Binding of the terephthalate dianion by di- tri- and tetrathiourea functionalised fused [3] and [5]polynorbornane based hosts

Article abstract of DOI:10.1039/b910522k

The affinity of new di-, tri- and tetrathiourea functionalised fused [3] and [5]polynorbornane based hosts 1-6 towards terephthalate^2- was proportional to the size of the preorganised cleft/cavity imparted by the polynorbornane scaffold. Recept

Full text of DOI:10.1039/b910522k

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Binding of the terephthalate dianion by di- tri- and tetrathiourea
functionalised fused [3] and [5]polynorbornane based hosts†
Adam J. Lowe and Frederick M. Pfeffer*
Received 29th May 2009, Accepted 15th July 2009
First published as an Advance Article on the web 17th August 2009
DOI: 10.1039/b910522k
The affinity of new di-, tri- and tetrathiourea functionalised fused [3] and [5]polynorbornane based
hosts 1–6 towards terephthalate2- was proportional to the size of the preorganised cleft/cavity imparted
by the polynorbornane scaffold. Receptors based on the [5]polynobornane framework had greater
affinities for the anion due to a higher degree of host:guest size complementarity. Hosts 1–5 formed 1:1
host:guest complexes with the rigid dianion, yet remarkably, host 6 was found to bind two terephthalate
guests.
groups in the field of anion recognition and a range of func-
Introduction
tionalised molecular frameworks have been specifically designed,
synthesised and evaluated (examples include calix[n]arenes,5
calix[n]pyroles,6 and cholapods7) to encapsulate anionic
guests.
Research in our laboratory is aimed at extending this arsenal
of preorganised hosts for anions through the use of suitably
functionalised fused [n]polynorbornane scaffolds containing mul-
tiple H-bond donors for strong recognition.8 Such frameworks
are readily constructed through the use of a thermal 1,3 dipolar
cycloaddition of a norbornene alkene with a cyclobutane epoxide
(ACE reaction).9
Anion recognition has become an active field of research1 due to
the crucial roles that anions play in many fundamental processes;
from biochemical transformations, such as the generation of high
energy polyphosphate bonds,2 to environmental pollutants such
as nitrates and phosphates which are linked to the eutrophication
of waterways.3
The key concept of conformational preorganisation as a means
to enhance recognition and binding strength was pioneered
by Cram.4 This concept has been adopted by several research
In the current study the size of the [n] polynorbornane
framework and also the number of H bond donors of the host
were varied (Fig. 1, compounds 1–6) and their interaction with a
single anion (terephthalate2-) probed using 1H NMR spectroscopic
titration analysis. In this full paper the synthesis of new hosts 3–6
is described, followed by the results of 1H NMR titration binding
studies.
School of Life and Environmental Sciences, Deakin University, Geelong,
VIC, Australia. E-mail: thefef@deakin.edu.au; Fax: + 61 3 5227 1040;
Tel: + 61 3 5227 1439
† Electronic supplementary information (ESI) available: Von-Baeyer num-
bering system used to name the [n]polynorbornane scaffolds, additional
experimental details, stack plots of 1H NMR titration spectra, and
titration isotherms for the internal framework C–H protons. See DOI:
10.1039/b910522k
Fig. 1 The structures of the 2, 3 or 4-armed [3] and [5]polynorbornane based hosts 1–6, * denotes the H-bond donor used for calculations.
This journal is The Royal Society of Chemistry 2009 Org. Biomol. Chem., 2009, 7, 4233–4240 | 4233
©

View Article Online
Results and discussion
Design
As illustrated in Fig. 1, all hosts 1–6 were based on either
a [3] or [5]polynorbornane scaffold with 2, 3 or 4 thiourea
functionalised anionophoric arms covalently attached through
flexible ethyl linkers. In the case of the 3 and 4-armed hosts 3–6,
amide functionality was also included with the design, potentially
providing additional anion recognition units.
Synthesis
The synthesis of [n]polynorbornane frameworks requires the
reaction of a norbornene alkene with a cyclobutane epoxide.9
The necessary epoxides are prepared from norbornene alkenes
using Ru catalysed Mitsudo reaction,10 followed by Weitz–Scheffer
epoxidation.11 The epoxides required herein (8, 10, 12 and 13,
Scheme 1) were manufactured according to this protocol using
known norbornene units.12 In the case of epoxide 12, cyclobutene
diester 11 was used and this was conveniently prepared from
quadricyclane and DMAD.9a,13
Scheme 2 Formation of a [n]polynorbornane framework using ACE
cycloaddition. Reagents and conditions: (i) THF, 140–150 ◦C, 12–72 h.
Scheme 1 Synthesis of cyclobutane epoxides. Reagents and conditions: (i)
DMAD, RuH2(CO)(PPh3)3, THF, 70–90 ◦C, 12–48 h, (ii) TBHP, KOtBu,
THF, 0 ◦C, 15–28 h.
Synthesis of the multi-armed [n] polynorbornane frameworks in-
volved in thermal ACE cycloaddition of the cyclobutane epoxides
with appropriately functionalised norbornene units (Scheme 2).
The reaction typically took between 12 and 72 hours and yields
ranged from 35–76%. This series of reactions highlights the
versatility of the methodology as from a limited number of
norbornenes and cyclobutane epoxides the full suite of desired
[3] and [5]polynorbornane scaffolds including the appropriate
number of arms were prepared.
