Translational isomerism and dynamics in multi-hydroquinone derived porphyrin [2]- and [3]-catenanes

Article abstract of DOI:10.1039/b308157e

A series of porphyrins strapped with polyether chains containing two or three 1,4-dioxybenzene units has been synthesised with a view to the production of porphyrin-containing [2] and [3]catenanes, where the porphyrin is strapped between ortho-positions o

Full text of DOI:10.1039/b308157e

View Article Online / Journal Homepage / Table of Contents for this issue
Translational isomerism and dynamics in multi-hydroquinone
derived porphyrin [2]- and [3]-catenanes
Maxwell J. Gunter,* Sandra M. Farquhar and Tyrone P. Jeynes
Chemistry, School of Biological, Biomedical and Molecular Sciences,
University of New England, Armidale, NSW 2351, Australia.
E-mail: mgunter@pobox.une.edu.au; Fax: 612 6773 3268; Tel: 612 6773 2767
Received 21st July 2003, Accepted 12th September 2003
First published as an Advance Article on the web 7th October 2003
A series of porphyrins strapped with polyether chains containing two or three 1,4-dioxybenzene units has been
synthesised with a view to the production of porphyrin-containing [2] and [3]catenanes, where the porphyrin is
strapped between ortho-positions of 5,15-(meso)-diaryl groups, and is interlinked with the bipyridinium macrocycle
cyclobis(paraquat-4,4Ј-biphenylene). The porphyrins were isolated as mixtures of atropisomers, where the linking
strap spans across the face of the porphyrin (α,α-isomer), or ‘twisted’ around its side (α,β-isomer). Their structures
were determined by detailed 1H NMR spectroscopy. The bis-1,4-dioxybenzene-strapped derivatives were shown
to undergo atropisomerisation on heating, to produce an equilibrium mixture. Catenation under high pressure
conditions of the mixture, or of the individual isomers, produced only a single catenane, that of the α,α-isomer.
Its structure was determined by mass spectral and dynamic NMR measurements. Rates were determined for:
(i) translational motion or ‘shuttling’ between 1,4-dioxybenzenes; (ii) ‘rotation’ of the macrocycle around the
1,4-dioxybenzene axis; and (iii) ‘rocking’ of the 1,4-dioxybenzene within the macrocycle. The atropisomers of
the strapped derivatives containing three 1,4-dioxybenzene units were also separated, and subjected to catenation.
Both [2]- and [3]catenanes were isolated, and were shown to be stable to further atropisomerisation. Their solution
structures were probed in detail by dynamic 1H NMR measurements. The rates for shuttling and rotation were
obtained in certain cases, although the complexity of the spectra of the [3]catenanes prevented a more detailed
investigation.
Introduction
The concept of using a catenane as a molecular device,1 or as a
component of a molecular-scale machine2–4 depends critically
on the ability to exercise precise control over its dynamics, such
as those involving translational motion. This can be achieved to
some extent by design, relying on differential thermodynamic
stability between isomers,5–7 or by higher order catenanes where
one component undergoes a degenerate translational between
equivalent sites;6,8,9 this is in essence the principle of molecular
‘trains’ or ‘shuttles’. In a more sophisticated approach, trans-
lational isomerism can be deliberately controlled by external
stimuli such as protonation, electrochemistry, or photo-
chemistry, for example. In these cases, the catenane is required
to have an in-built addressable component.4,10,11 Porphyrins
provide a particularly useful range of controlled functionality
that can be accessed through chemical, electrochemical, or
photochemical means, and there have been several examples
where these are incorporated into catenated structures.12–16
We have previously reported a series of strapped-porphyrin
[2]catenanes 1a and have examined in some detail the effects on
the dynamics, chemical and photophysical behaviour as both
the chain length of the strap and the nature of the central
π-donor unit (1,4-dioxybenzene or 1,5-dioxynaphthalene) were
varied.14–17 Indeed such variations proved to have a profound
influence on the observed dynamic behaviour of the catenanes
as evidenced by the wide range in the calculated thermo-
dynamic parameters for the diethylene to tetraethylene series of
strapped porphyrin catenanes.13,18
Although the [2]catenanes prepared thus far exhibit interest-
ing dynamic properties, their movement is restricted to rotation
of the tetracationic cyclophane around the 1,4-dioxybenzene
or 1,5-dioxynaphthalene axis, sometimes accompanied by a
stretching of the porphyrinic link of the catenane, as observed
in protonation studies.16 We now wished to extend the range of
molecular dynamics that might be possible in these systems by
introducing a further degree of freedom, viz translational
motion7,19 in the resultant porphyrin catenanes. To this end, the
synthesis of strapped-porphyrins which contain two (2HQ)
and three (3HQ) hydroquinone-derived π-donor units, capable
of forming catenanes, was seen as a logical extension. It was
envisioned that in subsequent protonation or photochemical
studies in these catenanes that the tetracationic cyclophane
could be driven between 1,4-dioxybenzene ‘stations’, effectively
acting as a molecular ‘shuttle’, which has been well illustrated in
the work of Stoddart2,7,8,11,20 for simpler non-porphyrinic sys-
tems derived from the classic BPP34-C-10-[2]catenane system
1b. In our systems however, the porphyrin subunit provides an
ideal trigger for control of the dynamics, as it can be addressed
by chemical, photochemical, or electrochemical stimuli.
T h i s j o u r n a l i s © T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 3
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 4 0 9 7 – 4 1 1 2
4097

View Article Online
Table 1 1H NMR resonance assignments for the free base porphyrin
isomers containing 2HQ (6a and 7a) or 3HQ rings (11a and 12a)
recorded at 300 MHz in CDCl3 at 298 K
Position
6a
7a
11a
12a
1
4
5
7
10.14
10.17
10.20
4.00
1.77
2.59
Ϫ2.27
7.79
7.40
7.82
7.34
4.17
3.20
2.06
1.63
5.20
5.86
3.70
3.70
3.70
3.70
3.70
3.70
3.87
4.10
6.89
10.21
4.01
1.79
2.60
Ϫ2.25
7.79
7.39
7.80
7.34
4.18
3.10
2.07
1.72
5.24
5.84
3.62
3.70
3.62
3.62
3.62
3.62
3.70
3.80
6.54
3.92
1.70
2.54
Ϫ2.36
7.77
7.36
7.74
7.29
4.15
3.20
2.19
2.19
5.94
6.44
4.00
3.85
3.80
3.80
—
3.98
1.75
2.56
Ϫ2.34
7.75
7.35
7.77
7.30
4.16
3.12
2.23
2.32
5.94
6.37
3.82
3.70
3.61
3.61
—
pyrrole NH
11
12
13
14
16
17
18
19
In these strapped-porphyrins (Scheme 1), the polyether strap
that we have chosen incorporates two and four ethyleneoxy
units in symmetrical combinations (ie, 2,4,2 in the case of 7
and 2,4,4,2 in the case of 12) between the porphyrin and the
1,4-dioxybenzene units. While this introduces a degree of
asymmetry in the 1,4-dioxybenzene ring protons closest to the
porphyrin, it still allows for some interaction between the
porphyrin and the tetracationic macrocycle as we have found in
previous studies.15
HQ-21
HQ-22
24
25
26
27
28
29
30
—
—
–
—
—
—
—
—
31
HQ-33
Furthermore, strapped porphyrins containing two or three
hydroquinone-derived units have the potential to undergo
self-assembly of higher order porphyrin [3]catenanes using
Stoddart’s templating methodology.19,21
[M]2ϩ. In addition, each pair of isomers have virtually identical
UV-vis spectra for both free base and zinc derivatives.
However, the 1H NMR spectra of each isomer are distinctive.
The spectra of the free base porphyrin isomers 6a and 7a,
as well as 11a and 12a, were assigned by a combination of
gradient COSY and NOESY two dimensional NMR experi-
ments; the data are collected in Table 1, and the non-systematic
Results and discussion
(a) Strapped-porphyrins containing two (2HQ) and three
(3HQ) hydroquinone-derived rings
1
Synthesis. The synthesis of the polyether strapped-
porphyrins 7 and 10 containing two and three 1,4-dioxybenzene
rings (2HQ and 3HQ) is outlined in Scheme 1 and follows a
similar methodology to that employed for the synthesis of
previously reported15,22,23 strapped-porphyrins. The dialdehydes
4 and 10 were produced from the corresponding 1,4-dioxy-
benzene polyethers7 as indicated in Scheme 1. The dialdehyde
derivatives were then condensed with dipyrrylmethane 5 by tri-
chloroacetic acid catalysis using a modification of Maruyama’s
numbering systems employed for the assignments of the H
NMR spectra are shown in Scheme 1.
1
Comparison of the H NMR resonances of the open chain
bis-hydroxy crown ethers 2 or 8 from which the free base
porphyrin 7a or 12a are formed shows that the methylene pro-
tons of the polyether chain, and the 1,4-dioxybenzene ring
protons (shielded ∆δ Ϫ0.88 and Ϫ0.45 for 7a, and ∆δ Ϫ1.59,
Ϫ1.29, and Ϫ0.26 for 12a), undergo significant upfield shifts on
porphyrin formation (Table 1), indicative of the ether chain
strapping the porphyrin ring. Presumably, the extra shielding
observed in 12a over that for 7a reflects the increased flexibility
provided by the third 1,4-dioxybenzene ring and longer poly-
ether strap, resulting in HQ-21 and HQ-22 being more centrally
located over the porphyrin macrocycle.
25
procedure,24 (with the addition of Cs2CO3 as template in the
case of the longer dialdehyde 10), followed by oxidation of the
porphyrinogen with o-chloranil to give the desired porphyrin
products. As we have previously observed for longer 1,4-
dioxybenzene- and 1,5-dioxynaphthalene-containing tetra-
ethylene strapped-porphyrins,23 and others for related long-
strapped porphyrin derivatives,26 the synthesis resulted in a
mixture in each case of both ‘twisted’ strapped or α,β-isomers
and the atropisomeric ‘over-the-face’ α,α-isomers; in these cases
the α,α-7a and its α,β-isomer 6a in 7% and 15% yields, respec-
tively, and for the longer 3HQ analogue, the α,α-isomeric
porphyrin 12a in 4% yield and its statistically more favoured
α,β-isomer 11a in 14%.
Also of interest are the chemical shift changes of the poly-
ether chain and 1,4-dioxybenzene ring protons for the α,β-iso-
mers 6a and 11a compared to their α,α-isomers 7a and 12a.
However, as expected the differences in resonances between
isomers is not as pronounced in the more flexible longer
strapped molecules in comparison to their shorter strapped
analogues.
1
Significant differences in H NMR chemical shifts between
The two atropisomers in each case could be separated
chromatographically, but each of the individual isomers inter-
converted in refluxing acetonitrile solution by rotation of the
meso-phenyl ring about the C(9)–(10) axis (see Scheme 1 for the
simplified numbering system), to form a 1:5 equilibrium
mixture of 6a and 7a, or a 3:2 mixture of 11a and 12a. Con-
firmation that these two sets of porphyrins are isomeric was
also provided by electrospray mass spectrometry (ESMS) where
both free base isomers 6a and 7a showed peaks at m/z 1182.1
(calcd 1181.6) [M]ϩ, and at 591.6 (calcd 591.3) [M]2ϩ. Similar
results confirmed the isomeric relationship of the porphyrins
containing the three hydroquinone-derived rings 11a and 12a
(peaks at 1450.4 (calcd 1449.7) [M]ϩ, and 725.7 (calcd 725.4)
the two atropisomers which support the structures shown are as
follows:
(i) 2HQ porphyrins 6 and 7
• the hydroquinone-derived (HQ-22) ring proton of 6a at
6.44 ppm is deshielded by ϩ0.07 ppm compared to that in 7a;
• the methylene protons of the polyether strap are shifted
downfield in the α,β-isomer 6a (∆δ ϩ0.18, ϩ0.15, ϩ0.19, ϩ0.19
ppm for H-24, -25, -26 and -27, respectively) compared to the
α,α-isomer 7a, as a result of being deshielded on twisting
around the side of the porphyrin;
• the H-18 and H-19 protons are shifted upfield in 6a com-
pared to 7a (∆δ Ϫ0.04 and Ϫ0.13 ppm, respectively), indicating
shielding by the porphyrin as a result of being held closer to the
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 4 0 9 7 – 4 1 1 2
4098

View Article Online
Scheme 1 Reagents and conditions: (i) p-toluenesulfonyl chloride, Et3N, CH2Cl2; (ii) salicylaldehyde, K2CO3, MeCN; (iii) CCl3CO2H, MeCN–THF,
then o-chloranil, then ZnOAc/MeOH/CH2Cl2 for 1b (iv) CCl3CO2H, Cs2CO3, MeCN–THF, then o-chloranil, then ZnOAc/MeOH/CH2Cl2 for 1b.
porphyrin nucleus as the strap is twisted around from one face
to the other.
(ii) 3HQ porphyrins 11 and 12
from one isomer to the other, (∆δ Ϫ0.04 and ϩ0.02 ppm,
respectively) as expected, whereas the HQ-33 proton is sig-
nificantly deshielded by ϩ0.35 ppm in 11a, indicating that the
HQ-33 ring is held at the side of the porphyrin;
• the HQ-21 and HQ-22 protons are only slightly shifted
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 4 0 9 7 – 4 1 1 2
4099

