Fluorescent PET chemosensors for lithium

Article abstract of DOI:10.1016/j.tet.2004.04.082

The synthesis of the chiral diaza-9-crown-3 derivatives 1 (S,S) and 2 (R,R) is described. These sensors were designed as luminescent chemosensors for lithium where the fluorescence emission from the naphthalene moieties was 'switched on' upon Li⁺

Full text of DOI:10.1016/j.tet.2004.04.082

Tetrahedron 60 (2004) 5799–5806
Fluorescent PET chemosensors for lithium
*
Thorfinnur Gunnlaugsson, Bastien Bichell and Claire Nolan
Department of Chemistry, Trinity College Dublin, Dublin 2, Ireland
Received 16 January 2004; revised 5 April 2004; accepted 29 April 2004
Abstract—The synthesis of the chiral diaza-9-crown-3 derivatives 1 (S,S) and 2 (R,R) is described. These sensors were designed as
luminescent chemosensors for lithium where the fluorescence emission from the naphthalene moieties was ‘switched on’ upon Liþ
recognition by the crown ether moiety in organic solvents, showing excellent selectivity over other group I and group II cations. Even though
the recognition of Liþ was not achieved in water (pH 7.4) or aqueous alcohol solution, the fluorescence (which was switched on at pH 7.4)
was substantially modulated by spherical anions, where the fluorescence emission was quenched in the presence of Br2 and I2 but less by
Cl2 and not by acetate. This indicates that the emission was quenched by heavy-atom affect. The recognition of Liþ was also investigated by
1H NMR in CD3CN and by observing the changes in the circular dicromism spectra. For the former, the resonances for the crown ether
moiety and a-methyl protons adjacent to the ring were sifted upfield and broadened, whereas for 1 the intensity of the CD signal for the p–p
transition was substantially modulated upon Liþ recognition.
q 2004 Elsevier Ltd. All rights reserved.
1. Introduction
(approximately 2 g) per day for the treatment of mental
illness. The therapeutic index for Liþ is narrow and should
lie between 0.4 and 0.8 mM in serum, 12 h after the dose has
been administered. If the serum concentration is too high
(around 1.5 mM), shaking, dizziness, drowsiness, vomiting
and diarrhoea are experienced by the patient. These
syndromes indicate serious toxicity effects and are usually
seen 4 h after the drug has been administered. Long-term
side effects include dermatological disorders, weight gain
and some problems with kidney and thyroid functions.8 In
medicine, Liþ determination in blood samples was
traditionally carried out using atomic absorption
spectroscopy and flame emission spectroscopy. The
impracticalities of measuring serum samples on site using
these methods led to the development of ion-selective
electrodes, and ionophores which are more practical.10–12
They work by measuring the activity in Liþ solutions and
are active within the clinical range (0.4–0.8 mM serum).
This technique provides immediate feedback without long
delays, high operation, instrumentation costs and bulkiness
of instruments. Determination of Liþ levels in serum must
also be monitored in the presence of 140 mM sodium,
4.3 mM potassium and 1.26 mM calcium. Furthermore, Liþ
is important from a synthetic as well as an industrial point of
view because of its use in batteries, but currently there is
real momentum away from cadmium-based batteries
towards the use of lithium in mobile phones. Lithium is
thus also a potential environmental hazard in the future.
The design and synthesis of chemosensors for ions and
neutral molecules has been an area of immeasurable study in
recent years.1 Chemical sensors have non-invasive and non-
destructive properties, which has made them an important
diagnostic tool in medicine and industry.2 Many are based
on the use of the ‘fluorophore–spacer–receptor’ model,
where the fluorescence emission at the fluorophore site is
modulated by ion or molecular recognition at the receptor
site, which usually results in the suppression of photo-
induced electron transfer (PET) quenching, operating
between the two moieties.3 Although PET sensors have
been developed for a wide range of analytes such as neutral
molecule, zwitterions, cations, Naþ, Kþ, Ca2þ, Mg2þ, Cd2þ
and Zn2þ anions such as acetates, halides, phosphates and
biologically active bis-carboxylates and pyrophosphate,5
then to the best of our knowledge, no Liþ selective PET
sensors have been reported in the literature prior to our
investigation.6,7 Liþ is unusual as it is one of the smallest
and lightest solid elements and has important clinical,
pharmacological and biochemical properties.8 Its beauty
lies in its simplicity, activating brain cells to regulate
abnormal mood cycles for the treatment of mentally ill
patients, such as manic-depressives.8,9 New uses of Liþ
include the treatment of skin diseases (such as dermatitis)
and autoimmune and immunological diseases. It is usually
administered orally as Li2CO3, at a total dose up to 30 mmol
We have developed both fluorescent and lanthanide
luminescent devices as luminescent switches,13 sensors14
and logic gate mimics.15 We have focused our efforts
particularly on the development of chemosensors for
Keywords: Fluorescent sensors; Chemosensors; Lithium; Photoinduced
electron transfer; Supramolecular chemistry.
