Heavy metal analysis using a Heck-catalyzed cyclization to create coumarin

Article abstract of DOI:10.1039/b502199p

This paper describes a new method for heavy metal analysis via catalytic signal amplification. This signal amplification protocol relies upon an exogenous inhibitor for deliberate deactivation of an organometallic reaction that catalytically creates a fluorophore. When the deactivation process is performed in the presence of the analyte, competitive binding of the inhibitor with the analyte and the catalyst occurs. A Heck reaction creating a coumarin fluorophore with a high quantum yield was studied. 1,4,7,10,13-Pentaaza- cyclopentadecane was chosen as the inhibiting ligand to selectively coordinate pre-catalyst Pd(II) and analytes Cd(II), Ni(II), Co(II). Monitoring product fluorescent intensity in real time revealed the concentration of analyte. The Royal Society of Chemistry 2005.

Full text of DOI:10.1039/b502199p

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/materials | Journal of Materials Chemistry
Heavy metal analysis using a Heck-catalyzed cyclization to create
coumarin
Qiaoyin Wu and Eric V. Anslyn*
Received 11th February 2005, Accepted 13th April 2005
First published as an Advance Article on the web 5th May 2005
DOI: 10.1039/b502199p
This paper describes a new method for heavy metal analysis via catalytic signal amplification. This
signal amplification protocol relies upon an exogenous inhibitor for deliberate deactivation of
an organometallic reaction that catalytically creates a fluorophore. When the deactivation process
is performed in the presence of the analyte, competitive binding of the inhibitor with the analyte
and the catalyst occurs. A Heck reaction creating a coumarin fluorophore with a high quantum
yield was studied. 1,4,7,10,13-Pentaaza-cyclopentadecane was chosen as the inhibiting ligand
to selectively coordinate pre-catalyst Pd(II) and analytes Cd(II), Ni(II), Co(II). Monitoring product
fluorescent intensity in real time revealed the concentration of analyte.
concentration of the analyte can be deciphered through
Introduction
monitoring of the fluorescence. In one example, reproducible
Over the last few years, optical chemical sensors1 and
results for Cu(II) were reported for both qualitative detection
biosensors2 have received increasing attention for chiral
and quantitative measurement (Scheme 1).15a Fluorophore
recognition,3 biological analyses,4 food analyses,5 as well as
environment protection.6 The detection of specific organic
3-methylindole (2) was generated through a Pd(0) catalyzed
Heck reaction of 2-iodo-N-allylaniline (1). Cyclam was used
molecules, ions, DNAs and antibodies, can benefit from signal
as the inhibitor, which possesses higher affinity to Cu(II)
amplification7 to enhance the readout to detectable levels.
compared to other metals (Cd(II), Co(II), Ni(II)). This study
Recently, many biosensors have been reported to use signal
showed that Cu(II) can successfully restore the Heck reaction,
enhancement. For example, using a specific enzyme (AlaDH)
and the limit of detection was 30 nM. As an extension of this
catalyzed dehydrogenation, alanine was measured down to
protocol, FRET induced by a Cu(I) catalyzed Huisgen
cycloaddition was exploited15b for investigation of other metal
7.2 mM in 2 seconds via an amperometric method that
monitored the consumption rate of O2.8 In addition, catalytic
analytes (e.g. Zn(II), Pb(II), Mg(II)). This study showed that
RNA cleavage has been used to give both detection and
the inhibitor EDTA has preferential coordination with Zn(II)
quantification of RNAs in situ.9 Due to its generality in
and Pb(II), thus releasing free Cu(II) to the catalytic cycle
industry, Enzyme-Linked Immunosorbent Assay (ELISA)
(Scheme 2).
remains a well renowned signal amplification technique
In this paper, we report a Heck reaction that is faster than
regardless of the requirement for different levels of antibodies
that of Scheme 1, producing 7-diethylaminocoumarin (4),
and multiple washing steps.10 In addition to enzyme catalyzed
which has a higher quantum yield of fluorescence than 2.
