Synthesis, characterization, solid-state structures, and spectroscopic properties of two catechol-based luminescent chemosensors for biologically relevant oxometalates

Article abstract of DOI:10.1021/ic700554n

The new heteroditopic ligand 2,3-dihydroxy-N-(1,10-phenanthroline-5-yl) benzamide (H₂-L³) was synthesized and coordinated to [Ru(bpy)₂(phen)]²⁺- and [ReBr(CO)₃(phen)]-type luminophores (bpy = 2,2′-bipyridine and phen = 1,10-phenanthroline). The resulting chemosensors [Ru(bpy)₂(H₂-L³)] ²⁺ and [ReBr(CO)₃(H₂-L³)] were fully characterized and their solid-state structures and spectroscopic properties were investigated to assess how the photophysical properties of the luminescent signaling units affect the performance of the sensors. [Ru(bpy) ₂(H₂-L³)]²⁺ and [ReBr(CO) ₃(H₂-L³)] both signal the presence and concentration of molybdate and vanadate in aqueous acetonitrile through a decrease in emission intensity. [ReBr(CO)₃(H₂-L ³)] also detects tungstate. Due to the higher emission intensity of the Ru-based sensor, its detection limits for molybdate (43 μg L ^-1) and vanadate (24 μg L^-1) are almost 1 order of magnitude lower than the ones achieved with the Re-based sensor. The optimum working pH of the chemosensors is determined by the pK_a values of the 2-hydroxy-groups of the receptor units: pH 4 for [ReBr(CO)₃(H ₂-L³)] and pH 3 for [Ru(bpy)₂(H ₂-L³)]²⁺. Both sensors are selective: equimolar amounts of PO₄^3-, SO₄^2-, ReO ₄^-, Mn(II), Fe(III), Co(II), Ni(II), Cu(II), and Zn(II) do not interfere with the detection of molybdate or vanadate.

Full text of DOI:10.1021/ic700554n

Inorg. Chem. 2007, 46, 6516−6528
Synthesis, Characterization, Solid-State Structures, and Spectroscopic
Properties of Two Catechol-Based Luminescent Chemosensors for
Biologically Relevant Oxometalates
Helen D. Batey, Adrian C. Whitwood, and Anne-K. Duhme-Klair*
Department of Chemistry, UniVersity of York, Heslington, York YO10 5DD, U.K.
Received March 23, 2007
The new heteroditopic ligand 2,3-dihydroxy-N-(1,10-phenanthroline-5-yl)benzamide (H2-L3) was synthesized and
coordinated to [Ru(bpy)2(phen)]2+- and [ReBr(CO)3(phen)]-type luminophores (bpy
) 2,2′-bipyridine and phen )
1,10-phenanthroline). The resulting chemosensors [Ru(bpy)2(H2-L3)]2+ and [ReBr(CO)3(H2-L3)] were fully characterized
and their solid-state structures and spectroscopic properties were investigated to assess how the photophysical
+
properties of the luminescent signaling units affect the performance of the sensors. [Ru(bpy)2(H2-L3)]2 and [ReBr-
(CO)3(H2-L3)] both signal the presence and concentration of molybdate and vanadate in aqueous acetonitrile through
a decrease in emission intensity. [ReBr(CO)3(H2-L3)] also detects tungstate. Due to the higher emission intensity
of the Ru-based sensor, its detection limits for molybdate (43 µ µ
g L-1) and vanadate (24 g L-1) are almost 1 order
of magnitude lower than the ones achieved with the Re-based sensor. The optimum working pH of the chemosensors
is determined by the pKa values of the 2-hydroxy-groups of the receptor units: pH 4 for [ReBr(CO)3(H2-L3)] and pH
3
2
3 for [Ru(bpy)2(H2-L3)]2+. Both sensors are selective: equimolar amounts of PO4 -, SO4 -, ReO4-, Mn(II), Fe(III),
Co(II), Ni(II), Cu(II), and Zn(II) do not interfere with the detection of molybdate or vanadate.
Introduction
Most receptor units for oxoanions are designed to take
advantage of host-guest interactions, such as hydrogen
bonding and electrostatic attraction.1,2 A supramolecular
approach, however, makes it difficult to differentiate oxo-
Recent advances in the field of oxoanion recognition have
led to the development of a new range of selective
chemosensors.1 The commonest oxoanions targeted are
nitrate, sulfate, phosphate, and phosphate derivatives, includ-
ing ATPn-.2 Studies addressing the structurally related
oxometalates molybdate, tungstate or vanadate, however, are
still rare.3 Despite the biological,4,5 industrial,6 and environ-
mental importance7 of such species, there are, as yet, no
luminescent chemosensors for oxometalates on the mar-
ket.8
metalates, such as MoO42- and HVO42-, from oxoanions of
2-
similar size, shape, and charge, such as SO42- and HPO4
.
Coordination chemistry, on the other hand, can be used to
achieve the required selectivity since the metal centers in
2-
MoO42-, WO42-, and HVO4 are able to bind additional
ligands to enhance their coordination number to six
2-
2-
(Scheme 1). SO4 and HPO4 cannot react in this way.
In our recent research, we thus based the selectivity of
our sensor molecules on a combination of coordinative
bonding and electrostatics by choosing 2,3-dihydroxyben-
zamides (catecholamides) as receptor units.9 Catechols are
known to react with oxometalates.10-12 Once deprotonated,
the negatively charged O donors have a high affinity for hard,
highly charged metal ions.13 Selectivity for oxometalates over
cations can be achieved by pH control, since the pH
dependence of the reaction of oxometalates with catechols
is fundamentally different from the binding of cations
(Scheme 1).
* To whom correspondence should be addressed. E-mail: akd1@
york.ac.uk.
(1) (a) Beer, P. D.; Gale, P. A. Angew. Chem., Int. Ed. 2001, 40, 486. (b)
Gale, P. A. Coord. Chem. ReV. 2003, 240, 1.
(2) Fabbrizzi, L.; Licchelli, M.; Rabaioli, G.; Taglietti, A. Coord. Chem.
ReV. 2000, 205, 85.
(3) (a) Pal, B. K.; Singh, K. A.; Dutta, K. Talanta 1992, 39, 971. (b)
Kawakubo, S.; Suzuki, H.; Iwatsuki, M. Anal. Sci. 1996, 12, 767. (c)
Jiang, C.; Wang, J.; He, F. Anal. Chim. Acta 2001, 439, 307 and
references therein.
(4) Pau, R. N.; Lawson D. M. In Metal Ions in Biological Systems; Sigel,
H., Sigel, A., Eds.; Marcel Dekker: New York, 2002; p 32.
(5) Kuper, J.; Llamas, A.; Hecht, H.-J.; Mendel, R. R.; Schwarz, G. Nature
2004, 430, 803.
(6) (a) Sato, K.; Aoki, A.; Noyori, R. Science 1998, 281, 1646. (b) Sato,
K.; Aoki, A.; Takagi, T.; Noyori, R. J. Am. Chem. Soc. 1997, 119,
12386.
(7) Pyrzynska, K.; Wierzbicki, T. Talanta 2004, 64, 823.
(8) Haugland, R. P. Handbook of Fluorescent Probes and Research
Products; Molecular Probes, 2004. http://www.probes.com.
