Towards mechanically linked polyrotaxanes by sequential deprotection-coupling steps of bifunctional rotaxanes

Article abstract of DOI:10.1039/a900914k

The monomer rotaxane 13, bearing two protected functional groups, is used to obtain a dimer, following a new approach of sequential deprotection-coupling steps which can lead to mechanically linked pslyrotaxanes.

Full text of DOI:10.1039/a900914k

Towards mechanically linked polyrotaxanes by sequential
deprotection–coupling steps of bifunctional rotaxanes
Michel P. L. Werts, Maarten van den Boogaard, Georges Hadziioannou and Gerasimos M. Tsivgoulis*
Department of Polymer Chemistry and Materials Science Centre, University of Groningen, Nijenborgh 4, 9747 AG
Groningen, The Netherlands. E-mail: g.tsivgoulis@chem.rug.nl
Received (in Liverpool, UK) 2nd February 1999, Accepted 25th February 1999
The monomer rotaxane 13, bearing two protected functional
groups, is used to obtain a dimer, following a new approach
of sequential deprotection–coupling steps which can lead to
mechanically linked polyrotaxanes.
rotaxane is used as the starting monomer. The dimer synthesis
and some preliminary studies are presented.
The rotaxane structure used as monomer (compound 13 on
Scheme 1) is based on dialkoxybenzene units with a cyclophane
containing two 4,4A-bipyridinium groups. Similar types of
systems have been studied extensively by the group of
Stoddart.7
The synthesis of polyrotaxanes and polycatenanes has proven
to be quite challenging. In the step-by-step approach, presented
in Scheme 2, the initial monomer rotaxane 13 was designed to
fulfil a number of requirements: (a) introduction of functional
units so that polymerization is possible, (b) protection of these
functional units for a step-by-step synthesis and (c) minimiza-
tion of electronic and/or steric problems.
The positions of the two functional groups in the rotaxane
were chosen in such a way that a dimerization and/or
polymerization reaction could lead to oligomers with the
repeating units connected non-covalently (Schemes 1 and 2).
These groups are a phenol and a carboxylic acid. In general,
between these two groups, a high-yield coupling reaction
(esterification) can take place in only one synthetic step. This is
important in order to minimize the number of reaction steps that
involve the rotaxane molecule.
Rotaxanes and catenanes as well as their polymer derivatives
appear increasingly in the literature. In the field of polymers, a
number of fascinating polyrotaxane structures have been
presented.1 It has been shown that the properties of these
compounds are in many cases very different compared to the
‘naked’ polymers.2
In the vast majority of the compounds synthesized so far, a
covalently connected polymer backbone is surrounded by
‘cyclic’ molecules, or alternatively, rotaxane subunits are
connected to the polymer backbone as side groups.1,2 The very
attractive concept of synthesizing oligomers and/or polymers
where the repeating units are connected to each other in a non-
covalent way was proposed some years ago,3 but progress
towards this aim has been achieved only very recently.4–6 Such
polymers and/or oligomers with non-covalent connections are
expected to give materials with new rheological and mechanical
properties. Until now, this research has mainly been focused on
polycatenanes.4 Reports on mechanically linked polyrotaxanes
have only described small assemblies5 and polypseudorotax-
anes.6
Although a number of protective groups are available for
sequential removal, in our case the choice decreased sig-
nificantly since selective deprotection of both groups is needed.
In addition, the various sensitive groups in the rotaxane
structure 13 should be inert to the deprotection conditions. This
compound contains an ester bond and a bis(4,4A-bipyridinium)
cyclophane derivative, which both show very limited stability in
Here we present the synthesis of a new mechanically linked
dimer rotaxane, where the repeating units are connected by non-
covalent bonds. The synthesis is based on a new step-by-step
approach consisting of sequential deprotection–coupling steps
that can lead to well defined oligomers. A bisfunctionalized
X
Si
TsO
O
O
O
O
O
CH
O
O
ii
O
C
O
6
iv
O
O
O
O
O
CH
HO
O
O
O
O
O
Y
CH
O
OH
O
7
HO
1
2
X = OH
5
i
X = OSiButPh2
3
4
Y = OH
CH Br
2
iii
O
Y = (OCH2CH2)4-OTs
C
O
Cl
HO
C
v
Si
O
O
CH2Br
9
10
O
O
O
O
CH
O
O
O
O
CH
O
O
O
O
O
O
O
vi
8
CH Br
2
O
O
O
C
CH2Br
N
O
C
Si
–
vii
4 PF6
O
11
+ N
N+
N
O
O
O
O
O
CH
O
O
O
CH
O
O
O
O
O
O
O
N+
12
+ N
13
+N
–
N +
C
O
O
–
C
O
PF6
O
PF6
Scheme 1 Synthesis of the bisprotected monomer 13. Reagents and conditions: i, TBDPSCl, Et3N, DMAP, Py, THF (71%); ii, NaBH4, THF (79%); iii, NaH,
Ts(OCH2CH2)4OTs, THF (65%); iv, K2CO3, DMF (58%); v, NaH, THF (38%); vi, Pri2EtN, CH2Cl2 (49%); vii, CH3CN, AgPF6 (4.8%).
Chem. Commun., 1999, 623–624
623