The final steps in the synthesis of the hosts 3–6 required a
three step protocol (Scheme 3) involving: (i) hydrogenation of the
norbornene double bonds using Pearlman’s catalyst (giving endo
product exclusively), (ii) deprotection using 20% trifluoroacetic
acid (TFA) in dichloromethane (DCM) to afford the free amines
and (iii) subsequent coupling with either 4-nitrophenyl or 4-
fluorophenylisothiocyanate to afford the desired hosts. In all cases
Scheme 3 Synthesis of hosts 3–6. Reagents and conditions: (i) H2,
Pd-OH/C, EtOH, 45 ◦C, 48 h, (ii) 20% TFA/DCM, RT, 2 h, (iii)
DIPEA, CHCl3, RT, 20 h, a 4-fluorophenylisothiocyanate, b 4-nitrophenyl-
isothiocyanate.
4234 | Org. Biomol. Chem., 2009, 7, 4233–4240
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
Table 1 Maximum observed chemical shifts, host:guest stoichiometries
and calculated association constants (logK) for hosts 1–6a
bound the anion
>
100 times more strongly than the
[3]polynorbornane based host 1b (logK = 3.8). As illustrated in
Fig. 2, this was due to the longer cleft width of host 2 experiencing
less stretching than host 1 when bound in the host–guest complex,
as such receptor 2 better complements the length of the rigid
dianionic guest (ca. 7.0 A).
max thiourea
NH Dd (ppm)
max internal
CH Dd (ppm)
H:G
logK (M-1)
8a
1a
1b
2a
2b
3a
3b
4a
4b
5a
5b
6a
2.22
3.02
3.26
3.61
3.11
3.47
3.64
3.84
2.14
2.81
3.38
-0.02
-0.05
0.14
0.20
0.04
-0.03
0.13
0.14
0.04
0.03
0.23
1:1
1:1
1:1
1:1
1:1
1:1
1:1
1:1
1:1
1:1
1:2
3.5 ( 0.23)
3.8 ( 0.52)
4.4 ( 0.34)
6.0 ( 0.49)
3.2 ( 0.35)
3.6 ( 0.34)
4.3 ( 0.28)
4.9 ( 0.47)
2.9 ( 0.17)
3.1 ( 0.10)
2.9 ( 0.22)
3.0 ( 0.29)b
3.0 ( 0.25)
4.5 ( 0.54)b
˚
6b
3.74
0.21
1:2
a In each case, the data refers to the thiourea H-bond donor marked with
an asterisk (*) in Fig. 1. Association constants were determined from 1H
NMR titration data using WinEQNMR,15 calculated errors < 14.0%, [H]i
~ 2.5 ¥ 10-3 M. b For 6a and 6b the binding constants are listed as logK1
and logK2.
the target compounds were purified by column chromatography
and/or recrystallisation and fully characterised using 1H and
13C NMR spectroscopy and high resolution mass spectrometry
(HRMS).
Terephthalate2- binding studies
1H NMR spectroscopy titration experiments were conducted to
determine how the increased number of prepositioned H-bond
donors and various cleft/cavity geometries of hosts 3–6 would
effect the ability of these hosts to bind the rigid dicarboxylate
guest, terephthalate2-. Solutions of terephthalate2- (prepared as a
tetrabutylammonium (TBA) salt)14 in DMSO-d6 (~ 3.0 ¥ 10-2 M)
were titrated against DMSO-d6 solutions of each host ([H]i ~ 2.5 ¥
10-3 M) while recording any change in chemical shift of the amide
and thiourea N–H resonances. The results of this study, and those
of the previous study involving hosts 1 and 2,8a are summarised
Fig. 2 Molecular models calculated at Hartree–Fock 3–21G(*) level
of theory depicting the 1:1 complexes formed between (a) host 1b and,
(b) host 2b and the rigid aryl dicarboxylate, terephthalate2-.
The interaction of host 2 and terephthalate2- was also monitored
by following the internal framework ‘endo’ C-H resonances during
1H NMR titrations (for example H2, 10, 12 and 20 in host 1,
Fig. 1). It was found that although the observed change in chemical
shift was relatively small (Dd ~ 0.2 ppm, Table 1,) a definite trend
existed,† where both the binding isotherm and Job plot confirmed
the formation of a 1:1 complex.8a As a downfield migration
occurred, it is likely that the CH protons were being deshielded
by the ring-current effect of the phenyl ring.16 In contrast, the
magnitude of Dd observed for the internal C–H protons of host 1
was negligible, this can be explained by considering the geometries
of the host:guest complexes in Fig. 2.
Calculated at Hartree–Fock 3–21G* level of theory, Fig. 2a
shows that in order to accommodate the rigid dianion, host 1
adopts a slightly distorted conformation in which the aromatic
ring of terephthalate2- is no longer in the same plane as the host
and as such has a reduced capacity to deshield the framework C–H
protons.
1
in Table 1, stacked H NMR spectra of these experiments are
supplied as ESI.†
Two armed hosts 1 and 2
The investigation of receptors 1 and 28a revealed that both
hosts formed strong 1:1 host:guest (H:G) complexes with
terephthalate2-, where the dianion spanned the cleft and was
bound cooperatively by both thiourea groups (4 H-bond donors
in total). Association constants were calculated for the single step
process (H + G ꢀ HG) using WinEQNMR15 and it was noted that
the 4-nitrophenyl functionalised hosts 1b and 2b both had a higher
affinity for terephthalate2- than their 4-fluorophenyl counterparts
1a and 2a. The difference in binding affinity was attributed to the
larger electron withdrawing effect of the nitrophenyl substituent
(versus the fluorophenyl) inductively increasing the acidity of the
adjacent thiourea protons, thereby enhancing their H-bonding
ability.