6516 Inorganic Chemistry, Vol. 46, No. 16, 2007
10.1021/ic700554n CCC: $37.00
© 2007 American Chemical Society
Published on Web 07/06/2007

Two Catechol-Based Luminescent Chemosensors
Scheme 1
nophores to obtain molecular sensors 1 and 2, which signal
the presence and concentration of oxometalates in solution
through a decrease in emission intensity.9 The high selectivity
of 1 and 2 for molybdate is consistent with the recent finding
that the siderophore aminochelin [N-(2, 3-dihydroxybenzoyl)-
diaminobutane] of Azotobacter Vinelandii, which was used
as the molybdenum-binding unit in our first sensor prototype
(1),9a is potentially a “molybdophore”.29
In this paper, we describe the synthesis and the properties
of two new molecular sensors, 3 and 4, which are based on
2,3-dihydroxy-N-(1,10-phenanthroline-5-yl)benzamide (H2-
L3), a ligand suitable for coordination to [Ru(bpy)2]2+ as well
as [ReX(CO)3] fragments. H2-L3 was designed to allow a
systematic investigation of the effect of the photophysical
properties of the luminescent signaling unit on the perfor-
mance of the chemosensors.
MoO42- + 2H2cat h [MoO2(cat)2]2- + 2H2O
Fe3+ + 3H2cat h [Fe(cat)3]3- + 6H+
Protocols for the quantitative analysis of molybdate based
on the colorimetric detection of its catecholate complexes
are well established.14 The sensitivity of these techniques,
however, is limited to the micromolar range by the low
intensity of the characteristic ligand-to-metal charge-transfer
band. In order to develop a more sensitive fluorimetric
technique, we are interested in luminescent signaling units
with high quantum yields and their ability to signal recogni-
tion events at appended catechol units via photoinduced
electron-transfer processes or excited-state switching.
Catechol units that are linked to luminescent signaling
units have previously been shown to function as sensors,15-17
redox switches,18 and photo-19 and electrocatalysts.20 [Ru-
(bpy)3]2+-type luminophores are the most commonly studied;
however, [ReX(CO)3(bpy)]-type luminophores (with bpy )
2,2′-bipyridine and X ) Br, Cl) are beginning to appear more
frequently in the field.21-24 In addition, complexes with
functionalized 1,10-phenanthroline ligands are being
developed.25-27 These often have significantly higher emis-
sion quantum yields and lifetimes than corresponding bipy-
ridine-based systems.28
In our previous work, we connected catechols as receptor
units to [Ru(bpy)3]2+- and [Re(CO)3(bpy)(py)]+-type lumi-
(9) (a) Jedner, S. B.; James, R. J.; Perutz, R. N.; Duhme-Klair, A.-K. J.
Chem. Soc., Dalton Trans. 2001, 2327. (b) Jedner, S. B.; Perutz, R.
N.; Duhme-Klair, A.-K. Z. Anorg. Allg. Chem. 2003, 629, 2421. (c)
Peacock, A. F. A.; Batey, H. D.; Raendler, C.; Whitwood, A. C.;
Perutz, R. N.; Duhme-Klair, A.-K. Angew. Chem., Int. Ed. 2005, 44,
1712.
(10) Kustin, K.; Liu, S.-T. J. Am. Chem. Soc. 1973, 95, 2487.
(11) Duhme-Klair, A.-K.; Vollmer, G.; Mars, C.; Fro¨hlich, R. Angew.
Chem., Int. Ed. 2000, 39, 1626.
(12) Liu, C.-M.; Nordlander, E.; Schmeh, D.; Shoemaker, R.; Pierpont, C.
G. Inorg. Chem. 2004, 43, 2114.
(13) (a) Pierpont, C. G.; Buchanan, R. M. Coord. Chem. ReV. 1981, 38,
45. (b) Pierpont, C. G.; Lange, C. W. Prog. Inorg. Chem. 1994, 41,
331.
(14) Snell, F. D. Photometric and Fluorometric Methods of Analysis, Metals
Part 2; John Wiley & Sons: New York, 1978; p 1296.
(15) Beer, P. D. Chem. Commun. 1996, 689.
(16) Costa, I.; Fabbrizzi, L.; Pallavicini, P.; Poggi, A.; Zani, A. Inorg. Chim.
Acta 1998, 275-276, 117.
(17) O’Bian, L.; Duati, M.; Rau, S.; Guckian, A. L.; Keyes, T. E.; O’Boyle,
N. M.; Serr, A.; Go¨rls, H.; Vos, J. G. J. Chem. Soc., Dalton Trans.
2004, 514.
(18) Goulle, V.; Harriman, A.; Lehn, J.-M. J. Chem. Soc., Chem. Commun.
1993, 1034.
(19) (a) Whittle, B.; Everest, N. S.; Howard, C.; Ward, M. D. Inorg. Chem.
1995, 34, 2025. (b) Shukla, A. D.; Whittle, B.; Bajaj, H. C.; Das, A.;
Ward, M. D. Inorg. Chim. Acta 1999, 285, 89.
(20) Storrier, G. D.; Takada, K.; Abruna, H. D. Inorg. Chem. 1999, 38,
559.
(21) Schanze, K. S.; MacQueen, D. B.; Perkins, T. A.; Cabana, L. A. Coord.
Chem. ReV. 1993, 122, 63.
(22) Sun, S.-S.; Lees, A. J. Coord. Chem. ReV. 2002, 230, 171.
(23) Lewis, J. D.; Perutz, R. N.; Moore, J. N. Chem. Commun. 2000, 1865.
(24) Busby, M.; Matousek, P.; Towrie, M.; Clark, I. P.; Motevalli, M.;
Hartl, F.; Vlcˇek, A., Jr. Inorg. Chem. 2004, 43, 4523.
(25) Bai, G.-Y.; Wang, K.-Z.; Duan, Z.-M.; Gao, L.-H. J. Inorg. Biochem.
2004, 98, 1017.
Experimental Section
Reagents and Instrumentation. Commercially available re-
agents were obtained from Aldrich, Fluka, or Lancaster and used
as supplied. Solvents were dried over molecular sieves where
necessary. NMR spectra were recorded on Jeol JNM-EX270 or
Bruker AMX500 instruments. Absorption spectra were measured
on a Hitachi U-300 spectrophotometer in 10 mm quartz cuvettes.
The emission and excitation spectra were recorded on a Hitachi
F-4500 fluorimeter and are uncorrected. The pH measurements were
carried out by using a WTW Profilab pH 597 pH meter equipped
(26) Lazarides, T.; Miller, T. A.; Jeffery, J. C.; Ronson, T. K.; Adams, H.;
Ward, M. D. Dalton Trans. 2005, 528.
(27) Higgins, B.; DeGraff, B. A. Inorg. Chem. 2005, 44, 6662.
(28) Kober, E. M.; Meyer, T. J. Inorg. Chem. 1985, 24, 106.
(29) Liermann, L. J.; Guynn, R. L.; Anbar, A.; Brantley, S. L. Chem. Geol.
2005, 220, 285.
Inorganic Chemistry, Vol. 46, No. 16, 2007 6517

Batey et al.
with a Metler Toledo InLab 422 pH electrode. The pH scale was
calibrated by using Gran’s method.30 Infrared spectra were recorded
on a Perkin-Elmer Lambda 7 FT-IR instrument. Electrospray
ionization mass spectra were obtained using a Finnigan LCT mass
spectrometer and CI and EI mass spectra were measured on a VG
Analytical Autospec mass spectrometer. HR-mass spectra were
obtained from the EPSRC National Mass Spectrometry Service
Centre in Swansea. Elemental analyses were performed by the
analytical services at the University of Manchester. Diffraction data
were collected on a Bruker Smart Apex diffractometer with an
Oxford cryostream cooling system and Mo KR radiation source
using a SMART CCD camera. Structures were solved by direct
methods using SHELXS-97 and refined by full-matrix least-squares
using SHELXL-97 (G. M. Sheldrick, University of Go¨ttingen,
Germany, 1997). All non-hydrogen atoms were refined anisotro-
pically. Hydrogen atoms were placed using a riding model and
included in the refinement at calculated positions.