*
Corresponding author. Tel.: þ353-1-608-3459; fax: þ353-1-671-2826;
e-mail address: gunnlaut@tcd.ie
0040–4020/$ - see front matter q 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2004.04.082

5800
T. Gunnlaugsson et al. / Tetrahedron 60 (2004) 5799–5806
biologically relevant targets such as ions, small molecules
and nucleic acids.16 Here we describe our efforts towards
developing novel PET sensors for Liþ.6 We demonstrate
that by using structurally simple motifs, whereby a small
crown ether and two amide moieties are combined in a
receptor, highly selective Liþ chemosensing can be
achieved in organic solvents such as CH3CN. Although
we were unable to achieve such sensing in an aqueous
environment, this work achieved the objectives outlined,
i.e. the selective sensing of Liþ by employing PET
chemosensing.
Williamson ether synthesis, from N,N-ditosyl diamino-
ethane and diethylene glycol ditosylate, which gave the
N,N0-ditosyl-1,4-diaza-9-crown ether. This product was de-
tosylated by refluxing in 48% HBr–AcOH solution for four
days, yielding the HBr salts 5 in 87% yield. The
a-chloroamide arms 3 and 4 were made in a single step
by peptide coupling of chloroacetic acid with either S- or
R-1-[1-naphtyl]ethylamine using EDCI and HOBt as
reactants, giving approximately 87% yield for each. This
compound and other related chiral derivatives have
previously been reported by Parker et al. using a different
synthesis.19 The final step in the above synthesis involved
the coupling of either 3 or 4 to the crown ether 5 in MeCN at
80 8C, giving 1 and 2, initially in good yields of ca. 70% as a
crude material, but in 33 and 30% yield respectively, after
final workup. An increase in the reaction times did not
improve the original yields nor prevent the one-armed
systems 6 and 7 being formed in a small amount, ca. 10% (as
crude material). No advantages were achieved by refluxing
the reaction in DMF, in fact this resulted in lower yields of
(,45%) with increased yield of the one-armed side product
6. Both the R,R- and S,S- sensors had to be purified via
alumina chromatography using DCM and 0!1% MeOH as
an eluent. When flash silica chromatography was used the
sensor became protonated. Consequently, it was necessary
to treat the resulting fractions with 1 M NaOH to yield the
free sensor. The final products were isolated as semi solids,
and titurated from ether to give both products as powders.
Both sensors were fully characterized by, 1H and 13C NMR,
ES-MS, IR and CHN or accurate mass analysis using
HRMS.
2. Results and discussion
2.1. Synthesis
The lithium-selective PET sensors 1 and 2 were designed
using the PET principle by combining a small crown ether
moiety with two naphthalene fluorophores, giving rise to a
‘fluorophore–spacer–receptor–spacer–fluorophore’3 – 7
model as developed by de Silva (Fig. 1). With the aim of
maximizing the Liþ selectivity over other competitive
cations, the rather small diaza-9-crown-3 (1-oxo-4,7-
diazacyclononane)17 crown ether was chosen, which was
functionalized with two amide pendent arms which could
participate directly in the coordination of the ion. Such
modifications have previously been demonstrated for
competitive Liþ recognition and transport across a liquid
membrane.18 We proposed that this design would ensure
that Liþ would not be complexed directly within the
macrocycle cavity, but just below it, where Liþ recognition
would be assisted by the amide arms, i.e. by direct
coordination to the oxygens of the carboxyamides. This
would ensure enhanced Liþ selectivity over Naþ. To
achieve this additional chelating effect we selected the
chiral amides 3 and 4, synthesized from the chiral S- and
R-1-(naphthyl)ethylamine, respectively.
The 1H NMR spectrum of either the S,S-isomer 1 or the R,R-
isomer of 2 showed that the axial and equatorial positions of
the aza-crown ring were not magnetically equivalent, with
several multiplets occurring around 2.3 and 3.6 ppm. The
diagnostic peak is a multiplet at 5.38 ppm representing the
chiral CH group of the spacer. It was expected that this
resonance would be a quartet, so its appearance as a
multiplet suggests that the environments around this proton
are non-equivalent. The electro-spray mass spectrum of 1
and 2 showed a single peak at 552.5 m/z for the molecular
ion.
2.2. Ground and excited state evaluation of 1 and 2
The photophysical properties of 1 and 2 were evaluated in
water, MeOH, in 50:50 MeOH/CH3CN mixture and in
CH3CN solutions in the presence of several metal cations
from groups IA and IIA. In water, using 160 mM NaCl to
maintain constant ionic strength, the pKa of both sensors
was determined by observing the changes in the fluor-
escence emission spectrum at lF 337 nm, when excited at
280 nm. However, the changes in the absorption spectrum,
which displayed typical naphthalene absorption bands with
fine structure at 271, 281 and 293 nm, over the same pH
range were only minor and not significant enough for
accurate binding constant determination. This is due to the
covalent spacers that separate the two naphthalene fluor-
ophores from the crown ether receptor and thus minimize
any p–n orbital interactions between the two moieties. In
alkaline solution the fluorescence of the naphthalene unit of
1 was almost fully ‘switched off’, signifying the quenching
Figure 1. Diagram illustrating the ‘fluorophore–spacer–receptor–spacer–
fluorophore’ model and the corresponding structure of 1.