reactions, a polymer based method is a well established
We use this system for general toxic transition metals
amplifying approach. The ‘‘Molecule wire’’ approach results
detection, specifically, Cd(II). The Heck reaction of acrylic
from energy migration through conjugated semi-conducting
acid 5-diethylamino-2-iodo-phenyl ester (3) produces the
polymer backbones.11 This technique has found applications
fluorophore 4 (Scheme 3). The electron withdrawing carbonyl
in toxic ion12 and explosive compound detection.13 As a last
group adjacent to the terminal alkene in 3 makes it more
electrophilic, therefore increasing the Heck coupling rate.16
example, discrimination of chirality can be achieved via
conformational changes in helical polymers as the result of
chiral molecular interactions.14
1,4,7,10,13 Pentaaza-cyclopentadecane (5) is a classic ligand
for complexation of Cd(II), the coordination complex crystal
structure of which has been previously reported.17 As reported
Very recently, we reported a signal amplification protocol
using organometallic reactions for transition metal detection
and measurement.15 This signal amplification protocol relies
study, and the increased quantum yield of the fluorophore
here, the method is faster than that reported for our original
upon deliberate deactivation by an inhibitor of an organome-
tallic reaction that catalytically creates a fluorophore or
chromophore. When the deactivation process is performed in
the presence of the analyte, competitive binding of the
inhibitor with the analyte and the catalyst occurs. The extent
that the deactivating ligand is sequestered by the analyte
directly affects the rate of the catalytic reaction, and the
Scheme 1 Fluorophore 3-methylindole (2) is generated from the non-
fluorescent compound 2-iodo-N-allylaniline (1).
*anslyn@ccwf.cc.utexas.edu
This journal is ß The Royal Society of Chemistry 2005
J. Mater. Chem., 2005, 15, 2815–2819 | 2815

View Article Online
6H); 13C NMR (100 MHz, CDCl3): d 164.7, 151.9, 149.0,
132.1, 129.8, 128.2, 109.2, 107.9, 104.5, 44.4, 12.5.
Acrylic acid 5-diethylamino-2-iodo-phenyl ester (3)
Acrylic acid 3-diethylamino-phenyl ester 7 (37 mg, 0.17 mmol)
was dissolved in 2 mL dry CH2Cl2 in a 25 mL round bottom
flask. To the resulting solution was added thallium(I) acetate
(53.3 mg, 0.203mmol) and a CH2Cl2 (5 mL) solution of iodine
(48.2 mg, 0.19 mmol). The mixture was allowed to stir at room
temperature for 48 h. After filtration through a small pad of
celite, the solution was concentrated under reduced pressure
and the residue was diluted in Et2O (6 mL). The resulting
solution was washed sequentially by water, 5% Na2S2O3, brine
and dried (MgSO4). The product was purified by chromato-
graphy on silica gel using 9 : 1 hexanes–EtOAc for elution to
give the title compound as a light yellow oil (53.8 mg, 82%). 1H
NMR (400 MHz, CDCl3): d 7.49 (d, J 5 8.8 Hz, 1H), 6.66 (dd,
J 5 17.6, 1.6 Hz, 1H), 6.41 (d, J 5 2.4 Hz, 1H), 6.37–6.30 (m,
2H), 6.03 (dd, J 5 10.4,1.2 Hz, 1H), 3.29 (q, J 5 14 Hz, 4H),
1.126 (t, J 5 7.2 Hz, 6H); 13C NMR (100 MHz, CDCl3): d
163.8, 151.9, 149.0, 132.9, 133.0, 127.9, 111.69, 109.7, 106.2,
44.5, 12.3; HRMS m/z calcd for C13H16NIO3 [M]+ 345.0226,
found: 345.0224.
Scheme 2 A FRET system constructed through an in situ-generated
Cu(I) catalyzed regiospecific Huisgen cycloaddition.
enhances sensitivity, but the new reaction is hampered by a
slow initiation step at low concentrations of analyte.
Experimental
All reactions were run under an atmosphere of argon unless
otherwise indicated. Reaction vessels were oven or flame-dried
and allowed to cool in a dry box or desiccator prior to use.
Solvents and reagents were purified by standard methods.18
The chemicals were obtained from Aldrich and Fischer, and
no further purification was done unless otherwise noted.
