Triphenylamine-based novel PET sensors in selective recognition of dicarboxylic acids

Article abstract of DOI:10.1016/j.tetlet.2006.02.008

The triphenylamine-based chemosensors 1 and 2 have been designed and synthesized, for the first time, for the selective recognition of dicarboxylic acids. Carboxylic acid binding takes place through charge neutral pyridyl amide receptor sites with concomitant quenching of fluorescence of the triphenylamine moiety. The bindings were examined using ¹H NMR, fluorescence and UV-vis spectroscopic methods. The receptor 1 was found to be selective for glutaric and adipic acids and the macrocycle 2 was specific for 2,2-dimethylmalonic acid.

Full text of DOI:10.1016/j.tetlet.2006.02.008

Tetrahedron Letters 47 (2006) 2365–2369
Triphenylamine-based novel PET sensors in selective recognition
of dicarboxylic acids
Kumaresh Ghosh* and Goutam Masanta
Department of Chemistry, University of Kalyani, Kalyani, Nadia 741235, India
Received 26 October 2005; revised 20 January 2006; accepted 1 February 2006
Available online 20 February 2006
Abstract—The triphenylamine-based chemosensors 1 and 2 have been designed and synthesized, for the first time, for the selective
recognition of dicarboxylic acids. Carboxylic acid binding takes place through charge neutral pyridyl amide receptor sites with con-
1
comitant quenching of fluorescence of the triphenylamine moiety. The bindings were examined using H NMR, fluorescence and
UV–vis spectroscopic methods. The receptor 1 was found to be selective for glutaric and adipic acids and the macrocycle 2 was spe-
cific for 2,2-dimethylmalonic acid.
ꢀ 2006 Elsevier Ltd. All rights reserved.
Interest in the selective recognition and sensing of neu-
tral species by synthetic receptors continues to attract
the attention of the supramolecular chemistry commu-
nity.1 Given the important role of dicarboxylic acids in
biology,2 the need for their selective recognition by syn-
thetic fluorescent sensors utilizing weak hydrogen bond-
ing interactions is of great importance in molecular
recognition research.3 During the last decade, consider-
able progress has been made for the recognition of
dicarboxylic acids by a number of synthetic receptors
of various architectures.4 Despite the development of
dicarboxylic acid receptors with signalling informa-
tion,5,3a there is continued interest in the search for
new fluorescent based molecular sensors for selective
recognition of dicarboxylic acids because of the many
advantages including multiple modes of detection (such
as quenching, enhancing, life time), extremely high sen-
sitivity, relatively low cost and easy availability. We are
interested in developing chromophores where the recog-
nition takes place at charge neutral recognition sites
with concomitant changes in the photophysical proper-
ties of a lumophore by modulation of the photoinduced
electron transfer (PET) mechanism.6,4d In this context,
we herein report the design and synthesis of triphenyl-
amine-based PET chemosensors 1 and 2, for the first
time, for size selective recognition of aliphatic dicarboxy-
lic acids. Although a number of receptors with a range
of spacer groups for neutral dicarboxylic acids of differ-
ent chain lengths are reported in the literature,4,7 triphen-
ylamine has not been used in this capacity until now.
N
N
O
O
O
N
O
N
HN
N
NH
N
NH
HN
O
O
1
2
Keywords: Triphenylamine; Selective recognition; Glutaric acid; 2,2-Dimethylmalonic acid; PET sensor.
*
Corresponding author. Fax: +91 33 2582 8282; e-mail: ghosh_k2003@yahoo.co.in
0040-4039/$ - see front matter ꢀ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2006.02.008

2366
K. Ghosh, G. Masanta / Tetrahedron Letters 47 (2006) 2365–2369
The triphenylamine moiety has been employed inten-
sely because of its structural rigidity8 and strong
fluorescence property. The peripherally substituted
binding sites are expected to modulate the electron den-
sity. The hydrogen bonding perturbation of the binding
sites alters the excited state properties of the receptor.
