appear in close proximity and
thus contribute to dipole–cation interactions with entrap-
pedNO +, the glycol chains shouldadopt the more com-
5) Complexation of NO+ apparently does not influence the
symmetry of nanotubes 1–4 (at 295Æ5 K). The number
of the propyl OCH2, glycol CH2OCH2 andArOCH , and
2
1
aromatic H NMR signals (Figure 2) for 1·
(NO+)2–
4·
(NO+)5 does not change, and this implies that the NO+
A
À
pact gauche C C conformation (Figure 3, right). As a
consequence, the fillednanotubes are shorter. This
shrinkage brings the aromatic rings of neighboring calix-
guests, with van der Waals dimensions of about 2 Š,
À
freely rotate along the N O axis andalso tumble within
the cavity at room temperature.[15] Similar dynamics were
Chem. Eur. J. 2007, 13, 4014 – 4023
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4017

D. M. Rudkevich et al.
tubes
A
(NO+)2 and
G
N
filledwith the help of NO
/
2
N2O4.[7] From the integration, it
was possible to quantitatively
estimate the concentration of
complexes 1·
(NO+)2 and 2·
A
(NO+)3. Two equivalents of tert-
butyl nitrite were needed to fill
dimeric tube 1, and adding
three equivalents of the nitrite
completely filledtrimeric nano-
tube 2. Analogously, 1:4 stoichi-
ometry was establishedfor
longer tube 3 having four calix-
Figure 3. Portions of the IR spectra (in (CHCl2)2, 295 K) of A) empty tube 1 andB) filledtube 1·
(NO+)2. The
A
À1
À
bands corresponding to the anti and gauche C conformations are marked. The band at n˜ =1370 cm belongs
to tert-butyl nitrite. Right: proposedconformational changes in the nanotubes upon complexation.
+
previously noticedfor simpler calixarene–NO
com-
plexes such as 14.[9,10]
Stoichiometry: Thus far, we usedNO /N2O4 gases to fill
2
nanotubes 1–4 with multiple NO+ guest species andfol-
lowedthe process by conventional spectroscopic techniques
in solution. NO2 andN 2O4 are aggressive anddifficult to
handle in small, precise quantities. Moreover, they react
with nanotubes upon standing and nitrosate/nitrate the aro-
matic rings. All this complicates the stoichiometric studies.
Initially, by molecular modeling
andanalogy with simpler calix-
arene–NO+ complexes (see 14,
for example),[9,10] it was suggest-
edthat one calixarene fragment
in the nanotubes can accommo-
date only one NO+. According-
ly,
4·
(NO+)5 shouldhave two,
three, four, andfive NO
nanotubes
1·
(NO+)2–
A
A
A
+
ions,
respectively. Higher stoichiometries of NO+ were ruledout:
there is simply no room to accept a larger number of mutu-
ally repulsive cations.
+
Recently, we foundmore reliable sources of NO for de-
termination of stoichiometry: alkyl nitrites (RON=O).[16]
Scheme 4. Filling synthetic nanotubes through supramolecular nitrosoni-
Alkyl nitrites are known as effective NO donors in medicine
um transfer with tert-butyl nitrite andSnCl .
4
[17]
andthey also act as nitrosating reagents.
We now report
that NO+ can be quantitatively transferredfrom alkyl ni-
trites into calixarene nanotubes 1–4. Alkyl nitrites are much
easier to handle than NO2/N2O4 gases. They are stable, non-
volatile liquids that are easy to transfer by conventional lab-
oratory pipettes.
Preliminary experiments with simple calix[4]arenes and
shorter tubes were recently published.[16] In short, addition
of tert-butyl nitrite to solutions of tetrakis(O-n-hexyloxy)ca-
lix[4]arene in CDCl3 or nanotubes 1 and 2 in (CDCl2)2 in
the presence of an excess of SnCl4 or trifluoroacetic acid
(TFA) ledto rapi,d quantitative formation of the corre-
sponding calixarene–NO+ complexes. These were identified
by means of NMR, FTIR, andUV/Vis spectra, which
showedfeatures similar to those of the previously described
arene units (Scheme 4). Further addition of the nitrite did
not change the NMR spectra. This confirms the stoichiome-
try of the nanotube complexes, which thus have one NO+
per calixarene unit. A similar trendis expectedfor longer
tube 4, although the titrations in this case are less accurate
due to broadening of signals and rather low solubility of the
complex.
Transfer of NO+ does not occur in the absence of SnCl4.
Most probably, the Lewis acidinteracts with the nitrite
À
CON oxygen atom andfacilitates breaking the O N bond
(Scheme 4, top). It may also stabilize the complexes by coor-
dinating to the counterion, similar to known arene nitrosoni-
um nitrate complexes.[18]
4018
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 4014 – 4023

On the binding strength: We noticedthat complexes
n˜
tube 2·
while in longer tubes 3·
(NO+)=1940 cmÀ1 dominates. This band was assigned to the
G
Nanotubes
FULL PAPER
À1
A
G
N
less than stoichiometric amounts of tert-butyl nitrite are
added. For example, addition of one equivalent of tert-butyl
nitrite to dimeric tube 1 produces 50Æ5% of fully occupied
AHCTREUNG
NO+ guest(s) situated in the middle of the tubes. Apparent-
ly, they are somewhat more strongly boundto the nanotube
walls. One possible explanation is participation of the glycol
CH2OCH2 oxygen atoms (Figure 3). As discussed earlier for
the anti–gauche conformational transition upon complexa-
tion, these oxygen atoms are in close proximity andthus
contribute to dipole–cation interactions with the entrapped
NO+. Interestingly, in calixarene–NO+ complex 14b, which
models the middle part of the filled tubes, the NO+ band
complex 1·
(NO+)2, and addition of two equivalents of tert-
E
butyl nitrite to trimeric tube 2 produces 55Æ5% of fully oc-
cupiedcomplex 2·
ACHTREUNG
simple statistical distribution. Furthermore, partially filled
complexes were not detected. This suggests cooperativity.
