A molecular gate: Control of free intramolecular rotation by application of an external signal

Article abstract of DOI:10.1002/poc.790

Control of internal movement in a conformationally free pentiptycene derivative is achieved by complexation of two phenanthroline units by copper(I) cation. Copyright

Full text of DOI:10.1002/poc.790

JOURNAL OF PHYSICAL ORGANIC CHEMISTRY
J. Phys. Org. Chem. 2004; 17: 749–751
Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/poc.790
A molecular gate: control of free intramolecular rotation
by application of an external signaly
Rita Annunziata, Maurizio Benaglia, Mauro Cinquini, Laura Raimondi and Franco Cozzi*
CNR-ISTM and Dipartimento di Chimica Organica e Industriale, Universita` di Milano, via Golgi 19, 20133 Milan, Italy
ABSTRACT: Control of internal movement in a conformationally free pentiptycene derivative is achieved by
complexation of two phenanthroline units by copper(I) cation. Copyright # 2004 John Wiley & Sons, Ltd.
KEYWORDS: molecular device; pentiptycene; phenanthroline; copper(I) complex; restricted rotation
Over the last three decades, control of molecular move-
ment has been a coveted goal for chemists interested in
the development of molecular devices (for reviews, see
Ref. 1). In this context, command of internal rotation has
been sought by two different approaches. One approach,
pioneered and mastered by Mislow, entailed the synthesis
of a sterically congested molecule that displays discrete
motion as the consequence of temperature variation (for
reviews and recent examples, see Ref. 2). In other
endeavours, conformationally free molecules were forced
to undergo discrete movements by application of an
external signal, the removal of which restored the original
situation3 (for recent reports describing the control of
other molecular movements, see Ref. 4).
In a first attempt to prepare a molecule suitable for this
study, hydroquinone 15 (for the use of 1 in the develop-
ment of chemosensors, see Refs 5b and c) (Scheme 1)
was alkylated with mesylate 3, obtained from 2-(4-
hydroxyphenyl)-1,10-phenanthroline 26 by O-alkylation
with 6-bromohexanol (Cs2CO3, acetonitrile, 60 ꢀC, 72 h,
90% yield), followed by mesylation (MsCl, triethyla-
mine, dichloromethane, 23 ꢀC, 12 h, 96% yield) (all
new compounds had spectral data in agreement with
the proposed structures). 1H NMR analysis of 4, obtained
in 21% unoptimized yield (Cs2CO3, acetonitrile, 80 ꢀC,
72 h), showed one set of signals for the eight aromatic
hydrogens shown as Ha in Scheme 1 and another set of
signals for those indicated as Hb. Thus, a clear demon-
stration of the equivalence of the four phenyl rings and,
hence, of the unhidered rotation of the pentiptycene
moiety around the C—O bond in this compound was
obtained.
Here, we report preliminary results on the development
of a molecular gate to control internal rotation in a
conformationally unrestricted molecule. The molecular
design (Fig. 1) is based on the connection of two
phenanthroline units to a pentiptycene skeleton repre-
senting the free-rotating rotor A (gate open), the move-
ment of which should be controlled by application of the
external signal (phenanthroline complexation) as in B or
Treatment of 4 with 1 mol equiv. of CuOTf in 1:1
chloroform–acetonitrile (23 ꢀC, 24 h) led to the formation
of a dark-red complex that was purified by short-path
chromatography to afford 4 ꢁ Cu (>90% yield). The
complex 4 ꢁ Cu was characterized as a single monomeric
species by mass spectrometry {fast atom bombardment
C (gate closed). [In a previous study by Kelly et al.,3a
a
similar approach was employed to block the internal
rotation of a triptycenyl bipyridine derivative. In that
case however, it was shown that, in the absence of an
external signal, internal rotation could also be slowed
(although not completely prevented) by decreasing the
temperature, the molecule under investigation being
considerably less conformationally mobile than those
reported in this work. We believe that the term ‘gate’
(a device that allows/prevents passage) descibes more
precisely than ‘brake’ (an apparatus for checking motion)
the sort of movement restriction reported here and in
previous work.3a In other words, the notion of ‘molecular
brake’ implies a much more refined movement control
than that of ‘molecular gate’.]
1
(FAB); m/z 1234/1236 [M ꢂ OTf]þ }. Also H and 13C
NMR spectra (CD3OD) were consistent with the forma-
tion of a single complex; the resonances of the H atoms
meta to nitrogen in the phenanthroline moieties under-
went upfield shifts of ꢃ0.5 ppm and those of the C atoms
in the same position were shifted downfield by ꢃ4 ppm
with respect to those of 4. The phenantroline protons gave
1
sharp H NMR signals, whereas those of the phenyl
groups of the pentiptycene skeleton were found to
be broad and unresolved in the temperature range from
ꢂ60 to þ60 ꢀC. Thus, even if complexation of 4 had
indeed occurred, this structural modification was not
sufficient to prevent the free rotation of the pentiptycene
rotor. Decomplexation with cyanide ions restored the
original ligand 4.
After molecular mechanics calculations suggested that
a more rigid tether between the pentiptycene and phe-
nanthroline groups could provide a better candidate to
*Correspondence to: F. Cozzi, Dipartimento di Chimica Organica e
`
Industriale, Universita di Milano, via Golgi 19, 20133 Milan, Italy.
E-mail: franco.cozzi@unimi.it
Contract/grant sponsors: MIUR; CNR.
yDedicated to Kurt Mislow on the occasion of his 80th birthday.
Copyright # 2004 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2004; 17: 749–751