Considering these points, we placed the pyridine amide,
a known carboxylic acid binding moiety,9 onto the
periphery of the rigid triphenylamine (1 and 2) unit,
which provides a conformationally well-defined V-shape
geometry for size selective complexation of dicarboxylic
acids.
binding occurs only when the cavity dimension matches
the chain length of the dicarboxylic acids.
The sensitivity and selectivity of these ditopic receptors
towards a series of aliphatic dicarboxylic acids of vari-
ous chain lengths was evaluated by observing the change
in the fluorescence emission spectra in CHCl3 and the
1H NMR upon dicarboxylic acid titration in CDCl3.
The fluorescence emission spectra of 1 and 2 consisted
of bands at 417 and 430 nm, respectively, when excited
at 360 nm. Upon addition of dicarboxylic acids of differ-
ent chain lengths (prepared in CHCl3 containing 0.02%
DMSO) the emission was drastically quenched due to
the formation of receptor–diacid complexes. During
the course of titration there was no other spectral
change in the emission spectra. Concurrently, the
changes in the absorption spectra (peaks at 360, 328
and 290 nm in the case of 1 and at 360, 323 and
291 nm in the case of 2) of both the sensors were only
minor during titration with dicarboxylic acids thus
indicating typical PET behaviour. This typical PET
behaviour is attributed due to separation of the triphen-
ylamine unit from the receptor sites by rigid amide
spacers; the only interaction between the two moieties
is through electron transfer. Once binding has occurred,
fluorescence quenching is therefore caused by electron
transfer between the pyridine amide-carboxylic acid
complexes and the chromophore triphenylamine units
in both 1 and 2.
The syntheses of 1 and 2 were accomplished according
to Scheme 1. Triphenylamine was initially formylated
using POCl3/DMF to yield mono-, di- and tri-formyl-
ated products. The desired 4,40-diformyltriphenylamine
3, isolated in 57% yield, was next oxidized to the diacid,
which on treatment with thionyl chloride gave diacid
chloride 4 in an overall 80% yield. Coupling of 2-amino-
6-methylpyridine with 4 in dry THF afforded the sensor
1 in 78% yield as yellowish solid,10 which was soluble in
common organic solvents such as CH2Cl2, CHCl3 and
CH3CN. The macrocycle 2 was synthesized by high
dilution coupling of diacid chloride 4 with diamine 5,
which was obtained by reaction of 2-pivaloylamino-6-
bromomethyl pyridine (obtained via NBS reaction of
2-pivaloylamino-6-methylpyridine in dry carbon tetra-
chloride) with 1,3-propanediol followed by alkaline
hydrolysis. This method gave 2 in 10–12% yield as
brownish solid. All the compounds were fully character-
ized by conventional methods.10
Upon addition of increasing concentrations of dicarboxy-
lic acids of various chain lengths, the fluorescence of
the open receptor 1 was essentially quenched to different
extents. The greater quenching for glutaric acid (ca.