With the complexation of the first NO+ guest(s), the nano-
tube structure becomes more rigidandpreorganizeddue to
conformational changes of the glycol linkers from anti to
gauche (see Figure 3). In such conformational transition, the
glycol oxygen atoms become more preorganizedfor further
only appearedat n˜
ACHTREUNG
rently, the tubular structures in 2·
A
ACHTREUNG
+
interactions with the entrappedNO species.[19,20]
more rigidity and enhance complexation. Possible coopera-
tivity through allosteric effects is currently under further in-
vestigation. It may be the result of multiple guests aligning
in one dimension, which brings an order that cannot be ach-
ievedfor shorter complexes.
Vibrational spectra allowed us to obtain independent,
unique information on the bonding of multiple NO+ species
inside tubes 1·
C
G
À
À
À
show a single stretching bandat n˜
(NO+)=2270 cmÀ1 in
CH3NO2 solution,[21] andthe stretching frequency of neutral
diatomic NO gas is n˜(NO)=1876 cmÀ1.[21] In calixarene–
NO+ complexes 14, the NO+ bandsignificantly shifted
Guest exchange and dynamics: Fillednanotubes 1·
A
4·
AHCTREUNG
hours, but readily dissociate upon addition of H2O or
MeOH, quantitatively producing free 1–4. The process, how-
ever, is not reversible: the releasedNO are now converted
(Dn=312 cmÀ1) to lower energies comparedto free NO
+
+
ion andappearedat
(Figure 4). This is due to strong electron donor–acceptor in-
to nitrous acidandcomplexes 1·
regenerated.
We further foundthat [18]crown-6 can remove the encap-
A
G
+
teractions between encapsulatedNO
andthe p-electron-
rich aromatic walls of the calixarene.[9] Dimeric complex
+
+
A
(NO+)2 also exhibitedsimilar shifts for the NO guests at
sulatedNO
species (Scheme 5). It is known that crown
R
ethers form stable complexes with NO+.[22] When about
four equivalents of [18]crown-6 were added to solutions of
A
N
ACHTREUNG
1·
A
G
N
(CDCl2)2, empty nanotubes 1
and 2, respectively, regenerated
within minutes, andthe edep
purple color disappeared. The
process can be followedby
1H NMR
spectroscopy
(Figure 5). Interestingly, further
addition of SnCl4 to the same
solutions fully restores com-
plexes 1·
A
G
N
That SnCl4 forms complexes
with crown ethers is well docu-
mented.[23] In our hands, the ad-
dition of four equivalents of
SnCl4 to a solution containing
only [18]crown-6 (d=3.6 ppm)
in (CDCl2)2 resultedin a new
[18]crown-6 singlet at d=
3.7 ppm. This indicates strong
interaction between SnCl4 and
[18]crown-6 in solution.
Apparently in our case, an
excess of SnCl4 displaces NO+
from the crown ether moiety,
Figure 4. Selectedportions of the IR spectra (in (CHCl 2)2, 295 K) of calixarene complex 14a andcompletely
fillednanotubes 1·
(NO+)2–4·(NO+)5. Calixarene complex 14b looks similar to 14a. For the filling experiments,
A
ACHTREUNG
tert-butyl nitrite andSnCl were used.
4
Chem. Eur. J. 2007, 13, 4014 – 4023
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4019

D. M. Rudkevich et al.
tween the calixarene mod-
ules.[24] Approach andexit
through the ends appears to be
less hindered. The middle gates
between the calixarene units
become narrower due to the
conformational change of the
glycol moieties from anti to
gauche upon complexation (see
Figure 3). The encapsulated
NO+ shouldalso avoidelectro-
static repulsions with each
other. Most probably, tube fill-
ing andrelease occurs through
the guest tunneling along the
interior, which may be common
to all nanotubes.
Conclusions and Outlook
Long synthetic nanotubes are
now readily available that can
be usedfor filling purposes.
The nanotubes reportedhere
are basedon calix[4]arenes and
Scheme 5. Nitrosonium exchange in calixarene nanotubes involving [18]crown-6 andSnCl .
4
take advantage of their extremely diverse, high-yield
chemistry. These nanotubes reversibly interact with NOx
gases andthus can be filledwith multiple nitrosonium
guests.[25] The major feature of these tubes is the ability to
align multiple guest species in one dimension. This brings an
order that cannot be achieved for simpler and shorter com-
plexes. It may also influence guest exchange into andout of
the nanotubes, as well as the binding strength. Among possi-
ble applications are the design of synthetic nanowires and
optical sensors for NOx. Chemical fixation of NOx is also of
great interest.[26] The tubes can be usedfor molecular stor-
age of active nitrosonium ions andact as size- andshape-se-
lective nitrosating reagents.[27] Generation of NO gas inside
[28]
these nanotubes andits release is also possible.
currently working in these directions.
We are
Finally, in contrast to other molecular containers,[4] supra-
molecular chemistry of synthetic nanotubes is still not ex-
plored. Their unique geometrical features and nanosize,
which allow for the simultaneous entrapment of multiple
guests in a one-dimensional fashion, places them in a unique
position to uncover novel phenomena in molecular encapsu-
lation.[29,30]
Figure 5. Nitrosonium exchange. Selectedportions of the 1H NMR spec-
tra (500 MHz, (CDCl2)2, 295 K) of A) nanotube 1 andB) fillednanotube
1·
(NO+)2. C) Same as B) with 1 equiv of [18]crown-6. D) Same as B)
G
with 2 equiv of [18]crown-6. E) same as D) with 2 equiv of SnCl4.
F) Same as D) with 4 equiv of SnCl4.
andthe latter goes back to the calixarene units of the nano-
tubes. This observation is important, since in this case for-
eign species can be replacedandreturnedwithout decompo-
sition or changing the polarity of the solution.