View Article Online
• the methylene protons in 11a are shifted only slightly
downfield (∆δ ϩ0.08, 0.00, ϩ0.08, ϩ0.08 ppm for H-24 to H-
27, respectively) upon twisting, reflecting the reduced influence
of the porphyrin in the longer strapped analogues compared to
the shorter strapped 6a;
• the H-18 and H-19 protons in 11a are shielded (∆δ Ϫ0.01
and Ϫ0.09 ppm, respectively), but as expected not to the same
extent as for the tighter strapped 6a. On the other hand, the
methylene protons of the strap adjacent to the central HQ-33
ring (H-30 and H-31) are most affected by atropisomerisation
(∆δ ϩ0.17 and ϩ0.30 ppm, respectively, from 12a to 11a),
indicating that these protons are held closest to the side of the
porphyrin ring and hence deshielded.
Interesting comparisons can also be made between the
porphyrin isomers containing 2HQ and 3HQ rings (Fig. 1 and
Table 1). Generally speaking, the methylene strap and HQ
ring protons in the longer strapped derivatives (3HQ) resonate
1
further upfield in the H NMR, especially for the HQ-21 and
HQ-22 protons and their adjacent methylene strap protons
either side of the HQ-21,22 ring. For example, a comparison of
the chemical shifts of the ‘twisted’ isomers 11a and 6a (Table 1
and Fig. 1), shows that the methylene strap protons, H-18,
-19, -24 and -25 (∆δ Ϫ0.13, Ϫ0.56, Ϫ0.30, Ϫ0.15 ppm), as well
as the HQ-21 and HQ-22 ring protons (∆δ Ϫ0.74, Ϫ0.58 ppm)
resonate significantly further upfield in 11a. Similar trends
are also observed for the α,α-isomers 12a and 7a, with the
equivalent methylene strap protons (∆δ Ϫ0.16, Ϫ0.60, Ϫ0.20,
0.00 ppm) and HQ ring protons (∆δ Ϫ0.70, Ϫ0.53 ppm)
resonating further upfield in the longer strapped 12a analogue.
However, in contrast to the above observations, inspection of
Table 1 reveals that the peripheral methyl and ethyl substituents
of the porphyrin, as well as the meso-phenyl ring protons are all
shifted further downfield in the 3HQ porphyrin series in com-
parison to the 2HQ porphyrin isomers. These two opposing
trends indicate that the HQ-21 and HQ-22 rings are held closer
to the peripheral groups in the tighter strapped 2HQ isomers,
whereas in the 3HQ porphyrins the inherent flexibility provided
by the third HQ ring and additional polyether chain allows the
HQ-21 and HQ-22 protons to be more centrally located over
the porphyrin face.
Scheme 2 Self-assembly of a strapped-porphyrin [2]catenane 14b
containing two hydroquinone-derived rings (2HQ) in DMF under
various reaction conditions. Yields improved substantially under high
pressure conditions (12 kbar), and with excess reagents.
one 1,4-dioxybenzene or 1,5-dioxynaphthalene unit.13,15,23 Very
low yields of catenated products were obtained at ambient tem-
perature and pressure, but by use of high pressures27 and using
excess reagents, the yields were increased markedly (Scheme 2).
The 2HQ porphyrin [2]catenane 14b was the only catenated
product identified and isolated when either of the individual
isomeric porphyrins 6b or 7b, or an equilibrated mixture of the
two, was used. From examination of models, it is apparent that
catenation of the ‘twisted’ isomer 6b would be unlikely because
of steric constraints. It is also significant however, that there
was no evidence for a [3]catenane from this reaction, although
molecular modelling suggests that it is sterically feasible,
although unlikely if there were to be π–π interactions involving
the porphyrin and the bipyridinium macrocycle (see below).
Electrospray mass spectrometry (ESMS) for the 2HQ [2]cat-
enane 14b showed a pattern which has come to be considered
indicative of catenated structures,19 where peaks were observed
for the successive loss of one, two, three and four PF6Ϫ counter-
ions, as well as peaks corresponding to the singly and doubly
charged parent porphyrin ions, as a result of the breaking of
the tetracationic macrocycle.
Fig. 1 Schematic representation of the differences in the chemical
shifts of the “twisted” α,β-strapped porphyrins compared to the α,α-
porphyrins: (A) 6a compared to its α,α-strapped atropisomer 7a;
(B) 11a compared to its α,α-strapped atropisomer 12a.
(b) [n]Catenane self-assembly
(i) [2]catenanes from 2HQ porphyrins 6 and 7. Synthesis.
The synthesis of the porphyrin catenanes is shown in Scheme 2
and employs a similar methodology and workup procedure to
that used previously for the porphyrin [2]catenanes containing
The assignment of the 1H NMR resonances of the 2HQ
strapped-porphyrin [2]catenane 14b under conditions of fast or
slow exchange was undertaken by a combination of gradient
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 4 0 9 7 – 4 1 1 2
4100

View Article Online
COSY, gradient NOESY, selective gradient NOE and satura-
tion transfer experiments. However, unlike previous porphyrin
[2]catenanes,15 no NOE interactions were observed between the
protons of the bound 1,4-dioxybenzene and tetracation com-
ponents. This may possibly be due to the extra exchange process
available in 14b, viz ‘shuttling’ or translational motion of the
tetracation between 1,4-dioxybenzene units, in addition to the
‘rocking’ and rotation of the tetracation, exchange processes
also observed in the single hydroquinone-derived catenanes.
Nevertheless, the usual NOE interactions that we have observed
between the protons of the tetracation cyclophane components
in previous porphyrin [2]catenanes were evident here in 14b
and allowed us to unambiguously assign these resonances.14–17
Furthermore, saturation transfer experiments, together with
gradient COSY and NOESY experiments were used to assign
the ‘inside’ and ‘outside’ 1,4-dioxybenzene (HQ) environments
(Table 2) in both (CD3)2CO and CD3CN solution at 30 ЊC.
In addition, four separate signals corresponding to the protons
H-16,16Ј and H-17,17Ј of the polyether chain were observed.
This indicates that shuttling between the two possible 1,4-
dioxybenzene units of this catenane is slow on the NMR time
scale at this temperature, and confirms the structure as that
of a [2]catenane 14b.
Selected 1H NMR resonances for 14b recorded in (CD3)2CO
or CD3CN solution at 30 ЊC are shown in Table 2. For com-
parison, 1H NMR chemical shifts for the tetraethylene strapped
1,4-dioxybenzene porphyrin [2]catenane (1a:n = 3), as well as 1H
NMR data for the BPP34C10-[2]catenane27 1b are included in
Table 2. Like the previously reported porphyrin [2]catenanes,15
the resonances of tetracation cyclophane components in 14b
have an increased chemical shift difference in comparison to 1b,
due to the influence of the large magnetic anisotropy of the
porphyrin macrocycle. Furthermore, as observed for previous15
1
porphyrin [2]catenanes the H NMR resonance shifts for 14b
Fig. 2 Cartoons indicating the possible conformations of the catenane
14b; conformation 14Ј is not supported by NMR evidence (see text).
support a structure in which the bipyridinium moiety, bound
1,4-dioxybenzene and porphyrin subunits are held co-parallel
(Fig. 2). For example, both the β- and α-bipyridinium protons
in (CD3)2CO solution for 14b (∆δ Ϫ1.85 and Ϫ1.09 ppm,
respectively) are more affected by catenane formation then
either the ϩNCH2 or phenylene (–C6H4–) protons (∆δ Ϫ0.84
and Ϫ0.70 ppm, respectively); similar trends are also observed
in CD3CN solution at 30 ЊC (Table 2). As before, this implies
that both the β- and α-bipyridinium protons of the bi-
pyridinium subunit are held closest to the porphyrin, and are
therefore most affected by the porphyrin ring current. Also,
the larger chemical shift difference experienced by the β-bi-
pyridinium protons upon catenane formation reflects the more
centralised position of these protons above the porphyrin.
Furthermore, the changes in chemical shift for the tetra-
cation and 1,4-dioxybenzene units upon catenation are of a
similar magnitude to that observed for (1a:n = 3) (Table 2), with
the exception of the phenylene (–C6H4–) resonance, which is
shifted further upfield in 14b (∆δ Ϫ0.70 ppm) compared to
(1a:n =3) (∆δ Ϫ0.33 ppm), and reflects the extra shielding
influence of the second hydroquinone-derived (HQ) unit in 14b.
Also, the change in chemical shift for the HQ (H-21 and -22)
ring protons in 14b for the ‘inside’ or bound protons (∆δ Ϫ3.27
and Ϫ3.42 ppm, respectively) in (CD3)2CO solution are of a
similar magnitude to those in both (1a:n =3) (∆δ Ϫ3.12 ppm)
and 1b (∆δ Ϫ3.01 ppm), further supporting the catenated struc-
ture. The downfield shifts observed for ‘outside’ or unbound
HQ-21 and HQ-22 ring protons on catenation (∆δ ϩ0.58
and ϩ0.29 ppm, respectively), as opposed to upfield shift in the
‘outside’ or ‘alongside’ HQ protons in 1b (∆δ Ϫ0.55 ppm),
imply that the second HQ unit undergoes a conformation
induced change in which this ring is moved away from the
effects of the porphyrin to allow π–π interactions to occur
between the bipyridinium moiety, the first 1,4-dioxybenzene
and the porphyrin subunits, as indicated in Fig. 2. The pattern
of chemical shifts clearly does not support an alternative struc-
ture 14Ј which would require close approach of the phenylene
groups on the macrocycle to the porphyrin ring.
Dynamic 1H NMR. Several different exchange processes were
identified for 14b in solution over a wide temperature range,
and selected 1H NMR chemical shifts under conditions of
both fast and slow exchange are collected in Table 3. The data
were evaluated using the coalescence method28 to determine the
kinetic and thermodynamic parameters (Table 4).
Further evidence for slow translational isomerism at 30 ЊC in
either CD3CN or (CD3)2CO solution for the 2HQ porphyrin
[2]catenane 14b (besides the distinct ‘inside’ and ‘outside’ HQ
resonances) were signals corresponding to the H-16,16Ј protons
of the polyether strap, two singlets for the methyl groups of the
porphyrin periphery (H-7), and two doublets corresponding
to the diastereotopic ϩNCH2 protons of the tetracationic
macrocycle.
On raising the temperature in CD3CN solution all these
resonances began to broaden and at 70 ЊC the HQ resonance
had disappeared; no peaks corresponding to the fast exchange
limit were observed for the HQ ring protons. Nevertheless, by
monitoring the coalescence of the porphyrin peripheral methyl
(H-7) and H-16,16Ј and ϩNCH2 resonances, the translational
motion of the tetracation was found (Table 4) to have kc = 23 sϪ1
at Tc of 50 ЊC in CD3CN (∆ν 10.4 Hz) with a ∆G ‡c of 70.7 kJ
molϪ1, which converts to around 1.3 times per second at 25 ЊC.
Hence the rate of translational motion between HQ ‘stations’ in
this catenane is slow, and is about two orders of magnitude
slower than that found by Stoddart8,27 for the BPP34C10-[2]-
catenane 1b of ca. 120 sϪ1 at ambient temperature (∆G ‡c 65.3 kJ
molϪ1), and reflects both the increased activation barrier for
translational motion imposed by the presence of the porphyrin
and the statistical difference arising from restriction to a back-
and-forth mode in these catenanes vs. full-circle possibilities in
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 4 0 9 7 – 4 1 1 2
4101