39%) as revealed in the Stern–Volmer plot (Fig. 3; inset)
at 417 nm, is considerable due to its ability to form a
tight and stable hydrogen bonded complex (Fig. 1) with
the ditopic receptor 1 compared to its higher and lower
homologues (quenching ca. 18–30%). The significant red
shift (Dk = 21 nm) along with the presence of an isobas-
tic point during the titration with glutaric acid (Fig. 5)
clearly indicates the formation of a 1:1 hydrogen bonded
complex. This is attributed to the complementary size of
the binding space of 1 with the chain length of glutaric
acid. A slightly greater quenching for 2,2-dimethyl-
malonic acid than adipic acid as observed in Figure 3
(inset), is possibly due to its higher acidity. In the case
In 1, the hydrogen bonding groups are conveniently
arranged in a concave face to bind dicarboxylic acids,
and can have three conformations of comparable energy
values (in–in, in–out and out–out) in the solution
phase. MMX calculations11 on the ‘in–in’ conformation
of 1 (Emin = 61.99 kcal/mol) indicates a separation of
˚
10.83 A between the two-pyridine ring nitrogens and
˚
7.92 A between two amide NH’s. The phenyl rings of
the triphenylamine moiety are slightly twisted from the
plane containing the central nitrogen atom. In 2 (Emin
=
84.06 kcal/mol), the pyridine ring nitrogens and amide
˚
NH’s are 8.80 and 7.54 A apart, respectively. Thus both
open and macrocyclic cavities can accommodate dicar-
boxylic acids of various chain lengths, and selective
i) KMnO4 in aq. acetone
ii) SOCl2 in dry benzene, reflux
2-amino-6-methylpyridine / Et3N
in dry THF
POCl3 / DMF
1
N
N
N
ClOC
OHC
COCl
CHO
4
3
i) NaH in dry THF,
4, high dilution coupling in dry THF
2
OH
HO
O
O
24 h, rt
2-pivaloylamino-6-bromomethyl pyridine
ii) 10 N KOH in aq. ethanol, reflux
20 h
N
NH2
H2N
N
5
Scheme 1. The syntheses of receptors 1 and 2.

K. Ghosh, G. Masanta / Tetrahedron Letters 47 (2006) 2365–2369
2367
700
600
500
7
6
5
4
3
2
1
N
O
N
O
400
NH
HN
300
O
O
N
200
100
0
OH
0
10
20
30
40
50
60
70
80
HO
( )n
-6
[G] x 10
M
R
R
n =1, R = Me; n = 3, 4, R = H
0
1
2
3
4
5
6
7
8
9
10 11 12 13 14 15
Figure 1. Hydrogen bonded complex of 1 with dicarboxylic acids.
[G]/[H]
Figure 4. Fluorescence titration curve for 2 (measured at 430 nm): 2,2-
dimethyl malonic acid ( ), glutaric acid ( ), adipic acid ( ), suberic
acid ( ); inset: Stern–Volmer plot at 430 nm.
N
1000
800
600
400
200
0
3.0
2.5
2.0
1.5
1.0
0.5
0.0
O
O
HN
NH
O
O
N
N
( ) n
R
OH
O
HO
R
O
n =1, R = Me; n = 3, 4, R = H
260 280 300 320 340 360 380 400
Wavelength (nm)
Figure 2. Hydrogen bonded complex of 2 with dicarboxylic acids.
400
2.0
1.9
350
390
420
450
480
510
540
570
600
1.8
1.7
1.6
1.5
1.4
1.3
1.2
1.1
1.0
300
250
200
150
100
50
Wavelength (nm)
Figure 5. Fluorescence spectra of 1 (c = 0.775 · 10ꢀ6 M) in CHCl3 and
the change in the UV–vis spectra of 1 (1.55 · 10ꢀ5 M) (inset) upon
addition of glutaric acid.
the shorter chain 2,2-dimethylmalonic acid. This was
evident from the Stern–Volmer plot (Fig. 4; inset) at
430 nm. Upon addition of 2,2-dimethylmalonic acid
the emission of 2 was ca. 84% ‘switched off’ or quenched
(Fig. 6) due to the formation of a strong 1:1 complex
analogous to that shown in Figure 2. In comparison,
for higher homologues, the fluorescence emission of 2
was ‘switched off’ by ca. 17–20% (Fig. 4; inset). Figure
4 shows the change in the fluorescence intensity profile
as a function of the concentrations of dicarboxylic acids
to the ditopic receptor 2 and presents the different bind-
ing potencies of the dicarboxylic acids studied. How-
ever, in both cases of 1 and 2, 1:1 stoichiometric
binding with size selective glutaric and 2,2-dimethyl-
malonic acids, respectively, is attributed from the break
of the titration curves (see Figs. 3 and 4).