(m, 4H, bpy CH), 8.17 (d, 1H, phen CH), 8.11 (m, 2H, bpy CH),
8.01 (m, 3H, bpy CH and phen CH), 7.84 (m, 3H, bpy CH and
phen CH), 7.72 (m, 3H, phen CH and cat CH), 7.58 (m, 2H, bpy
CH), 7.46 (m, 2H, bpy CH), 7.32 (m, 2H, cat CH), 7.26 (m, 2H,
bpy CH), 3.93 (s, 6H, CH3). 13C NMR (125 MHz, CD3CN): δ
165.4 (CdO), 153.9 (CH bpy), 153.6 (CH phen), 153.0 (CH bpy),
152.3 (CH phen), 138.9 (CH bpy), 137.4 (CH phen), 132.1(CH
phen), 128.6 (CH bpy), 128.4 (CH bpy), 127.3 (CH cat), 127.0
(CH phen), 126.0 (CH cat), 125.3 (CH bpy), 123.1 (CH cat), 118.6
(CH phen), 117.9 (CH phen), 62.6 (-OCH3), 56.9 (-OCH3). UV-
vis (CH3CN/H2O 20:1): ꢀ450
)16 500 dm3 mol-1 cm-1
.
nm
Mp: 274-276 °C. HR-ESI-MS: m/z 912.1357 (calculated for
C41H33O3N7F6P96Ru: 912.1358).
Synthesis of [Ru(bpy)2(H2-L3)](PF6)2, 3 (PF6)2. Under an inert
atmosphere, [Ru(bpy)2(Me2-L3)](PF6)2 (0.25 g, 0.24 mmol) was
dissolved in 15 mL anhydrous dichloromethane. The solution was
cooled to -78 °C, and a 10-fold molar excess of 1.0 M boron
tribromide in dichloromethane was added slowly. The reaction was
stirred at -78 °C for 1 h and then allowed to warm to room
temperature over 12 h. Water was added carefully until no more
HBr evolved. The mixture was evaporated to dryness, and the
residue was taken up in methanol, which was evaporated to remove
boron-containing byproducts. The residue was then dissolved in a
minimum amount of methanol, and the solution was brought to
neutral pH with 3 M NaOH to precipitate the product. To ensure
that all 3 precipitated, a few drops of a saturated ammonium
hexafluorophosphate solution were added and the mixture was
cooled to 4 °C for 12 h. The precipitate was isolated, washed with
cold water, and dried in vacuo. Yield: 0.17 g (0.16 mmol, 68%).
1H NMR (500 MHz, CD3CN): δ 8.923 (d, 1H, phen CH), 8.695
(m, 2H, phen CH), 8.534 (m, 5H, bpy CH), 8.160 (m, 2H, phen
CH), 8.107 (m, 2H, bpy CH), 8.014 (m, 4H, bpy CH), 7.841 (m,
3H bpy CH and phen CH), 7.741 (m, 1H, phen CH, 7.31 (d, 1H,
cat CH), 7.584 (d, 1H, cat CH), 7.465 (m, 2H, bpy CH), 7.286 (m,
2H, bpy CH and cat CH). 13C NMR (125 MHz, CD3CN): δ 170.0
(CdO), 154.5 (phen CH), 154.4 (phen CH), 153.2 (cat CH), 153.0
(cat CH), 153.9 (bpy CH), 141.1 (bpy CH), 138.9 (bpy CH), 138.2
(phen CH), 134.8 (phen CH), 131.5 (phen CH), 128.6 (bpy CH),
128.5 (cat CH), 127.9 (phen CH), 127.6 (phen CH), 125.3 (bpy
CH). UV-vis (CH3CN/H2O, 20:1; pH 6): ꢀ450 nm ) 17 100 dm3
mol-1 cm-1. Mp: 279-280 °C. HR-ESI-MS: m/z ) 890.1012
(calculated for C39H29O3N7F6P102Ru: 890.1012).
Synthesis of [ReBr(CO)3(Me2-L3)]. Me2-L3 (0.34 g, 0.94 mmol)
and rhenium pentacarbonyl bromide (0.19 g, 0.46 mmol) were
dissolved in 80 mL dry toluene and refluxed for 4 h under an inert
atmosphere. Once the completeness of the reaction was confirmed
by IR spectroscopy, the volume of the reaction mixture was reduced
to half. A yellow precipitate formed, which was isolated, washed
with cold toluene, and dried. Yield: 0.29 g (0.41 mmol, 90%). 1H
NMR: (500 MHz, CDCl3): δ 11.104 (s, 1H, NH), 9.460 (d, 1H,
phen CH), 9.286 (d, 1H, phen CH), 9.143 (s, 1H, phen CH), 8.666
(d, 1H, phen CH), 8.505 (d, 1H, phen CH), 7.911 (m, 1H, phen
CH), 7.866 (d, 1H, cat CH), 7.816 (m, 1H, phen CH), 7.364 (t,
1H, cat CH), 7.209 (d, 1H, cat CH), 4.101 (s, 3H, CH3), 3.993 (s,
3H, OCH3). 13C NMR (125 MHz, CDCl3): δ 196.6 (Re-CO),
163.9 (CdO), 153.1 (phen CH), 152.0 (phen CH), 137.8 (phen CH),
131.0 (phen CH), 126.2 (phen CH), 125.6 (phen CH), 125.5 (cat
CH), 123.2 (cat CH), 116.9 (cat CH), 116.5 (phen CH), 62.4
Synthesis of 2,3-Dimethoxy-N-(1,10-phenanthroline-5-yl)ben-
zamide, Me2-L3. Dicyclohexylcarbodiimide (1.13 g, 5.5 mmol) was
added at 0 °C to a stirred solution of 2,3-dimethoxybenzoic acid
(2.00 g, 11.0 mmol) in 50 mL of dry acetonitrile. After 15 min,
the reaction was brought to room temperature and stirred until thin-
layer chromatography (TLC) analysis (10:1 chloroform/methanol
v/v) confirmed that the reaction had gone to completion. After
removal of the precipitated byproduct by filtration, the filtrate was
evaporated to dryness. 2,3-Dimethoxybenzoic anhydride was
obtained as an oily residue, which was redissolved in 20 mL
anhydrous acetonitrile. 1,10-Phenanthroline-5-yl amine (0.80 g,
4.1 mmol) was suspended in 50 mL anhydrous chloroform, and
the insoluble components were removed by filtration. The resulting
solution was added to the 2,3-dimethoxybenzoic anhydride solution,
and the mixture was refluxed for 6 h under N2 and under exclusion
of light. After the removal of the solvents, the residual oil was
taken up in chloroform. The product was extracted from the organic
phase with 2 M HCl. The pH of the aqueous phase was then
increased to 10 with 3 M NaOH, and the product was re-extracted
into chloroform. The solvent was removed in vacuo, and the
remaining white solid was vacuum-dried. Recrystallization from
chloroform/petroleum ether gave the pure product. Yield: 1.23 g
(3.4 mmol, 83%). 1H NMR (500 MHz, CD3CN): δ 10.511 (s, 1H,
NH), 9.155 (d, 1H, phen CH), 9.042 (d, 1H, phen CH), 8.605 (s,
1H, phen CH), 8.515 (d, 1H, phen CH), 8.356 (d, 1H, phen CH),
7.804 (m, 1H, phen CH), 7.679 (m, 2H, 1 phen CH, 1 cat CH),
7.274 (m, 2H, cat CH), 4.089 (s, 3H, CH3), 3.953 (s, 3H, CH3).
13C NMR (125 MHz, CD3CN): δ 165.2 (CdO), 150.9 (phen CH),
150.3 (phen CH), 136.8 (phen CH), 131.0 (phen CH), 125.6 (cat
CH), 124.5 (cat CH), 124.1 (phen CH), 123.1 (phen CH), 119.4
(phen CH), 117.4 (cat CH), 62.4 (-O-CH3), 56.9 (-OCH3). UV-
vis (CH3CN/H2O 20:1): ꢀ270
) 26 000 dm3 mol-1 cm-1. Mp:
nm
108-110 °C. HR-CI-MS: m/z 360.1347 (calculated for
C21H18N3O3: 360.1348).