O
Table 2 Selected 1H NMR resonances (δ ppm) and chemical shift differences (∆δ ppm) for the 2HQ porphyrin [2]catenane 14b, isomeric 3HQ α,β-16b and α,α-15b [2]catenanes, and isomeric 3HQ α,β-18b and α,α-17b
[3]catenanes
α
β
–C6H4–
ϩNCH2
outside HQ-21
outside HQ-22
inside HQ-21
inside HQ-22
HQ-33
Macrocyclic tetracationa
9.38
8.85
9.28
(Ϫ0.10)
8.28
8.58
7.66
8.18
(Ϫ0.40)
6.66
7.76
7.79
8.04
(ϩ0.28)
7.43
(Ϫ0.33)
6.15
5.67
6.01
(Ϫ0.14)
5.29
(Ϫ0.86)
Macrocyclic tetracationb
1b (BPP34C10 [2]catenane)a
6.22
(Ϫ0.55)
3.76
(Ϫ3.01)
3.00
(∆δ)c
1:n = 3d
(∆δ)c
7be
(Ϫ1.10)
(Ϫ1.92)
(Ϫ3.12)
5.98
6.56
(ϩ0.58)
6.68
6.44
6.73
(ϩ0.29)
6.91
14bf
8.29
(Ϫ1.09)
7.71
6.73
(Ϫ1.85)
5.99
7.06
(Ϫ0.70)
7.10
5.31
(Ϫ0.84)
5.03/4.77
(Ϫ0.64/–0.90)
2.71i
(Ϫ3.27)
2.36
3.02i
(Ϫ3.42)
2.69
(∆δ)c
14bg
(∆δ)c
11be
(Ϫ1.14)
(Ϫ1.67)
(Ϫ0.69)
(ϩ0.70)
5.34
6.22
(ϩ0.47)
6.06
6.22
(Ϫ3.62)
(Ϫ3.75)
6.88
6.32k/5.83k
(Ϫ0.56)/(Ϫ1.05)
6.60
6.94k
(ϩ0.34)
—
16bh
(∆δ)c
12be
8.19
(Ϫ1.19)
6.44
(Ϫ2.14)
7.56
(Ϫ0.20)
5.36
(Ϫ0.79)
3.44j
3.58j
(ϩ0.78)
5.48
6.85
(ϩ0.16)
6.06
7.05
(Ϫ1.90)
(Ϫ2.48)
15bf
8.12
(Ϫ1.26)
8.70
(Ϫ0.68)
8.63
6.08
(Ϫ2.50)
7.24
(Ϫ1.34)
7.17
7.32
(Ϫ0.44)
7.64
(Ϫ0.02)
7.61
5.36
(Ϫ0.79)
5.49
(Ϫ0.66)
5.49
2.58i
(Ϫ2.90)
2.68i
(Ϫ3.38)
(∆δ)c
18bh
(∆δ)c
17bh
(∆δ)c
(ϩ1.37)
(ϩ0.99)
—
—
—
—
—
—
—
—
—
(Ϫ0.75)
(Ϫ1.41)
(Ϫ0.05)
(Ϫ0.66)
a Macrocyclic tetracation is cyclobis(paraquat-4,4Ј-biphenylene); data is from Stoddart et al.19 and was recorded in (CD3)2CO. b Data is from Stoddart et al.27 and was recorded in CD3CN. c (∆δ) represents the change in
chemical shift on formation of the [2]catenane from its component parts, viz. macrocyclic tetracation and the corresponding crown ether, 1b (BPP34C10), or strapped porphyrin containing 1HQ (1: n = 3), 2HQ 7b, or
3HQ (α,β-11b and α,α-12b) units. d Spectrum recorded in (CD3)2CO at 25 ЊC. e Recorded in CDCl3 at 30 ЊC. f Recorded in (CD3)2CO at 30 ЊC. g Recorded in CD3CN at 30 ЊC. h Recorded in (CD3)2SO at 60 ЊC; ‘inside and
‘outside’ HQ resonances not assigned under slow exchange conditions, but identified at 90 ЊC under fast exchange, see text. i Identified by a saturation transfer experiment performed at 30 ЊC in (CD3)2CO. j Identified by
gradient NOESY and COSY experiments performed at 30 ЊC in (CD3)2CO. k Refers to the ‘outside’ HQ-33 protons – the ‘inside’ HQ-33 protons hidden at 30 ЊC; ‘inside’ HQ-33 protons observed at δ 4.11 and 4.34 ppm at
70 ЊC in CD3CN.

View Article Online
Table 3 Selected proton chemical shift values (δ ppm) for the 2HQ porphyrin [2]catenane 14b under conditions of both fast and slow exchangea
1H
Fast rotation, CD3CN, 343 K
Slow rotation,b (CD3)2CO, 253 K
∆δc
α
β
7.75
6.17
‘outside’ 8.53
‘inside’ 8.07
‘outside’ 6.88
‘inside’ 6.19
7.20
0.46
0.69
–C6H4–
7.08
4.89
—
0.33
ϩNCH2
5.41
5.08
d
HQ-21
–
‘outside’ 6.75
‘inside’ 2.87e
‘outside’ 6.52
‘inside’ 2.64e
4.82
4.68
2.80
3.88
3.88
0.14
0.16
d
HQ-22
–
OCH2 (H-16)
Me (H-7)
4.58
2.62
2.64
a Spectra were recorded at 300 MHz using the appropriate residual solvent as reference. b This is not the ‘slowest’ exchange limit, with a further
splitting of the α resonances observed at temperatures below 208 K, which is attributed to the ‘rocking’ process observed in 1,4-dioxybenzene
catenanes (see Table 4 and text) c Chemical shift differences (∆δ ppm) between protons in slow exchange environments. d Peaks disappeared at this
temperature and were not identified. e Identified by saturation transfer techniques in (CD3)2CO at 30 ЊC.
Table 4 Kinetic and thermodynamic parameters from the dynamic 1H NMR studies on the porphyrin [2]catenanes containing one (1a:n = 3), two
(14b) and three (16b and 15b) 1,4-dioxybenzene rings
Catenane
Probe protons
Solvent
∆ν/Hz
Tc/ЊC
kc/sϪ1
Rotation rates at 25 ЊC/Hz
∆Gc‡/kJ molϪ1
1a:n = 3b
14bc
α-CH
(CD3)2CO
CD3CN
CD3CN
62
31
31.5
10.4
108
38
Ϫ80
65
55
50
3
Ϫ65
70
70
138
69
70
23
241
85
3.4 × 105
5.2 × 10Ϫ1
2.7
39
71
69
71
55
44
63
62
ϩNCH2
OCH2 (H-16)
Me (H-7)
α-CH
14bc
14bc
CD3CN
1.3
14bb
(CD3)2CO
(CD3)2CO
CD3CN
7.0 × 102
9.3 × 104
3.5 × 101
4.9 × 101
14bd
α-CH
16bc, e
15bc, e
HQ-33f
HQ-33f
613
828
1926
2601
CD3CN
a Data obtained by the coalescence method27 as previously reported.15 b Refers to the spinning process of the tetracation around the 1,4-dioxybenzene
unit and involves exchange of the bipyridinium units between ‘inside’ and outside’ environments. c Refers to the ‘threading’ or translational motion
of the tetracation between the 1,4-dioxybenzene units. d Refers to ‘rocking’ process of the 1,4-dioxybenzene unit between different orientations
within the tetracation e Data obtained by the exchange method as described by Stoddart27 f Assignment of the ‘inside’ and ‘outside’ environments by
saturation transfer experiments was not unequivocal in 15b, and not observed in 16b (see text).
the non-porphyrinic analogues. Nevertheless, it indicates that
translational motion in this catenane is either sterically
restricted because of the short diethylene strap linking the
porphyrin and the 2HQ rings, or may reflect a stronger π–π
interaction between the tetracation, bound HQ ring, and
porphyrin moieties, or a combination of both.
and disappear from the 1H NMR spectrum at ca. Ϫ50 ЊC, as do
the ‘inside’ HQ resonances. At temperatures below Ϫ60 ЊC the
broadened α-bipyridinium resonances begin to reappear and
split into two equal intensity ‘outside’ peaks (the two ‘inside’
α-bipyridinium resonances were hidden under several Ar–H
protons at Ϫ85 ЊC at ca. 7.8 ppm) at temperatures below Ϫ65
ЊC before starting to sharpen again. These results may be inter-
preted as a slowing down of the ‘rocking’ process of the ‘inside’
1,4-dioxybenzene ring between two different orientations
within the tetracation, leading to the further observable
splitting in the α-bipyridinium resonances (∆δ 0.15 ppm) and
observed substantial broadening of the –C6H4– and ϩNCH2
peaks at Ϫ85 ЊC. This process was found (Table 4) to have kc
= 85 sϪ1 at Tc of Ϫ65 ЊC in (CD3)2CO (∆ν 38 Hz) with a ∆G ‡c of
44 kJ molϪ1, which equates to around 9.3 × 104 reorientations
per second of the ‘inside’ HQ ring at 25 ЊC.29
At lower temperatures in (CD3)2CO solution two other
dynamic processes were also observed. At temperatures below
30 ЊC in the 1H NMR spectrum both the α- and β-bipyridinium
resonances of the tetracation began to broaden, and then split
into two equal intensity peaks at temperatures below ca. 0 ЊC.
This is interpreted as an interchange of the bipyridyl rings (C
and D in Fig. 2) between the ‘inside’ and ‘outside’ environ-
ments,8,27 a process which can be considered as rotating of the
tetracation about an axis created by the 1,4-dioxybenzene ring.
This process was found (Table 4) to have kc = 241 sϪ1 at Tc of 3
ЊC in (CD3)2CO (∆ν 108 Hz) with a ∆G ‡c of 55 kJ molϪ1, which
equates to ca. 700 rotations per second at 25 ЊC. This value for
the rotation of the tetracation is significantly smaller than that
found for the tetraethylene strapped-porphyrin [2]catenane
containing one HQ ring (1a: n = 3) (Table 4) (3.4 × 105 Hz), and
the non-porphyrinic BPP34C10 [2]catenane8,27 1b (2000 Hz),
but is larger than the diethylene strapped-porphyrin [2]catenane
containing one HQ ring (1a:n = 1) (50 Hz).15 These trends indi-
cate that the short diethylene straps linking the porphyrin and
the 2HQ rings in 14b cause a degree of steric restriction, despite
the presence of the longer tetraethylene strap between the 2HQ
rings.
(ii) [2]catenanes from 3HQ porphyrins 11 and 12. Under
similar reaction conditions to those used for the catenane
formation of 14 from 2HQ porphyrins 6 and 7, and using the
equilibrated mixture of porphyrins containing three hydro-
quinone-derived rings (3HQ) 11a and 12a, a total of four
catenated products was identified. NMR analysis indicated that
these were formed in pairs in the same ratios (approximately
3:2) as the equilibrated mixture of starting porphyrins; they
were subsequently identified as two [2]catenanes α,β-15b and
α,α-16b, and two [3]catenanes α,β-18b and α,α-17b as shown in
Scheme 3, in 21 and 6% overall yield, respectively.30 Subsequent
reactions on the isolated single isomers 11a and 12a indeed
resulted in only two products in each case, the corresponding
Secondly, at temperatures below ca. Ϫ30 ЊC, the already split
α- and β-bipyridinium resonances begin to broaden further still
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 4 0 9 7 – 4 1 1 2
4103

View Article Online
Scheme 3 Self-assembly of the isomeric strapped-porphyrin [2]catenanes 15 and 16, and isomeric strapped-porphyrin [3]catenanes 17 and 18,
containing three hydroquinone-derived rings (3HQ). Reagents and conditions: DMF, 12 kbar, 6 days, then ZnOAc/MeOH/H2O/acetone, followed by
NH4PF6/H2O.
atropisomeric [2] and [3]catenanes. Furthermore it was estab-
lished that none of the isolated catenanes interconverted either
under catenation conditions or subsequently during heating for
several hours in refluxing acetonitrile.
Fig. 3 shows the spectrum of the [2]catenane 15b compared to
its corresponding starting porphyrin 12b; the non-systematic
numbering system used for proton assignments follows that in
Scheme 1.
FAB-MS of the individual isomers or mixtures of the 3HQ
[2]catenanes (16b and 15b) and [3]catenanes (18b and 17b) were
indicative of their proposed structures (Table 5). For the iso-
meric [2]catenanes (16b and 15b), peaks corresponding to the
successive loss of each counterion and the parent porphyrin ion
peak were observed, as well as a small intensity peak corre-
sponding to the parent catenane ion. For the isomeric [3]cat-
enanes (18b and 17b), peaks were observed for the successive
Catenation is indicated by a shielding (0.2ppm) of the
meso-hydrogen compared to the starting porphyrin 12b.15 The
resonances of the tetracationic moiety are shifted upfield com-
pared to those of the original tetracation. Of significant interest
are the shifts of the 1,4-dioxybenzene subunits (particularly
positions H-21 and H-22 and H-21Ј and H-22Ј on the side of
the strap) in the catenated compound 15b, and particularly the
fact that these show two distinct environments. The typical AB
quartet resonances which are downfield (7.0 and 6.81 ppm
respectively) are deshielded compared to those of the starting
porphyrin 12b and are assigned to H-21 and H-22. With the
aid of gradient NOE experiments, the protons for the 1,4-
dioxybenzene unit on the other side of the porphyrin (H-21Ј
and H-22Ј) were found at ca. 2.69 and 2.5 ppm which are
Ϫ
loss of one to five PF6 counterions, as well as peaks corre-
sponding to the loss of one tetracation together with successive
loss of five to seven counterions. In addition, a peak corre-
sponding to the parent porphyrin was also observed due to
the unlinking of both tetracations and eight counterions, con-
firming the existence of the isomeric porphyrin [3]catenanes.
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 4 0 9 7 – 4 1 1 2
4104