Three armed hosts 3 and 4
In the case of the new 3-armed [n]polynorbornane based hosts 3
and 4, Job plots (Fig. 3) immediately indicated the formation of
1:1 H:G complexes in both cases. Binding isotherms revealed that
the terephthalate2- guest was bound solely through the thiourea
Nevertheless, the dominant factor influencing binding affinity
was that of the preorganisation imparted by the [n]polynorbornane
scaffold; where the [5]polynorbornane based host 2b (logK = 6.0)
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 4233–4240 | 4235
©

View Article Online
Fig. 5 Proposed binding conformations of the 1:1 complexes formed by
hosts 3 and 4 when binding terephthalate2-.
of host 4 (logK = 4.9) for terephthalate2- was greater than that
of host 3 (logK = 3.6). Monitoring the internal framework, C–H
resonances of host 4†also revealed a similar trend to that of host
2 where a more pronounced downfield shift was observed due to
better accommodation of the guest.
Fig. 3 Job plots of hosts 3b–6b upon the addition of terephthalate2-.
H-bond donors with no assistance from the amide H-bond donors
(Fig. 4). Unlike hosts 1 and 2, receptors 3 and 4 are not C2
symmetric and as such a total of five resonances (one for each
H-bond donor) could be followed during 1H NMR titration anal-
ysis providing additional information regarding possible binding
modes.
Four armed hosts 5 and 6
In the case of the new 4-armed [n]polynorbornane based hosts 5
and 6, it was immediately apparent from Job plot data (Fig. 3)
that host 5 exhibited a 1:1 host:guest stoichiometry while host
6 was arranged in a 1:2 H:G complex with terephthalate2-.
Binding isotherms (Fig. 6) clearly indicated that both hosts bound
the dicarboxylate guests symmetrically through all 4 thiourea
groups (8 H-bonds in total), whilst the chemical shifts of the
amide H-bond donors remained unchanged indicating little or
no participation in the binding event. It was interesting that host
5 bound a single terephthalate2- guest, whereas host 6 bound two
terephthalate2- dianions, forming four H-bonds with each (Fig. 7).
Fig. 4 Titration isotherms of hosts 3b (top), and 4b (bottom) upon the
addition of terephthalate2- (the coloured curves represent each of the
coloured H-bond donors shown in Fig. 5).
These binding isotherms (Fig. 4) clearly show that in both cases
the terephthalate2- is bound cooperatively through all 3 thiourea
groups (6 H-bond donors in total: two from the single armed side
of the host (orange and green) and four from the 2-armed side of
the host (red and pink) as shown in Fig. 5).
As observed for hosts 1 and 2 the dominant factor governing
the binding event was the larger cleft width provided by the
[5]polynorbonane host 4 which better accommodated the tereph-
thalate guest than the [3]polynorbornane based host 3. This was
reflected by the calculated association constants where the affinity
Fig. 6 Titration isotherms of hosts 5b (top), and 6b (bottom) upon the
addition of terephthalate2- (the coloured curves represent each of the
coloured H-bond donors shown in Fig. 7).
The fact that the shifts associated with the thiourea H-bond
donors of host 6 were approximately double that of host 5 (at the
4236 | Org. Biomol. Chem., 2009, 7, 4233–4240
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
Conclusions
In summary, the successful synthesis of 2, 3 and 4 arm func-
tionalised [3] and [5]polynorbornane scaffolds 1–6 was accom-
plished through the successful application of 1,3 dipolar ACE
cycloaddition methodology. In doing so the array of preorganised
[n]polynorbornane based anion hosts has been expanded to
include those with binding pockets that provide 4, 8 or 12 H-
bond donors arranged around cavities of customisable geometric
lengths.
Hosts 1–5 bound terephthalate in a 1:1 H:G arrangement, where
the 2-armed hosts 1 and 2 exhibited symmetrical binding through
four cooperative thiourea H-bonds (Fig. 2), the 3-armed hosts
3 and 4 bound the guest unsymmetrically through six H-bonds
(Fig. 5) and the 4-armed host 5 used eight thiourea H-bonds
(Fig. 7) to bind the dianion. The only exception to this general
trend was host 6 which exhibited a 1:2 H:G stoichiometry; the
increased number of H-bond donors coupled with the larger
[5]polynorbornane scaffold and increased flexibility of the arms
enabled two terephthalate2- guests to ‘coinhabit’ the cavity.
In all cases the 4-nitrophenyl derivative of each host had a
slightly higher affinity for the guest than the 4-fluoro derivative
due to the enhanced H-bonding ability of the thiourea H-bond
donors. However, the prevailing factor controlling host–guest
affinity was the size of the preorganised cleft/cavity imparted by
the [n]polynorbornane scaffold, in all cases the [5]polynorbornane
scaffold provided a cleft/cavity length that better complemented
the guest, resulting in increased binding affinities.
Fig. 7 Proposed binding conformations of the 1:1 and 1:2 arrangements
exhibited by hosts 5 and 6, respectively, when binding terephthalate2-.
equivalence point Dd6 = 3.5 and Dd5 = 1.8 ppm), also supported
the 1:2 stoichiometry determined from Job plots. When complexed
with host 5 the electron density of each of the two carboxylates
of the single terephthalate dianion was being shared between four
thiourea H-bond donors, compared with host 6, where each of
the four carboxylates of the two guests were being shared between
only two thiourea H-bond donors (Fig. 7). As such, the electron
density withdrawn from each thiourea H-bond donor of host 6
was approximately double that of host 5, resulting in a larger
1
downfield shift being observed in the H NMR spectrum. Upon
re-examination of the isotherm for host 4b (Fig. 4) a similar
trend was observed where the thiourea N–H resonances of the
2 armed side of the host experience a change in chemical shift
approximately half that (at the equivalence point) of the N–H
resonances at the single armed side. Thus a consistent ‘dilutive’
effect in the change of chemical shift occurred when a single
carboxylate was ‘shared’ amongst multiple co-operating H-bond
donors.