Acetonitrile (CH3CN) was chromatography grade; triethyl-
amine (Et3N) was distilled from calcium hydride right before
1
usage. H NMR and 13C NMR were obtained from a Varian
1
Unity Plus 300 or 400 spectrometer. For H NMR, chemical
shifts are reported in parts per million (ppm) and referenced
to the residual proton resonance peaks: d 7.24 for CHCl3.
Emission spectra were obtained on a PTI fluorimeter. A
Finnigan VG analytical ZAB2-E spectrometer was used to
obtain high resolution mass spectra.
7-Diethylaminocoumarin (4)
Acrylic acid 5-diethylamino-2-iodo-phenyl ester 3 (0.0398 g,
0.115 mmol) was dissolved in a mixed solvent of CH3CN–H2O
(5 mL : 0.5 mL) in a 25 ml round bottom flask under argon.
To the flask was sequentially added Pd(OAc)2 (0.64 mg,
0.0029 mmol), Et3N (30 mL, 0.23 mmol), and tri-o-tolylphos-
phine (1.75 mg, 0.0057 mmol). The solution was stirred at 70 uC
for 1 h. The reaction mixture was partitioned with Et2O and
H2O, and the organic layer was dried over anhydrous MgSO4.
After filtration through a small pad of celite, volatiles were
removed under reduced pressure. The product was purified by
chromatography on silica gel using 7 : 1 hexanes–EtOAc for
elution to give the title compound as a light yellow solid
Acrylic acid 3-diethylamino-phenyl ester (7)
To a 25 mL round bottom flask was added 3-diethylamino-
phenol 6 (343 mg, 2.1 mmol), Et3N (0.28 mL, 4 mmol),
4-dimethylaminopyridine (DMAP, 6 mg, 0.1 mmol.), and
CH2Cl2 (7 mL). The resulting solution was cooled to 0 uC
followed by slow addition of acryloyl chloride (0.1 mL,
2.4 mmol). After 1.5 h, the reaction solution was poured into
aqueous saturated NaHCO3 (8 mL) and the resulting mixture
was extracted with chloroform (3 6 5 mL). The combined
organic solution was washed with H2O (2 6 2 mL), brine
(3 mL) and dried (MgSO4). After filtration through a small
pad of celite, volatiles were removed under reduced pressure.
The product was purified by chromatography on silica gel
using 9 : 1 hexanes–EtOAc for elution to give the title
compound as a light yellow oil (0. 27 g, 61%). 1H NMR
(400 MHz, CDCl3): d 7.18 (dd, J 5 8.8, 2.4 Hz, 1H), 6.58 (dd,
J 5 16.8, 0.8 Hz, 2H), 6.52 (d, J 5 7.6 Hz, 1H), 6.37 (d, J 5
6.8 Hz, 1H), 6.30 (dd, J 5 17.2, 10.4 Hz, 2H), 5.97 (dd, J 5 10,
1.2 Hz, 1H), 3.32 (q, J 5 14, 6.8 Hz, 4H), 1.14 (t, J 5 7.2 Hz,
1
(20 mg, 80%). Mp 90–94 uC; H NMR (400 MHz, CDCl3): d
7.51 (d, J 5 9.2 Hz, 1H), 7.22 (d, J 5 8.8 Hz, 1H), 6.54 (dd,
J 5 8.8,2.4 Hz, 1H), 6.47 (d, J 5 2.4 Hz, 1H), 6.01 (d,
J 5 9.2 Hz, 1H), 3.39 (q, J 5 14 Hz, 4H), 1.19 (dd, J 5 6.8,
2.8 Hz, 6H); HRMS m/z calcd for C13H15NO2 [M + H]+
218.1181, found: 218.1174.
Measurements
Solutions of acrylic acid 5-diethylamino-2-iodo-phenyl ester
(3, 0.5 mM), tri-o-tolyphosphine (0.05 mM), triethylamine
(1 mM) were prepared in CH3CN–H2O (10 : 1) in 3.5 mL
Scheme 3 Synthesis of 4.
2816 | J. Mater. Chem., 2005, 15, 2815–2819
This journal is ß The Royal Society of Chemistry 2005

View Article Online
Scheme 4 Synthesis of 3.
quartz cuvettes, followed by the addition of 1,4,7,10,13-
pentaaza-cyclopentadecane (5, 0.025 mM) and the analyte.