View Article Online
Table 5 Positive ion FAB mass spectrometry selected peaks for the
catenanes 15b, 16b, 17b, and 18b
Peak
15b
16b
17b
18b
[M]ϩ
2613.1
2467.5
2323
2177.7
—
1511.2
—
—
—
1511.2
2612.5
2468
2323
2177.9
—
1511.3
—
—
—
1511.3
3712.6
3567.2
3422.9
3279.6
3132.8
—
2987.8
2467.8
2323
3712.4
3567.7
3423.2
3279.3
3133.3
—
2988.4
2467.5
2323
[M–PF6]ϩ
[M-2PF6]ϩ
[M-3PF6]ϩ
[M-4PF6]ϩ
[M-tetracation-4PF6]ϩ
[M-5PF6]ϩ
[M-tetracation-5PF6]ϩ
[M-tetracation-6PF6]ϩ
[Parent porphyrin]ϩ
1511.4
1511.7
Fig. 3 1H NMR spectra of the porphyrin 12b compared to its
[2]catenane 15b in acetone-d6 at 303 K.
significantly shielded compared to those in 12b. Thus, at 30 ЊC
the tetracationic moiety is ‘shuttling’ between the two side 1,4-
dioxybenzene units at a slow enough rate to observe the two
environments on the NMR timescale at this frequency.
The downfield shifts for the ‘unbound’ protons H-21 and H-
22 imply that this 1,4-dioxybenzene ring undergoes a change in
conformation which reduces or counters the original shielding
effect of the porphyrin in 12b to allow π–π interactions between
the bipyridinium moiety, the ‘bound’ 1,4-dioxybenzene unit
(H-21Ј, H-22Ј) and the porphyrin, in a conformation as indi-
cated in Fig. 4. The slow translational motion observed is
further supported by the inequivalence of the ethylene glycol
resonances H-16 to H-24, with H-16 having the smallest
difference in shift. The largest chemical shift difference is
seen for H-24 which indicates that the tetracationic macro-
cycle sits closer to the top side (over H-22Ј) of this 1,4-dioxy-
benzene unit. The central 1,4-dioxybenzene (B) unit is also
shielded (∆δ 0.5 ppm), consistent with such a conformation.
These results do not support an alternative structure in which
the HQ-33 ring (ring B) is bound within the cavity of the tetra-
cation and sandwiched between the two ‘side’ HQ rings
(rings A), and held orthogonal to the porphyrin (structure 15Ј in
Fig. 4).
Fig. 4 Conformation of [3]catenane 15 as deduced from NMR
spectra; an alternative possible structure 15Ј, in which the HQ-33 ring
(ring B) is bound within the cavity of the tetracation and sandwiched
between the two “side” HQ rings (rings A), and held orthogonal with
the porphyrin, is not supported by the NMR evidence.
The broadness of the α- and β-bipyridinium resonances
indicates that the tetracationic macrocycle is rotating as well as
shuttling. At 30 ЊC this process is fast on the NMR time scale
as there is no evidence for inside and outside environments
for these resonances. The multiplicity of the Nϩ-methylene res-
onance is a commonly observed feature of related porphyrin
catenanes, resulting from diastereotopicity of these protons.15
Fig. 5 1H NMR spectra of the porphyrin 11b compared to its
[2]catenane 16b in acetone-d6 at 303 K.
1
The H NMR spectra of the twisted [2]catenane 16b and its
precusor porphyrin 11b are shown in Fig. 5. Although the peaks
are broadened at 30 ЊC, it was apparent from the exchange cross
peaks that the ‘shuttling’ process observed for 15b is also slow
enough to be observed in this catenane at 30 ЊC. Interestingly,
the 1,4-dioxybenzene protons H-33 do not shift on formation
of the [2]catenane 16b, indicating that this subunit’s proximity
to the edge of the porphyrin is little changed.
Dynamic NMR. At higher temperatures, both [2]catenanes
undergo fast exchange in both the translational and rotational
motions of the tetracationic macrocycle. Fig. 6 shows the
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 4 0 9 7 – 4 1 1 2
4105

View Article Online
meso-hydrogen (∆δ 0.2 ppm) also indicates catenation, and the
sharpness of the peripheral ethyl H-7 and methyl protons H-5,
and the ethylene glycol resonances, is indicative of higher sym-
metry in the structure compared to the [2]catenane 16b.
Fig. 7 indicates a dynamic process expected for the [3]-
catenane where the two macrocycles shuttle back and forth,
exchanging between the 1,4-dioxybenzene units A and B.
The macrocycles may move alternately or in tandem along the
polyether chain. The upfield shift of the 1,4-dioxybenzene pro-
tons H-33 (B) into the region of δ 4.0–3.6 ppm is significant and
indicates that this moiety is preferentially located ‘inside’ the
macrocyclic cavity. This is also supported by the net deshielding
effect observed for the α (∆δ 0.55 ppm) and β bipyridinium
(∆δ 0.91 ppm) protons of the macrocycle compared to those
in the [2]catenane 16b because they, in turn, would experience
more deshielding effects on average from the edge of the
porphyrin subunit.
Fig. 6 1H NMR spectral comparison of the [2]catenanes 15b and 16b
in DMSO-d6 at 363 K.
comparative shifts of the protons for both 15b and 16b. It is
apparent that at fast exchange, the difference in resonance shifts
of both spectra have been greatly reduced, indicating similar
rates of exchange (both translational and rotational) for both.
Nevertheless, the 1,4-dioxybenzene protons at position H-33
in 16b are significantly deshielded (∆δ Ϫ0.49 ppm) compared
to those of 15b, accounting for their location at the edge of the
porphyrin.
There are several dynamic processes observed in both of
these [2]catenanes: (i) translational motion or ‘shuttling’
between the 1,4-dioxybenzene rings (A and C), with perhaps
an intermediate ‘stopover’ on the central 1,4-dioxybenzene (B)
subunit; (ii) tetracation rotation; and (iii) ‘rocking’ of the 1,4-
dioxybenzene units. Because of this complexity, only the trans-
lational motion of the tetracationic macrocycle in compound
15b between 1,4-dioxybenzene ‘stations’ (A) and (C) could be
estimated by using the H-16,16Ј protons and the coalescence
method (Table 4). For 15b a k of 2601 sϪ1 at T of 70 ЊC in
CD3CN (∆ν 828 Hz) with a ∆G ‡ of 62 kJ molϪ1 was calculated,
which equates to around 50 translations per second at 25 ЊC,
and which is on par with the value determined by Stoddart for
the translational motion for the BPP34C10-based [2]catenane
1b of 120 sϪ1 with a ∆G ‡c of 65.3 kJ molϪ1. Likewise, evaluation
of the peak separation observed in 16b gave similar values, with
k = 1926 sϪ1 at T of 70 ЊC in CD3CN (∆ν 613 Hz) with a ∆G ‡
of 63 kJ molϪ1, which converts to around 35 translations per
second at ambient temperature.
The rate is also about one order of magnitude faster than
that calculated for the [2]catenane 14b, indicating that the trans-
lational motion in the [2]catenane 15b is less sterically restricted
than in 14b. In comparison to the Stoddart system 1b however,
it may reflect a stronger π–π interaction between the tetra-
cation, bound 1,4-dioxybenzene ring and porphyrin moieties or
a combination of all factors.
(iii) [3]catenanes from 3HQ porphyrins 11 and 12. The
[3]catenane 18b was also isolated from the reaction mixture of
catenation of porphyrin isomer 16b. When this was re-subjected
to the catenation reaction conditions, no further change was
observed, confirming that the porphyrin does not atropisomer-
ise during the catenation reaction. Likewise, on prolonged heat-
ing above 90 ЊC in DMSO, no interconversion was observed,
but some decomposition was evident at higher temperatures
(>130 ЊC), resulting in the formation of an equilibrated mixture
of the uncatenated porphyrins.
Fig. 7 Cartoon indicating the tandem or alternate shuttling which is
possible in the [3]catenane 18.
A clearer picture of the possible dynamics of 18b can be
gained by comparing it to the analogous [2]catenane 16b at
higher temperature (Fig. 8). The similarities of the 1H NMR of
18b and 16b are quite obvious, although there are some impor-
tant differences. Firstly, the α and β bipyridinium resonances
At 30 ЊC the spectrum was not as broad as observed in 16b,
and integration of the peaks for the bipyridinium macrocycle
confirmed a [3]catenane. The diagnostic upfield shift of the
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 4 0 9 7 – 4 1 1 2
4106

View Article Online
Electronic spectra
UV-vis spectra of the 2HQ [2]catenane 14b and 3HQ [2]cat-
enanes (16b and 15b) and [3]catenanes (18b and 17b) showed
the usual red shifts (∆λ ∼ 8 nm) observed before in previous15
porphyrin [2]catenanes. This indicates some form of electronic
interaction between the porphyrin and tetracation subunits,
despite the fact that any π–π charge transfer absorptions that
may be present would be hidden under the large Soret absorp-
tion band of the porphyrin. However, the Soret and Q bands of
the [3]catenane isomers were very broad with the usual Q band
absorption at ca. 575 nm observed in all previous catenanes
appearing as a very broad shoulder, clearly a direct effect of the
second macrocycle.
Conclusions
A series of porphyrins strapped with polyether chains contain-
ing two or three 1,4-dioxybenzene units has been synthesised
with a view to the production of porphyrin-containing [2]
and [3]catenanes, where the porphyrin is strapped between
ortho-positions of 5,15-(meso)-diaryl groups, and is interlinked
with the bipyridinium macrocycle cyclobis(paraquat-4,4Ј-bi-
phenylene). The porphyrins were isolated as mixtures of atrop-
isomers, where the linking strap spans across the face of the
porphyrin (α,α-isomer), or ‘twisted’ around its side (α,β-iso-
Fig. 8 1H NMR spectra of the [2]catenane 16b compared to its
[3]catenane counterpart 18b at 363 K in DMSO-d6.
are deshielded in the [3]catenane 18b in comparison to the
[2]catenane 16b by 0.35 and 0.5 ppm, respectively. This can be
attributed to the translational motion where there are now
two charged macrocycles on the strap which means that these
protons now have a longer time-averaged occupancy near the
edge of the porphyrin encircling 1,4-dioxybenzene (B) and
hence there is a net deshielding effect.31
1
mer). Their structures have been determined by detailed H
NMR spectroscopy.
The bis-(1,4-dioxybenzene)-strapped derivatives have been
shown to undergo atropisomerisation on heating, to produce
an equilibrium mixture. Catenation under high pressure con-
ditions of the mixture, or of the individual isomers, produced
only a single catenane, that of the α,α-isomer. Its dynamics were
measured, and rates were determined for (i) translational
motion or ‘shuttling’ between 1,4-dioxybenzenes; (ii) ‘rotation’
of the macrocycle around the hydroquuinol axis; and (iii)
‘rocking’ of the 1,4-dioxybenzene within the macrocycle.
The atropisomers of the strapped derivatives containing
three 1,4-dioxybenzene units were also separated, and subjected
to catenation. Both [2]- and [3]catenanes have been isolated,
and shown to be stable to further atropisomerisation. Their
solution structures were probed in detail by dynamic 1H NMR
measurements. The rates for shuttling and rotation were
obtained in certain cases, although the complexity of the spec-
tra of the [3]catenanes precluded a more detailed investigation.
Thus we have demonstrated the viability of the synthesis of
these porphyrin catenanes, in which the porphyrin subunit pro-
vides an ideal trigger for control of the dynamics, as it can
be addressed by chemical, photochemical, or electrochemical
stimuli. In this way, the assemblies can be regarded as molecular
machines, in which mechanical motion can be controlled in a
systematic and predictable manner. The way is now clear for the
next generation of experiments.
Conversely, the 1,4-dioxybenzene protons H-33 are shielded
(∆δ 0.8 ppm) which is as expected; the same translational
motion of the two macrocycles also means that in fast exchange
this subunit (B) would experience more time ‘inside’ one of the
two charged cavities as demonstrated by Fig. 7. However, due to
the presence of several processes (tetracation rotation, ‘rocking’
of the 1,4-dioxybenzene units and two possible modes of trans-
lational motion) and the broadness of the resonances at low
temperatures, we were unable to evaluate exchange rates for this
[3]catenane.32
Unfortunately, the second [3]catenane 17b could not be iso-
lated pure from the reaction mixture, and it was always con-
taminated with non-porphyrinic polymeric material. This,
1
together with broadened H NMR spectra, has prevented a
full characterisation and evaluation of kinetic parameters.
It was also clear that this catenane was less stable than its
twisted isomer 18b, and decomposition was observed when
heated above 90 ЊC in DMSO solution, although no atropiso-
merisation to 18b was observed. Mass and visible spectra were
essentially identical to 18b.
Nevertheless, it was clear that there are certain NMR spectral
similarities and critical differences to the twisted isomer 18b.
At 30 ЊC in CD3CN solution, broad resonances were observed
corresponding to the HQ-33 (ring B) protons at δ 6.35 and 5.77
ppm for 17b, compared to δ 6.94 and 6.86 ppm for the other
isomer 18b, indicating an exchange process which is slow on the
NMR time scale at this temperature. The more upfield position
(ie less deshielding of these protons in 17b compared to 18b)
is consistent with the structures indicated, where the ‘outside’
HQ 33 would be deshielded in the twisted isomer 18b compared
to 17b. On raising the temperature to 70 ЊC, these protons
began to sharpen into one singlet in each case at δ 6.38 (17b)
and 6.87 ppm (18b), although the resonance for 17b was still
somewhat broad. In addition, saturation transfer experiments
at 70 ЊC revealed an ‘inside’ environment for these protons at
δ 4.34 ppm (cf. 4.11 for 18b), and this allowed an estimation of
the translational motion, which was similar to that for the
equivalent process in 18b. It was also noticeable that the signals
due to the methyl and ethyl groups on the porphyrin periphery
were more symmetrical at lower temperatures than those of the
twisted isomer 18b.
Experimental
All solvents were distilled before use, using standard pro-
cedures: tetrahydrofuran (THF) was distilled over benzo-
phenone and sodium under N2; triethylamine (Et3N) was dis-
tilled over CaH2; dimethylformamide (DMF) was dried over
type 4 Å molecular sieves. Column chromatography used
Aldrich silica gel (grade 9385, 230–400 mesh). Preparative TLC
was performed on 20 × 20 cm plates coated with 0.5 mm thick
Art. 7731 Kieselgel 60 G Merck silica. Analytical TLC was
carried out on Merck Silica Gel 60 G254 pre-coated aluminium
sheets.
1H NMR spectra were acquired using a 300 MHz Bruker
Avance 300 spectrometer at 303 K, unless otherwise stated.
Chemical shifts (δ) are reported in parts per million relative to
residual solvent. Coupling constants (J) are reported in Hz.
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 4 0 9 7 – 4 1 1 2
4107