The longer cavity of [5]polynorbornane host 6 was able
to encapsulate two rigid terephthalate dianions as the guests
can both span the cleft parallel to one another (Fig. 8). The
4-armed receptors 6a and 6b are more flexible than their 2-armed
counterparts 2a and 2b; the thiourea functionalised arms of 2a
and 2b are attached through a rigid five-membered cyclic imide,
whereas the arms of 6a and 6b are linked through a more flexible
amide bond. This enables greater variation in possible binding
conformations including a conformation in which the anions can
be held sufficiently far enough apart to allow a 1:2 H:G complex
to exist with little or no electrostatic repulsion (Fig. 8). This
1:2 binding mode was further supported by re-examination of
the internal framework C–H resonances of host 6b where the
magnitude of the downfield shift was approximately double that
observed for host 4b.†
Experimental
NMR spectra were collected on either a JEOL EX 270 MHz
FT-NMR spectrometer (Tokyo, Japan), or a JEOL EX 400 MHz
FT-NMR spectrometer (Tokyo, Japan) where indicated. HRMS
was performed with an Agilent 6210 LC/MSDTOF instrument
using CH3CN as the mobile phase. Melting points were deter-
R
mined on a digital Electrothermalꢀ 9200 (UK) heated-block
melting point apparatus and are uncorrected. Microanalysis was
performed by Chemical and Microanalytical Services Pty Ltd,
Belmont, Geelong, 3216. TLC was performed using Merck 60
F254 aluminium backed silica plates. Visualisation employed
a UVP Mineralight 254 NM UV lamp or an oxidising dip
containing KMnO4 (1.0 g), K2CO3 (1.0 g) and H2O (100 mL).
Flash chromatography was performed using Merck Kieselgel 60
(70–230 mesh). General reagents were analytical grade and used as
supplied unless otherwise stated. Isothiocyanates were purchased
from Aldrich Chemical Co.
General two step protocol for epoxide synthesis
Step
1
Mitsudo reaction. Freshly distilled dimethyl
acetylenedicarboxylate (DMAD) (1.3 eq.) was added to a
pressure vessel containing the starting norbornene unit (1.0 eq.)
and RuH2(CO)(PPh3)3 (0.10 eq.) in dry THF. The tube was sealed
and the mixture stirred at 70–90 ◦C for up to 48 h. Following
cooling, the pressure vessel was opened, the reaction mixture
filtered and the solvent removed under reduced pressure. The
resultant crude solid was subject to column chromatography to
afford the cyclobutene diester.
Fig. 8 Illustration of the 1:2 binding of host 6b with terephthalate2-.
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 4233–4240 | 4237
©

View Article Online
Step 2 Epoxidation. A nitrogen-flushed solution of alkene
diester (1.0 eq.) in dry THF was cooled to 0 ◦C in an ice bath
before tert-butyl hydroperoxide (TBHP) (1.2 eq. of a 1.14 M
toluene solution) was added by syringe. Following rapid stirring
for 10 min, potassium tertbutoxide (KOtBu) (0.25 eq.) was added
and the reaction stirred for another 10 min. The ice bath was
removed and the solution stirred at RT overnight whereupon
aqueous sodium thiosulfate solution (10%) was added and the
mixture stirred for a further 30 min. The resultant two-phase
mixture was concentrated to 1/3 of its volume, transferred to a
separatory funnel and extracted with CHCl3 (¥3), the organics
combined, dried (MgSO4), filtered and evaporated to dryness
under reduced pressure. The crude solid was purified by column
chromatography and/or recrystallisation resulting in the requisite
cyclobutane epoxide as a white solid.
Dimethyl (2b,3a,4a,8a,9a,10b,12b,13a,16a,17b) 6-tertbutoxy-
carbonylaminoethyl-14,15-di[(tertbutoxycarbonylamino)ethyl
carboximido]-6-aza-19-oxa-5,7-dioxoheptacyclo [9.6.1.13,9.113,16
02,10.04,8.012,17] icosa-14-ene-1,11-dicarboxylate 14
.
Equimolar amounts of epoxide 88a (902 mg, 1.94 mmol) and
norbornene 912 (901 mg, 1.94 mmol) were heated (150 ◦C) in
THF (4 ml) for 70 h. Column chromatography (EtOAc, Rf =
0.37) gave 14 (1.601 g, 1.72 mmol, 89%) as an off-white powder;
◦
mp 147.2–150.7 C; dH(270 MHz; d6-DMSO; Me4Si) 1.02 (1 H,
d, J = 7.8 Hz, H20), 1.18 (1 H, d, J = 9.8 Hz, H18), 1.34 (9
H, s, 1 ¥ C(CH3)3), 1.36 (18 H, s, 2 ¥ C(CH3)3), 2.02 (2 H, s,
H2,10), 2.24 (1 H, d, J = 8.2 Hz, H20), 2.31 (2 H, s, H13,16),
2.33 (1 H, d, J = 8.2 Hz, H18), 2.43 (2 H, s, H3,9), 3.02 (4
H, s, 2 ¥ CH2NHCO), 3.04 (4 H, s, H4,8,12,17), 3.11 (4 H, s,
2 ¥ CH2NHCOO), 3.14 (2 H, s, 1 ¥ CH2NHCOO), 3.41 (2 H, s,
1 ¥ CH2N), 3.84 (6 H, s, 2 ¥ COOCH3), 6.78 (1 H, s, 1 ¥ NHCOO),
6.84 (2 H, t, J = 5.1 Hz, 2 ¥ NHCOO) and 8.89 (2 H, s, 2 ¥ NHCO);
dC(270 MHz; CDCl3; Me4Si) 28.3, 28.4, 38.1, 39.0, 40.0, 41.0, 48.4,
48.5, 50.9, 52.8, 53.8, 79.7, 89.1, 148.4, 156.1, 156.7, 164.3, 168.2,
176.9; HRMS: m/z = 929.45020 [M + H]+, C45H65N6O15 requires
929.45024.