After stirring at rt for 5 min, Pd(OAc)2 (0.025 mM) was
added into the solution. The product was excited at 377 nm
(absorption maximum), emission was monitored at 450 nm
(emission maximum) in real time at 60 uC, under argon.
the Heck reaction was not completely shut down. As
represented by curve (e), very slow product formation was
observed. The observed result could be attributed to a
thermodynamic binding equilibrium. In our previous study,
the fluorescence was monitored once per 10 minutes to
minimize possible side photo reactions.15a Curve (d) in Fig. 1
indicates that photo reactions are negligible in the current
Cd(II) detection study.
Results and discussion
The substrate acrylic acid 5-diethylamino-2-iodo-phenyl ester
(3) was prepared from commercially available 3-diethyl-
aminophenol (6) through acylation19 with acryloyl chloride
followed by regioselective iodination20 (Scheme 4). The non-
fluorescent product (3) was subjected to an intramolecular
Heck coupling condition under the mediation of Pd(OAc)2
to afford fluorophore product 7-diethylaminocoumarin (4)
(Scheme 3). Ligand 1,4,7,10,13-pentaaza-cyclopentadecane (5)
was prepared according to the literature procedure.21
Cd(II) kinetic study by monitoring the fluorescence of coumarin
(4)
As anticipated, introducing one equivalent of Cd(II) and an
equivalent of inhibiting ligand does not change the Heck
reaction rate (curve (c) vs. (a), Fig. 1). This result reveals that
one Cd(II) releases one Pd(II), and by thus monitoring the
fluorescence intensity one can obtain the Cd(II) concentration
in the system. Hence, a series of reactions were performed in
the fluorimeter. The reaction progress was monitored in real
time by tracking the fluorescence intensity of coumarin.
Various reaction rates were collected for concentrations of
Cd(II) varying from 25 mM to 0 mM (Fig. 2). If the conversion
of starting compound into product is monitored over only the
first 10% of reaction, pseudo-first order initial rate kinetics can
be applied. By dividing the rate of the reaction by the Cd(II)
concentration, the observed rate constant was obtained. A plot
of this rate constant as a function of [Cd(II)] is given in Fig. 3.
The rate constants were measured between 300 seconds and
700 seconds in Fig. 2. It is well established that Pd(II) has to be
reduced by tri-o-tolyphosphine into Pd(0) to catalyze the
coupling reaction, and this initiation process was complete
Control experimental analysis
The Heck reaction is a Pd(0) catalyzed chemical process, and
it has not been reported to be influenced by Cd(II). As
demonstrated in Fig. 1, curve (a) represents a Heck reaction
without exogenous Cd(II) and inhibiting ligand. In the
presence of Cd(II), the reaction curve (b) perfectly overlapped
with the former, indicating no interference with the Heck
reaction. Theoretically, when an equal amount of Pd(II) and
the inhibiting ligand are present in the solution, complete
complexation would be established and no active palladium
would flow into the Heck reaction cycle. However, in this case
Fig. 2 Kinetic study of Cd(II). Real time monitoring of Pd(0)
catalyzed Heck reactions with various concentrations of Cd(II) being
subjected to the system. Coumarin was excited at 377 nm, emission was
monitored at 451 nm. Reaction conditions: [3] 5 0.5 mM, [Pd(II)] 5
0.025 mM, [IL] 5 0.025 mM, at 60 uC, under Ar. Co-solvent CH3CN–
H2O (10 : 1 v/v).
Fig. 1 Control experiments. Coumarin was excited at 377 nm,
emission was monitored at 451 nm. Reaction conditions: [3] 5
0.5 mM, [Pd(II)] 5 0.025 mM, [IL] 5 0.025 mM, at 60 uC, CH3CN–
H2O (10 : 1 v/v), Ar.
This journal is ß The Royal Society of Chemistry 2005
J. Mater. Chem., 2005, 15, 2815–2819 | 2817

View Article Online
Fig. 4 Sensitivity analysis. Reaction conditions: [5] 5 0.5 mM;
[Pd(II)] 5 0.025 mM; [IL] 5 0.025 mM; [Ni] 5 [Co(II)] 5 [Cd] 5
0.025 mM; 60 uC; Solvent: CH3CN/H2O (V:V/10:1). Coumarin 6 was
excited at 377 nm, emission was monitored at 451 nm.