Scheme 2 Schematic representation of the synthesis of the bisprotected
dimer 16.
alkaline pH. Furthermore, the two blocking groups are of the
diphenylmethyl ether type and are unstable in acidic conditions.
The hindered TBDPS protective group appears to be a good
protective group for the phenol unit. It is stable under a variety
of basic and acidic conditions8 (necessary in steps ii, iii and v in
the rotaxane synthesis), while it can be removed specifically
with fluoride anions under very mild conditions. In the
carboxylic acid case, our initial attempts were focused on the
tetrahydropyranyl (THP) group. Unfortunately, despite encour-
aging results obtained with model compounds, it was found that
this group was too labile, and was removed during simple work-
up operations of the rotaxane 13. In contrast, the allyl group was
found to be suitable. The allyl ester bond is stable during steps
vi and vii (Scheme 1) as well as during the reactions used in the
synthesis of oligomers. At the same time, it can be selectively
removed in the presence of other ester bonds by Pd catalysts
under mild conditions.
The last point in the design of rotaxane 13 concerns the
reactivity of the carboxylic acid group. It has been reported9 that
a carboxylic group directly connected to the bis(4,4A-bipyr-
idinium) cyclophane is not active towards esterification. In
preliminary experiments we found similar behaviour. There-
fore, in molecule 13 the protected carboxylic group was moved
to a position that is not in close proximity to the tetracationic
cyclophane.
Fig. 1 MALDI-TOF spectra in 2,5-dihydroxybenzoic acid matrix: (a)
monomer 13, (b) dimer 16.
4159, 4015 and 3869 mass units (mu), corresponding to the ions
[M 2 4PF6]+, [M 2 5PF6]+, [M 2 6PF6]+ and [M 2 7PF6]+,
respectively. The peaks at lower masses can be easily identified
as fragments of the molecular ion. By using 5-methoxysalicylic
acid instead of 2,5-dihydroxybenzoic acid as matrix, a different
series of ions can be identified [M 2 2PF6]+, [M 2 3PF6]+, [M
2 4PF6]+ and [M 2 5PF6]+. Similar results were found for the
monomer 13. The three peaks at 2299, 2154 and 2009 mass
units (mu) correspond to the ions [M 2 2PF6]+, [M 2 3PF6]+
and [M 2 4PF6]+, respectively [Fig. 1(a)].
In conclusion, we were able to synthesize the monomer
rotaxane 13, which bears two protected functional groups and
can give oligomers in a controlled way for the first time, by
sequential deprotection–coupling steps. Coupling between
derivatives of 13 gives a dimer 16 which can be used for the
synthesis of longer rotaxanes by applying the same approach.
The connection of the repeating units in the dimer 16 provides
a new way to obtain polyrotaxanes based on non-covalent
bonds. Synthesis and study of the rheological properties of
longer oligorotaxanes based on monomer 13 are in progress.
We are grateful to B. Kwant for help in the synthetic work
and to Dr A. Kievit for his advice concerning MALDI-TOF
spectrometry. This work was financially supported by the
Netherlands Foundation for Chemical Research (NWO-CW).
The synthesis of the rotaxane 13 was accomplished by
reacting 8, 11 and 12 in the presence of excess of AgPF6 in 4.8%
yield.† The low yield may be attributed5 to the steric hindrance
of the carboxylic ester group of 11. Despite this low yield, at the
end of the reaction most of the unreacted compound 8 can be
recovered and used again in step vii. Compounds 5, 6, 10 and 12