0
1
2
3
4
5
6
[G] x 10-6
M
0
0
2
4
6
8
10
[G]/[H]
12
14
16
18
20
Figure 3. Fluorescence titration curve for 1 (measured at 417 nm): 2,2-
dimethyl malonic acid ( ), glutaric acid ( ), adipic acid ( ), suberic
acid ( ); inset: Stern–Volmer plot at 417 nm.
of long chain suberic acid, the quenching was nearly
similar in magnitude to that of monocarboxylic myristic
acid. This underlines the fact that the long chain diacid
binds only one pyridine amide instead of bridging both
sites and the binding is thus 2:1 (diacid: sensor 1). This is
indeed found to be the case as can be seen in Figure 3 for
the change in fluorescence at 417 nm as a function of the
receptor 1 to the concentrations of dicarboxylic acids.
To investigate further the binding interactions and stoi-
chiometries of the complexes, we carried out NMR titra-
tions on both 1 and 2. 1H NMR studies in CDCl3
revealed a considerable downfield shift of the pyridine
Similar experiments were performed on the macrocyclic
ditopic receptor 2, which showed a high specificity for

2368
K. Ghosh, G. Masanta / Tetrahedron Letters 47 (2006) 2365–2369
Table 1. Binding constants (ka) based on 1H NMR
1.0
0.8
0.6
0.4
0.2
0.0
800
Dicarboxylic acids
Sensor 1 (Mꢀ1
)
Sensor 2 (Mꢀ1
)
700
600
500
400
300
200
100
0
2,2-Dimethylmalonic
Glutaric
Adipic
43.9
>105
3.31 · 103
2.19 · 103
1.18 · 102
1.80 · 102
a
a
Suberic
a Not determined due to their overlong nature in respect of cavity
dimension.
250
300
350
400
Wavelength (nm)
390
420
450
480
510
540
570
600
1.5
1.0
Wavelength (nm)
Figure 6. Fluorescence spectra of 2 (c = 0.360 · 10ꢀ5 M) in CHCl3 and
the change in the UV–vis spectra of 2 (1.36 · 10ꢀ5 M) (inset) upon
addition of 2,2-dimethylmalonic acid.
amide protons of 1 (Dd = 1.80–2.61 ppm) upon addition
of dicarboxylic acids (dissolved in CDCl3 containing
3–4% DMSO-d6 in each case due to only the partial
solubility of the diacids in pure CDCl3), suggesting that
amido pyridyl moieties serve as binding sites for the
dicarboxylic acid. The dicarboxylic acid–receptor stoi-
chiometry was confirmed from the break of the titration
curve12 (Dd vs Cguest/Chost) in Figure 7.
0.5
0.0
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
[G]/[H]
Titration of 1 with a series of dicarboxylic acids showed
appreciable binding constants (Table 1).13 It should be
mentioned that a higher association constant with 1:1
stoichiometry was observed for glutaric acid as com-
pared with 2,2-dimethylmalonic (1:1) and suberic acids,
which showed 2:1 (diacid:receptor) binding stoichio-
metry. The association constant for glutaric acid is only
slightly higher than its close homologue adipic acid.
The precise selectivity between two purely aliphatic
dicarboxylic acids is difficult to predict due to their free
bond rotation in open chain form. On the other hand,
the macrocyclic receptor 2 is specific for 2,2-di-
methylmalonic acid in preference to its higher homo-
Figure 8. NMR titration curve for 2: 2,2-dimethylmalonic acid ( ),
glutaric acid ( ).
logues. During titration with 2,2-dimethylmalonic acid
the amide protons of 2 showed a large downfield shift
(Dd = 1.63 ppm) in the 1:1 complex (Fig. 8), which on
further dilution did not exhibit practically any change,
suggesting an association constant of >105 Mꢀ1. This
is due to the perfect match of the cavity with the chain
length of 2,2-dimethylmalonic acid. It is obvious that
the presence of DMSO-d6 as a competitive binding part-
ner will reduce the binding to some extent in each case.