Synthesis of [Ru(bpy)2(Me2-L3)](PF6)2. Under an inert atmo-
sphere, Me2-L3 (0.20 g, 0.55 mmol) and cis-[RuCl2(bpy)2] × 2H2O
(0.25 g, 0.55 mmol) were dissolved in 35 mL of a EtOH/H2O 3:1
mixture and refluxed for 6 h. After cooling, the volume of the
solution was reduced by half, and the product was precipitated by
the addition of a saturated solution of ammonium hexafluorophos-
phate in water. The precipitate was isolated, washed with cold water
and ethanol, and dried in vacuo. Yield: 0.51 g (0.47 mmol, 94%).
1H NMR (500 MHz, CD3CN): δ 10.90 (s, 1H, NH), 9.07 (s, 1H,
phen CH), 8.68 (d, 1H, phen CH), 8.61 (d, 1H, phen CH), 8.53
(-OCH3), 56.4 (-OCH3). UV-vis (CH3CN/H2O, 20:1): ꢀ390
nm
) 3140 dm3 mol-1 cm-1. IR (THF): ν(CO) 2019, 1918, 1893 cm-1
.
Mp: 248-250 °C. HR-FAB-MS: m/z 686.0240 (calculated for
(30) Gran, G. Analyst 1952, 77, 661.
C24H17N3O6185Re Na: 686.0233).
6518 Inorganic Chemistry, Vol. 46, No. 16, 2007

Two Catechol-Based Luminescent Chemosensors
Scheme 2
N.B.: [ReCl(CO)3(Me2-L3)] can be synthesized from rhenium
pentacarbonyl chloride, analogue to the procedure described for
[ReBr(CO)3(Me2-L3)]. Upon deprotection with BBr3, however, a
partial exchange of the chloride ligands with bromide ligands was
observed.
Metal-to-Sensor Titrations. Titrations for the determination of
the composition of the complexes were conducted using the
following procedure. A volume of 3 mL of the sample was buffered
with 3.6 µL of lutidine, and the pH was adjusted to the required
value with standard acid and base solutions. To the 0.02 mM
solution of [Ru(bpy)2(H2-L3)]2+, 5 µL aliquots of a metal-containing
solution of concentration 0.75 mM were added. After each addition,
the sample was stirred for ca. 3 min to allow the solution to
equilibrate before the emission spectrum was recorded. Aliquots
of the metal-containing solution were added until an approximate
1:1 ratio was reached. For the analogous titrations with [ReBr-
(CO)3(H2-L3)], the sample solutions were 0.1 mM and the solutions
of metal salts were 3.75 M.
Competitive Selectivity. To test the selectivity of the sensors,
their emission intensity at λmax was recorded in the absence and
presence of each potentially interfering ion. The experiments were
conducted at a pH value, which was within the active range of the
particular sensor. First, the emission intensity of the free sensors
was recorded at the standard concentration. Then, a stoichiometric
quantity of the competing ion was added, the solution stirred, and
a second intensity reading taken. A final reading was taken 3 min
after the addition of a stoichiometric quantity of molybdate (or
vanadate).
Synthesis of [ReBr(CO)3(H2-L3)], 4. Under an inert atmosphere,
[ReBr(CO)3(Me2-L3)] (0.20 g, 0.029 mmol) was dissolved in 15
mL anhydrous dichloromethane and cooled to -78 °C. A 10-fold
molar excess of 1.0 M boron tribromide in dichloromethane was
slowly added. The reaction was stirred at -78 °C for 1 h and then
allowed to warm to room temperature. The mixture was stirred at
room temperature for a further 12 h. Water was added until additions
no longer caused the evolution of HBr. The reaction mixture was
evaporated to dryness, and the residue was taken up in methanol
(approximately 200 mL). The volume of the solution was reduced
until a yellow precipitate formed, which was isolated and vacuum-
dried. Yield: 0.14 g (0.21 mmol, 72%). 1H NMR (500 MHz, CD3-
CN): δ 11.37 (s, 1H, OH), 9.77 (s, 1H, OH), 9.46 (d, 1H, phen
CH), 9.39 (d, 1H, phen CH), 8.82 (d, 1H, phen CH), 8.75 (d, 1H,
phen CH), 8.46 (s, 1H, phen CH), 7.97 (m, 2H, phen CH), 7.57 (d,
1H, cat CH), 7.15 (d, 1H, cat CH), 6.97 (m, 2H, cat CH and NH).
IR (CH3CN): ν(CO) 2022, 1918 and 1896 cm-1. Mp: 263-
265 °C. Elemental Anal. Calcd for C22H13O6N3ReBr: C, 38.5; H,
2.2; N, 5.8. Found: C, 38.7; H, 1.8; N, 6.1.
General Titration Protocol. The solvent system used consisted
of a mixture of acetonitrile and water in 20:1 ratio. Adjustments to
the pH were carried out with a 0.6 M solution of HCl in this solvent
system and a 0.6 M solution of tetramethyl ammonium hydroxide
in water. Solutions of ruthenium-based compounds were made to
a concentration of 0.02 mM, and rhenium-based compounds were
made to a concentration of 0.1 mM. These concentrations gave an
absorbance within the Beer-Lambert range. For the ruthenium-
based samples, the excitation wavelength was set to 450 nm and
uncorrected spectra were recorded between 490 and 800 nm. The
rhenium-based samples were excited at 390 nm, and uncorrected
spectra were recorded between 450 and 800 nm. In all cases, 3 mL
amounts of the solutions were pipetted into a 10 mm quartz cuvette,
which remained sealed to minimize solvent evaporation.
Determination of Relative Quantum Yields. The relative
quantum yields were calculated according to eq 1 by using an
aerated aqueous solution of [Ru(bpy)3]Cl2 as the standard (φstd
)
0.02831). Φsam and Φstd are the quantum yields of the sample and
the standard, respectively. Asam and Astd are the absorbances of the
sample (in acetonitrile/water 20:1) and the standard (in water) at
the excitation wavelength (λexc ) 400 nm). Isam and Istd are the
integrated emission intensities obtained from the corrected emission
spectra, recorded with aerated solutions at room temperature.
Astd Isam
Φsam ) Φstd
(1)
(
)(I
)
Asam
std
Results and Discussion
Determination of pH Profiles. The solution of the sensors was
adjusted to the starting pH value using the standard acid or base
solution, and a spectrum was recorded. Depending on the direction
of the titration, small aliquots of either acid or base were added to
the sample. The pH of the solution was allowed to stabilize before
a spectrum was recorded. The spectra were recorded at intervals
of approximately 0.5 pH units, across the pH range of ca. 1-10.
Analogous titrations were conducted in the presence of stoichio-
metric quantities of the oxometalates of interest as well as
potentially interfering anions and cations. For this purpose, aqueous
standard solutions of the corresponding metal salts were added in
stoichiometric quantities in a 5 µL aliquot, so as not to significantly
alter the solvent composition. Standard solutions of oxoanions were
prepared by using Na2MoO4‚2H2O, Na2WO4‚2H2O, NH4VO3, NH4-
ReO4, Na2SO4, and Na2HPO4. Metal cations were used as chloride
or nitrate salts.