Fig. 3 Plot of slope vs. [Cd(II)].
within the first 300 seconds for the more concentrated Cd(II)
solutions. However, by inspection of the kinetic plots for the
lower concentrations of Cd(II), it does not appear that the
initiation step is complete by 300 s.
initiation associated with the Heck catalyst limits the
sensitivity. The general protocol is associated with easy reagent
accessibility, and convenient operation. As we and others
expand the general principle, we foresee high potential for
practical applications.
In the proposed mechanism, the relationship between the
initial rate and the corresponding analyte (Cd(II)) concentra-
tion should be constant. In practice, however, this was not
observed. As shown in Fig. 3, when the Cd(II) concentration is
lower than 10 mM the rate is much slower than that with higher
concentration. This appears to be due to the incomplete
initiation of Pd(II) to Pd(0), as required in the catalytic Heck
cycle, when lower Cd(II) concentrations are used. Initiation is a
second order process, and hence with a large fraction of Pd(II)
sequestered by the cyclam the free Pd(II) is negligible, resulting
in very slow initiation by the added tri-o-tolylphosphine. In the
lowest concentration samples, the initiation does not appear
complete even after 1000 seconds. Therefore, the initial rates
plotted in Fig. 3 near or below 10 mM do not correspond to the
same steps that are plotted at the higher concentrations. This
limits the sensitivity of the method we report here, and
indicates that future directions for this strategy will focus on
catalytic cycles that do not require initiation steps.
Acknowledgements
This research was supported by the NIH (EB00549-5)
Qiaoyin Wu and Eric V. Anslyn*
Department of Chemistry and Biochemistry, The University of Texas at
Austin, Austin, TX, USA. E-mail: anslyn@ccwf.cc.utexas.edu
References
1 (a) K. R. A. S. Sandanayake and I. O. Sutherland, Sens. Actuators
B: Chem., 1993, B11, 331; (b) D. R. Johnson and L. G. Bachas,
Anal. Bioanal. Chem., 2003, 376, 328; (c) M. Ando, Chem. Sens.,
2001, 17, 123; (d) B. Mizaikoff, Anal. Chem., 2003, 75, 258A.
2 (a) I. Willner and E. Katz, Angew. Chem., Int. Ed., 2000, 39, 1180;
(b) D. R. The´venot, K. Toth and G. S. Wilson, Pure Appl. Chem.,
1999, 71, 2333; (c) G. S. Wilson and Y. Hu, Chem. Rev., 2000, 100,
2693; (d) T. Chen, K. A. Friedman, I. Lei and A. Heller, Anal.
Chem., 2000, 72, 3757.
3 (a) E. Yashima, k. Maeda and T. Nishimura, Chem. Eur. J., 2004,
10, 4; (b) L. Zhu and E. V. Anslyn, J. Am. Chem. Soc., 2004, 126,
3676–3677; (c) Z.-B. Li, J. Lin, H.-C. Zhang, M. Sabat,
M. Hyacinth and L. Pu, J. Org. Chem., 2004, 69, 6284.
4 A. Miyawaki, J. Llopis, R. Helm, J. M. McCaffery, J. A. Adams,
M. Ikura and R. Y. Teisen, Nature, 1997, 388, 882.
5 (a) A. K. Deisingh, D. C. Stone and M. Thompson, Int. J. Food
Sci. Technol., 2004, 39, 587–604; (b) W. Khampha, V. meevootisom
and S. Wiyakrutta, Anal. Chim. Acta, 2004, 520, 133.
Selectivity analysis
Many transition metals have high affinities to poly-aza macro-
cyclic ligands. We were interested in testing the selectivity of
this sensing protocol. Hence, the transition metals Co(II) and
Ni(II) were analyzed using the same procedure as applied
to Cd(II). Fig. 4 shows that the catalytic production of
fluorophore 4 tracks the affinities of the inhibiting ligand
for Ni(II), Co(II), Cd(II). The higher the binding affinity, the
greater the initial rate observed.
6 (a) R. Pohl, D. Aldakov, P. Kuba´t, K. Jurs´ıkova´, M. Marquez and
P. Anzenbacher, Jr., Chem. Commum., 2004, 1282; (b) T. Vo-Dinh,
Sens. Actuators B, 1995, 29, 183.