The synthesis of 1 and 2 is shown in Scheme 1.† The
synthesis of both was achieved in two sequences. The first
sequence was the synthesis of the diaza-9-crown-3-ether
receptor 5, which is a known compound, using conventional
†
In our earlier communication (Ref. 6), the naphthalene part of 1, was
shown to be connected through position two on the aromatic ring.
However, the description of the synthesis in the text was correct. We
apologize for this mistake.

T. Gunnlaugsson et al. / Tetrahedron 60 (2004) 5799–5806
5801
Scheme 1. Synthesis of chemosensors 1 and 2 from the a-chloroamides 3, 4 and the diaza-9-crown-3 (1-oxo-4,7-diazacyclononane) crown ether 5. The mono
products 6 and 7 were also observed in about 84–10% yields.
of the naphthalene excited state by efficient electron transfer
from the amino moieties of the crown ether receptor to the
naphthalene-excited state. This is typical PET quenching, as
on protonation of the amino moiety, the oxidation potential
of the receptor was increased, removing the thermodynamic
pathway for the excited state quenching.1,4 The correspond-
ing changes in the fluorescence emission of 1 at 337 nm as a
function of pH are shown in Figure 2. As the spacer
between the amino moiety and the naphthalene moiety is
made of four carbon bond lengths, which is significantly
Figure 2. The changes in the fluorescence emission spectra of 1 at 337 nm upon pH titration in water. Excitation at 280 nm.

5802
T. Gunnlaugsson et al. / Tetrahedron 60 (2004) 5799–5806
larger then the classical one or two carbon spacers usually
used for PET sensors,1,4 the efficiency of the quenching
process is not ideal as the PET is distance dependent
(function of 1/r6). Hence the emission is not fully switched
off in alkaline solution. However, upon protonation of the
amino receptor the emission was ‘switched on’ with a large
order of magnitude enhancement in fluorescence, without
any other substantial spectral shifts. This process was fully
reversible, as upon addition of base to 1 in acidic solution,
the emission was switched off. The changes in Figure 2
occur over ca. two pH units indicating that the protonation is
due to 1:1 binding and a simple equilibrium.4 We were
unable to determine the second protonation of the amino
crown ether, which suggests that it is either extremely low,
due to repulsive effects by the first protonation, or that the
small crown ether is operating like a pseudo proton sponge,
where the proton is shared between the two amino moieties.
It could also be the case that the second protonation simply
does not affect the fluorescent properties of the molecule.
We were however, unable to prove this explanation. From
the above change, a pKa of 7.2^0.1 was determined (Fig. 1).
Similarly, upon carrying out a pH titration of 2, the same
pKa value of 7.2 was determined. From these changes it is
clear that pKa’s are too high for determination of Liþ in
water at pH 7.4, as the partial protonation of the amino
crown ether would prevent the Liþ sensing due to charge
repulsion. This was indeed found to be the case as upon
titration of 1 or 2 at this pH using LiCl, NaCl, KCl and
CaCl2 salts. The emission of these two sensors was not
enhanced, which would have signified the suppression of the
PET process. However, substantial quenching was
observed. This indicated that the PET mechanism was not
active, or at least not observed, due to other active
quenching mechanisms. To investigate this further we
carried out a series of titrations at pH 7.4 and 8.5, where the
receptor would be expected to be in its ‘not fully protonated’
or ‘free-form’ respectively, using a range of lithium salts.
quenching by Cl2, Br2 and I2 was most likely due to a
heavy-atom effect by these halide counter ion, since it was
largely noticeable for spherical anions, in the concentration
range of 04–2.5 mM. This is strengthened by the fact that
other anions, such as perchlorate ClO4- or sulfate, did not
give rise to such quenching. To investigate this further we
also carried out anion titrations using the a-chloro-
naphthalene ligand 4. They strongly suggested that the
anion was quenching fluorescence due to the heavy atom
effect even in the absence of the crown ether, suggesting that
the anion was not binding to the receptor. However, these
changes were somewhat smaller than those seen for 1,
suggesting that some cooperative effect by the crown ether
moiety. When the fluorescence emission was evaluated at
pH 8.5, using the acetate salt of Liþ, Naþ and Kþ, no
fluorescence enhancement were observed, signifying that
the receptor was unable to extract the Liþ from the highly
solvated aqueous environment. Because of this, the Liþ
recognition was evaluated in less polar protic and aprotic
solvents.
The recognition of Liþ using 1 was evaluated in MeOH,
MeCN and in a mixture of both. In MeOH there were minor
changes in fluorescence when titrated with Liþ acetate.