Dimethyl (1a,2a,6a,7a,8b,12b)-4-(2¢-tertbutoxycarbonylamino
ethyl)-4-aza-10-oxapentacyclo [5.5.1.02,6.08,12.09,11]trideca-3,5-
dione-9,11-dicarboxylate 88a
Prepared according to a standard two step protocol; Step
1: (1a,2a,6a,7a)-4-(2¢-tert-Butoxycarbonylaminoethyl)-4-azatri-
cyclo[5.2.1.02,6]deca-8-ene-3,5-dione 78a (5.002 g, 16.34 mmol),
DMAD (2.5 ml, 20 mmol) and RuH2(CO)(PPh3)3 (401 mg,
See ESI for the synthesis of [n]polynorbornenes 15, 16, 17
and 18.
◦
0.436 mmol) in THF (20 ml) at 70 C for 48 h to give (1.823 g,
3.005 mmol, 94%) the cyclobutene diester as a white powder
following chromatographic purification (50% EtOAc/Pet Sp, Rf =
0.26); mp 153.8–154.7 ◦C; dH(400 MHz; CDCl3; Me4Si) 1.37
(9H, s, C(CH3)3), 1.47 (1 H, d, J = 11.6 Hz, H12), 1.74 (1 H,
d, J = 11.6 Hz, H12), 2.79 (2 H, s, H8,11), 2.83 (2 H, s, H1,7),
3.22 (2 H, s, H2,6), 3.33 (2 H, m, NHCH2), 3.61 (2 H, m, NCH2),
3.76 (6 H, s, 2 ¥ COOCH3) and 4.70 (1 H, br s, NH); dC(400 MHz;
CDCl3; Me4Si) 27.9, 33.7, 35.5, 37.8, 38.4, 42.0, 47.0, 51.5, 79.0,
140.7, 155.8, 160.1 and 176.5; HRMS: m/z = 449.19030 [M + H]+,
C22H29N2O8 requires 449.19240.
General hydrogenation procedure
A
round-bottomed flask was charged with the required
[n]polynorbornene scaffold, catalytic Pd(OH)2/C, EtOH and
equipped with a reflux condenser. This reaction mixture was stirred
at 45 ◦C under a H2 atmosphere for up to 48 h. After cooling to
RT the solution was diluted with EtOH, filtered through celite and
solvent removed under reduced pressure. The product was purified
using column chromatography.
Dimethyl (2b,3a,4a,8a,9a,10b,12b,13a,14a,15a,16a,17b)
6-tertbutoxycarbonylaminoethyl-14b,15b-di[(tertbutoxy-
carbonylamino)ethylcarboximido]-6-aza-19-oxa-5,7-dioxo-
heptacyclo [9.6.1.13,9.113,16.02,10.04,8.012,17] icosa-1,11-dicarboxylate
Step 2. The above diester (1.197 g, 2.670 mmol) was epoxidised
with TBHP (1.14M, 3.0 ml, 3.4 mmol) and KOtBu (81 mg,
0.72 mmol) in dry THF (150 ml) for 28 h. Column chromatography
(50% EtOAc/Pet Sp, Rf = 0.22) afforded 88a (854 mg, 1.84 mmol,
69%) as a white solid; mp 190.8–192.2 ◦C; dH(270 MHz; CDCl3;
Me4Si) 1.39 (9 H, s, C(CH3)3), 1.72 (1 H, d, J = 11.5 Hz, H13), 2.12
(1 H, d, J = 11.5 Hz, H13), 2.37 (2 H, s, H8,12), 3.17 (2 H, s, H1,7),
3.27 (2 H, m, NHCH2), 3.29 (2 H, s, H2,6), 3.55 (2 H, m, NCH2),
3.79 (6 H, s, 2 ¥ COOCH3) and 4.68 (1 H, br s, NH); dC(400 MHz;
CDCl3; Me4Si) 27.9, 34.2, 36.4, 38.6, 38.7, 45.6, 47.1, 52.5, 79.1,
141.2, 155.5, 163.3 and 176.2; HRMS: m/z = 465.18621 [M + H]+,
C22H29N2O9 requires 465.18732.