7 (a) J. S. Hartig, I. Gru¨ne, S. H. Najafi-Shoushtari and M. Famulok,
J. Am. Chem. Soc., 2004, 126, 722; (b) B. Liu and G. C. Bazan,
Chem. Mater., 2004, in press; (c) J. J. Harvey, S. P. Lee, E. K. Chan,
J. H. Kim, E.-S. Hwang, C.-Y. Cha, J. R. Knutson and M. K. Han,
Anal. Biochem., 2004, 333, 246; (d) T. M. Swager, Acc. Chem. Res.,
1998, 31, 201; (e) A. Ehret, M. T. Spitler and L. S. Stuhl, Comments
Inorg. Chem., 2002, 23, 275; (f) W. J. Blaedel and R. C. Boguslaski,
Anal. Chem., 1978, 50, 1026; (g) T. Shibata, J. Yamamoto,
N. Matsumoto, S. Yonekabo, S. Osanai and K. Soai, J. Am. Chem.
Soc., 1998, 120, 12157.
Conclusion
In conclusion, a new signal amplification protocol for Cd(II)
recognition has been developed. This methodology involves a
Pd(II) catalyzed Heck transformation of acrylic acid 5-diethyl-
amino-2-iodo-phenyl ester (5) to fluorescent product 7-diethyl-
aminocoumarin (6). This reaction shortens the detection time
compared to the former one shown in Scheme 1, but the slow
8 R. C. H. Kwan, P. Y. T. Hon and R. Renneberg, Anal. Chim. Acta,
2004, 523, 81.
2818 | J. Mater. Chem., 2005, 15, 2815–2819
This journal is ß The Royal Society of Chemistry 2005

View Article Online
9 J. S. Hartig, I. Gru¨ne, S. H. Najafi-Shoushtari and M. Famulok,
J. Am. Chem. Soc., 2004, 126, 722.
15 (a) Q. Wu and E. V. Anslyn, J. Am. Chem. Soc., 2004, 126, 14682;
(b) L. Zhu, V. M. Lynch and E. V. Anslyn, Tetrahedron, 2004, 60,
7267.
16 G. T. Crisp, Chem. Soc. Rev., 1998, 27, 427.
17 G. W. Franklin, D. P. Riley and W. L. Neumann, Coord. Chem.
Rev., 1998, 174, 133.
18 W. L. F. Armarego and D. D. Perrin, Purification of Laboratory
Chemicals, 4th edn., Butterworth Heinemann, Oxford and Boston,
1996.
19 K. Hiroya, N. Suzuki, A. Yasuhara, Y. Egawa, A. Kasano and
T Sakamoto, J. Chem. Soc., Perkin Trans. 1, 2000, 4339.
20 R. C. Cambie, P. S. Rutledge, T. Smith-Palmer and
P. D. Woodgate, J. Chem. Soc., Perkin Trans. 1, 1976, 1161.
21 Z. Kovacs, E. A. Archer, M. K. Russell and A. D. Sherry, Synth.
Commun., 1999, 29, 2817.
10 (a) H. C. Hang, C. Yu, M. R. Pratt and C. R. Bertozzi, J. Am.
Chem. Soc., 2004, 126, 6; (b) P. Geymayer, N. Bahr and
J.-L. Reymond, Chem. Eur. J., 1999, 5, 1006; (c) Y. Weizmann,
F. Patolsky, E. Katz and I. Willner, J. Am. Chem. Soc., 2003, 125,
3452; (d) M. P. Robertson and A. D. Ellington, Nat. Biotechnol.,
1999, 17, 62.
11 Y. Kim, Z. Zhu and T. M. Swager, J. Am. Chem. Soc., 2004, 126,
452.
12 T. Kim and T. M. Swager, Angew. Chem., Int. Ed., 2003, 42,
4803.
13 K. Kuroda and T. M. Swager, Macromol. Symp., 2003, 201, 127.
14 E. Yashima, K. Maeda and T. Nishimura, Chem. Eur. J., 2004,
10, 42.
This journal is ß The Royal Society of Chemistry 2005
J. Mater. Chem., 2005, 15, 2815–2819 | 2819