Again, no significant changes were seen in 50:50 MeCN/
MeOH solution. However, in 80:20 MeCN/MeOH a slight
increase in fluorescence was observed but this was
insignificant, giving mere 0.8-fold enhancement in fluor-
escence. The above titrations were repeated in 100% MeCN
using both lithium acetate and lithium perchlorate salts.
Unlike previously observed, the fluorescent emission was
greatly enhanced upon titration with Liþ, leading to a 9-fold
increase in fluorescent intensity, as shown in Figure 4. The
fluorescence quantum yield, FF at high Liþ concentration
was measured to be 0.11, whereas in the absences of Liþ it
was 0.022. This is an order of magnitude fluorescent
enhancement upon ion recognition, which is quite signifi-
cant given the fact that the crown ether receptor (i.e. the
electron donor) is separated form the naphthalene fluoro-
phore by a four-carbon spacer. This would be expected to
reduce the efficiency of the PET quenching, which is highly
distance dependent, as previously discussed. Moreover, no
significant spectral changes were observed in the fluor-
escence spectra when titrated with any other group I and II
cations, as is evident from Figure 5, which shows the
changes in the fluorescence emission of 1 at 337 nm as the
function of the concentration of these ions. From Figures 4
and 5, it is clear that the fluorescence emission of 2 is highly
dependent on the Liþ concentration, which upon
recognition by the receptor increases the oxidation potential
of the receptor in a similar manner to that observed for the
pH titration earlier. Furthermore, no other significant
spectral changes were seen in the emission spectra (Fig. 4),
e.g. no changes were seen in the lmax, and no excimer
emission was observed at longer wavelength. One can thus
conclude that 1 is behaving like an ideal PET sensor.1,4
The changes in the fluorescence emission for LiBr, LiI and
LiOAc are shown in Figure 3, for the changes in 1 at pH 7.4
(0.1 M buffer and 0.1 M ionic strength). No major changes
were observed when LiOAc, NaOAc or KOAc were used.
However, when using the above LiCl, LiBr or LiI solutions,
substantial quenching was observed. As acetate did not give
rise to any significant quenching, we propose that the
From the changes in the 337 nm wavelength we were able to
determine the binding constant log b as 5.4 (^01) for 1
using the equation:
Figure 3. The changes in the fluorescence emission spectra of 1 at 338 nm
in pH 7.4 titration in water.
log b ¼ log½ðImax 2 IÞ=ðI 2 IminÞꢀ 2 log½Liþꢀ

T. Gunnlaugsson et al. / Tetrahedron 60 (2004) 5799–5806
5803
Figure 4. The changes in the fluorescence emission spectra of 1 upon titration with Liþ in CH3CN. Excitation at 280 nm.
where I is the fluorescent intensity at 337 nm, Imax is the
maximum intensity observed at 337 nm and Imin is the
minimum intensity observed at 337 nm. As can be seen
from Figure 5, the selectivity of 1 towards Liþ is very good,
as only at very high concentration of other competitive
group I and group II ions is the emission modulated. It is
thus clear that 1 is highly selective and sensitive to Liþ
recognition in a non-aqueous environment. To the best of
our knowledge, this is the first example of such a highly
selective and sensitive PET sensor for Liþ. When these
titrations were repeated for 2, similar results were observed,
as can be seen in Figure 6. Again, the fluorescence emission
spectra were switched on, with no other major spectral
changes. The quantum yield for the free sensor 2, and with
Liþ was determined to be 0.11 and 0.22, respectively,
mirroring that of 1 earlier. Furthermore, these changes
occurred over two logarithmic units, indicating that the
recognition was due to 1:1 binding and a simple
equilibrium.
We also investigated the changes in the 1H NMR of 1 upon
titration with Liþ in CD3CN. We foresaw that the largest
changes would be expected to occur for the resonances of
the crown ether moiety and the a-position of the pendent
arm. Indeed this was found to be the case, as upon titration
of 1 with Liþ, the largest changes were encountered for the
ring protons, causing minor upfield shift and broadening of
these resonances. It was also noticeable that the aromatic
resonances became broadened, which might suggest some
minor interactions between the two naphthalene moieties
upon Liþ recognition or perhaps some contribution from
cation–p type interactions.
To investigate the Liþ recognition further we observed the
changes in the circlar dicromism (CD) spectrum as a
function of Liþ concentration. Since the two pendent arms
are chiral they could cause an enhancement in the CD
spectra, where Liþ recognition gives rise to significant
changes in the CD intensity. We investigated these features
Figure 5. Titration profiles for 1 using perchlorate salts in CH3CN. V¼Liþ, £¼Ca2þ, X¼Kþ, K¼Naþ. Excitation at 280 nm.