The guest-exchange mechanism is currently under investi-
gation. Modeling suggests that NO+ can enter andleave the
nanotube through either its ends or the middle gates be-
Experimental Section
General: Melting points were determined on a Mel-Temp apparatus
(Laboratory Devices, Inc.) andare uncorrected. 1H, 13C NMR, COSY,
andNOESY spectra were recoreddat 295
Æ18C on JEOL 300 and
500 MHz spectrometers. Chemical shifts were measuredrelative to resid-
ual undeuterated solvent resonances. FTIR spectra were recorded on a
Bruker Vector 22 FTIR spectrometer. UV/Vis spectra were measuredon
4020
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 4014 – 4023

Nanotubes
FULL PAPER
a Varian Cary-50 spectrophotometer. Mass spectra were recorded at the
Scripps Center for Mass Spectrometry (La Jolla, CA). High-resolution
MALDI FT mass spectra were obtainedon an IonSpec Ultima FTMS.
MALDI TOF mass spectra were obtainedon an AppliedBiosytems Voy-
ager STR (2). Elemental analysis was performedon a Perkin–Elmer 2400
CHN analyzer. All experiments with moisture- and/or air-sensitive com-
pounds were run under a dried nitrogen atmosphere. For column chro-
matography, silica gel 60 Š (Sorbent Technologies, 200–425 mesh) was
used. Parent tetrahydroxycalix[4]arene[31] andcalixarene 14b[32] were pre-
pared according to the published procedures. NO2/N2O4 was generated
from copper andconcentratedHNO 3. Molecular modeling was per-
formed using commercial MacroModel 7.1 with Amber* Force Field.[33]
4.38 (s, 4H; ArCH2O), 3.85 (AB q, J=16.9 Hz, 8H; ArCH2Ar), 3.6 (m,
8H; OCH2CH2O), 3.30 (t, J=6.0 Hz, 4H; OCH2CH2O), 2.91 (t, J=
6.0 Hz, 4H; OCH2CH2O), 1.68 ppm (brs, 2H; OH); 13C NMR (CDCl3):
d=156.3, 156.1, 138.4, 133.8, 133.7, 129.5, 129.4, 128.5, 127.9, 127.8, 127.6,
123.2, 123.0, 73.2, 71.4, 68.1, 61.3, 38.0 ppm; FTIR (CHCl3): n˜ =3529,
3421, 3066, 3033, 2927, 2872, 1468, 1365, 1324, 1247, 1215, 1092,
1037 cmÀ1; elemental analysis calcd(%) for C 50H52O8: C 76.90, H 6.71;
found: C 76.82, H 6.66.
Bis(1-propyloxy)bis[2-benzyloxy(ethyloxy)]calix[4]tube (13): Calix[4]ar-
N
ene 11 (0.12 g, 0.153 mmol) was dissolved in dry THF (40 mL), and NaH
(60% suspension in mineral oil, 0.61 g, 1.5 mmol) was added. The mix-
ture was stirredat 45 8C for 1 h, after which a solution of calix[4]arene
12[12] (0.14 g, 0.15 mmol) in THF (20 mL) was added slowly over 4 h. The
reaction mixture was refluxedfor 24 h. After cooling, the solvent was
evaporated under reduced pressure and CH2Cl2 (30 mL) and10% aque-
ous HCl (10 mL) were added. The organic layer was washed with 10%
aqueous NaHCO3 (10 mL) anda saturatedsolution of NaCl (2”10 mL),
dried over MgSO4, andevaporatedin vacuo. The residue was purifiedby
column chromatography with CH2Cl2/MeOH (99/1, Rf =0.3) as eluent to
Caution: NO2 has an irritating odor and is very toxic!
25,27-Bis[2-benzyloxy)ethyloxy]-26,28-dihydroxycalix[4]arene (9): A sus-
pension of parent calix[4]arene (1.0 g, 2.4 mmol), 2-(benzyloxy)ethanol p-
toluenesulfonate (1.51 g, 4.94 mmol), andK 2CO3 (0.67 g, 4.8 mmol) in
MeCN (20 mL) was refluxedfor 24 h. After cooling to RT, the solvent
was removed under reduced pressure, and the residue was treated with
10% aqueous HCl (20 mL) andCH 2Cl2 (40 mL). The organic layer was
afford 13 as
a colorless solid. Yield: 0.064 g (32%); m.p. >2708C
separatedandwashedwith 10% aqueous NaHCO
(2”15 mL) and
3
(decomp); 1H NMR (500 MHz, CDCl3): d=7.3 (m, 10H; ArH), 7.17,
7.16 (2”d, J=7.5 Hz, 8H; ArH), 7.02 (m, 12H; ArH), 6.85, 6.72 (2”t,
J=7.5 Hz, 4H; ArH), 4.31 (s, 4H; ArCH2O), 3.90 (AB q, J=16.5 Hz,
8H; ArCH2Ar), 3.88 (AB q, J=16.5 Hz, 8H; ArCH2Ar), 3.56 (m, 8H;
OCH2CH2O), 3.48, 3.31, 2.92 (3”t, J=6.5 Hz, 12H; OCH2CH2O, OCH2),
2.57 (m, 8H; OCH2CH2O), 1.06 (m, 4H; CH2), 0.56 ppm (t, J=7.3 Hz,
6H; CH3); 13C NMR (CDCl3): d=157.3, 156.6, 155.9, 155.8, 134.2, 134.1,
134.0, 129.1, 129.0, 128.9, 128.8, 128.5, 127.7, 127.6, 122.9, 122.8, 122.5,
122.0, 73.1, 71.5, 69.1, 68.5, 68.2, 66.1, 66.0, 38.3, 38.1, 22.7, 10.2 ppm;
FTIR (CHCl3): n=3065, 3033, 2925, 2875, 1462, 1247, 1215, 1124, 1094,
1034, 1009 cmÀ1; MALDI-FTMS: m/z: 1363.6439 [M+Na]+; calcdfor
C88H92O12Na: 1363.6486.