The [3]polynorbornene 14 (0.891 mg, 0.959 mmol) was subject
to hydrogenation conditions in EtOH (30 ml) for 48 h, column
chromatography (10% EtOH/EtOAc, Rf = 0.42) gave the 3-
armed [3]polynorbornane scaffold (0.879 mg, 0.944 mmol, 98%)
as a white powder; mp 152.1–154.2 ◦C; dH(270 MHz; d6-DMSO;
Me4Si) 0.81 (1 H, d, J = 9.2 Hz, H20), 1.17 (1 H, d, J = 6.7 Hz,
H18), 1.35 (9 H, s, 1 ¥ C(CH3)3), 1.37 (18 H, s, 2 ¥ C(CH3)3),
1.81 (2 H, s, H13,16), 2.10 (2 H, s, H2,10), 2.18 (1 H, d, J =
9.4 Hz, H20), 2.33 (1 H, d, J = 8.2 Hz, H18), 2.38 (2 H, s,
H3,9), 2.59 (2 H, s, H14,15), 2.72 (2 H, s, H12,17), 2.94 (6 H, s,
2 ¥ CH2NHCO, H4,8), 3.01 (6 H, br s, 3 ¥ CH2NHCOO), 3.47
(2 H, br s, 1 ¥ CH2N), 3.77 (6 H, s, 2 ¥ COOCH3), 6.59 (2
H, s, 2 ¥ NHCOO), 6.85 (1 H, t, J = 5.4 Hz, 1 ¥ NHCOO)
and 7.56 (2 H, s, 2 ¥ NHCO); dC(270 MHz; d6-DMSO; Me4Si)
28.8, 29.6, 29.9, 34.8, 37.5, 37.8, 38.8, 39.4, 40.8, 43.1, 46.8, 48.3,
49.7, 50.7, 52.7, 78.2, 78.3, 90.4, 156.1, 156.3, 169.0, 171.6 and
177.4; HRMS: m/z = 931.46566 [M + H]+, C45H67N6O15 requires
931.46589.
See ESI for the syntheis of epoxides 10, 12 and 13.
General ACE coupling reaction
A pressure vessel was charged with equimolar amounts of the
desired epoxide and norbornene unit (two eq. in the case of bis
epoxides), the minimum amount of tetrahydrofuran (THF) added
for dissolution then the vessel sealed and heated at 140–150 ◦C
for up to 72 h. The vessel was then cooled, opened, the solvent
removed under reduced pressure, and the crude purified using flash
chromatography.
See ESI for the hydrogenation of compounds 16, 17 and 18.
4238 | Org. Biomol. Chem., 2009, 7, 4233–4240
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
General Boc deprotection and isothiocyanate coupling
1H NMR spectroscopy titration technique
The required [n]polynorbornane was stirred in 20% TFA/DCM
for 2–4 h, then both TFA and DCM were removed under reduced
pressure. The resulting crude solid was dissolved in CHCl3 then
evaporated to dryness (twice) to ensure complete removal of
TFA. The resultant free amine was dissolved in CHCl3, then
diisopropylethylamine (DIPEA) and the required equivalents of
isothiocyanate added. Following stirring at RT for 18–46 h,
the solvent and excess DIPEA were removed under reduced
pressure and the resultant solid purified using flash chromato-
graphy.
A JEOL EX 270 MHz FT-NMR spectrometer (Tokyo, Japan)
was employed to conduct the H NMR spectroscopy titration
1
experiments. In general, a stock solution of each host was made
up to 2.5 ¥ 10-3 M in d6-DMSO, 600 ml of this solution was
then transferred to an NMR tube and the spectrum collected.
The chemical shift (ppm) of the resonances corresponding to the
amide and thiourea H-bond donors, as well as the internal C–H
protons, were recorded. A 5 ml aliquot of the stock guest solution
(TBA terephthalate, ~3.0 ¥ 10-2 M in d6-DMSO) containing 0.1
equivalents (eq.) of terephthalate2- was then added to the host
solution by auto-pipette, this solution was briefly mixed in the
1
NMR tube then the H NMR spectrum collected. Again, the
chemical shift of the resonances corresponding to the amide and
thiourea H-bond donors, as well as the internal C–H protons, were
recorded, this process was repeated until 2.0 eq. of guest had been
added. The aliquot was then increased (10 ml) so that it contained
0.2 eq. of guest, and the procedure repeated until a total of 4.0 eq.
of guest had been added. The final additions were made using
20 ml aliquots (containing 0.4 eq. of guest) until a total of 6.0 eq.
of terephthalate2- had been added. The data was then plotted
as a titration isotherm and the association constant determined
through a non-linear regression analysis using WinEQNMR.15
In cases where the resonances became significantly broadened a
smoothing function was used to more accurately determine the
Dimethyl (2b,3a,4a,8a,9a,10b,12b,13a,14a,15a,16a,17b)
6-(4-fluorophenylthioureido)ethyl-14b,15b-di[(4-fluorophenyl-
thioureido)ethylcarboximido]-6-aza-19-oxa-5,7-dioxohepta-cyclo
[9.6.1.13,9.113,16.02,10.04,8.012,17] icosa-1,11-dicarboxylate 3a
Dimethyl (2b,3a,4a,8a,9a,10b,12b,13a,16a,17b) 6-tert-buto-
xycarbonylaminoethyl-14b,15b-di[(tertbutoxycarbonyl-amino)
ethylcarboximido]-6-aza-19-oxa-5,7-dioxoheptacyclo [9.6.1.13,9.