In contrast, the higher homologue glutaric acid showed
very weak binding, indicating an impressive steric mis-
match, therefore, we did not study the other higher
homologues further.
2.8
2.4
2.0
1.6
1.2
0.8
0.4
0.0
In conclusion, we have used triphenylamine as a PET
sensor, for the first time, in reporting the size selective
recognition of aliphatic dicarboxylic acids within the
concave cavity. The recognition events indicate that it
is possible to regulate the signal output of the triphenyl-
amine fluorophore with dicarboxylic acids of different
chain lengths. The macrocyclic sensor 2 shows a greater
quenching response for 2,2-dimethylmalonic acid
(specific fit) than the open form 1 with size selective
glutaric acid. Further work along this direction is in
progress in our laboratory.
0
1
2
3
4
Acknowledgements
[G]/[H]
We thank DST (SR/FTP/CS-18/2004), Government of
India for financial support and DST-FIST for providing
Figure 7. NMR titration curve for 1: 2,2-dimethylmalonic acid ( ),
glutaric acid ( ), adipic acid ( ), suberic acid ( ).

K. Ghosh, G. Masanta / Tetrahedron Letters 47 (2006) 2365–2369
2369
6. Ghosh, K.; Masanta, G. Supramol. Chem. 2005, 17, 331–
334.
the facilities in the Department. G.M. thanks CSIR,
Govt. of India for a fellowship.
7. (a) Collinson, S. R.; Gelbrich, T.; Hursthouse, M. B.;
Tucker, J. H. R. Chem. Commun. 2001, 555–556; (b)
Fitzmaurice, R. J.; Kyne, G. M.; Douheret, D.; Kilburn, J.
D. J. Chem. Soc., Perkin Trans. 1 2002, 841–864.
8. (a) Wei, P.; Bi, X.; Wu, Z.; Xu, Z. Org. Lett. 2005, 7,
3199–3202; (b) Reinhardt, B. A.; Brott, L. L.; Clarson, S.
J. Chem. Mater. 1998, 10, 1863–1874.
References and notes
1. (a) Comprehensive Supramolecular Chemistry; Atwood, J.
L., Davies, J. E. D., MacNicol, D. D., Vogtle, F., Eds.;
Elsevier: Exeter, 1996, and references cited therein; (b)
Tellado, F. G.; Geib, S. J.; Goswami, S.; Hamilton, A. D.
J. Am. Chem. Soc. 1991, 113, 9265–9269; For a general
discussion of neutral molecule recognition see: (c) van
Doorn, A. R.; Verboom, W.; Reinhoudt, D. N. In
Advances in Supramolecular Chemistry; Gokel, G. W.,
Ed.; JAI Press: Greenwich, CT, 1993; Vol. 3, pp 159–206.
2. (a) Dugas, H. Bioorganic Chemistry; Springer: New York,
1996; (b) Stryer, L. Biochemistry; W.H. Freeman and
Company: New York, 1998.
9. Etter, M. C.; Adsmond, D. A. J. Chem. Soc., Chem.
Commun. 1990, 589–591.