Synthesis and Characterization of [Ru(bpy)2(H2-L3)]-
(PF6)2, 3(PF6)2, and [ReBr(CO)3(H2-L3)], 4. 2,3-Dimethoxy-
N-(1,10-phenanthroline-5-yl)benzamide, Me2-L3, was syn-
thesized from commercially available 2,3-dimethoxybenzoic
acid and 1,10-phenanthroline-5-yl amine, as shown in
Scheme 2. 2,3-Dimethoxybenzoic acid anhydride was formed
using DCC32 (N,N′-dicyclohexylcarbodiimide) and reacted
with 1,10-phenanthroline-5-yl amine. Since 1,10-phenan-
throline-5-yl amine was observed to form insoluble decom-
position products upon exposure to light, the second step of
the reaction was carried out in the dark.
(31) Nakamaru, K. Bull. Chem. Soc. Jpn. 1982, 2697.
(32) Deroo, S.; Defrancq, E.; Moucheron, C.; Mesmaeker, A. K.-D.; Dumy,
P. Tetrahedron Lett. 2003, 44, 8379.
Inorganic Chemistry, Vol. 46, No. 16, 2007 6519

Batey et al.
Figure 2. Intramolecular hydrogen bonds in 2,3-dihydroxybenzamides.
the rhenium center is distorted octahedral, as is typical for
diimine-coordinated fac-ReI(CO)3 units. The small N(1)-
Re-N(2) bond angle of 75.28° is determined by the bite
angle of the bidentate phenanthroline ligand. The Re-C and
Re-N distances are within the expected range.38-41
Figure 1. ORTEP plot (50% probability ellipsoids) of the molecular
structure of 4.
Table 1. Selected Bond Distances (Å) and Angles (deg) of 4‚CH3CN
Re(1)-C(1)
1.920(8)
1.934(8)
1.894(8)
2.649(7)
2.671(6)
Re(1)-Br(1)
Re(1)-N(1)
Re(1)-N(2)
O(6)-Br(1)#1a
C(16)-O(4)
2.6254(8)
2.195(5)
2.156(5)
3.317(5)
1.207(8)
1.353(8)
100.1(2)
92.5(2)
97.8(2)
86.18(12)
85.15(13)
75.28(18)
122.7(6)
Re(1)-C(2)
The conformation of the ligand H2-L3 in 4 is of interest,
since it determines the degree of electronic coupling between
the metal-binding unit and the signaling unit. In 4, the
dihedral angle between the fitted planes of the catechol and
the phenanthroline deviates by only 4.1(2)° from coplanarity.
The catecholamide unit is held planar by an intramolecular
hydrogen bond formed between the amide N-H and the
ortho O atom of the catechol unit (dN(3)-O(5) ) 2.649(7) Å).
A second intramolecular hydrogen bond is formed between
the ortho O-H and the meta O atoms of the catechol ring
(dO(5)-O(6) ) 2.671(6) Å).
Re(1)-C(3)
N(3)-O(5)
O(5)-O(6)
O(5)-N(4)
2.823(10) C(16)-N(3)
C(1)-Re(1)-C(2)
C(3)-Re(1)-C(1)
C(3)-Re(1)-C(2)
C(1)-Re(1)-Br(1)
90.0(3)
91.7(3)
87.3(4)
86.9(2)
C(1)-Re(1)-N(1)
C(2)-Re(1)-N(1)
C(2)-Re(1)-N(2)
N(1)-Re(1)-Br(1)
N(2)-Re(1)-Br(1)
N(2)-Re(1)-N(1)
N(3)-C(16)-O(4)
C(2)-Re(1)-Br(1) 176.4(2)
C(3)-Re(1)-Br(1)
C(8)-N(3)-C(16)
94.7(3)
128.7(6)
a Symmetry transformation used to generate equivalent atom: #1, - x,
-y + 1, -z + 1.
A survey of the Cambridge Structural Database revealed
that this hydrogen-bond orientation is unusual for a 2,3-
dihydroxybenzamide. In 9 out of the 10 structures reported,
it is the carbonyl oxygen of the amide link that interacts with
the ortho-hydroxyl group, as shown as II in Figure 2. Recent
DFT calculations confirm that hydrogen-bond orientation II
is significantly more stable than orientation I.42
In the structure of 4, however, the unusual hydrogen bond
orientation I allows the formation of an additional intermo-
lecular hydrogen bond between the O-H group in the meta
position of the catechol and the bromide of an adjacent
rhenium luminophore (Figure 3 and Supporting Information
S1). The observed O(6)-Br(1′) distance of 3.317(5) Å is
consistent with those described in the literature.43 In addition,
a weak hydrogen bond is formed between the ortho-hydroxyl
The complexes [Ru(bpy)2(Me2-L3)]2+ and [ReBr(CO)3-
(Me2-L3)] were formed by refluxing the methyl-protected
ligand with the respective luminophore precursors, [Ru-
(bpy)2Cl2]33 and [Re(CO)5Br],34 according to published
procedures.35,36 The neutral rhenium-based sensor was iso-
lated directly from the reaction mixture while the cationic
ruthenium complex was precipitated as the hexafluorophos-
1
phate salt. Both complexes were characterized by H and
13C NMR spectroscopy and ESI mass spectrometry. The
aromatic region in the spectrum of [Ru(bpy)2(Me2-L3)](PF6)2
1
was assigned based on a H-1H COSY NMR spectrum.
The quantitative demethylation of both complexes was
37
achieved with BBr3 and confirmed by TLC, mass spec-
trometry, NMR spectroscopy, and X-ray crystallography. In
the proton NMR spectra of both compounds, the complete-
ness of the deprotection was verified by the absence of the
two characteristic methyl resonances.
Crystal Structure Determinations. The Crystal Struc-
ture of [ReBr(CO)3(H2-L3)]. Single crystals of [ReBr(CO)3-
(H2-L3)] (4) were obtained by the slow evaporation of a
concentrated solution of 4 in acetonitrile. The molecular
structure of the complex is shown in Figure 1, selected bond
angles and distances are given in Table 1 and crystal data
are summarized in Table 2. The coordination geometry of
group and an acetonitrile solvent molecule (dO(5)-N(4)
)
2.823 Å). Interestingly, the only other structure reported to
date to adopt hydrogen-bond orientation I also contains
additional intermolecular hydrogen-bonding interactions that
involve polar solvent molecules.44 This demonstrates that the
presence of intermolecular hydrogen-bond acceptors can
change the orientation of the intramolecular hydrogen bonds
(38) Mart´ı, A. A.; Mezei, G.; Maldonado, L.; Paralitici, G.; Raptis, R. G.;
Colo´n, J. L. Eur. J. Inorg. Chem. 2005, 118.
(39) Lazarides, T.; Miller, T. A.; Jeffery, J. C.; Ronson, T. K.; Adams, H.;
Ward, M. D. Dalton Trans. 2005, 528.
(33) Sullivan, B. P.; Salmon, D. J.; Meyer, T. J. Inorg. Chem. 1978, 17,
3334.
(34) Kutal, C.; Weber, M. A.; Ferraudi, G.; Geiger, D. Organomet. 1985,
4, 2161.
(35) Sacksteder, L.; Zipp, A. P.; Brown, E. A.; Streich, J.; Demas, J. N.;
DeGraff, B. A. Inorg. Chem. 1990, 29, 4335.
(40) Yam, V. W.-W.; Chong, S. H.-F; Cheung, K.-K. Organomet. 2000,
19, 5092.
(41) Busby, M.; Gabrielsson, A.; Matousek, P.; Towrie, M.; Di Bilio, A.
J.; Gray, H. B.; Vlcˇek, A., Jr. Inorg. Chem. 2004, 43, 4994.
(42) Hay, B. P.; Dixon, D. A.; Vargas, R.; Garza, J.; Raymond, K. N. Inorg.
Chem. 2001, 40, 3922.
(36) Wu, F.; Riesgo, E.; Pavalova, A.; Kipp, R. A.; Schmehl, R. H.;
Thummel, R. P. Inorg. Chem. 1999, 38, 5620.