brine. The organic solution was dried over MgSO4, filtered, and evaporat-
ed in vacuo. The resulting pale yellow oil solidified upon addition of
MeOH (3”15 mL) to give 9 as a white powder. Yield: 1.158 g (71%);
m.p. 138–1408C; 1H NMR (500 MHz, CDCl3): d=7.87 (s, 2H; ArOH),
7.33 (m, 10H; ArH), 7.06, 6.89 (2”d, J=7.5 Hz, 8H; ArHm), 6.72, 6.65
(2”t, J=7.5 Hz, 4H; ArHp), 4.65 (s, 4H; ArCH2O), 4.45, 3.37 (2”d,
J=12.8 Hz, 8H; ArCH2Ar), 4.18 (t, J=5.1 Hz, 4H; OCH2CH2O),
3.92 ppm (t, J=5.1 Hz, 4H; OCH2CH2O); 13C NMR (CDCl3): d=153.4,
152.0, 138.2, 133.4, 129.0, 128.6, 128.5, 128.2, 128.0, 127.7, 125.4, 118.9,
75.7, 73.7, 69.2, 31.4 ppm; FTIR (KBr): n=3333, 3062, 3028, 2928, 2861,
1467, 1453, 1355, 1250, 1200, 1123, 1090, 1045 cmÀ1; elemental analysis
calcd(%) for C 46H46O6: C 79.51, H 6.67; found: C 79.23, H 6.41.
Bis(1-propyloxy)bis[2-hydroxy(ethyloxy)]calix[4]tube (7): A mixture of
A
25,27-Bis[2-benzyloxy)ethyloxy]-26,28-bis[2-(tert-butyldimethylsiloxy)e-
thyloxy]calix[4]arene (1,3-alternate conformer 10): A suspension of cal-
ix[4]arene 9 (1.15 g, 1.65 mmol) andCs 2CO3 (2.7 g, 8.3 mmol) in DMF
(30 mL) was heatedat 90 8C for 1 h, then a solution of 2-(tert-butyldime-
thylsiloxy)ethanol p-toluenesulfonate (2.74 g, 8.29 mmol) in DMF
(10 mL) was added. The reaction mixture was stirred at 908C for 24 h.
DMF was completely removedin vacuo, andthe reaction mixture was
treatedwith 10% aqueous acetic acid(50 mL) andCH 2Cl2 (50 mL). The
organic layer was separated, washed with 10% aqueous NaHCO3
10 wt% Pd/C (0.03 g), the above-described benzylated calix[4]tube 13
(0.06 g, 0.044 mmol), andAcOH (0.1 mL) in THF (5 mL) was stirred
under H2 (1 atm) at RT until all starting material disappeared (TLC,
CH2Cl2/MeOH, 99/1). The suspension was filteredthrough Celite andthe
clear solution was concentratedin vacuo. The residue was trituratedwith
MeOH (3”2 mL) to afford 7 as a white powder. Yield: 0.042 g (82%);
1
m.p.>3008C (decomp); H NMR (500 MHz, CDCl3): d=7.21, 7.15 (2”d,
J=7.6 Hz, 8H; ArH), 7.05 (m, 12H; ArH), 6.95, 6.86 (2”t, J=7.6 Hz,
4H; ArH), 3.95 (AB q, J=17.2 Hz, 8H; ArCH2Ar), 3.88 (AB q,
J=16.5 Hz, 8H; ArCH2Ar), 3.63 (t, J=7.0 Hz, 4H; OCH2CH2O), 3.56
(m, 8H; OCH2CH2O), 3.31 (t, J=7.0 Hz, 4H; OCH2), 3.21(m, 4H;
OCH2CH2O), 2.56 (m, 8H; OCH2CH2O), 2.27 (brs, 2H; OH), 1.06 (m,
4H; CH2), 0.56 ppm (t, J=7.6 Hz, 6H; CH3); 13C NMR (CDCl3):
d=157.3, 156.4, 156.0, 155.8, 134.2, 134.0, 133.9, 133.5, 129.2, 129.1, 128.9,
128.8, 123.6, 123.0, 122.8, 121.9 ppm; FTIR (KBr): n=3529, 3453, 3060,
3030, 3011, 2924, 2875, 1459, 1216, 1132, 1094, 1035, 1011 cmÀ1; MALDI-
FTMS: m/z 1183.5426 [M+Na]+; calcdfor C 74H80O12Na: 1183.5547.
(15 mL) andbrine, (2”15 mL), anddriedover MgSO
4. Evaporation of
CH2Cl2 gave a light yellow solid, which was refluxed with KI (1 g) and
Et3N (1 mL) in MeCN (30 mL) for 1 h. The solvent was evaporatedin
vacuo, and the residue solidified upon addition of MeOH (3”15 mL).
The resulting brownish yellow solidwas then refluxedwith MeOH
(25 mL), filteredhot, andwashedwith MeOH (3”15 mL) to give 10 as a
1
white powder. Yield: 1.12 g (67%); m.p. 116–1208C; H NMR (500 MHz,
CDCl3): d=7.36 (m, 10H; ArH), 7.07, 7.04 (2”d, J=7.3 Hz, 8H; ArHm),
6.62, 6.56 (2”t, J=7.3 Hz, 4H; ArHp), 4.66 (s, 4H; ArCH2O), 3.87 (m,
8H; OCH2CH2O), 3.77, 3.71 (2”m, 8H; OCH2CH2O), 3.55 (m, 8H;
ArCH2Ar), 0.96 (s, 18H; tBu), 0.15 ppm (s, 12H; CH3); 13C NMR
(CDCl3): d=155.8, 155.6, 138.4, 133.6, 130.0, 129.8, 128.5, 127.8, 121.9,
74.0, 73.4, 71.6, 69.7, 62.7, 34.9, 26.2, 18.5, À5.1 ppm; FTIR (CHCl3):
n˜ =3068, 3032, 2931, 2714, 1452, 1361, 1250, 1198, 1098, 1029 cmÀ1; ele-
mental analysis calcd(%) for C 62H80O8Si2: C 73.77, H 7.99; found: C
73.52, H 7.93.