1
10. Compound 1: Mp 210 ꢁC; yield: 78%; H NMR (CDCl3,
500 MHz): d 8.46 (s, 2H, –NH), 8.17 (d, 2H, J = 10 Hz),
7.84 (d, 4H, J = 10 Hz), 7.63 (t, 2H, J = 10 Hz), 7.38–
7.31 (m, 2H, J = 10 Hz), 7.20 (t, 1H, J = 10 Hz), 7.17–
6.93 (m, 6H), 6.92 (d, 2H, J = 10 Hz), 2.47 (s, 6H). 13C
NMR (CDCl3 75 MHz): 164.9, 156.5, 150.9, 150.3, 145.8,
138.7, 130.4, 128.7, 128.0, 126.3, 125.3, 122.5, 119.1,
111.0, 23.7. FTIR: m cmꢀ1 (KBr): 3428, 1674, 1591, 1574,
1496, 1448. UV (CHCl3): (c = 1.55 · 10ꢀ5 M) kmax (nm)
290, 328, 360. Mass: 514 (M+1), 440, 422, 406. Com-
pound 2: Mp 204 ꢁC; 1H NMR (CDCl3, 400 MHz): d
8.40 (s, 2H, –NH), 7.89 (d, 2H, J = 8 Hz), 7.73–7.66 (m,
6H), 7.30 (t, 2H, J = 8 Hz), 7.15 (d, 2H, J = 8 Hz), 7.08–
7.06 (m, 5H), 6.97 (d, 2H, J = 8 Hz), 4.53 (s, 4H), 3.73 (t,
4H, J = 4 Hz), 2.15 (t, 2H, J = 4 Hz). 13C NMR (CDCl3,
125 MHz): 166.9, 156.7, 151.9, 145.4, 139.3, 130.7, 129.9,
129.5, 129.4, 126.3, 123.3, 121.8, 117.6, 112.9, 73.2, 69.2,
46.3. FTIR: m cmꢀ1 (KBr): 3319, 2864, 1676, 1593, 1542,
1502, 1457, 1107, UV (CHCl3): (c = 1.36 · 10ꢀ5 M) kmax
(nm) 290, 324, 360, Mass: 586.1 (M+1), 442.2, 433.1,
392.4.
3. (a) Bell, T. W.; Hext, N. M. Chem. Soc. Rev. 2004, 33,
589–598; (b) Gunnlaugsson, T.; Davis, A. P.; O’Brien, J.
E.; Glynn, M. Org. Lett. 2002, 4, 2449–2452.
4. (a) Tellado, F. G.; Goswami, S.; Chang, S. K.; Geib, S. J.;
Hamilton, A. D. J. Am. Chem. Soc. 1990, 112, 7393–7394;
(b) Vincent, C.; Fan, E.; Hamilton, A. D. Tetrahedron
Lett. 1992, 33, 4269–4272; (c) Karle, I. L.; Ranganathan,
D.; Haridas, V. J. Am. Chem. Soc. 1997, 119, 2777–2783;
(d) Goswami, S.; Ghosh, K.; Dasgupta, S. J. Org. Chem.
2000, 65, 1907–1914, and references cited therein; (e)
Goswami, S.; Ghosh, K.; Halder, M. Tetrahedron Lett.
1999, 40, 1735–1738; (f) Goswami, S.; Ghosh, K.; Muk-
herjee, M. Tetrahedron 2001, 57, 4987–4993; (g) Moriuchi,
T.; Yoshida, K.; Hirao, T. Org. Lett. 2003, 5, 4285–4288;
(h) Rebek, J., Jr.; Nemeth, D.; Ballester, P.; Lin, F. T. J.
Am. Chem. Soc. 1987, 109, 3474–3475; (i) Miyaji, H.;
Dudic, M.; Tucker, J. H. R.; Prokes, I.; Light, M. E.;
Hursthouse, M. B.; Stibor, I.; Lhotak, P. Tetrahedron
Lett. 2002, 43, 873–878.
11. Energy minimization was carried out by MMX (PC Model
Serena Software 1993). Molecular modelling was per-
formed using standard constants, and the dielectric
constant was maintained at 1.5.
12. Blands, M. T.; Horner, J. H.; Newcomb, M. J. J. Org.
Chem. 1989, 54, 4626–4636.
5. (a) Zhao, J.; James, T. D. Chem. Commun. 2005, 1889–
1891; (b) Martinez-Manez, R.; Scancenon, F. Chem. Rev.
2003, 103, 4419–4476.
13. Foster, R.; Fyfe, C. A. Prog. Nucl. Magn. Reson.
Spectrosc. 1969, 4, 1–89.