5804
T. Gunnlaugsson et al. / Tetrahedron 60 (2004) 5799–5806
3. Conclusion
In summary, we have synthesized two amide based crown
ether ligands 1 and 2 for the detection of Liþ. Both were
synthesized in two step synthesis form the corresponding S
or R 1-(-1-naphtyl)ethylamine, and the 1,4-diaza-9-crown-
3-ether that was made in short step synthesis. The 1H NMR
of the resulting sensors showed that the axial and the
equatorial crown ether protons were magnetically
inequivalent. The ability of these two sensors to detect
Liþ was investigated in various solutions. In water, the pKa
of 7.2 was determined for both from the changes in the
fluorescence emission changes upon pH titration. Unfortu-
nately, we were unable to show Liþ detection in aqueous
solution. However, in CH3CN the fluorescence emissions
were significantly modulated upon Liþ recognition, where
the emission was switched on with almost an order of
magnitude enhancement in the fluorescence. In aqueous
solutions, the emission was quenched, most likely due to
vibrational heavy atom affect quenching by spherical anions
such as Cl2, Br2 and I2. However, this quenching was only
very minor using acetate or perchloride salts. The
Figure 6. The changes in the fluoþrescence emission spectra of 2 (being
switched on) upon titration with Li in CH3CN. Excitation at 280 nm.
for both 1 and 2. The CD analysis of 1 and 2, gave opposite
spectra, where 1 had a negative absorption band and 2,
positive absorption for the p!pp transition. The results for
1 are shown in Figure 7. Here it can be seen that 1 gave rise
to three transitions in the CD spectra, centered at 230, 250
and 280 nm, respectively. Of these the 225 nm and the
280 nm transitions were substantially affected by the Liþ
recognition. Hence, the long 280 nm transition doubled in
intensity and the short wavelength transition reduced in
intensity by half. These changes were not investigated any
further, but they do strongly suggest that upon Liþ sensing,
which is clearly observed by the fluorescence enhancements
discussed above, conformational changes occur in 1.
Similar, but not identical changes were observed for 2,
where enhancements were seen in the 280 nm absorption
band, indicating that the Liþ complexation might be give
rise to slightly different conformation within the molecule.
Even though we are unable to support this further, then these
results indicate that the amide functions are participating in
the Liþ coordination. We deduce this from the fact that the
two chromophores in 1 are separated at a relatively long
distance from the crown ether moiety to be actively affected
if the ion was only coordinating to the nitrogens and the
oxygens of the ring. Consequently, if the lithium ion was
coordinating to the two oxygen of the carboxylic amides
such a conformational effect would more likely to be
observed strongly by the chromophores. However, we were
unable to support this theory by X-crystallographic
evidence, as various attempts to obtain crystals of 1 or 2
suitable for X-ray diffraction, or their corresponding Liþ
complexes failed.
1
recognition was also evident from the H NMR and the
CD spectra which indicated that ion recognition was most
likely involving the crown ether nitrogen and the oxygen
moieties as well as the two oxygens of the carboxylic
amides. We are currently initiating a research programme
into the developments of PET sensors for Liþ for use in
aqueous solution.
4. Experimental
4.1. General
Starting materials were obtained from Sigma Aldrich, Strem
Chemicals and Fluka. Solvents were used at GPR grade
unless otherwise stated. Infrared spectra were recorded on a
Mattson Genesis II FTIR spectrophotometer equipped with
a Gateway 2000 4DX2-66 workstation. Oils were analyzed
using NaCl plates, solid samples were dispersed in KBr and
recorded as clear pressed discs. 1H NMR spectra were
recorded at 400 MHz using a Bruker Spectrospin DPX-400
instrument. Tetramethylsilane (TMS) was used as an
internal reference standard, with chemical shifts expressed
in parts per million (ppm or d) downfield from the standard.
13C NMR were recorded at 100 MHz using a Bruker
Spectrospin DPX-400 instrument. Mass spectroscopy was
carried out using HPLC grade solvents. Mass spectra were
determined by detection using Electrospray on a Micromass
LCT spectrometer, using a Shimadzu HPLC or Water’s
9360 to pump solvent. The whole system was controlled by
MassLynx 3.5 on a Compaq Deskpro workstation.
4.2. Modified peptide synthesis of 2-chloro-N-(2-
naphthyl)ethylethanamide 3 and 4
1-(-1-Naphtyl)ethylamine (1 g, 0.63 mmol) and HOBt
(0.8 g, 0.63 mmol), chloroacetic acid (0.6 g, 0.63 mmol)
were stirred in DCM (25 mL) at 210 8C for 20 min under
inert atmosphere. EDCl (1.2 g, 0.63 mmol) was then added
and the reaction stirred overnight at room temperature under
inert atmosphere. The solution was washed with 1 M
Figure 7. Theþchanges in the CD spectrum of 1 upon titration with 0, 1 and
5 equiv. of Li in CD3CN.

T. Gunnlaugsson et al. / Tetrahedron 60 (2004) 5799–5806
5805
NaHCO3 and brine. The organic layer was collected, dried
over MgSO4, filtered and evaporated to give a white solid.