Bis(1-propoxy)bis[2-(p-toluenesulfonyloxy)ethoxy]calix[4]tube (8): cal-
ix[4]tube 7 (0.11 g, 0.095 mmol), DMAP (0.046 g, 0.38 mmol), and p-tol-
uenesulfonyl chloride (0.070 g, 0.38 mmol) were dissolved in CH2Cl2
(15 mL) andcooledto À58C. Triethylamine (0.20 mL) was added, and
the solution was stirredovernight at RT. The solution was evaporatedto
dryness, the residue redissolved in CH2Cl2, andthe resulting solution
washedwith 5% HCl, water, andbrine. The organic layer was driedwith
Na2SO4 andevaporatedto dryness. Treatment with MeOH produced
a
8 as
25,27-Bis[2-benzyloxy)ethyloxy]-26,28-bis[2-(hydroxy)ethyloxy]calix[4]ar-
G
colorless solid. Yield: 0.12 g (86%); m.p. 242–2438C; 1H NMR
ene (1,3-alternate conformer 11): Acetyl chloride (0.51 g, 0.46 mL,
6.5 mmol) was added dropwise to ice-cold CH2Cl2/MeOH (10 mL,
9/1 v/v), andafter 30 min a solution of calix[4]arene 10 (1.1 g, 1.1 mmol)
in CH2Cl2 (5 mL) was added. After the starting material was consumed
(TLC, hexane/AcOEt, 7/3, ca. 1 h), the reaction mixture was washedwith
(500 MHz, CDCl3): d=7.79 (d, J=8.4 Hz, 4H; ArHtosyl), 7.40 (d,
J=8.4 Hz, 4H; ArHtosyl), 7.17, 7.15 (2”d, J=1.7 Hz, 8H; ArH), 7.06–6.97
(t+d+t, 8H; ArH), 6.94 (d, J=7.3 Hz, 4H; ArH), 6.86 (t, J=7.3 Hz,
2H; ArHp), 6.69 (t, J=7.3 Hz, 2H; ArHp), 3.89–3.80 (m, 16H;
ArCH2Ar), 3.57 (m, 12H; OCH2CH2O), 3.39 (m, 4H; OCH2CH2O), 3.32
(t, J=7.3 Hz, 4H; OCH2), 2.57 (m, 8H; OCH2CH2O), 2.48 (s, 6H;
ArCH3), 1.10 (m, 4H; CH2), 0.58 ppm (t, J=7.3 Hz, 6H; CH3); 13C
(CDCl3): d=157.3, 155.9, 155.8, 155.6, 145.0, 134.2, 134.0, 133.9, 133.8,
133.2, 130.0, 129.1, 129.0, 128.8, 128.0, 123.3, 122.9, 122.8, 121.9, 71.5,
10% aqueous NaHCO3 (5 mL) andbrine (5 mL) anddriedover MgSO
.
4
The residue was purified by column chromatography (CH2Cl2/MeOH,
96/4, Rf =0.2) to give 11 as a colorless solid. Yield: 0.74 g (87%); m.p.
174–1768C; 1H NMR (500 MHz, CDCl3): d=7.33 (m, 10H; ArH), 7.07,
7.00 (2”d, J=7.3 Hz, 8H; ArHm), 6.91, 6.68 (2”t, J=7.3 Hz, 4H; ArHp),
Chem. Eur. J. 2007, 13, 4014 – 4023
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4021

D. M. Rudkevich et al.
69.1, 68.9, 67.8, 66.8, 66.2, 66.0, 38.3, 38.0, 22.7, 21.7, 10.2 ppm; FTIR
(KBr): n˜ =3061, 3031, 3014, 2959, 2918, 2874, 1459, 1214, 1177, 1095,
1033, 1011 cmÀ1; elemental analysis calcd(%) for C 88H96O16S2: C 71.71,
H 6.57; found: C 71.78, H 6.46.
containing a solution of calix[4]tube 3 (ca. 4 mm) andSnCl or TFA
4
(8 equiv) in (CDCl2)2, andafter homogenization the spectrum was re-
corded. Additional aliquots of the nitrite were added and the spectrum
was recorded after each addition. The tert-butyl nitrite concentration
rangedbetween 2 and25 m m. The concentration of complex 3·
(NO+)4
G
Trimeric tube 2: A solution of calix[4]arenes 5[11] (100 mg, 0.08 mmol)
and 6[7b] (98 mg, 0.16 mmol) in THF (25 mL) was added dropwise to a
suspension of NaH (60% in mineral oil, 26 mg, 1.3 mmol) in THF
(150 mL) at 708C over 8 h. The mixture was further refluxedfor 4 d,
evaporated to dryness, and the residue suspended in CH2Cl2 (50 mL) and
neutralizedat 0 8C with 5% aqueous HCl (50 mL). The organic layer was
washedwith water (3”5 mL), driedover Na 2SO4, andthen evaporatedto
dryness. The residue was treated with MeCN to produce nanotube 2 as a
white solid. Yield: 0.10 g (70%); m.p. >3008C (decomp); 1H NMR
(500 MHz, CDCl3): d=7.20, 7.18 (2”d, J=7.3 Hz, 16H; ArHm), 7.08 (t,
J=7.3 Hz, 4H; ArHp), 7.02 (d, J=7.3 Hz, 8H; ArHm), 6.98, 6.85 (2”t,
J=7.3 Hz, 8H; ArHp), 3.94 (s, 8H; ArCH2Ar), 3.88 (AB q, J=16.5 Hz,
16H; ArCH2Ar), 3.57 (m, 16H; ArOCH2CH2O), 3.30 (t, J=7.3 Hz, 8H;
OCH2), 2.55 (m, 16H; ArOCH2CH2O), 1.06 (m, 8H; CH2), 0.56 ppm (t,
J=7.3 Hz, 12H; CH3); 13C NMR (C2D2Cl4): d=157.2, 156.2, 155.9, 134.2,
134.1, 134.0, 129.3, 129.1, 128.8, 122.9, 122.7, 122.0, 71.8, 70.9, 69.2, 66.7,
66.5, 38.3, 38.1, 22.8, 10.4 ppm; FTIR (CCl4): n˜ =3064, 3033, 3016, 2959,
2921, 2873, 1459, 1216, 1125, 1093, 1035, 1010 cmÀ1; MALDI-FTMS: m/z:
1743.8443 [M+Na]+; calcdfor C 112H120O16Na: 1743.8473. This modified
procedure reproducibly gave much better yields than the previously pub-
was determined by integration of the aromatic CH protons and/or the
OCH2 methylene protons versus the corresponding signals of free tube 3.