(37) O’Brien, L.; Duati, M.; Rau, S.; Guckian, A. L.; Keyes, T. E.; O’Boyle,
N. M.; Serr, A.; Go¨rls, H.; Vos, J. G. Dalton Trans. 2004, 514.
(43) Depree, C. V.; Ainscough, E. W.; Brodie, A. M.; Gainsford, G. J.;
Lensink, C. Acta Crystallogr. 2000, C56, 17.
(44) Zimmer, B.; Bulach, V.; Drexler, C.; Erhardt, S.; Hosseini, M. W.;
De Cian, A. New J. Chem. 2002, 26, 43.
6520 Inorganic Chemistry, Vol. 46, No. 16, 2007

Two Catechol-Based Luminescent Chemosensors
Table 2. Crystal Data and Summary of Data Collection and
Refinement Details
[ReBr(CO)3(H2-L3)]-
‚CH3CN
[Ru(bpy)2(H1.51-L3)]-
Br1.49‚6H2O
empirical formula
fw
C24H16BrN4O6Re
722.52
C39H29Br1.49N7O9Ru
959.73
T (K)
298(2)
115(2)
cryst syst
space group
a (Å)
b (Å)
c (Å)
R (deg)
â (deg)
γ (deg)
V (Å3)
triclinic
P1h
monoclinic
C2/c
21.159(6)
17.582(5)
22.842(7)
90
98.204(7)
90
8411(4)
7.3389(6)
12.8693(10)
13.8173(11)
70.464(2)
86.758(2)
76.554(2)
1195.84(17)
2
Figure 3. Ball-and-stick representation of two adjacent [ReBr(CO)3(H2-
L3)] molecules. O atoms are shown in red, N atoms in blue, and H atoms
except NH and OH have been omitted for clarity.
Z
F
8
calcd (mg/m3)
2.007
6.800
692
1.516
1.848
3840.9
µ (mm-1
F(000)
)
θ range (deg)
index ranges
1.56-25.10
-8 eh e 8
-15 e k e 14
-16 e l e 12
6780
1.69-19.04
-19 e h e 19
-12 e k e 16
-20 e l e 20
12 308
reflns collected
independent reflns
4202 [R(int) )
0.0281]
3416 [R(int) )
0.0498]
data/restraints/params
GOF on F 2
R indices [I > 2σ(I)]
4202/0/327
1.023
3416/0/543
1.055
Figure 4. ORTEP plot (50% probability ellipsoids) of the molecular
structure of 3 (H atoms, except those of NH and OH, bromide counterions,
and lattice water molecules, are not shown).
R1 ) 0.0356,
R2 ) 0.0799
R1 ) 0.0503,
R2 ) 0.0855
1.972/-0.585
R1 ) 0.0483,
R2 ) 0.1230
R1 ) 0.0677,
R2 ) 0.1349
1.171/-0.437
R indices (all data)
largest diff peak/hole
in 2,3-dihydroxybenzamides in the solid state. The proton
NMR data reveal, however, that 4 changes to the usual
hydrogen-bond orientation II in solution. Upon deprotection,
the signal of the amide proton shifts dramatically from
11.10 ppm in [ReBr(CO)3(Me2-L3)] to 6.97 ppm in 4. This
increase in shielding indicates that the amide proton in the
latter is not involved in hydrogen bonding.
(e Å-3
)
Table 3. Selected Bond Distances (Å) and Angles (deg) of
[Ru(bpy)2(H1.51-L3)]Br1.49‚6H2O
Ru(1)-N(1)
2.048(7) Ru(1)-N(4)
2.051(6) Ru(1)-N(5)
2.045(7) Ru(1)-N(6)
2.641(9) O(3)-Br(2)#2a
2.654(8) C(33)-N(7)
3.063(7) C(33)-O(1)
96.8(3)
92.7(2)
78.6(3)
90.1(2)
94.7(3)
79.0(3)
2.060(7)
2.065(7)
2.058(6)
3.071(6)
1.358(11)
1.225(10)
97.4(3)
97.4(3)
94.6(3)
93.7(3)
79.3(3)
Ru(1)-N(2)
Ru(1)-N(3)
N(7)-O(2)
The two sensor molecules in the unit cell are further held
together through π-π-stacking interactions between the H2-
L3 ligands (Figures 3 and Supporting Information S1).
Carbon atom C5 of one of the electron-poor pyridine rings
of the phenanthroline is positioned over the center of the π
electrons of the electron-rich catechol ring. The intermo-
lecular distance between the center of the catechol ring and
C5 of 3.3 Å indicates significant aromatic-aromatic interac-
tions. Parallel-displaced aromatic-aromatic interactions of
this type are often observed in π-stacked systems that involve
nitrogen-containing heterocycles.45
O(2)-O(3)
O(2)-Br(1)#1a
N(3)-Ru(1)-N(1)
N(3)-Ru(1)-N(2)
N(1)-Ru(1)-N(2)
N(1)-Ru(1)-N(6)
N(2)-Ru(1)-N(6)
N(3)-Ru(1)-N(4)
N(2)-Ru(1)-N(4)
N(2)-Ru(1)-N(4)
N(6)-Ru(1)-N(4)
N(3)-Ru(1)-N(5)
N(6)-Ru(1)-N(5)
N(4)-Ru(1)-N(5)
87.9(3)
C(33)-N(7)-C(32) 129.6(8)
N(7)-C(33)-O(1) 121.8(9)
a Symmetry transformations used to generate equivalent atoms: #1, -x
+ 1, -y + 2, -z + 1; #2, x, y + 1, z - 1.
The conformation of H2-L3 in 3 is very similar to the one
observed in 4. Again, the ligand is nearly planar. The dihedral
angle between the fitted planes of the catechol and the
phenanthroline refined to 4.8(4)°. Intramolecular hydrogen
bonds are formed between the amide N-H and the ortho O
atoms of the catechol (dN(7)-O(2) ) 2.641(9) Å) and the ortho
O-H and the meta O atoms of the catechol ring (dO(5)-O(6)
) 2.654(8) Å), again corresponding to orientation I in
Figure 2. This orientation appears to be favored in this case
through additional intermolecular interactions between theO-H
groups of the catechol with the bromide counterions (dO(2)-Br(1′)
) 3.317(5) Å and dO(3)-Br(2′′) ) 3.317(5) Å).
The Crystal Structure of [Ru(bpy)2(H1.51-L3)]Br1.49
.
Attempts to grow crystals of the PF6 salt of [Ru(bpy)2(H2-
L3)]2+ (3) were unsuccessful. After anion exchange, however,
single crystals of the bromide salt could be obtained by slow
evaporation of a concentrated solution in aqueous acetonitrile.
The structure of the complex is shown in Figure 4, crystal
data are summarized in Table 2 and selected bond angles
and distances are given in Table 3. The ruthenium-based
luminophore has approximately octahedral geometry, as
expected for such systems.46
(45) Janiak, C. J. Chem. Soc., Dalton Trans. 2000, 3885.
(46) Rillema, D. P.; Jones, D. S.; Levy, H. A. J. Chem. Soc., Chem. Comm.
1979, 849.
The bromide counterions in the structure are disordered
over four sites, one of which lies on a special position. The
Inorganic Chemistry, Vol. 46, No. 16, 2007 6521

Batey et al.
Table 4. Selected Spectroscopic Data for [Ru(bpy)2(Me2-L3)](PF6)2
and [ReCl(CO)3(Me2-L3)] (Aerated Solutions at Room Temperature,
Solvent acetonitrile/water in ratio of 20:1)
absorption
λ
abs/nm (10-3 ꢀ/ emission quantum yield
b
compound
M-1 cm-1
)
λ
max/nm
Φem
[Ru(bpy)2(Me2-L3)]-
(PF6)2
400 (7.46),
450 (16.50)
390 (4.00),
400 (3.68)
625
0.011
[ReCl(CO)3(Me2-L3)]a
635
0.004
a [ReCl(CO)3(Me2-L3)] was synthesized to allow a direct comparison
withthereportedspectroscopicdataof[ReCl(CO)3(phen)]-typecomplexes.53-55
b Relative luminescence quantum yield determined by using [Ru(bpy)3]Cl2
as the standard (Φ ) 0.028), λexc ) 400 nm.