750
R. ANNUNZIATA ET AL.
Figure 1. Schematic represenatation of (A) free and (B, C)
blocked rotors. Lobes of the same shade indicate symmetry-
equivalent rings in pentiptycene
Scheme 1. Synthesis of ligand 4. Reagents and conditions:
a, 6-bromohexanol, Cs2CO3, CH3CN, 60 ꢀC, 72 h; b, MsCl,
Et3N, CH2Cl2, 23 ꢀC, 12 h; c, 3, Cs2CO3, CH3CN, 80 ꢀC, 72 h
observe the desired phenomenon, 9 was synthesized as
described in Scheme 2. Thus, 5, prepared in 75% yield
from 1,10-phenanthroline and 3-bromotoluene (tBuLi,
THF, ꢂ78 to 23 ꢀC, 40 h, then MnO2, dichloromethane,
23 ꢀC, 24 h), was sequentially converted into monobro-
mide 6 (NBS, refluxing carbontetrachloride, 2 h, 33%
yield at 50% conversion), alcohol 7 (4-hydroxymethyl-
phenol, Cs2CO3, acetonitrile, 60 ꢀC, 72 h, 96% yield), and
bromide 8 (PBr3, dichloromethane, 0–23 ꢀC, 12 h, 75%).
Alkylation of 1 with 8 (Cs2CO3, acetonitrile, 80 ꢀC, 72 h)
afforded ligand 9 in 45% unoptimized yield. Also for this
groups was more complicated. Indeed, for one adduct
the 2D NMR [H,H] COSY experiment carried out at
25 ꢀC revealed two distinct, well-resolved pairs of corre-
lated protons, with four of the eight Ha protons resonating
at 6.83 ppm and four at 6.91 ppm, and four of the eight Hb
protons resonating at 7.12 ppm and four at 7.25 ppm,
respectively. [It must be noted that when pentiptycene
rotation is frozen, the 9 ꢁ Cu complex has C2 symmetry.
Hence the four protons of each pentiptycene phenyl ring
1
0 0
should give rise to four resonances (Ha, Hb, Hb , Ha ). De
compound, H NMR analysis showed one set of signals
0
facto, accidental isochrony occurs between Ha/Ha , and
0
Hb/ Hb which will be considered equivalent in the
for the eight aromatic protons indicated as Ha and another
signal for those indicated as Hb in 9, confirming the free
rotation of the pentiptycene residue that persisted even at
ꢂ70 ꢀC.
following discussion. We thank professor Siegel for
calling our attention to this problem.] In addition, a third
set of poorly resolved signals was observed for the same
protons, with Ha resonating at 7.26 ppm and Hb at
7.65 ppm. On cooling to ꢂ45 ꢀC, the signals of the first
two sets remained almost unchanged (shifting to 6.81,
6.87, 7.07, and 7.29 ppm, respectively), whereas those of
the third set became sharper and shifted to 6.95 and
7.35 ppm, respectively. Decomplexation with cyanide
ions quantitatively restored the original ligand 9.
The formation of the corresponding Cu complex was
performed as described above for 4 ꢁ Cu to afford a >90%
isolated yield of 9 ꢁ Cu. Mass spectrometry indicated that
only monomeric species were formed (FAB; m/z 1453/
1
13
1455 [M ꢂ OTf]þ ). H and C NMR analysis at 25 ꢀC
(CD3CN) showed the presence of two species in a
roughly equimolar ratio. For both, the sharp shape and
the observed chemical shifts of the H and C signals of the
phenanthroline residues clearly indicated that these
groups are firmly involved in Cu(I) complexation. The
pattern of the hydrogens of the pentiptycene phenyl
These results can be interpreted as follows. The ap-
pearance of distinct sets of signals for Ha and Hb in
proximal and distal phenyl rings in 9 ꢁ Cu strongly
Copyright # 2004 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2004; 17: 749–751