The experiments were performedat least in duplicate.
Acknowledgements
This research was supportedby the National Science Foundation (CHE-
0350958), the Texas Advanced Technology Program (003656-0146-2003),
andthe AlfredP. Sloan Foundation. We are also grateful to the NSF for
funding the purchase of the 300 MHz NMR spectrometer (CHE-
0234811). Finally, we thank Dr. Christopher Matranga (National Energy
Technology Laboratory, US DOE), Dr. Alex V. Leontiev, Dr. Vaclav
Stastny, andAlbert Wong for valuable experimental advice.
[1] Self-assembling nanotubes: a) D. T. Bong, T. D. Clark, J. R. Granja,
M. R. Ghadiri, Angew. Chem. 2001, 113, 1016–1041; Angew. Chem.
Int. Ed. 2001, 40, 988–1011; b) S. J. Dalgarno, G. W. V. Cave, J. L.
Atwood, Angew. Chem. 2006, 118, 584–588; Angew. Chem. Int. Ed.
2006, 45, 570–574; c) T. Yamaguchi, S. Tashiro, M. Tominaga, M.
Kawano, T. Ozeki, M. Fujita, J. Am. Chem. Soc. 2004, 126, 10818–
10819; d) S. Tashiro, M. Tominaga, T. Kusukawa, M. Kawano, S. Sa-
kamoto, K. Yamaguchi, M. Fujita, Angew. Chem. 2003, 115, 3389–
3392; Angew. Chem. Int. Ed. 2003, 42, 3267–3270; e) M. Tominaga,
S. Tashiro, M. Aoyagi, M. Fujita, Chem. Commun. 2002, 2038–2039;
f) V. Sidorov, F. W. Kotch, G. Abdrakhmanova, R. Mizani, J. C. Fet-
tinger, J. T. Davis, J. Am. Chem. Soc. 2002, 124, 2267–2278; g) L.
Baldini, F. Sansone, A. Casnati, F. Ugozzoli, R. Ungaro, J. Supra-
mol. Chem. 2002, 219–226.
[2] Calixarene nanotubes: a) A. Ikeda, S. Shinkai, J. Chem. Soc. Chem.
Commun. 1994, 2375–2376; b) A. Ikeda, M. Kawaguchi, S. Shinkai,
An. Quim. Int. Ed. 1997, 93, 408–414; c) S. K. Kim, W. Sim, J.
Vicens, J. S. Kim, Tetrahedron Lett. 2003, 44, 805–809; d) S. K. Kim,
J. Vicens, K.-M. Park, S. S. Lee, J. S. Kim, Tetrahedron Lett. 2003, 44,
993–997; e) S. K. Kim, J. K. Lee, S. H. Lee, M. S. Lim, S. W. Lee, W.
Sim, J. S. Kim, J. Org. Chem. 2004, 69, 2877–2880; f) J.-A. Perez-
Adelmar, H. Abraham, C. Sanchez, K. Rissanen, P. Prados, J. de
Mendoza, Angew. Chem. 1996, 108, 1088–1090; Angew. Chem. Int.
Ed. Engl. 1996, 35, 1009–1011; g) S. Le Gac, X. Zeng, O. Reinaud,
I. Jabin, J. Org. Chem. 2005, 70, 1204–1210; other covalently linked
nanotubes: h) A. Harada, J. Li, M. Kamachi, Nature 1993, 364, 516–
518; this was the first covalently linkedsynthetic nanotube; the tub-
ular polymer was preparedby cross-linking of polyrotaxane-thread-
ed a-cyclodextrins; i) Y. Kim, M. F. Mayer, S. C. Zimmerman,
Angew. Chem. 2003, 115, 1153–1158; Angew. Chem. Int. Ed. 2003,
42, 1121–1126; j) T. Iwanaga, R. Nakamoto, M. Yasutake, H. Take-
mura, K. Sako, T. Shinmyozu, Angew. Chem. 2006, 118, 3725–3729;
Angew. Chem. Int. Ed. 2006, 45, 3643–3647.
[7a]
lishedprotocol.
Tetrameric nanotube 3: A solution of calix[4]tubes 8 (57 mg, 0.039 mmol)
and 7 (45 mg, 0.039 mmol) in THF (25 mL) was added dropwise over 1 h
to a mixture of NaH (60% suspension in mineral oil, 12 mg, 0.31 mmol)
andK 2CO3 (22 mg, 0.16 mmol) in THF (100 mL) at reflux temperature.
The mixture was further refluxedfor 24 h, evaporatedto dryness, sus-
pended in CH2Cl2 (50 mL), andneutralizedat
0
8C with 5% HCl
(25 mL). The organic layer was washedwith water (2”5 mL), driedover
Na2SO4, and then evaporated to dryness. The residue was treated with
MeCN to produce 3 as a colorless solid. Yield: 56 mg (64%); m.p.
>3008C (decomp); 1H NMR (500 MHz, (C2D2Cl4): d=7.22, 7.19 (2”d,
J=7.3 Hz, 24H; ArHm), 7.07–6.97 (m, 20H; ArH), 6.87 (t, J=7.3 Hz,
4H; ArHp), 3.94 (s, 16H; ArCH2Ar), 3.88 (AB q, J=17.6 Hz, 16H;
ArCH2Ar), 3.58 (m, 24H; ArOCH2CH2O), 3.28 (t, J=7.3 Hz, 8H;
OCH2), 2.58 (m, 24H; ArOCH2CH2O), 1.09 (m, 8H; CH2), 0.58 ppm (t,
J=7.3 Hz, 12H; CH3); FTIR (CCl4): n=3061, 3032, 3010, 2958, 2925,
2873, 1459, 1216, 1128, 1094, 1033, 1012 cmÀ1; MS (MALDI-TOF): m/z:
2308 [M+Na]+; calcdfor C 148H156O22Na: 2308.