Figure 5. Ball-and-stick representation of two selected [Ru(bpy)2(H2-
L3)]2+ complexes and of the Ru-bpy fragment of an adjacent complex,
illustrating aromatic-aromatic interactions. O atoms are shown in red, N
atoms in blue, and H atoms have been omitted for clarity.
solvent system and a 0.6 M solution of tetramethyl am-
monium hydroxide in water.
The spectroscopic properties of the methyl-protected
derivative [Ru(bpy)2(Me2-L3)](PF6)2 are summarized in Table
4. They agree well with the properties reported for similar
[Ru(bpy)2(phen)]2+-type complexes.26,47 While the absorption
and emission spectra of [Ru(bpy)2(Me2-L3)](PF6)2 (Support-
ing Information, Figure S3) are pH-independent, the absorp-
tion spectrum of [Ru(bpy)2(H2-L3)](PF6)2, 3(PF6)2, changes
with pH, as shown in Figure 6. By comparison with the
reported spectrum of [Ru(bpy)3]2+ and the spectra of related
polypyridine complexes,48 the lowest energy band in the
absorption spectra can be assigned to a Ru(dπ) f bpy or
phen(π*) charge transfer (metal-to-ligand charge-transfer,
MLCT) transition. While the MLCT absorbance at around
450 nm shows only little pH dependence in 3(PF6)2, the
ligand-based absorbance at 365 nm increases significantly
with pH. With reference to the pH dependence of the
absorption spectra of 2,3-dihydroxybenzamides, which gen-
erally show an increase in absorbance upon deprotonation
of the OH group in the 2-position,49 this increase can be
attributed to the formation of [Ru(bpy)2(H-L3)]+. From the
inflection point, a pKa value of approximately 4.5 can be
estimated. This pKa value is comparatively low for a 2,3-
dihydroxybenzamide50 due to the electron-withdrawing effect
of the electron-deficient phenanthroline unit that is enhanced
by the coordinated positively charged luminophore.
occupancy of the bromides was refined and gave a total of
1.49 bromides per asymmetric unit. Attempts to restrain the
total number of bromide ions in the asymmetric unit to two,
in order to balance the 2+ charge of the ruthenium, resulted
in a significant increase in the R value. Since the crystals
were obtained from a solution at pH 4.5, which is close to
the pKa value of the 2-hydroxy-group (vide infra), ap-
proximately half of the catechol units are likely to be
deprotonated, giving rise to an empirical formula of [Ru-
(bpy)2(H1.51-L3)]Br1.49. Although the data were collected at
low temperature to increase the resolution of the structure,
the positions of the hydrogen atoms could not be located in
the Fourier difference map. The hydrogen atoms of the [Ru-
(bpy)2(H2-L3)]2+ component were thus calculated using a
riding model.
The structure revealed that H2-L3 is again involved in
intermolecular π-π-stacking interactions. Its catechol ring
is sandwiched between the phenanthroline unit and a
bipyridine ligand of the adjacent [Ru(bpy)2(H2-L3)]2+ com-
plexes (Figures 5 and Supporting Information S2). The
intermolecular distance between the centroid of the catechol
ring and the closest atom of the bipyridine ligand (C9) at
3.35 Å is very similar to the distance of 3.4 Å between the
centroid and the closest atom of the phenanthroline system
(C27) .
Upon excitation into the MLCT band at λmax ) 450 nm,
acidic solutions of 3(PF6)2 in air-equilibrated aqueous ac-
etonitrile at room temperature show an intense emission from
3
In summary, the crystal structures of 3 and 4 reveal that
the conformation of H2-L3 in the solid state does not change
significantly on moving from a [ReBr(CO)3(phen)]- to a [Ru-
(bpy)2(phen)]2+-type luminophore. The catechol and phenan-
throline rings are essentially coplanar in both structures and
allow π-overlap and electronic coupling between the metal-
binding units and the luminophores.
the corresponding MLCT state with a λmax of 610 nm
(Figure 7). The intensity of the emission decreases sigmoi-
dally with increasing pH. The inflection point is located at
pH 4.4. For similar ruthenium complexes, it was shown that
the inflection point of the emission titration (pHi) can be
-
related to the excited-state pKa* via eq 2, where τHS and τS
are the lifetimes of the protonated and deprotonated species,
Electronic Absorption and Emission Properties of [Ru-
(bpy)2(H2-L3)](PF6)2, 3(PF6)2. The electronic absorption and
emission properties of [Ru(bpy)2(Me2-L3)](PF6)2 and [Ru-
(bpy)2(H2-L3)](PF6)2 were investigated between pH 0.1 and
10. Due to the limited water solubility of the hexafluoro-
phosphate salts, a mixed solvent system consisting of
acetonitrile and water (20:1) was used. Adjustments to the
pH were carried out with a 0.6 M solution of HCl in this
(47) Li, M.-J.; Chu, W.-K.; Zhu, N.; Yam, V. W.-W. Inorg. Chem. 2007,
46, 720.
(48) Juris, A; Balzani, V.; Barigelletti, F.; Campagne, S.; Belser, P.; von
Zelewsky, A. Coord. Chem. ReV. 1988, 84, 85.
(49) (a) Garett, T. M.; McMurry, T. J.; Hosseini, M. W.; Reyes, Z. E.;
Hahn, F. E.; Raymond, K. N. J. Am. Chem. Soc. 1991, 113, 2965. (b)
Cohen, S. M.; Meyer, M.; Raymond, K. N. J. Am. Chem. Soc. 1998,
120, 6277.
(50) Raymond, K. N.; Mueller, G.; Matzanke, B. F. Top. Curr. Chem. 1984,
123, 49.
6522 Inorganic Chemistry, Vol. 46, No. 16, 2007

Two Catechol-Based Luminescent Chemosensors
Figure 6. Left: Absorption spectra recorded during the titration of an acidic solution (0.02 mM) of [Ru(bpy)2(H2-L3)](PF6)2 in aqueous acetonitrile with
an aqueous tetramethylammonium hydroxide solution. The baseline rises at pH 12 since the added quaternary ammonium salt makes acetonitrile and water
less miscible. Right: Plot of the absorbance at 365 nm (9) and 450 nm (2) as a function of pH.
Figure 7. Left: Emission spectra (uncorrected, λexc ) 450 nm) recorded during the titration of an acidic 0.02 mM solution of [Ru(bpy)2(H2-L3)](PF6)2 in
aqueous acetonitrile with an aqueous tetramethylammonium hydroxide solution. Right: Emission intensity at maximum (610 nm) versus pH for [Ru(bpy)2-
(H2-L3)](PF6)2 (gray circles) and [Ru(bpy)2(Me2-L3)](PF6)2 (black circles).