7.03 (2H, d, N–H), 5.59 (2H, d, C H(CH3)NH), 3.22 (4H,
m), 2.77 (4H, q), 2.40 (8H, m), 1.67 (6H, d, J¼6.5 Hz,
CH(CH3)NH); 13C NMR (100 MHz, CD3CN): 169.8
(CvO), 137.7 (Ar-H), 133.4 (Ar-H), 130.9 (Ar-H), 128.3
(Ar-H), 128.0 (Ar-H), 126.1 (Ar-H), 125.5 (Ar-H), 124.7
(Ar-H), 123.1 (Ar-H), 122.3 (Ar-H), 71.6, 60.2, 56.0, 54.8,
43.5, 19.7 (CH(C H3); ES-MS: m/z 552.9 (Mþ); D1/8¼225
(1/dm3 mol21 cm21 10399) 281.6 (11750) 293.2 (6889.5);
n/cm21 (KBr) 3272 (N–H); 2499, 2554 (ar CvH); 1656
(CvO amide); 1607, 1528 (ar CvC); 1468 (C–N amide);
1357 (C–N crown ether); 1132 (C–O–C crown ether).
4.2.1. 2-Chloro-N-[(S)-1-naphthyl]ethylethanamide 3.
0.58 g, 86.9% yield. Mp 140 8C; calcd for C14H14NOCl: C
67.88; H, 5.70; N, 5.65. Found C 67.63, H 5.77, N 5.86; 1H
NMR (400 MHz, CD3CN): d 8.16 (1H, d, Ar-H, J¼8.5 Hz),
7.96 (1H, d, Ar-H, J¼7.5 Hz), 7.87 (1H, d, Ar-H,
J¼7.5 Hz), 7.59 (1H, m, Ar-H), 7.25 (1H, sb, Ar-H,), 5.83
(1H, m, J¼7.0 Hz), 4.04 (2H, s), 1.63 (3H, d, J¼7.0 Hz);
13C NMR (100 MHz, CD3CN): 164.8, 138.8, 133.4, 130.2,
128.31, 127.3, 125.0, 125.3, 125.0, 122.6, 122.1, 42.2, 20.1;
ES-MS: m/z 247.9 (Mþ); n/cm21 (KBr) 3295 (N–H), 1648
(CvO), 1541 (CvC), 780 (C–Cl).
(at OD¼0.1, 225 nm)
lmax/nm (CH3CN) 260.7
Acknowledgements
4.2.2. 2-Chloro-N-[(R)-1-naphthyl]ethylethanamide 4.
0.60 g, 87.3% yield. Mp 140 8C; calcd for C14H14NOCl:
C, 67.88;H, 5.70; N, 14.31. Found: C; 67.67;H, 5.81; N,
5.83; 1H NMR (400 MHz, CDCl3): d 8.11 (1H, d,
J¼8.5 Hz), 7.92 (1H, d, J¼8.6 Hz), 7.86 (1H, d,
J¼7.6 Hz), 7.56 (4H, m), 6.81 (1H, s, N–H), 4.15 (2H, q),
1.74 (3H, d, J¼7.0 Hz); 13C NMR (100 MHz, CDCl3):
164.4 (qC), 137.0 (qC), 133.5 (qC), 130.5 (qC), 128.5 (Ar-H),
128.2 (Ar-H), 126.2 (Ar-H), 125.5 (Ar-H), 124.8 (Ar-H),
122.6 (Ar-H), 122.1 (Ar-H), 44.8 (C H2), 42.2 (C H2), 20.4
(C H3); ES-MS: m/z 247.9 (Mþ); n/cm21 (KBr) 3290 (N–
H), 1652 (CvO), 1540 (CvC), 782 (C–Cl).
We would like to thank Trinity College Dublin (Krieble
Fund 2001-2002), and HEA (Higher Education Authority in
Irealnd) under the PRTLI 98 (Molecular Cell Biology
Programme) for financial support, Dr. John E. O’Brien for
running NMR spectra, Dr. Hazel Moncrieff and Professor
D. Clive Williams (Biochemistry TCD) for their valuable
discussion.
References and notes
4.3. General synthesis of the sensors 1 and 2
1. de Silva, A. P.; McCaughan, B.; McKinney, B. O. F.; Querol,
M. Dalton Trans. 2003, 1902. De Silva, A. P.; Fox, D. B.;
Huxley, A. J. M.; Moody, T. S. Coord. Chem. Rev. 2000, 205,
41. Lavigne, J. J.; Anslyn, E. V. Angew. Chem., Int. Ed. 2001,
40, 3119. de Silva, A. P.; Fox, D. B.; Huxley, A. J. M.; Moody,
T. S. Coord. Chem. Rev. 2000, 205, 41. Fabbrizzi, L.;
Licchelli, M.; Rabaioli, G.; Taglietti, A. F. Coord. Chem. Rev.
2000, 205, 85. de Silva, A. P.; Fox, D. B.; Huxley, A. J. M.;
McClenaghan, N. D.; Roiron, J. Coord. Chem. Rev. 1999, 186,
297.
1,4-Diaza-9-crown-3-ether (1 equiv.), Cs2CO3 (6 equiv.)
and KI (0.1 equiv.) were stirred in MeCN under inert
atmosphere. The appropriate chromophore (2.1 equiv.) in
MeCN was added via a pressure equalized dropping funnel.