CONTROL OF FREE INTRAMOLECULAR ROTATION
751
signals whose chemical shift does not change with
temperature can tentatively be assigned to the complex
in which rotation is more effectively blocked.
In conclusion, the free intramolecular rotation in a
pentiptycene derivative suitably modified by insertion of
two phenanthroline residues was blocked by complexa-
tion of the latter by Cu(I) cation. Removal of the metal
restored free rotation. As a whole, the system described
here represents an equivalent of a molecular gate. Work is
in progress to develop other molecules in which internal
rotation can be controlled at will.
Acknowledgment
This work was supported by MIUR and CNR (Rome).
REFERENCES
1. (a) Lehn J-M. In Supramolecular Chemistry—Concepts and
Perspectives. VCH: Weinheim, 1995; 124–138; (b) Balzani V,
Credi A, Raymo FM, Stoddart JF. Angew. Chem., Int. Ed. 2000;
39: 3348–3391; (c) Acc. Chem. Res. 2001; 34: special issue on
molecular machines; (d) Balzani V, Credi A, Venturi M. Chem.
Eur. J. 2002; 8: 5525–5530; (e) Jimenez-Molero MC, Dietrich-
Buchecker C, Sauvage J-P. Chem. Commun. 2003; 1613–1616.
2. (a) Mislow K. Chemtracts Org. Chem. 1989; 2: 151–174; (b)
Bedard TC, Moore JS. J. Am. Chem. Soc. 1995; 117: 10662–
10671; (c) Stevens AM, Richards CJ. Tetrahedron Lett. 1997; 38:
7805–7808; (d) Kelly TR, Tellitu I, Sestelo JP. Angew. Chem., Int.
Ed. Engl. 1997; 36: 1866–1868; (e) Kelly TR, Sestelo JP, Tellitu I.
J. Org. Chem. 1998; 63: 3655–3665; (f) Clayden J, Pink JH.
Angew. Chem., Int. Ed. 1998; 37: 1937–1939; (g) Dominguez Z,
Dang H, Strouse MJ, Garcia-Garibay MA. J. Am. Chem. Soc.
2002; 124: 7719–7727.
3. (a) Kelly TR, Bowyer MC, Bhaskar KV, Babbington D, Garcia A,
Lang F, Kim MH, Jette MP. J. Am. Chem. Soc. 1994; 116: 3657–
3658; (b) Kelly TR, De Silva H, Silva RA. Nature (London) 1999;
401: 150–152; (c) Koumura N, Zijlstra RWJ, van Delden RA,
Harada N, Feringa BL. Nature (London) 1999; 401: 152–155; (d)
van Delden RA, Koumura N, Skoevaars A, Meetsma A, Feringa
BL. Org. Biomol. Chem. 2003; 1: 33–35, and references cited
therein.
4. (a) Lane AS, Leigh DA, Murphy A. J. Am. Chem. Soc. 1997;
119: 11092–11093; (b) Jimenez MC, Dietrich-Buchecker CO,
Sauvage J-P. Angew. Chem., Int. Ed. 2000; 39: 3284–3287; (c)
Weinig H-G, Krauss R, Seydack M, Bendig J, Koert U. Chem.
Eur. J. 2001; 7: 2075–2088; (d) Fabbrizzi L, Foti F, Licchelli M,
Maccarini PM, Sacchi D, Zema M. Chem. Eur. J. 2002; 8: 4965–
4972; (e) Barboiu M, Vaughan G, Kyritsakas N, Lehn J-M. Chem.
Eur. J. 2003; 9: 763–769, and references cited therein.
5. (a) Yang J-S, Lee C-C, Yau S-L, Chang C-C, Lee C-C, Leu J-M.
J. Org. Chem. 2000; 65: 871–877; (b) Yang J-S, Swager TM.
J. Am. Chem. Soc. 1998; 120: 11864–11873; (c) Yang J-S,
Lin C-S, Hwang C-Y. Org. Lett. 2001; 3: 889–892.
Scheme 2. Synthesis of ligand 9. Reagents and conditions:
a, m-methylphenyllithium, THF, ꢂ78 to 23 ꢀC, 40 h, then
MnO2, CH2Cl2, 23 ꢀC, 24 h; b, NBS, CCl4, 80 ꢀC, 2 h; c,
HOC6H4CH2OH, Cs2CO3, CH3CN, 60 ꢀC, 72 h; d, PBr3,
CH2Cl2, 0–23 ꢀC, 12 h; e, 8, Cs2CO3, CH3CN, 80 ꢀC, 72 h
suggests that the free rotation of the pentiptycene rotor in
9 has been blocked by complexation. Therefore, this
event acts as a gate at the molecular level. The restricted
rotation results in the non-equivalence of the phenyl
groups of the pentiptycene residues, consistent with the
NMR evidence (see Fig. 1). The observation of more than
two sets of signals shows that more than one complex
is formed, possibly two structures similar to B and C
(Fig. 1). Accidental isochrony can be invoked to explain
why, in one case, only one and not two sets of signals are
observed. Although it is not possible at present to assign
unequivocally the B or C structure to the observed
species, it seems possible that these diastereoisomeric
complexes display different hindrances to rotation. The
6. Dietrich-Buchecker CO, Nierengarten J-F, Sauvage J-P, Armaroli
N, Balzani V, De Cola L. J. Am. Chem. Soc. 1993; 115: 11237–
11244.
Copyright # 2004 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2004; 17: 749–751