113,16.02,10.04,8.012,17]icosa-1,11-dicarboxylate (690 mg, 0.741 mmol)
was deprotected (5 ml, 20% TFA/DCM) in 2 h to yield the
free triamine which underwent coupling with 4-fluorophenyl-
isothiocyanate (341 mg, 2.23 mmol) and DIPEA (1.2 ml,
6.9 mmol) in CHCl3 (15 ml) for 20 h. The resultant crude off-white
solid was purified by column chromatography (5% EtOH/EtOAc,
Rf = 0.34) to afford host 3a (752 mg, 0.689 mmol, 93%) as a white
powder; mp 181.1–185.6 ◦C; dH(270 MHz; d6-DMSO; Me4Si) 0.83
(1 H, d, J = 9.1 Hz, H20), 1.17 (1 H, d, J = 9.2 Hz, H18), 1.85
(2 H, s, H2,10), 2.15 (2 H, s, H13,16), 2.21 (1 H, d, J = 10.2 Hz,
H20), 2.29 (1 H, d, J = 10.1 Hz, H18), 2.40 (2 H, s, H3,9), 2.62
(2 H, s, H14,15), 2.84 (2 H, s, H12,17), 3.01 (2 H, s, H4,8), 3.14 (4
H, s, 2 ¥ CH2NHCO), 3.46 (4 H, s, 2 ¥ CH2NHCS), 3.60 (2 H, s,
1 ¥ CH2NHCS), 3.62 (2 H, s, NCH2), 3.77 (6 H, s, 2 ¥ COOCH3),
7.17 (6 H, q, J = 7.2 Hz, 6 ¥ ArCHCF), 7.27 (2 H, t, J = 6.7 Hz,
2 ¥ ArCHCNH), 7.38 (4 H, t, J = 6.6 Hz, 4 ¥ ArCHCNH), 7.67
(5 H, br s, 2 ¥ CONH, 3 ¥ CH2NHCS), 9.51 (1 H, s, 1 ¥ ArNH)
and 9.53 (2 H, s, 2 ¥ ArNH); dC(270 MHz; d6-DMSO; Me4Si)
35.1, 37.7, 38.4, 38.6, 40.9, 42.2, 43.2, 44.1, 46.8, 48.6, 49.6, 50.8,
52.8, 90.5, 115.9, 126.8, 135.5, 159.6, 169.1, 171.8, 177.6 and 181.5;
HRMS: m/z = 1090.33996 [M + H]+, C51H55N9O9F3S3 requires
1090.32315.
1
centre of the peak. Stack plots of the H NMR titration spectra
are provided as ESI.†
Hartree–Fock molecular model calculations
The molecular models of Fig. 2 were calculated using Spartan
’04 (Wavefunction Inc., California, 2004). Both the hosts and
guest were constructed using the builder function, the guest was
then arranged within the cleft so that each of the carboxylates
was in close proximity to the thiourea H-bonding motifs. The
energy minimize function, which employs Merck Molecular Force
Field (MMFF) molecular mechanics to minimize the steric energy,
was used to optimize the geometry before the Hartree–Fock
calculation was performed using the following constraints and
conditions. Properties: total charge–dianion; multiplicity–singlet;
Calculation: calculate–equilibrium geometry; at–ground state;
with–Hartree–Fock 3–21G(*); start from–initial (MMFF) geome-
try; subject to–symmetry. Results: host 1b; CPU time–021:23:53.1;
Energy = -4399.62 Eh; host 2b; CPU time–051:38:45.7; Energy =
-5269.56 Eh.
See ESI for the synthesis of hosts 3b and 4a,b–6a,b.
Tetrabutylammonium terephthalate
Notes and references
1 For reviews see: (a) P. A. Gale, J. L. Sessler, W.-S. Cho, Anion receptor
chemistry, Wiley-VCH, Weinheim, 2007; (b) A. Bianchi, K. Bowman-
James, and E. Garcia´, Supramolecular Chemistry of Anions, Wiley-
VCH, New York, 1997; (c) T. Gunnlaugsson, M. Glynn, G. M. Tocci,
P. E. Kruger and F. M. Pfeffer, Coord. Chem. Rev., 2006, 250, 3094–
3117; (d) P. D. Beer and P. A. Gale, Angew. Chem., Int. Ed., 2001, 40,
486–516; (e) P. A. Gale, S. E. Garcia-Garrido and J. Garric, Chem.
Soc. Rev., 2008, 37, 151–190; (f) J. W. Steed, Chem. Soc. Rev., 2009, 38,
506–519; (g) B. T. Nguyen and E. V. Anslyn, Coord. Chem. Rev., 2006,
250, 3118–3127.
One equivalent of terephthalic acid (150 mg, 0.903 mmol) was
stirred in two equivalents of tetrabutylammonium hydroxide in
MeOH (1.0 M, 1.8 ml, 1.8 mmol) for 48 hours. Excess MeOH
was then removed under reduced pressure, complete dryness was
obtained by heating (70 ◦C) the crude white solid under vacuum for
2 days to yield (568 mg, 96.9%) a white powder; mp 98.2–102.1 ◦C;
dH(400 MHz; DMSO; TMS) 0.93 (24 H, t, J 6.5, CH3), 1.30 (16
H, sextet, J 7.2, CH2CH3); 1.57 (16 H, q, J 7.2, CH2CH2CH3);
3.17 (16 H, t, J 7.8, CH2CH2CH2CH3) and 7.62 (4 H, d, J 1.7,
Ar-CH); dC(400 MHz; DMSO; TMS) 14.06, 19.78, 23.65, 58.13,
128.06, 142.28 and 169.15.
2 M. Cox and D. Nelson, Lehninger Principles of Biochemistry, 3rd edn.,
Worth Publishers, New York, 2000; J.-L. Wu, Y.-B. He, Z.-Y. Zeng,
L.-H. Wei, L.-Z. Meng and T.-X. Yang, Tetrahedron, 2004, 60, 4309–
4314.
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 4233–4240 | 4239
©

View Article Online
3 (a) T. Addiscott, A. Whitmore and D. Powlson, ed. Farming, Fertilisers
and the Nitrate Problem, CAB International, Wallingford, 1991; (b) D.
Ballantine, D. E. Walling and G. J. L. Leeks, Water, Air, Soil Pollut.,
2009, 196, 311–320.
4 (a) D. J. Cram, Angew. Chem., 1986, 98, 1041–1060 (Angew. Chem.
Int. Ed., 1986, 25, 1039–1057); D. J. Cram, Angew. Chem., 1988, 100,
1041–1052 (Angew. Chem. Int. Ed., 1988, 27, 1009–1020).