Pentameric nanotube 4:
A
solution of calix[4]arene 5[11] (50 mg,
0.041 mmol) andcalix[4]tube 7 (95 mg, 0.082 mmol) in THF (25 mL) was
added dropwise over 1 h to a mixture of NaH (60% suspension in miner-
al oil, 13 mg, 0.33 mmol) andK 2CO3 (22 mg, 0.16 mmol) in THF
(100 mL) at reflux temperature. The mixture was further refluxedfor
24 h, evaporated to dryness, suspended in CH2Cl2 (50 mL), andneutral-
izedat 0 8C with 5% HCl (25 mL). The organic layer was washedwith
water (2”5 mL), dried over Na2SO4, andthen evaporatedto dryness.
The residue was treated with MeCN to produce pentacalix[4]tube 4 as a
colorless solid. Yield: 96 mg (82%); m.p. >3008C (decomp); 1H NMR
(500 MHz, CDCl3): d=7.20, 7.17 (2”d, J=7.0 Hz, 32H; ArH), 7.09–6.98
(m, 24H; ArH), 6.87 (t, J=7.3 Hz, 4H; ArHp), 3.94 (s, 24H; ArCH2Ar),
3.93 (AB q, J=16.9 Hz, 16H; ArCH2Ar), 3.62 (m, 32H; ArOCH2CH2O),
3.33 (t, J=7.3 Hz, 8H; OCH2), 2.58 (m, 32H; ArCH2CH2O), 1.11 (m,
J=7.3 Hz, 8H; CH2), 0.58 ppm (t, J=7.3 Hz, 12H; CH3); FTIR (CCl4):
[3] For timely overviews on synthesis andapplications of calixarenes,
see: a) Calixarene 2001 (Eds.: Z. Asfari, V. Bçhmer, J. Harrowfield,
J. Vicens), Kluwer Academic, Dordrecht, 2001; b) Calixarenes in the
Nanoworld (Eds.: J. Vicens, J. Harrowfield) Springer, Dordrecht,
2006.
ACHTREUNG
;
C
[4] a) D. J. Cram, J. M. Cram, Container Molecules andtheir Guests ,
Royal Society of Chemistry, Cambridge, 1994; cavitands: b) D. M.
Rudkevich, J. Rebek, Jr., Eur. J. Org. Chem., 1999, 1991–2005;
c) B. W. Purse, J. Rebek, Jr., Proc. Natnl. Acad. Sci. USA 2005, 102,
10777–10782; carcerands and hemicarcerands: d) A. Jasat, J. C.
Sherman, Chem. Rev. 1999, 99, 931–967; e) R. Warmuth, J. Yoon,
Acc. Chem. Res. 2001, 34, 95–105; f) R. Warmuth, Eur. J. Org.
Chem. 2001, 423–437; cryptophanes: g) A. Collet, J.-P. Dutasta, B.
Lozach, J. Canceill, Top. Curr. Chem. 1993, 165, 103–129. Self-as-
)
2
2
lents of SnCl4 andtwo equivalents of tBuONO per calixarene unit. The
spectra were recorded in solution by using an NaCl amalgamated sealed
cell (1.0 mm) by coaddition of 20 scans, back and forward, at a resolution
of 4 cmÀ1. For each measurement, the solvent was usedfor background.
Titration experiments, typical procedure: An aliquot from the stock solu-
tion of tert-butyl nitrite (1.66m) in (CDCl2)2 was added to an NMR tube
4022
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 4014 – 4023

Nanotubes
FULL PAPER
sembling capsules: h) J. de Mendoza, Chem. Eur. J. 1998, 4, 1373–
[17] a) E. Iglesias, J. Casado, Int. Rev. Phys. Chem. 2002, 21, 37–74;
b) P. G. Wang, M. Xian, X. Tang, X. Wu, Z. Wen, T. Cai, A. J. Janc-
zuk, Chem. Rev. 2002, 102, 1091–1134.
[18] E. Bosch, J. K. Kochi, J. Org. Chem. 1994, 59, 3314–3325.
[19] A somewhat similar effect was described elsewhere: N. Chopra, C.
Naumann, J. C. Sherman, Angew. Chem. 2000, 112, 200–202;
Angew. Chem. Int. Ed. 2000, 39, 194–196.
[20] For allosteric effects in supramolecular chemistry, see a) M. Takeu-
chi, M. Ikeda, A. Sugasaki, S. Shinkai, Acc. Chem. Res. 2001, 34,
865–873; b) S. Shinkai, M. Ikeda, A. Sugasaki, M. Takeuchi, Acc.
Chem. Res. 2001, 34, 494–503.
[21] a) E. K. Kim, J. K. Kochi, J. Am. Chem. Soc. 1991, 113, 4962–4974;
b) D. B. Keck, C. D. Hause, J. Mol. Spectrosc. 1968, 26, 163–174; for
spectroscopic properties of arene–nitrosonium complexes, see G. I.
Borodkin, V. G. Shubin, Russ. Chem. Rev. 2001, 70, 211–230.
[22] a) G. S. Heo, P. E. Hillman, R. A. Bartsch, J. Heterocycl. Chem.
1982, 19, 1099–1103; b) S. Ricard, P. Audet, R. Savoie, J. Mol.
Struct. 1988, 178, 135–140; c) K. Y. Lee, D. J. Kuchynka, J. K. Kochi,
Inorg. Chem. 1990, 29, 4196–4204.
1377; i) F. Hof, S. L. Craig, C. Nuckolls, J. Rebek, Jr., Angew. Chem.
2002, 114, 1556–1578; Angew. Chem. Int. Ed. 2002, 41, 1488–1508;
j) J. Rebek, Jr., Chem. Commun. 2000, 637–643; k) D. M. Rudke-
vich in Calixarene 2001 (Eds.: Z. Asfari, V. Bçhmer, J. Harrowfield,
J. Vicens), Kluwer Academic, Dordrecht, 2001, p. 155–180; l) M.
Fujita, K. Umemoto, M. Yoshizawa, N. Fujita, T. Kusukawa, K. Bir-
adha, Chem. Commun. 2001, 509–518; m) D. M. Rudkevich in
Functional Artificial Receptors (Eds.: T. Shrader, A. D. Hamilton,
Wiley-VCH, 2005, pp. 257–298.