View Article Online
Deuterated solvents were purchased from Aldrich and stored
over type 3 Å molecular sieves after opening. COSY-45,
gradient COSY, one-bond C–H correlation (HMQC), long-
range C–H correlation (HMBC), NOESY, gradient NOESY
two-dimensional NMR experiments employed the standard
Bruker parameters. DEPT, NOE difference and saturation
transfer experiments, as well selective gradient NOE experi-
ments utilised standard Bruker pulse programs.
UV-Vis spectra were recorded on a Varian Cary IE UV-VIS
spectrophotometer. Melting points were determined using a
Reichert microscopic hot-stage apparatus.
FAB and ESI mass spectrometry was carried out by CSIRO
Molecular Science at the Ian Wark Laboratory, Clayton and the
Australian National University, Canberra, and high resolution
ESI was performed at the Centre for Molecular Architecture,
Rockhampton.
amounts of CCl3COOH (ca. 60 mg) added, and the solution
bubbled with N2 again for 3 min. The reaction mixture was then
stirred for 5 h (RT, N2, dark), before adding o-chloranil (0.33 g,
1.34 × 10Ϫ3 mol) in THF (11 mL) all at once and stirring was
continued overnight (RT, dark). Then added Et3N (2 mL) and
stirred for 30 min, before taking the solution to dryness by
rotary evaporation. An initial purification was carried out on
an alumina column using CH2Cl2and then CH2Cl2/Et2O (2%)
as the eluent to obtain several porphyrin fractions. Further
purification was carried out by column chromatography (silica),
eluting initially with CH2Cl2/Et2O (2%) to remove impurities
followed by CH2Cl2/Et2O (5%) to obtain the porphyrin isomers
6a and 7a. A final purification on a fraction containing a
mixture of 6a and 7a was carried out by preparative TLC plates
using CH2Cl2/Et2O (10%) as the eluent.
6a was recrystallised to obtain a purple/red solid (117 mg,
15%), mp 180–183 ЊC (from CH2Cl2/MeOH) (Found: C, 72.48;
H, 7.21; N, 4.66. C72H84N4O11.H2O requires C, 72.09; H, 7.23;
N, 4.67%); m/z (ESMS) [M ϩ H]ϩ 1182.1 (calcd. 1181.6), [M ϩ
2H]2ϩ 591.6 (calcd. 591.3); UV (λ nm, CHCl3) 410, 507, 541,
574, 626; δH (300 MHz; CDCl3, 298 K) 10.14 (2H, s, H-meso),
7.77 (2H, dd, J 7, 2,Ar–H), 7.74 (2H, dt, J 8, 2, Ar–H), 7.36
(2H, t, J 7, Ar–H), 7.29 (2H, d, J 8, Ar–H), 6.44 (4H, dd, J 7, 2,
Ar–H), 5.94 (4H, dd, J 7, 2, Ar–H), 4.15 (4H, t, J 4, OCH2),
4.01–3.84 (16H, m, CH2CH3(8H), OCH2(8H)), 3.80 (8H, t, J 4,
OCH2), 3.20 (4H, t, J 4, OCH2), 2.54 (12H, s, CH3), 2.19 (8H, s,
OCH2), 1.70 (12H, t, J 8, CH3CH2), 1.26 (2H, s, H2O), Ϫ2.36
(2H, br s, pyrrole NH).
7a was recrystallised to obtain a purple/red solid (55 mg,
7%), mp 184–186 ЊC (from CH2Cl2/MeOH) (Found: C, 71.85;
H, 7.37; N, 4.43. C72H84N4O11.H2O requires C, 72.09; H, 7.23;
N, 4.67%); m/z (ESMS) [M ϩ H]ϩ 1182.1 (calcd. 1181.6), [M ϩ
2H]2ϩ 591.6 (calcd. 591.3); UV (λ nm, CHCl3) 410, 507, 542,
575, 626; δH (300 MHz; CDCl3, 298 K) 10.17 (2H, s, H-meso),
7.79–7.74 (4H, m Ar–H), 7.35 (2H, t, J 7, Ar–H), 7.30 (2H, d,
J 8, Ar–H), 6.37 (4H, d, J 9, Ar–H), 5.94 (4H, d, J 9, Ar–H),
4.16 (4H, t, J 4, OCH2), 3.98 (8H, m, CH2CH3), 3.82 (4H, m,
OCH2), 3.70 (4H, m, OCH2), 3.61 (8H, s, OCH2), 3.12 (4H, t,
J 4, OCH2), 2.56 (12H, s, CH3), 2.32 (4H, m, OCH2), 2.23 (4H,
m, OCH2), 1.75 (12H, t, J 8, CH3CH2), 1.25 (2H, s, H2O), Ϫ2.34
(2H, br s, pyrrole NH).
1,11-Bis{4-[2-(2-(toluene-p-sulfonyl)ethoxy)ethoxy]phenoxy}-
3,6,9-trioxaundecane 3
1,11-Bis{4-[2-(2-hydroxyethoxy)ethoxy]-phenoxy}-3,6,9-triox-
aundecane7 2 (4.64 g, 8.37 mmol) was dissolved with stirring
in dry CH2Cl2 (40 mL) and Et3N (2.12 g, 20.9 mmol). Then
toluene-p-sulfonyl chloride (3.99 g, 20.9 mmol) and Et3N
(1.27 g, 12.6 mmol) in dry CH2Cl2 (40 mL), were added drop-
wise over 20 min, and the resulting solution was stirred at
room temperature for 40 h. The solution was then poured
into crushed ice and diluted with H2O. The organic layer was
separated, washed (2% HCl solution and H2O), and dried
(MgSO4). Purification was carried out by column chromato-
graphy (silica) by eluting initially with CH2Cl2/petroleum spirit
(1:1) and then (2:1) to remove impurities, followed by CH2Cl2
and then CH2Cl2/MeOH (1%) to obtain 3 as a yellow oil that
solidified upon standing (5.79 g, 80%); δH (300 MHz; CDCl3)
7.78 (4H, dd, J 8, 2, Ar–H), 7.29 (4H, d, J 8, Ar–H), 6.84–6.77
(8H, m, Ar–H), 4.18 (4H, t, J 5, OCH2), 4.06 (4H, t, J 5, OCH2),
3.97 (4H, t, J 5, OCH2), 3.82 (4H, t, J 5, OCH2), 3.75–3.66
(16H, m, OCH2), 2.40 (6H, s, CH3).
1,11-Bis{4-[2-(2-(o-formylphenoxy)ethoxy)ethoxy]phenoxy}-
3,6,9-trioxaundecane 4
Salicylaldehyde (1.74 g, 1.43 × 10Ϫ2 mol) and K2CO3 (3.59 g,
2.60 × 10Ϫ2 mol) were stirred in dry CH3CN (55 mL) with heat-
ing under an atmosphere of N2 for 1 h. Then 1,4-dioxybenzene
(2,4,2)-bistosylate 2 (5.6 g, 6.49 mmol) in dry CH3CN (200 mL)
was added all at once, and the resulting solution refluxed under
N2 for 4 days. Upon cooling, the reaction mixture was filtered,
the solid washed (CH2Cl2), the solvent removed by rotary
evaporation, and the residue partitioned between CH2Cl2and
H2O. The organic layer was separated, washed (H2O), and dried
(MgSO4). The product was purified using column chromato-
graphy (silica) by eluting with CH2Cl2/petroleum spirit (1:1),
(2:1), and then CH2Cl2 to remove impurities, followed by
CH2Cl2/Et2O (2%), increasing to CH2Cl2/Et2O (10%) to afford
4 as a yellow oil that solidified upon standing. Recrystallised
to obtain a white solid (4.11 g, 83%), mp 69–70 ЊC (from hot
acetone with slow evaporation) (Found: C, 66.18; H, 6.48.
C42H50O13 requires C, 66.13; H, 6.61%); δH (300 MHz; CDCl3)
10.51 (2H, s, CHO), 7.82 (2H, dd, J 8, 2, Ar–H), 7.52 (2H, dt,
J 8, 2, Ar–H), 7.02 (2H, t, J 8, Ar–H), 6.98 (2H, d, J 8, Ar–H),
6.82 (8H, d, J 2, Ar–H), 4.26 (4H, t, J 5, OCH2), 4.10–4.04 (8H,
m, OCH2), 3.97 (4H, t, J 5, OCH2), 3.89 (4H, t, J 5, OCH2), 3.81
(4H, t, J 5, OCH2), 3.69 (8H, m, OCH2).
Zinc 1,4-dioxybenzene (2,4,2)-porphyrin 6b
Zinc was inserted into the 1,4-dioxybenzene (2,4,2)-porphyrin
isomer 6a to give 6b which was recrystallised to obtain a purple/
pink solid, mp 143–145 ЊC (from CH2Cl2/MeOH) (Found: C,
67.08; H, 6.30; N, 3.94. C72H82N4O11.2H2O requires C, 67.51; H,
6.77; N, 4.38%); UV (λ nm, CHCl3) 412, 541, 576; δH (300 MHz;
CDCl3) 10.09 (2H, s, CHO), 7.74 (2H, d, J 8, Ar–H), 7.73 (2H,
t, J 8, Ar–H), 7.35 (2H, t, J 7, Ar–H), 7.27 (2H, d, J 8, Ar–H),
6.52 (4H, dd, J 7, 2, Ar–H), 5.97 (4H, dd, J 7, 2, Ar–H), 4.12
(4H, t, J 4, OCH2), 4.03 (4H, t, J 5, OCH2), 3.94 (4H, m,
CH2CH3), 3.85 (8H, m, CH2CH3 (4H), OCH2 (4H)), 3.79 (8H,
t, J 5, OCH2), 3.17 (4H, t, J 4, OCH2), 2.51 (12H, s, CH3), 2.21
(4H, m, OCH2), 2.10 (4H, m, OCH2), 1.69 (12H, t, J 8,
CH3CH2), 1.24 (4H, s, 2H2O).
Zinc 1,4-dioxybenzene (2,4,2)-porphyrin 7b
Zinc was inserted into the 1,4-dioxybenzene (2,4,2)-porphyrin
7a to give 7b, which was recrystallised to obtain a purple/pink
solid, mp 176–180 ЊC (from CH2Cl2/MeOH) (Found: C, 68.73;
H, 6.66; N, 4.17. C72H82N4O11.H2O requires C, 68.48; H, 6.70;
N, 4.44%); UV (λ nm, CHCl3) 412, 540, 575; δH (300 MHz;
CDCl3) 10.15 (2H, s, meso-H), 7.80 (2H, d, J 8, Ar–H), 7.79
(2H, t, J 8, Ar–H), 7.39 (2H, t, J 8, Ar–H), 7.32 (2H, d, J 9,
Ar–H), 6.44 (4H, d, J 9, Ar–H), 5.98 (4H, d, J 9, Ar–H), 4.17
(4H, t, J 4, OCH2), 4.03–3.98 (8H, m, CH2CH3), 3.86 (4H, t,
J 4, OCH2), 3.71 (4H, m, OCH2), 3.60 (8H, s, OCH2), 3.15 (4H,
Hydroquinone-derived (2,4,2)-porphyrin 6a and 7a
1,4-Dioxybenzene (2,4,2)-dialdehyde 4 (0.51 g, 6.69 × 10Ϫ4mol)
and 3,3Ј-diethyl-4,4Ј-dimethyl-2,2Ј-dipyrrylmethane (0.31 g,
1.34 × 10Ϫ3 mol) were dissolved with stirring in dry CH3CN
(70 mL). The solution was bubbled with N2for 10 min, catalytic
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 4 0 9 7 – 4 1 1 2
4108