B. Branchi, V. Balzani, P. Ceroni, M. C. Kuchenbrandt, F. G.
Klarner, D. Blaser and R. Boeso, J. Org. Chem., 2008, 73, 5839–
5851.
9 (a) R. N. Warrener, A. C. Schultz, D. N. Butler, S. D. Wang, I. B.
Mahadeban and R. A. Russell, Chem. Commun., 1997, 1023–1024;
(b) R. C. Foitzik, A. J. Lowe and F. M. Pfeffer, Tetrahedron Lett., 2009,
50, 2583–2584 For examples see: L. D. V. Vliet, T. Ellis, P. J. Foley, L.
Liu, F. M. Pfeffer, R. A. Russell, R. N. Warrener, F. Hollfelder and
M. J. Waring, J. Med. Chem., 2007, 50, 2326–2340; F. M. Pfeffer and
R. A. Russell, Org. Biomol. Chem., 2003, 1, 1845–1851; R. N. Warrener,
D. N. Butler, D. Margetic, F. M. Pfeffer and R. A. Russell, Tetrahedron
Lett., 2000, 41, 4671–4675.
10 (a) T. Mitsudo, T. Shinsugi, Y. Nakagawa and Y. J. Takegami, J. Org.
Chem., 1979, 44, 4492; (b) T. Mitsudo, T. Kondo, Y. Ozaki and Y.
Watanabe, Angew. Chem. Int. Ed., 1994, 33, 580–581; (c) T. Mitsudo
and T. Kondo, Proc. Jpn. Acad., Ser. B, 2007, 83, 65–76.
11 (a) E. Weitz and A. Scheffer, Ber. Dtsch. Chem. Ges., 1921, 54, 2327;
(b) C. Clark, P. Hermans, O. Meth-Cohn, C. Moore, H. C. Taljaard and
G. van Vuuren, Chem. Commun., 1986, 1378–1380; (c) B. K. Sharpless,
B. E. Rossiter and G. Hill, J. Org. Chem., 1983, 48, 3607–3608.
12 (a) A. J. Lowe, G. A. Dyson and F. M. Pfeffer, Eur. J. Org,. Chem.,
2008, 1559–1567; (b) A. J. Lowe, G. A. Dyson and F. M. Pfeffer, Org.
Biomol. Chem., 2007, 5, 1343–1346.
5 (a) C. Lee, H. Miyaji, D. Yoon and J. L. Sessler, Chem. Commun., 2008,
24–34; (b) J. P. Anzenbacher, K. Jursikova´ and J. L. Sessler, J. Am.
Chem. Soc., 2000, 122, 9350–9351; (c) E. Quinlan, S. E. Matthews
and T. Gunnlaugsson, J. Org. Chem., 2007, 72, 7497–7503; (d) S. E.
Matthews and P. D. Beer, Supramol. Chem., 2005, 17, 411–435.
6 (a) J. L. Sessler, S. Camiolo and P. A. Gale, Coord. Chem. Rev., 2003,
240, 17–55; (b) D. W. Yoon, D. E. Gross, V. M. Lynch, C. H. Lee, P. C.
Bennett and J. L. Sessler, Chem. Commun., 2009, 1109–1111; (c) J. L.
Sessler, D. E. Gross, W. S. Cho, V. M. Lynch, F. P. Schmidtchen, G. W.
Bates, M. E. Light and P. A. Gale, J. Am. Chem. Soc., 2006, 128,
12281–12288.
7 (a) A. P. Davis, Coord. Chem. Rev., 2006, 250, 2939–2951; (b) A. J. P.
Clare, J.-B. Joos, A. L. Sisson, G. Magro, M. N. Pe´rez-Paya´n, T. N.
Lambert, R. Shukla, B. D. Smith and A. P. Davis, J. Am. Chem. Soc.,
2005, 127, 10739–70746; (c) Y. J. Kim, H. Kwak, S. J. Lee, J. S. Lee,
H. J. Kwon, S. H. Nam, K. Lee and C. Kim, Tetrahedron, 2006, 62,
9635–9640; (d) A. P. Davis and J. B. Joos, Coord. Chem. Rev., 2003, 240,
143–156.
13 (a) C. D. Smith, J. Am. Chem. Soc., 1966, 88, 4273–4274; (b) R. W.
Warrener and D. N. Butler, Aldrichimica Acta, 1997, 30, 119–130.
14 See: S.-Y. Liu, Y.-B. He, W. H. Chan and A. W. M. Lee, Tetrahedron,
2006, 62, 11687–11696.
15 M. J. Hynes, J. Chem,. Soc., Dalton Trans., 1993, 311.
16 G. C. Bassler, R. M. Silverstein, T. C. Morrill, Spectrometric Identifi-
cation of Organic Compounds, 4th edn., John Wiley & Sons, Inc., New
York, 1981.
8 (a) A. J. Lowe and F. M. Pfeffer, Chem. Commun., 2008, 1871–1873;
(b) F. M. Pfeffer, T. Gunnlaugsson, P. Jensen and P. E. Kruger, Org.
Lett., 2005, 7, 5357–5360; (c) F. M. Pfeffer, P. E. Kruger and T.
Gunnlaugsson, Org. Biomol. Chem., 2007, 5, 1894–1902. For related
supramolecular hosts see: F. G. Klarner, M. Lobert, U. Naatz,
H. Bandmann and R. Boese, Chem.–Eur. J., 2003, 9, 5036–5047;
4240 | Org. Biomol. Chem., 2009, 7, 4233–4240
This journal is
The Royal Society of Chemistry 2009
©