[5] a) N. Sakai, J. Mareda, S. Matile, Acc. Chem. Res. 2005, 38, 79–87;
b) S. Matile, A. Som, N. Sorde, Tetrahedron 2004, 60, 6405–6435;
c) S. Matile, Chem. Soc. Rev. 2001, 30, 158–167.
[6] a) A. N. Khlobystov, D. A. Britz, G. A. Briggs, Acc. Chem. Res.
2005, 38, 901–909; b) M. Monthioux, Carbon 2002, 40, 1809–1823;
c) O. Vostrowsky, A. Hirsch, Angew. Chem. 2004, 116, 2380–2383;
Angew. Chem. Int. Ed. 2004, 43, 2326–2329.
[7] For our preliminary communications on shorter synthetic tubes, see
a) V. G. Organo, A. V. Leontiev, V. Sgarlata, H. V. R. Dias, D. M.
Rudkevich, Angew. Chem. 2005, 117, 3103–3107; Angew. Chem. Int.
Ed. 2005, 44, 3043–3047; b) G. V. Zyryanov, D. M. Rudkevich, J.
Am. Chem. Soc. 2004, 126, 4264–4270.
[23] a) S. G. Bott, H. Prinz, A. Alvanipour, J. L. Atwood, J. Coord.
Chem. 1987, 16, 303–309; b) U. Russo, A. Cassol, A. Silvestri, J. Or-
ganomet. Chem. 1984, 260, 69–72.
[8] Molecular encapsulation of gases: a) D. M. Rudkevich, A. V. Leon-
tiev, Aust. J. Chem. 2004, 57, 713–722; b) D. M. Rudkevich, Angew.
Chem. 2004, 116, 568–581; Angew. Chem. Int. Ed. 2004, 43, 558–
571.
[9] R. Rathore, S. V. Lindeman, K. S. S. Rao, D. Sun, J. K. Kochi,
Angew. Chem. 2000, 112, 2207–2211; Angew. Chem. Int. Ed. 2000,
39, 2123–2127.
[10] a) G. V. Zyryanov, Y. Kang, D. M. Rudkevich, J. Am. Chem. Soc.
2003, 125, 2997–3007; b) Y. Kang, D. M. Rudkevich, Tetrahedron
2004, 60, 11219–11225; experiments with wider and more flexible
O-peralkylatedcalix[5]-, calix[6]-, andcalix[8]arenes didnot leadto
NO+ entrapment. No characteristic color change was detected upon
mixing with NO2/N2O4.
[11] S. Kim, J. S. Kim, S. K. Kim, I.-H. Suh, S. O. Kang, J. Ko, Inorg.
Chem. 2005, 44, 1846–1851.
[12] J. S. Kim, O. J. Shon, J. W. Ko, M. H. Cho, I. Y. Yu, J. Vicens, J. Org.
Chem. 2000, 65, 2386–2392.
[13] a) R. Begum, T. Sagawa, S. Masatoki, H. Matsuura, J. Mol. Struct.
1998, 442, 243–250; b) A. L. Hofacker, J. R. Parquette, Angew.
Chem. 2005, 117, 1077–1081; Angew. Chem. Int. Ed. 2005, 44, 1053–
1057.
[24] For a detailed discussion of oscillation of a single metal cation in
nanotubes, see reference [2b]. Two possible scenarios were pro-
posed: intracalixarene metal tunneling and intercalixarene metal
hopping.
[25] Preliminary experiments with Li+PicÀ andNa +PicÀ in (CDCl2)2 re-
sultedin no complexation with the nanotubes.
[26] The U.S. EPA site on NOx gases: http://www.epa.gov/air/urbanair/
nox/index.html
[27] Y. Kang, G. V. Zyryanov, D. M. Rudkevich, Chem. Eur. J. 2005, 11,
1924–1932.
[28] For reviews on nitric oxide, see a) S. Pfeiffer, B. Mayer, B. Hem-
mens, Angew. Chem. 1999, 111, 1824–1844; Angew. Chem. Int. Ed.
1999, 38, 1714–1731; b) A. R. Butler, D. L. H. Williams, Chem. Soc.
Rev. 1993, 233–241.
[29] Simultaneous encapsulation: J. Rebek, Jr., Angew. Chem. 2005, 117,
2104–2115; Angew. Chem. Int. Ed. 2005, 44, 2068–2078.
[30] Nanoscale molecular containers: D. M. Rudkevich, Bull. Chem. Soc.
Jpn. 2002, 75, 393–413.
[31] a) C. D. Gutsche, M. Iqbal, Org. Synth. 1990, 68, 234–237; b) C. D.
Gutsche, J. A. Levine, P. K. Sujeeth, J. Org. Chem. 1985, 50, 5802–
5806.
[14] This was confirmedby NOESY for the nitrosonium-fillednonsym-
metrical, O,O’-dimethylated derivative of 7.
[32] W. Verboom, S. Datta, Z. Asfari, S. Harkema, D. N. Reinhoudt, J.
Org. Chem. 1992, 57, 5394–5398.
[15] At this stage, the location of nitrate counterions in complexes
[33] MacroModel 7.1; MM2 and Amber* Force Field. See: F. Mohamadi,
N. G. Richards, W. C. Guida, R. Liskamp, M. Lipton, C. Caufield, G.
Chang, T. Hendrickson, W. C. Still, J. Comput. Chem. 1990, 11, 440–
467.
1·
N
N
taining single crystals for X-ray analysis has proveddifficult, as cal-
ixarene–nitrosonium complexes tendto decompose upon prolonged
standing. Molecular modeling strongly suggests that these anions,
which are too bulky to enter the tube interior, stay outside, in rather
close proximity to the exterior in apolar solvent.
Received: October 29, 2006
Revised: December 12, 2006
Publishedonline: February 13, 2007
[16] V. Sgarlata, V. G. Organo, D. M. Rudkevich, Chem. Commun. 2005,
5630–5632.
Chem. Eur. J. 2007, 13, 4014 – 4023
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4023