View Article Online
t, J 4, OCH2), 2.57 (12H, s, CH3), 2.36 (4H, m, OCH2), 2.20
(4H, m, OCH2), 1.78 (12H, t, J 8, CH3CH2).
(bromomethyl)benzene (77 mg, 2.93 × 10Ϫ4 mol), as well as
catalytic amounts of NaI and NH4PF6 with stirring at ambient
temperature and pressure for 17 days in DMF (6 mL), was also
carried out using the same workup procedure to that above to
give 14b (3.5 mg, 1%).
Catenane 14b
Zinc 1,4-dioxybenzene (2,4,2) porphyrin isomers [6b (ca. 20%)
and 7b (ca. 80%), estimated from NMR] (110 mg, 8.84 × 10Ϫ5
mol), 1,1Ј-[1,4-phenylenebis(methylene)]bis(4,4Ј-bipyridinium)-
bis-(hexafluorophosphate) 1327 (187 mg, 2.65 × 10Ϫ4 mol), 1,4-
bis(bromomethyl)benzene (87 mg, 3.31 × 10Ϫ4 mol), and cata-
lytic amounts of NaI and NH4PF6 were dissolved in DMF
(7 mL), deoxygenated with N2 for 20 min, and then stirred at
room temperature at 12 Kbar for 3 days. The reaction mixture
was then filtered and the solvent removed under vacuum with
heating. The combined solid residues were then washed well
with CH2Cl2, followed by hot H2O. Purification was carried out
by column chromatography (silica) by eluting initially with
CH2Cl2/MeOH (10%) then MeOH, followed by MeOH/2M
NH4Cl solution/MeNO2 (7:2:1 v/v). The product fractions
were taken to dryness, the residue dissolved in a minimum
amount of MeOH/H2O/acetone, and a solution of Zn(OAc)2
in MeOH was added and allowed to stir at room temperature
for 3 h. The acetone was then removed, the solution filtered, the
solid dissolved in a minimum amount of MeOH/H2O/acetone,
and the product taken to dryness again. The residue was then
dissolved in a minimum amount of MeOH/H2O, and saturated
NH4PF6 solution added until no further precipitation was
evident. The catenane product was collected by filtration,
washed (H2O), and pumped dry. The solid was then washed
with acetone and the filtrate recrystallised to obtain 14b as a
dark purple solid (82 mg, 40%), mp 223–225 ЊC (from acetone/
iPr2O) (Found: C, 54.24; H, 4.75; N, 4.58. C108H114N8O11-
Zn(PF6)4ؒ2H2O requires C, 54.46; H, 4.99; N, 4.71%);
m/z (ESMS) [M ϩ H Ϫ PF6]ϩ 2198.7 (calcd. 2198.7), [M ϩ H Ϫ
tetracation Ϫ 4PF6]ϩ1243.2 (calcd. 1243.5), [M Ϫ 2PF6]2ϩ
1026.8 (calcd. 1026.4), [M Ϫ 3PF6]3ϩ636.2 (calcd. 635.9), [M Ϫ
tetracation Ϫ 4PF6]2ϩ621.4 (calcd. 621.3), [M ϩ 4H]4ϩ587.1
(calcd. 586.7), [M Ϫ 4PF6]4ϩ 440.7 (calcd. 440.7); UV (λ nm,
acetone) 421, 540, 574; δH (300 MHz; acetone) 10.03 (2H, s,
H-meso), 8.29 (8H, br s, α-bipy), 7.90 (2H, m, Ar–H), 7.69 (3H,
m, Ar–H), 7.46 (1H, t, J 6, Ar–H), 7.39 (2H, m, Ar–H), 7.06
(8H, s, –C6H4–), 6.73 (10H, m, β-bipy (8H), Ar–H (2H)), 6.56
(2H, d, J 9, Ar–H), 5.31 (4H, d, J 13, ϩN–CH2), 4.98 (4H, d,
J 13, ϩN–CH2), 4.77 (2H, s, OCH2), 4.63 (2H, s, OCH2), 4.17
(4H, m, CH2CH3), 4.06 (4H, m, CH2CH3), 4.00 (2H, s, OCH2),
3.95 (2H, s, OCH2), 3.86–3.81 (6H, m, OCH2), 3.74 (8H, s,
OCH2), 3.66 (4H, s, OCH2), 3.22 (2H, s, OCH2), 3.11 (2H, s,
OCH2), 3.02 (2H, d, J 9, Ar–H), 2.85 (2H, m, OCH2), 2.71 (2H,
m, Ar–H), 2.67 (6H, s, CH3), 2.59 (6H, s, CH3), 1.80 (12H, m,
CH3CH2); δH (300 MHz; CD3CN) 9.90 (2H, s, H-meso), 7.93
(2H, m, Ar-H), 7.71 (8H, d, J 4, α-bipy), 7.68 (3H, m, Ar–H),
7.45 (1H, t, J 6, Ar–H), 7.38 (2H, m, Ar–H), 7.10 (8H, s, –C6H4–),
6.91 (2H, d, J 8, Ar–H), 6.68 (2H, d, J 9, Ar–H), 5.99 (8H, d,
J 4, β-bipy), 5.03 (4H, d, J 13, ϩN–CH2), 4.77 (4H, d, J 13,
ϩN–CH2), 4.61 (2H, s, OCH2), 4.53 (2H, s, OCH2), 4.21 (4H, m,
CH2CH3), 4.05 (2H, s, OCH2), 3.94–3.84 (12H, m, CH2CH3
(4H), OCH2 (8H)), 3.75–3.72 (4H, m, OCH2), 3.63 (4H, s,
OCH2), 3.53 (4H, m, OCH2), 3.04 (2H, s, OCH2), 2.92 (2H,
s, OCH2), 2.71 (2H, s, OCH2), 2.69 (2H, m, Ar–H), 2.63 (6H, s,
CH3), 2.56 (6H, s, CH3), 2.36 (2H, d, J 7, Ar–H), 1.80 (12H, m,
CH3CH2).
1,4-Bis{2-{2-{2-{2-{[2-(2-(toluene-p-sulfonyl)ethoxy)ethoxy]-
phenoxy}ethoxy}-ethoxy}ethoxy}ethoxy}benzene 9
1,4-Bis{2-{2-{2-{2-{[2-(2-hydroxyethoxy)ethoxy]phenoxy}-
ethoxy}ethoxy}ethoxy}ethoxy}benzene 87 (5.91 g, 7.18 mmol)
was dissolved with stirring in dry CH2Cl2 (50 mL) and Et3N
(1.82 g, 18.0 mmol). Then toluene-p-sulfonyl chloride (3.42 g,
18.0 mmol) and Et3N (1.09 g, 10.8 mmol) in dry CH2Cl2
(50 mL), were added dropwise over 30 min, a drying tube
connected, and the resulting solution stirred for 36 h at room
temperature. The reaction mixture was then poured into crushed
ice and diluted with H2O. The organic layer was separated,
washed (2% HCl solution and H2O), and dried (Na2SO4). Puri-
fication was carried out by column chromatography (silica)
by eluting initially with CH2Cl2/petroleum spirit (1:1) and then
CH2Cl2 to remove impurities, followed by CH2Cl2/Et2O (20%)
to obtain 9 as a yellow oil, which was recrystallised to form a
white solid (4.71 g, 58%), mp 81–83 ЊC (from cold acetone)
(Found: C, 59.02; H, 6.70. C56H74O20S2 requires C, 59.45; H,
6.59%); δH (300 MHz; CDCl3) 7.77 (4H, dd, J 7, 2, Ar–H), 7.28
(4H, d, J 8, Ar–H), 6.80 (8H, s, Ar–H), 6.79 (4H, s, Ar–H), 4.17
(4H, t, J 5, OCH2), 4.04 (8H, m, OCH2), 3.96 (4H, t, J 5,
OCH2), 3.80 (8H, dt, J 5, 2, OCH2), 3.74–3.67 (24H, m, OCH2),
2.39 (6H, s, CH3).
1,4-Bis{2-{2-{2-{2-{[2-(2-(o-formylphenoxy)ethoxy)ethoxy]-
phenoxy}ethoxy}ethoxy}-ethoxy}ethoxy}benzene 10
Salicylaldehyde (1.07 g, 8.73 mmol) and K2CO3 (2.74 g, 1.98 ×
10Ϫ2 mol) were stirred in dry CH3CN (50 mL) with heating
under an atmosphere of N2 for 20 min. Then 1,4-dioxybenzene
(2,4,4,2)-bistosylate 9 (4.49 g, 3.97 mmol) in dry CH3CN
(130 mL) was added all at once, and the resulting solution
refluxed under N2 for 60 h. Upon cooling, the reaction mixture
was filtered, the solid washed (CHCl3), the solvent removed
by rotary evaporation, and the residue partitioned between
CHCl3and H2O. The organic layer was separated, washed (2M
NaOH solution, H2O), and dried (Na2SO4). Purification was
carried out by multiple fractional recrystallisations to obtain 10
as a white solid (1.38 g, 33%), mp 79–80 ЊC (from CH2Cl2/
MeOH with slow evaporation) (Found: C, 64.87; H, 6.64.
C56H70O18 requires C, 65.23; H, 6.84%); δH (300 MHz; CDCl3)
10.49 (2H, s, CHO), 7.80 (2H, d, J 9, Ar–H), 7.50 (2H, t, J 8,
Ar–H), 7.01 (2H, t, J 8, Ar–H), 6.96 (2H, d, J 9, Ar–H), 6.80
(12H, d, J 3, Ar–H), 4.25 (4H, t, J 4, OCH2), 4.08–4.03 (12H,
m, OCH2), 3.95 (4H, t, J 4, OCH2), 3.88 (4H, t, J 4, OCH2), 3.79
(8H, t, J 4, OCH2), 3.68 (16H, m, OCH2).
1,4-dioxybenzene (2,4,4,2)-porphyrin 11a and 12a
1,4-dioxybenzene (2,4,4,2)-dialdehyde 10 (0.24 g, 2.25 × 10Ϫ4
mol) and 3,3Ј-diethyl-4,4Ј-dimethyl-2,2Ј-dipyrrylmethane
5
(0.11 g, 4.50 × 10Ϫ4 mol) were dissolved with stirring in dry
CH3CN (20 mL), and the solution bubbled with N2 for 20 min.
Then Cs2CO3 (90 mg, 2.70 × 10Ϫ4 mol) and CCl3COOH (90 mg,
5.40 × 10Ϫ4 mol) in dry CH3CN (5 mL), together with addi-
tional catalytic amounts of CCl3COOH were added all at once,
and the solution bubbled with N2 again for 3 min. The reaction
mixture was then stirred for 5 h (RT, N2, dark), before adding
o-chloranil (0.11 g, 4.50 × 10Ϫ4 mol) in dry THF (5 mL) all at
once and stirring was continued overnight (RT, dark). Then
Et3N (1 mL) was added and the mixture was stirred for 30 min,
before taking the solution to dryness by rotary evaporation. An
initial purification was carried out on an alumina column using
CH2Cl2/MeOH (5%) as the eluent to obtain several porphyrin
The above reaction was also performed with equilibrium
mixtures of 6b and 7b (158 mg, 1.27 × 10Ϫ4 mol), and 2.5 mole
equivalents less of 13 (107 mg, 1.52 × 10Ϫ4 mol) and 1,4-bis-
(bromomethyl)benzene (50 mg, 1.91 × 10Ϫ4 mol), as well as
catalytic amounts of NaI and NH4PF6 using the same reaction
conditions and workup procedure to that above, gave 14b
(58 mg, 20%).
Reaction with equilibrium mixtures of 6b and 7b (230 mg,
1.95 × 10Ϫ4 mol), 13 (165 mg, 2.34 × 10Ϫ4 mol) and 1,4-bis-
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 4 0 9 7 – 4 1 1 2
4109