The mixture was left to reflux at 80 8C overnight under inert
atmosphere. The reaction was filtered and the solvent
evaporated. The yellow residue was dissolved in CHCl3 and
washed with 10% K2CO3 (3£20 mL). The organic layer was
collected, dried over MgSO4, filtered and evaporated to give
a white solid. After purification by alumina chromatography
with DCM: 0!5% MeOH, the product was washed and
recrystallized from diethyl ether to yield a white solid.
2. Chemical Sensors and Biosensors for Medical and Biological
Applications. Spichiger-Keller, U. S., Ed.; Wiley-VCH:
Weinheim; Germany, 1998. Chemosensors of Ion and
Molecular Recognition. Desvergne, J. P., Czarnik, A. W.,
Eds.; Kluwer Academic: Dordrecht, Netherland, 1997.
3. de Silva, A. P.; Gunaratne, H. Q. N.; Gunnlaugsson, T.;
Huxley, A. J. M.; McCoy, C. P.; Rademacher, J. T.; Rice, T. E.
Chem. Rev. 1997, 97, 1515. Czarnik, A. W. Acc. Chem. Res.
1994, 27, 302. Fluorescent Chemosensors for Ion and
Molecular Recognition. Czarnik, A. W., Ed.; ACS Books:
Washington, 1993.
4.3.1. Compound 1 (S,S). 0.22 g, 32.7% yield. Mp 107 8C;
calcd for C34H40N4O3: C 73.88; H, 7.29; N, 10.14. Found C
73.39, H 7.14, N 9.66; 1H NMR (400 MHz, CD3CN): d 8.12
(2H, d, Ar-H, J¼8.5 Hz), 7.91 (2H, d, Ar-H, J¼7.5 Hz),
7.81 (2H, d, Ar-H, J¼8.0 Hz), 7.53 (8H, m, J¼8.0 Hz), 5.80
(2H, m, J¼7.0 Hz), 3.24 (4H, m), 3.05 (4H, s), 2.57 (8H, m),
1.68 (6H, d, J¼7.0 Hz); 13C NMR (100 MHz, CD3CN):
139.2 133.4, 130.5, 128.2, 127.2, 125.8, 125.3, 124.9, 122.8,
122.2, 71.7, 60.2, 56.3, 55.1, 43.7, 19.8; ES-MS: m/z 553.5
(Mþ), D1/8¼þ25 (at OD¼0.1, 225 nm); lmax/nm (CH3CN)
260.8 (1/dm3 mol21 cm21 10427) 281.6 (11801) 293.2
(6801.5); n/cm21 (KBr) 3287 (N–H); 2525, 2554 (ar
CvH); 1650 (CvO amide); 1602, 1511 (ar CvC); 1450
(C–N amide); 1357 (C–N crown ether); 1126 (C–O–C
crown ether).
4. Examples include: Gunnlaugsson, T.; Lee, C. T.; Parkesh, R.
Org. Lett. 2003, 5, 4065. Gunnlaugsson, T.; Lee, C. T.;
Parkesh, R. Org. Biomol. Chem. 2003, 1, 3265. He, H.;
Mortellaro, M. A.; Leiner, M. J. P.; Fraatz, R. J.; Tusa, J. K.
J. Am. Chem. Soc. 2003, 125, 1468. Tusa, J. K.; Leiner, M. J.
P. Ann. Biol. Clin-Paris 2003, 61, 183. Gunnlaugsson, T.;
Nieuwenhuyzen, M.; Richard, L.; Thoss, V. J. Chem. Soc.,
Perkin Trans. 2 2002, 141. de Silva, A. P.; Gunaratne, H. Q.
N.; Maguire, G. E. M. J. Chem. Soc., Chem. Commun. 1994,
1213. Bissell, R. A.; de Silva, A. P.; Gunaratne, H. Q. N.;
Lynch, P. L. M.; Maguire, G. E. M.; McCoy, C. P.;
Sandanayake, K. R. A. S. Top. Curr. Chem. 1993, 168, 223.
Huston, M. E.; Czarnik, A. W.; Engleman, C. J. Am. Chem.
4.3.2. Compound 2 (R,R). 0.18 g, 30.1% yield. Mp 107 8C.
Accurate mass: calcd for C34H40N4O3: found C34H41N4O3;
1H NMR (400 MHz, CDCl3): d 8.02 (2 ArH, d, J¼8.4 Hz),
7.67 (5 ArH, dd, J¼8.0 Hz), 7.53 (4H, d, Ar-H, J¼8 Hz),

5806
T. Gunnlaugsson et al. / Tetrahedron 60 (2004) 5799–5806
Soc. 1990, 112, 7054. de Silva, A. P.; Sandanayake, K. R. A. S.
Angew. Chem., Int. Ed. 1990, 29, 1173.
Nicholson, P. E.; Parker, D.; Covington, A. K. Analyst 1991,
116, 135–140. Kataky, R.; Nicholson, P. E.; Parker, D.