were obtained according to literature procedures while mole-
cule 9 was accessed by ring opening of g-butyrolactone with
allyl alcohol under acidic conditions. The procedures followed
for the synthesis of molecule 8 are delineated in Scheme 1.
Selective deprotection of the two functional groups in two
different batches was the first step towards the synthesis of the
dimer. Thus, rotaxane 14 contained a free carboxylic acid group
obtained by deprotection of 13 with Pd(PPh3)4, while rotaxane
15 contained a free phenol group, obtained by deprotection of
13 with Bu4NF. In the last step, esterification (DCC/Py)
between the acid 14 and the alcohol 15 gave the dimer 16 in a
30–40% yield. This compound also contains a protected phenol
and a protected carboxylic acid, like the monomer 13.
Therefore, longer derivatives (tetramer, octamer etc.) can be
obtained by sequential deprotection–esterification steps.
Both the monomer 13 and the dimer 16 were characterized by
Notes and references
† All new compounds have been characterized by mass and 1H-NMR
spectroscopy.
1 Large ring molecules, ed. J. A. Semlyen, Wiley-Interscience, New
York, 1996, pp. 155–263.
2 H. W. Gibson and H. Marand, Adv. Mater., 1993, 5, 11.
3 J. A. Preece and J. F. Stoddart, Macromol. Symp., 1995, 98, 527.
4 For example: R. Ja¨ger and F. Vo¨gtle, Angew. Chem., Int. Ed. Engl.,
1997, 36, 930; C. Hamers, F. M. Raymo and J. F. Stoddart, Eur. J. Org.
Chem., 1998, 2109; S. Shimada, K. Ishikawa and N. Tamaoki, Acta
Chem. Scand., 1998, 52, 374.
5 T. Du¨nnwald, R. Ja¨ger and F. Vo¨gtle, Chem. Eur. J., 1997, 3, 2043; N.
Tamaoki and S. Shimada, Acta Chem. Scand., 1997, 51, 1138.
6 N. Yamaguchi, D. S. Nagvekar and H. W. Gibson, Angew. Chem., Int.
Ed. Engl., 1998, 37, 2361; P. R. Ashton, I. Baxter, S. J. Cantrill,
M. C. T. Fyte, P. T. Glink, J. F. Stoddart, A. J. P. White and D. J.
Williams, Angew. Chem., Int. Ed. Engl., 1998, 37, 1294; P. R. Ashton,
I. W. Parsons, F. M. Raymo, J. F. Stoddart, A. J. P. White, D. J. Williams
and R. Wolf, Angew. Chem., Int. Ed. Engl., 1998, 37, 1913.
7 V. Balzani, M. Gomez-Lopez and J. F. Stoddart, Acc. Chem. Res., 1998,
31, 405 and references therein.
8 Protective Groups in Organic Synthesis, ed. T. W. Greene, Wiley-
Interscience, New York, 1981, pp. 47–48.
9 S. Menzer, A. J. P. White, D. J. Williams, M. Belohradsky, C. Hamers,
F. M. Raymo, A. N. Shipway and J. F. Stoddart, Macromolecules, 1998,
31, 295.
10 R. Ballardini, V. Balzani, M. T. Gandolfi, L. Prodi, M. Venturi, D.
Philp, H. G. Ricketts and J. F. Stoddart, Angew. Chem., Int. Ed. Engl.,
1993, 32, 1301.
1
MALDI-TOF mass spectrometry and H NMR and UV-VIS
absorption spectroscopy. The 1H NMR spectra of 13 and 16 in
CD3COCD3 are consistent with the assigned structures.
The hexafluorophosphate salts of the rotaxanes 13 and 16 are
red solids, insoluble in H2O but soluble in acetone, CH3CN and
CH2Cl2. Also, they are insoluble in solvents of lower polarity
such as Et2O, CHCl3 and hexane. The UV-VIS spectra of both
13 and 16 in CH3CN and in CH2Cl2 show two absorption
maxima; an intense peak at 265 nm and a very weak peak at 480
nm. The last one corresponds to the charge transfer interaction
known in rotaxanes of similar structure.10
The MALDI-TOF mass spectrum of 16 with 2,5-dihydroxy-
benzoic acid as matrix [Fig. 1(b)] shows four peaks at 4303,
Communication 9/00914K
624
Chem. Commun., 1999, 623–624