View Article Online
fractions. Further purification was carried out by column
chromatography (silica), eluting with CH2Cl2/Et2O (30%) to
obtain the porphyrin isomer 11a and with CH2Cl2/Et2O (40%)
to obtain the porphyrin isomer 12a. A final purification on both
fractions was carried out by preparative TLC plates using
CH2Cl2/Et2O (20%) as the eluent.
solution filtered, and saturated NH4PF6 solution added to the
filtrate until no further precipitation was evident. The [2]-
catenane product was collected by filtration, washed (water),
and pumped dry. The solid was then washed with acetone and
the filtrate recrystallised (from acetone/iPr2O) to obtain the
porphyrin [2]-catenane 16b as a dark purple solid (39 mg, 21%),
mp 200–204 ЊC; m/z (FAB) [M]ϩ 2612.0 (calcd 2612.79), [M–
PF6]ϩ 2467.0 (calcd 2467.83), [M-2PF6]ϩ 2322.0 (calcd 2322.86),
[M-3PF6]ϩ 2177.0 (calcd 2177.90), [M-tetracation-PF6]ϩ 1511.3
(calcd 1512.67); δH (300 MHz; DMSO-d6; 363K) 9.74 (2H, s,
H-1), 8.22 (8H, d, J 6, α-bipy), 7.82 (2H, t, J 9, H-13), 7.62 (2H,
d, J 8, H-14), 7.55 (2H, d, J 8, H-11), 7.46 (8H, s, phenylene),
7.39 (2H, t, J 8, 12), 6.59 (8H, d, J 6, β-bipy), 6.43 (4H, s, H-33),
5.21 (8H, s, Nϩ–CH2), 4.28 (4H, m, H-16), 3.99 (4H, m, OCH2),
3.84 (8H, m, H-4), 3.67–3.62 (24H, m, H-21, H-22, 4 × OCH2),
3.61 (4H, m, OCH2), 3.57 (4H, m, OCH2), 3.37 (4H, m, H-17),
3.00 (4H, m, OCH2), 2.93 (16H, s, H-7, OCH2), 2.88 (4H, m,
OCH2), 1.68 (12H, t, J 8, H-5).
18b was also collected and recrystallised in the same manner
(15 mg, 6%), mp 234–237 ЊC; m/z (FAB)- [M]ϩ 3712.4 (calcd
3712.91), [M–PF6]ϩ 3567.7 (calcd 3567.95), [M-2PF6]ϩ 3423.2
(calcd 3422.98), [M-3PF6]ϩ 3279.3 (calcd 3278.02),
[M-4PF6]ϩ 3133.3 (calcd 3133.06), [M-5PF6]ϩ 2988.4 (calcd
2988.09), [M-tetracation-5PF6]ϩ 2467.5 (calcd 2467.83), [M-
tetracation-5PF6]ϩ 2323 (calcd 2322.86), [parent porphyrin]ϩ
1511.7 (calcd 1512.67); δH (300 MHz; DMSO-d6; 363 K)
9.70 (2H, s, H-1), 8.58 (16H, d, J 6, α-bipy), 7.87 (2H, t, J 9,
H-13), 7.67–7.63 (4H, m, H-14, H-11), 7.60 (16H, s, phenylene),
7.46 (2H, t, J 8, H-12), 7.12 (16H, d, J 6, β-bipy), 5.45 (16H, m,
Nϩ–CH2), 4.40 (4H, m, H-16), 3.87 (8H, m, H-4), 3.80–3.68
(28H, m, H-21, H-22, H-33, 4 × OCH2), 3.60 (4H, m, OCH2),
3.51 (4H, m, H-17), 3.40 (4H, m, OCH2), 3.31–3.26 (8H, m,
OCH2), 2.93 (20H, s, H-7, OCH2), 1.70 (12H, t, J 8, H-5).
11a was recrystallised to obtain a purple/red solid (45 mg,
14%), m.p. 155–158 ЊC (from CH2Cl2/MeOH) (Found: C, 69.66;
H, 7.03; N, 3.80. C86H104O16N4.2H2O requires C, 69.52; H, 7.33;
N, 3.77%); m/z (ESMS) [M ϩ H]ϩ 1450.4 (calcd. 1449.7), [M ϩ
2H]2ϩ 725.7 (calcd. 725.4); UV (λ nm, CHCl3) 410, 507, 541,
574, 626; δH (300 MHz; CDCl3) 10.20 (2H, s, H-meso), 7.83–
7.77 (4H, m, Ar–H), 7.40 (2H, t, J 8, Ar–H), 7.34 (2H, d, J 9,
Ar–H), 6.89 (4H, s, Ar–H), 5.86 (4H, d, J 9, Ar–H), 5.20 (4H, d,
J 9, Ar–H), 4.17 (4H, t, J 4, OCH2), 4.10 (4H, t, J 4, OCH2),
4.00 (8H, m, CH2CH3), 3.87 (4H, t, J 4, OCH2), 3.75–3.66
(24H, m, OCH2), 3.20 (4H, t, J 4, OCH2), 2.59 (12H, s, CH3),
2.06 (4H, t, J 4, OCH2), 1.77 (12H, t, J 8, CH3CH2), 1.63 (4H, t,
J 4, OCH2), 1.28 (4H, s, 2H2O), Ϫ2.27 (2H, br s, pyrrole NH).
12a was recrystallised to obtain a purple/red solid (12 mg,
4%), mp 142–145 ЊC (from CH2Cl2/MeOH) (Found: C, 69.98;
H, 6.93; N, 3.85. C86H104O16N4.H2O requires C, 70.37; H, 7.28;
N, 3.82%); m/z (ESMS) [M ϩ H]ϩ 1450.4 (calcd. 1449.7), [M ϩ
2H]2ϩ 725.7 (calcd. 725.4); UV (λ nm, CHCl3) 410, 507, 541,
573, 626; δH (300 MHz; CDCl3) 10.21 (2H, s, H-meso), 7.82–
7.77 (4H, m, Ar–H), 7.39 (2H, t, J 8, Ar–H), 7.32 (2H, d, J 9,
Ar–H), 6.54 (4H, s, Ar–H), 5.84 (4H, d, J 9, Ar–H), 5.24 (4H, d,
J 9, Ar–H), 4.18 (4H, t, J 4, OCH2), 4.01 (8H, m, CH2CH3),
3.80 (4H, t, J 4, OCH2), 3.70 (8H, m, OCH2), 3.62 (20H, m,
OCH2), 3.10 (4H, t, J 4, OCH2), 2.60 (12H, s, CH3), 2.07 (4H, t,
J 4, OCH2), 1.79 (12H, t, J 8, CH3CH2), 1.72 (4H, t, J 4, OCH2),
1.28 (2H, s, H2O), Ϫ2.25 (2H, br s, pyrrole NH); δC (CDCl3)
158.6, 152.9, 152.1, 152.0, 145.4, 144.4, 140.9, 135.8, 134.5,
131.3, 130.3, 121.1, 115.4, 114.6, 114.2, 111.8, 96.1, 70.9, 70.8,
69.8, 69.3, 68.7, 67.9, 67.7, 65.6, 19.9, 17.7 and 13.7.
[2] catenane 15b and [3]catenane 17b. The free base 1,4-dioxy-
benzene (2,4,4,2) α,α-porphyrin 12a (183 mg, 1.26 × 10Ϫ4mol)
1,1Ј-[1,4-phenylenebis(methylene)]bis(4,4Ј-bipyridinium)bis-
(hexafluorophosphate)27 13 (370 mg, 5.30 × 10Ϫ4mol), 1,4-
bis(bromomethyl)benzene (170 mg, 6.62 × 10Ϫ4mol), and cata-
lytic amounts of NaI and NH4PF6 were dissolved in DMF
(7 mL), then stirred at room temperature at 12 kbar for 6 days.
The solvent was removed under vacuum with heating, and the
residue washed well with CH2Cl2, followed by hot water. Purifi-
cation was carried out by column chromatography (silica) by
eluting initially with CH2Cl2/MeOH (10%), then MeOH/2M
NH4Cl solution/MeNO2 (7:2:1 v/v) to obtain the [2]-catenane
fraction 15b, followed by MeOH/2M NH4Cl solution/DMF
(4:5:2 v/v) to obtain the [3]-catenane 17b. The [2]-catenane
fraction was taken to dryness, the residue dissolved in a mini-
mum amount of MeOH/water/acetone, and a solution of
Zn(OAc)2 in MeOH was added and allowed to stir at room
temperature for 1 h. The acetone was then removed, the solu-
tion filtered, and saturated NH4PF6 solution added to the fil-
trate until no further precipitation was evident. The product
was collected by filtration, washed (water), and pumped dry.
The solid was then washed with acetone and the filtrate
recrystallised to obtain the porphyrin [2]catenane 15b as a dark
purple solid (24 mg 13%), mp 200–204 ЊC (from acetone/iPr2O);
m/z (FAB) [M]ϩ 2612.0 (calcd 2612.79), [M–PF6]ϩ 2467.5 (calcd
2467.83), [M-2PF6]ϩ 2323.0 (calcd 2322.86), [M-3PF6]ϩ 2177.0
(calcd 2177.90), [M-tetracation-4PF6]ϩ 1511.7 (calcd 1512.67);
δH (300 MHz; CD3COCD3; 300 K) 9.94 (2H, s, H-1), 8.08 (8H,
d, J 6, α-bipy), 7.89 (2H, t, J 9, H-13), 7.72–7.65 (2H, d, J 8,
H-14, H-14Ј), 7.46 (2H, d, J 8, H-11), 7.41 (2H, t, J 8, H-12),
7.26 (8H, s, phenylene), 7.00 (2H, d, J 8, H-21), 6.81 (2H, d, J 8,
H-22), 6.17 (4H, d, J 8, H-33), 6.06 (8H, bs, β-bipy), 5.32 (8H,
m, Nϩ–CH2), 4.79 (2H, m, H-16), 4.61 (2H, m, H-16Ј), 4.17
(4H, m, H-4), 3.82–3.45 (24H, m, H-4Ј, H-17, H-24, H-19,
H-18, H-31, 2 × OCH2), 3.29 (6H, m, H-19Ј, OCH2), 2.71 (12H,
m, H-24Ј, H-18Ј, 2 × OCH2), 2.59 (16H, bs, H-7, H-21Ј, H-22Ј),
Zinc 1,4-dioxybenzene(2,4,4,2)porphyrin 12b
Zinc was inserted into the 1,4-dioxybenzene (2,4,4,2) porphyrin
H2(3HQ)2,4,4,2P 12a to give 12b, which was recrystallised to
obtain a purple/pink solid, mp 73–77 ЊC (from CH2Cl2/
heptane); UV (λ nm, CHCl3) 412, 540, 575; δH (300 MHz;
CDCl3) 10.16 (2H, s, H-meso), 7.80 (4H, m, Ar–H), 7.39 (2H, t,
J 8, Ar–H), 7.34 (2H, d, J 7, Ar–H), 6.60 (4H, s, Ar–H), 6.06
(4H, d, J 9, Ar–H), 5.48 (4H, d, J 9, Ar–H), 4.18 (4H, t, J 4,
OCH2), 4.02 (8H, m, CH2CH3), 3.82 (4H, t, J 4, OCH2), 3.68
(12H, m, OCH2), 3.60 (16H, m, OCH2), 3.11 (4H, t, J 4, OCH2),
2.58 (12H, s, CH3), 2.15 (4H, t, J 4, OCH2), 2.03 (4H, m,
OCH2), 1.78 (12H, t, J 8, CH3CH2).
Catenanes 15, 16, 17 and 18
[2]catenane 16b and [3]catenane 18b. The free base 1,4-di-
oxybenzene (2,4,4,2) α,β-porphyrin (161 mg, 1.11 × 10Ϫ1 mmol)
1,1Ј-[1,4-phenylenebis(methylene)]bis(4,4Ј-bipyridinium)bis-
(hexa-fluorophosphate) 13 (330 mg, 4.67 × 10Ϫ1 mmol), 1,4-
bis(bromomethyl)benzene (150 mg, 5.83 × 10Ϫ1 mmol), and
catalytic amounts of NaI and NH4PF6 were dissolved in DMF
(7 mL), then stirred at room temperature at 12 kbar for 6 days.
The solvent was removed under vacuum with heating, and the
residue washed well with CH2Cl2, followed by hot water. Purifi-
cation was carried out by column chromatography (silica) by
eluting initially with CH2Cl2/MeOH (10%), then MeOH/2M
NH4Cl solution/MeNO2 (7:2:1 v/v) to obtain the [2]-catenane
fraction 16a, followed by MeOH/2M NH4Cl solution/DMF
(4:5:2 v/v) to obtain the [3]-catenane 18a. The [2]-catenane
fraction was taken to dryness, the residue dissolved in a
minimum amount of MeOH/water/acetone, and a solution of
Zn(OAc)2 in MeOH was added and allowed to stir at room
temperature for 1 h. The acetone was then removed, the
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 4 0 9 7 – 4 1 1 2
4110