Tetrahedron Lett. 1989, 30, 4559.
5. Gunnlaugsson, T.; Kruger, P. E.; Lee, T. C.; Parkesh, R.;
Pfeffer, F. M.; Hussey, M. G. Tetrhedron Lett. 2003, 44, 6575.
Gunnlaugsson, T.; Davis, A. P.; Glynn, M. Org. Lett. 2002, 4,
2449. Gunnlaugsson, T.; Davis, A. P.; Glynn, M. Chem.
Commun. 2001, 2556.
12. Kobiro, K. Coord. Chem. Rev. 1996, 148, 135.
13. Gunnlaugsson, T.; Harte, A. J.; Leonard, J. P.; Senechal, K.
J. Am. Chem. Soc. 2003, 125, 12062. Gunnlaugsson, T.;
Leonard, J. P. J. Chem. Soc., Perkin Trans. 2 2002, 1980.
Gunnlaugsson, T.; Nieuwenhuyzen, M.; Richard, L.; Thoss, V.
J. Chem. Soc., Perkin Trans. 2 2002, 141. Gunnlaugsson, T.;
Nieuwenhuyzen, M.; Richard, L.; Thoss, V. Tetrahedron Lett.
2001, 42, 4725.
6. For a communicaton of this work see: Gunnlaugsson, T.;
Bichell, B.; Nolan, C. Tetrahedron Lett. 2002, 43, 4989.
7. Examples of other types of fluorescent or optically based Liþ
sensors include: Obare, S. O.; Murphy, C. J. New. J. Chem.
2001, 25, 1600. Ajayaghosh, A.; Chenthamarkshan, C. R.
Tetrahedron Lett. 1998, 39, 1795. Tsuchiya, S.; Ogawa, S.
Chem. Lett. 1996, 709. Bochen˜ska, M. J. Incl. Phenom. 1995,
22, 269. Blackburn, C.; LeCompte, M. B. K.; Langmuir, M. E.
Tetrahedron Lett. 1994, 35, 1994. McCarrick, M.; Wu, B.;
Harris, S. J.; Diamond, D.; Barrett, G.; McKervey, M. A.
J. Chem. Soc., Perkin Trans. 2 1993, 1963. Hiratani, K.;
Nomoto, M.; Sugihara, H.; Okada, T. Analyst 1992, 117, 1491.
Sholl, A. F.; Sutherland, I. O. J. Chem. Soc., Chem. Commun.
1992, 1716. Hiratani, K. Analyst 1988, 113, 1065. Nakashima,
K.; Nagaoka, Y.; Nakatsuji, S.; Kaneda, T.; Tanigawa, I. Bull.
Chem. Soc. Jpn 1987, 60, 3219.
14. Gunnlaugsson, T.; Leonard, J. P. Chem. Commun. 2003, 2424.
´
Gunnlaugsson, T.; Mac Donaill, D. A.; Parker, D. Chem.
Commun. 2000, 93. Gunnlaugsson, T. Tetrahedron Lett. 2001,
42, 8901.
´
15. Gunnlaugsson, T.; Mac Donaill, D. A.; Parker, D. J. Am.
Chem. Soc. 2001, 123, 12866.
16. Gunnlaugsson, T.; Kelly, J. M.; Nieuwenhuyzen, M.; O’Brien,
A. M. K. Tetrahedron Lett. 2003, 44, 8571. Gunnlaugsson, T.;
Harte, A. J.; Leonard, J. P.; Nieuwenhuyzen, M. Supramol.
Chem. 2003, 15, 505. Gunnlaugsson, T.; Harte, A. J.; Leonard,
J. P.; Nieuwenhuyzen, M. Chem. Commun. 2002, 2134.
17. Ewin, G.; Hill, J. Chem. Res. 1985, 3501.
8. Christian, G. D. J. Pharm. Biomed. Anal. 1996, 14, 899.
9. Birch, N. J. Chem. Rev. 1999, 99, 2659.
18. Dickens, R. S.; Howard, J. A. K.; Maupin, C. L.; Moloney,
J. M.; Parker, D.; Peacock, R. D.; Riehl, J. P.; Siligardi, G.
New J. Chem. 1998, 22, 891.
10. Lindoy, L. F. The Chemistry of Macrocyclic Ligand
Complexes. Cambridge University Press: New York, 1989.
11. Hori, K.; Tsukbe, H. J. Incl. Phenom. Mol. Recognit. Chem.
1998, 32, 311. Kataky, R.; Nicholson, P. E.; Parker, D.
J. Chem. Soc., Perkin Trans. 2. 1990, 321. Kataky, R.;
19. Bruce, J. I.; Dickins, R. S.; Govenlock, L. J.; Gunnlaugsson,
T.; Lopinski, S.; Lowe, M. P.; Parker, D.; Peacock, R. D.;
Perry, J. J. B.; Aime, S.; Botta, M. J. Am. Chem. Soc. 2000,
122, 9674.