View Article Online
M. Pietraszkiewicz, N. Spencer and J. F. Stoddart, Angew. Chem.,
Int. Ed. Engl., 1991, 30, 1042.
1.76 (12H, t, J 8, H-5); δH (300 MHz; DMSO-d6; 363 K) 9.80
(2H, s, H-1), 8.36 (8H, d, J 6, α-bipy), 7.83 (2H, t, J 9, H-13),
7.57 (2H, d, J 8, H-14), 7.54 (2H, d, J 8, H-11), 7.51 (8H, s,
phenylene), 7.36 (2H, t, J 8, H-12), 6.80 (8H, d, J 6, β-bipy),
5.95 (4H, s, H-33), 5.29 (8H, s, Nϩ–CH2), 4.37 (4H, m, H-16),
3.97 (4H, m, H-4), 3.80 (8H, m, H-4, OCH2), 3.68–3.55 (32H,
m, H-21, H-22, 6 × OCH2), 3.45 (4H, m, H-17), 3.13 (4H, m,
OCH2), 2.93 (20H, s, H-7, 2 × OCH2), 1.71 (12H, t, J 8, H-5).
Catenane 17b was obtained similarly (10 mg, 5%). NMR
analysis indicated very broad peaks, and that this fraction was
contaminated with polymeric bipyridinium-based material,
which could not be separated. Nevertheless, certain peaks
belonging to the catenane could be assigned, as their ratio did
9 D. B. Amabilino, P. R. Ashton, A. S. Reder, N. Spencer and
J. F. Stoddart, Angew. Chem., Int. Ed. Engl., 1994, 33(4), 433;
D. B. Amabilino, P. R. Ashton, A. S. Reder, N. Spencer and
J. F. Stoddart, Angew. Chem., Int. Ed. Engl., 1994, 33(12), 1286; P. R.
Ashton, R. Ballardini, V. Balzani, M. T. Gandolfi, D. J. F. Marquis,
L. Pérez-García, L. Prodi, J. F. Stoddart and M. Venturi, J. Chem.
Soc., Chem. Commun., 1994, 177.
10 P. R. Ashton, V. Baldoni, V. Balzani, A. Credi, H. D. A. Hoffmann,
M.-V. Martinez-Diaz, F. M. Raymo, J. F. Stoddart and M. Venturi,
Chem. Eur. J., 2001, 7, 3482; P. R. Ashton, S. E. Boyd, A. Brindle,
S. J. Langford, S. Menzer, L. Pérez-García, J. A. Preece,
F. M. Raymo, N. Spencer, J. F. Stoddart, A. J. P. White
and D. J. Williams, New J. Chem., 1999, 23, 587; M. Asakawa,
P. R. Ashton, V. Balzani, A. Credi, C. Hamers, G. Mattersteig,
M. Montalti, A. M. Shipway, N. Spencer, J. F. Stoddart,
M. S. Tolley, M. Venturi, A. J. P. White and D. J. Williams, Angew.
Chem., Int. Ed. Engl., 1998, 37, 333; P. R. Ashton, J. A. Preece,
J. F. Stoddart, M. S. Tolley, A. J. P. White and D. J. Williams,
Synthesis, 1994, 1344; Z.-T. Li, P. C. Stein, J. Becher, D. Jensen,
P. Mørk and N. Svenstrup, Chem. Eur. J., 1996, 2, 624;
D. J. Cárdenas, A. Livoreil and J.-P. Sauvage, J. Am. Chem. Soc.,
1996, 118, 11980.
not vary after each attempted purification step; m/z (FAB)
Ϫ ϩ
[M-PF6
]
3566.8 (calcd 3565.9), [M-2PF6]ϩ 3421.7 (calcd
3421.0), [M-3PF6]ϩ 3276.8 (calcd 3276.0), [M-4PF6]ϩ 3131.9
(calcd 3131.0), [M-5PF6]ϩ 2986.9 (calcd 2986.1), [M-tetra-
cation-5PF6]ϩ 2465.9 (calcd 2465.8), [M-tetracation-6PF6]ϩ
2320.9 (calcd 2320.8), [M-tetracation-7PF6]ϩ, [M-2(tetra-
cation)-8PF6]ϩ 1511.3 (calcd 1510.7); UV (λ nm, acetone) 420
vbr, 544 br.
11 V. Balzani, A. Credi, S. J. Langford, F. M. Raymo, J. F. Stoddart and
M. Venturi, J. Am. Chem. Soc., 2000, 122, 3542.
Using a mixture of the free base 1,4-dioxybenzene (2,4,4,2)
porphyrin isomers 11a (61 mg, 4.21 × 10Ϫ5 mol) and 12a (40 mg,
2.76 × 10Ϫ5 mol), together with 13 (188 mg, 2.66 × 10Ϫ4 mol),
1,4-bis(bromomethyl)benzene (88 mg, 3.33 × 10Ϫ4 mol), under
the same reaction conditions as above, produced a mixture of
products. Attempted purification was carried out by column
chromatography (silica) by eluting initially with CH2Cl2/MeOH
(10%), then MeOH/2M NH4Cl solution/MeNO2 (7:2:1 v/v) to
obtain isomeric mixtures of α,β- and α,α-[2]-catenane fractions
15a and 16a in a ratio of ca. 3:2 (from NMR) as a dark purple
solid (42 mg, 23%), mp 200–204 ЊC (from acetone/iPr2O). The
NMR spectrum of the mixture was a sum of the spectra of the
individual components isolated above. This was followed by
MeOH/2M NH4Cl solution/DMF (4:5:2 v/v) to obtain iso-
meric mixtures of α,β- and α,α-[3]-catenane fractions 17a and
18a in a ratio of ca. 3:2 (from NMR) as a dark purple solid
(10 mg, 4%), mp 234–237 ЊC (from acetone/iPrOH). The
NMR spectrum contained all of the peaks of the individual
components isolated as above, together with broad peaks
presumed as arising from polymeric non-porphyrinic material.
12 D. B. Amabilino and J.-P. Sauvage, New J. Chem., 1998, 395;
M. Linke, N. Fujita, J. C. Chambron, V. Heitz and J. P. Sauvage, New
J. Chem., 2001, 25, 790; J.-C. Chambron, C. O. Dietrich-Buchecker,
V. Heitz, J.-F. Nierengarten, J.-P. Sauvage, C. Pascard and
J. Guilhem, Pure Appl. Chem., 1995, 67, 233; D. B. Amabilino and
J.-P. Sauvage, Chem. Commun., 1996, 2441.
13 L. Flamigni, A. M. Talarico, S. Serroni, F. Puntoriero, M. J. Gunter,
M. R. Johnston and T. P. Jeynes, Chem. Eur. J., 2003, 9, 2649.
14 M. J. Gunter and M. R. Johnston, J. Chem. Soc., Chem. Commun.,
1992, 1163.
15 M. J. Gunter, D. C. R. Hockless, M. R. Johnston, B. W. Skelton and
A. H. White, J. Am. Chem. Soc., 1994, 116, 4810.
16 M. J. Gunter and M. R. Johnston, J. Chem. Soc., Chem. Commun.,
1994, 829.
17 M. J. Gunter and R. N. Warrener, Chem. Aust., 1997, 25.
18 I. Willner, E. Kaganer, E. Joselevich, H. Durr, E. David,
M. J. Gunter and M. R. Johnston, Coord. Chem. Rev., 1998, 171,
261; L. Flamigni, A. M. Talarico, M. J. Gunter, M. R. Johnston and
T. P. Jeynes, New J. Chem., 2003, 27, 551.
19 D. B. Amabilino, P. R. Ashton, C. L. Brown, E. Córdova,
L. A. Godínez, T. T. Goodnow, A. E. Kaifer, S. P. Newton,
M. Pietraszkiewicz, D. Philp, F. M. Raymo, A. S. Reder,
M. T. Rutland, A. M. Z. Slawin, N. Spencer, J. F. Stoddart and
D. J. Williams, J. Am. Chem. Soc., 1995, 117, 1271.
Acknowledgements
20 P. R. Ashton, D. Philp, N. Spencer and J. F. Stoddart, J. Chem. Soc.,
Chem. Commun., 1992, 1124.
We thank the Australian Research Council for funding for this
project, and the Royal Society of Chemistry for funding
through a Journals Grant for International Authors.
21 D. B. Amabilino, P. R. Ashton, J. F. Stoddart, A. J. P. White and
D. J. Williams, Chem. Eur. J., 1998, 4, 460; P. R. Ashton,
C. L. Brown, E. J. T. Chrystal, T. T. Goodnow, A. E. Kaifer,
K. P. Parry, A. M. Z. Slawin, N. Spencer, J. F. Stoddart
and D. J. Williams, Angew. Chem., Int. Ed. Engl., 1991, 30, 1039;
P. R. Ashton, S. E. Boyd, C. G. Claessens, R. E. Gillard, S. Menzer,
J. F. Stoddart, M. S. Tolley, A. J. P. White and D. J. Williams, Chem.
Eur. J., 1997, 3, 788.
22 M. J. Gunter, M. R. Johnston, B. W. Skelton and A. H. White,
J. Chem. Soc., Perkin Trans. 1, 1994, 1009.
23 M. J. Gunter, T. P. Jeynes, M. R. Johnston, P. Turner and Z. P. Chen,
J. Chem. Soc., Perkin Trans. 2, 1998, 21, 1945.
24 A. Osuka, F. Kobayashi and K. Maruyama, Bull. Chem. Soc. Jpn.,
1991, 64, 1213; A. Osuka, T. Nagata, F. Kobayashi and
K. Maruyama, J. Heterocycl. Chem., 1990, 27, 1657.
25 The addition of Cs2CO3 to the reaction mixture was used to aid in
the formation of this large porphyrin macrocycle via a cation
templating effect—the so called ‘cesium effect’.
26 J. E. Redman and J. K. M. Sanders, Org. Lett., 2000, 2, 4141.
27 P. L. Anelli, P. R. Ashton, R. Ballardini, V. Balzani, M. Delgado,
M. T. Gandolfi, T. T. Goodnow, A. E. Kaifer, D. Philp,
M. Pietraszkiewicz, L. Prodi, M. V. Reddington, A. M. Z. Slawin,
N. Spencer, J. F. Stoddart, C. Vicent and D. J. Williams, J. Am.
Chem. Soc., 1992, 114, 193.
References
1 A. R. Pease, J. O. Jeppesen, J. F. Stoddart, Y. Luo, C. P. Collier and
J. R. Heath, Acc. Chem. Res., 2001, 34, 433; J.-M. Lehn, Angew.
Chem., Int. Ed. Engl., 1988, 27, 89; V. Balzani, A. Credi and
M. Venturi, Chem. Eur. J., 2002, 8, 5525.
2 V. Balzani, M. Gómez-Lopez and J. F. Stoddart, Acc. Chem. Res.,
1998, 31, 405; V. Balzani, A. Credi, F. M. Raymo and J. F. Stoddart,
Angew. Chem. Int. Ed., 2000, 39, 3349.
3 B. L. Feringa, Acc. Chem. Res., 2001, 34, 504; J. F. Stoddart, Chem.
Aust., 1992, 576; J. F. Stoddart, Acc. Chem. Res., 2001, 34, 410.
4 J.-P. Sauvage, Acc. Chem. Res., 1998, 31, 611.
5 D. B. Amabilino, P. R. Ashton, M. Belohradsky, F. M. Raymo and
J. F. Stoddart, Chem. Commun., 1995, 751; P. R. Ashton, M. Blower,
D. Philp, N. Spencer, J. F. Stoddart, M. S. Tolley, R. Ballardini,
M. Ciano, V. Balzani, M. T. Gandolfi, L. Prodi and C. H. McLean,
New J. Chem., 1993, 17, 689.
6 P. R. Ashton, L. Pérez-García, J. F. Stoddart, A. J. P. White and
D. J. Williams, Angew. Chem., Int. Ed. Engl., 1995, 34(5), 571.
7 D. B. Amabilino, P.-L. Anelli, P. R. Ashton, G. R. Brown,
E. Cordova, L. A. Godinez, W. Hayes, A. E. Kaifer, D. Philp,
A. M. Z. Slawin, N. Spencer, J. F. Stoddart, M. S. Tolley and
D. J. Williams, J. Am. Chem. Soc., 1995, 117, 11142.
28 Dynamic NMR Spectroscopy, ed. J. Sandstrom, Academic Press,
New York, 1982, ch. 6; I. O. Sutherland, Annu. Rep. NMR
Spectrosc., 1971, 4, 71.
29 This process was too fast to evaluate in the previous porphyrin
8 P. R. Ashton, C. L. Brown, E. J. T. Chrystal, K. P. Parry,
[2]catenanes, but Stoddart19 has found that this process occurs at a
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 4 0 9 7 – 4 1 1 2
4111

View Article Online
rate of approximately 2.5 × 105 reorientations per second at ambient
temperature for both the BPP40C12- and BPP46C14-based
[2]catenanes, which are of a similar magnitude to 14b. However,
no value has been reported for the BPP34C10 [2]catenane 1b,
presumably indicating that the ‘rocking’ process occurs too fast in
this catenane to be evaluated. The larger activation barrier for 1,4-
dioxybenzene reorientation observed in 14b over that observed for
both the BPP40C12- and BPP46C14-based [2]catenanes may again
reflect the short nature of the diethylene straps of the polyether
chain linking the porphyrin and the 2HQ rings which would result
in less flexibility, particularly in the presence of the porphyrin, to
reorientate within the tetracation cavity.
supports the structure shown in Fig. 7. Hence, while one tetracation
is shielded by the porphyrin and the other is deshielded, and because
this would obviously be a dynamic process, then the average of
these two environments would lead to a net smaller shielding of the
tetracation protons for the [3]catenane isomers compared to their
[2]catenane counterparts.
32 Additional processes that may be expected to occur include (i) a
‘stacking, unstacking, shuttling and restacking’ process involving an
unstacking of the HQ-33 (ring B in Fig. 7) from the tetracation,
bound HQ (ring A) and porphyrin subunits, followed by ‘shuttling’
of the tetracation to the second HQ ring A, and then restacking
of the HQ ring B to the π-stacked superstructure; (ii) processes
involving possible π–π interactions between the tetracation, HQ ring
B, and porphyrin subunits (although this possible exchange process
may not involve π–π interactions with the porphyrin, in order for
process (ii) to occur, process (i) must also be involved, as the
‘shuttling’ between HQ rings A requires the intermediate ‘shuttling’
from HQ ring A to B as a ‘stopover’); (iii) other processes with
possible π–π stacking interactions between the tetracation, bound
HQ ring B, and both HQ rings A (Fig. 7), without interaction with
the porphyrin, although the NMR evidence suggests that such an
arrangement is unlikely.
30 However, we were not able to separate the catenane isomers given
the solvent required for chromatography, viz. MeOH–2M NH4Cl–
MeNO2 (7:2:1 v/v) for [2]catenanes and MeOH–2M NH4Cl–DMF
(4:5:2 v/v) for [3]catenanes.
31 Furthermore, the smaller shifts for the tetracation resonances for the
[3]catenane isomers in comparison to the [2]catenane isomers upon
catenane formation make good sense given that in order for the
tetracation and bound HQ ring A to maintain π–π interactions with
the porphyrin, then the second bound HQ and tetracation must
move away from the porphyrin to allow this interaction, and
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 4 0 9 7 – 4 1 1 2
4112