respectively.51 We have determined the lifetimes of the
protonated and deprotonated form of 3 from time-resolved
absorption and emission data and will report the results and
the pKa* of 3 in a follow-up paper. Since the inflection point
pHi, however, is similar to the ground-state pKa value of the
2-hydroxy-group, it can already be concluded here that the
intensity decrease is caused by the deprotonation of the
catechol unit. This assertion is further supported by the
observation that the emission intensity of the protected sensor
derivative [Ru(bpy)2(Me2-L3)](PF6)2 is pH-independent (Fig-
ure 7). In addition, emission quenching upon deprotonation
of appended phenolic OH groups was reported for similar
systems.52,53
reported for similar [ReCl(CO)3(diimine)] complexes.53-55
The absorption and emission spectra of [ReCl(CO)3(Me2-
L3)] (Supporting Information, Figure S4) and [ReBr(CO)3-
(Me2-L3)] are similar and independent of the pH of the
solution. In contrast, the titration of an acidic solution of 4
with base revealed a significant pH dependence (Figure 8),
indicating that the MLCT transition with λmax ) 395 nm
overlaps with pH-dependent ligand-based transitions. This
is not unusual; the Re(dπ) f phen(π*) charge-transfer
transitions in rhenium(I) tricarbonyl complexes are often
found very close to the low-energy side of intense ligand-
based π f π* absorption bands.24,52 In 4, the addition of
base leads to an increase in the absorbance at 390 nm, which
is again due to the deprotonation of the ortho-OH group of
the catecholamide unit. In this case, [ReBr(CO)3(H-L3)]- is
formed. A sigmoidal fit of the corresponding pH profile gives
a midpoint value of 5.75 ( 0.05, which is significantly higher
than the estimated midpoint value of 4.5 for [Ru(bpy)2(H2-
L3)]2+. This difference can be attributed to the fact that the
cationic RuII complex can stabilize the negative charge
resulting from the deprotonation more effectively than the
ReI-based sensor. Similar differences in pKa values have been
observed for 1,10-phenanthroline-5-carboxylic acid when
coordinated to related RuII- and ReI-containing chro-
mophores.27
pKa* ) pHi + log τHS/τS
(2)
-
The ∼20% drop in emission intensity upon deprotection
of [Ru(bpy)2(Me2-L3)]2+ can be attributed to interactions of
the free catechol unit with the solvent. Since the hydroxyl
groups form hydrogen bonds with acetonitrile and water, the
number of nonradiative decay pathways increases and the
emission intensity is reduced.
Electronic Absorption and Emission Properties of
[ReBr(CO)3(H2-L3)], 4. The spectroscopic properties of the
methyl-protected derivative [ReCl(CO)3(Me2-L3)] are sum-
marized in Table 4. They agree well with the properties
(51) Vos, J. G. Polyhedron 1992, 11, 2285.
(52) Burdinski, D.; Bothe, E.; Wieghardt, K. Inorg. Chem. 2000, 39, 105.
(53) Reece, S. Y.; Nocera, D. G. J. Am. Chem. Soc. 2005, 127, 9448.
(54) Kalyanasundaram, K. Photochemistry of Polypyridine and Porphyrin
Complexes; Academic Press: London, 1992; Chapter 10.
(55) Wrighton, M.; Morse, D. L. J. Am. Chem. Soc. 1974, 96, 998.
Inorganic Chemistry, Vol. 46, No. 16, 2007 6523

Batey et al.
Figure 8. Left: Absorption spectra recorded during the titration of an acidic solution (0.1 mM) of [ReBr(CO)3(H2-L3)] in aqueous acetonitrile with an
aqueous tetramethylammonium hydroxide solution. The baseline rises at high pH since the added quaternary ammonium salt makes acetonitrile and water
less miscible. Right: Plot of the absorbance at 390 nm as a function of pH.
Figure 9. Left: Selected emission spectra (uncorrected, λexc ) 390 nm) recorded during the titration of an acidic 0.1 mM solution of [ReBr(CO)3(H2-L3)]
in aqueous acetonitrile with an aqueous tetramethylammonium hydroxide solution. Right: Emission intensity at maximum (605 nm) versus pH for [ReBr-
(CO)3(H2-L3)] (gray circles) and [ReBr(CO)3(Me2-L3)] (black circles).
Acidic solutions of 4 in aqueous acetonitrile emit with a
λmax at 605 nm when excited at 390 nm. The emission
spectrum of the acidic solution resembles that of the methyl-
protected complex [ReBr(CO)3(Me2-L3)]; however, the
intensity of the emission of 4 decreases upon the addition
of base (Figure 9). At pH 10, only ca. 5% of the initial
emission intensity is observed. This decrease can again be
ascribed to the deprotonation of the ortho-OH group. The
midpoint of the pH-response was fitted to a value of 5.35 (
0.05. As expected, the emission intensity of the methyl-
protected sensor [ReBr(CO)3(Me2-L3)] remains constant
between pH 0.1 and 10.
Figure 10. Emission intensity at maximum (605 nm) versus pH for 4
Oxometalate Detection. pH-Profiles. The addition of
(gray circles), 4 + 0.5 equiv molybdate (black squares), 4 + 0.5 equiv
vanadate (black triangles), and [ReBr(CO)3(Me2-L3)] + 0.5 equiv molybdate
0.5 equiv of molybdate, tungstate, or vanadate to solutions
of 4 results in a decrease of the emission intensity in the
acidic pH-range (Figures 10 and 11). Consequently, 4 can
be used as a chemosensor for these species. The decrease in
emission intensity in the presence of the oxometalates can
be attributed to the deprotonation of the catechol units upon
metal-ion coordination. The observation that the emission
intensity of the methyl-protected derivative [ReBr(CO)3(Me2-
L3)] is not influenced by the presence of molybdate supports
this assertion and demonstrates that the decrease in emission
intensity is not due to intermolecular quenching processes.
Molybdate and tungstate as well as vanadate tend to
condense in concentrated acidic solutions to form polyoxo-
metalates. Since the 0.05 mM oxometalate concentrations
used in the study with 4 are comparatively high, we were
interested in investigating whether polyoxometalate formation
(black circles) (0.1 mM solutions in aqueous acetonitrile).
would be indicated by 4. For this purpose, the pH profiles
were recorded again after pre-acidification of the solutions
to pH 1. It emerged that the pH profile of 4 in the presence
of tungstate shows only a very slight deviation from the
profile of the free 4 upon titration of the pre-acidified solution
with base. The two combined pH profiles obtained in the
presence of tungstate thus displayed a pronounced hysteresis
loop (Figure 11). A subsequent investigation of the revers-
ibility of the titrations with molybdate and vanadate revealed
that these also show some degree of hysteresis (Supporting
Information, Figure S5). This behavior is consistent with
polyoxometalate formation in the pre-acidified solutions since
in the condensed species, the actual concentration of reactive
6524 Inorganic Chemistry, Vol. 46, No. 16, 2007

Two Catechol-Based Luminescent Chemosensors
pH, the emission intensity already starts to increase at ca.
pH 7. The reasons for this behavior, which is unusual but
reproducible, are still under investigation. Irrespective of the
direction of the titration, however, tungstate does not quench
the emission significantly below pH 3.5. Consequently, the
presence of tungstate does not interfere with the sensing of
molybdate or vanadate by 3.
Selectivity Studies. The selectivity of [Ru(bpy)2(H2-L3)]2+
and [ReBr(CO)3(H2-L3)] was investigated by repeating the
titrations in the presence of potentially interfering ions. The
pH profiles obtained in the presence of biologically relevant
metal cations, such as Mn(II), Fe(III), Co(II), Ni(II), and Zn-
(II) are very similar to the pH profiles of the free sensors
(Supporting Information, Figures S5 and S6). Since [Ru-
(bpy)2(H2-L3)]2+ and [ReBr(CO)3(H2-L3)] emit only in the
protonated state, cation binding has to take place in acidic
media for this interaction to be signaled by emission
quenching. The coordination of cations by catechols, how-
ever, liberates protons and is thus disfavored at low pH
(Scheme 1).
Figure 11. pH profile of [ReBr(CO)3(H2-L3)] (0.1 mM in aqueous
acetonitrile) in the presence of 0.5 equiv of tungstate obtained upon titration
from low-to-high pH (hollow diamonds) and high-to-low pH (black
diamonds) in comparison with the pH profile of [ReBr(CO)3(H2-L3)](gray
circles).
The emission intensities in the presence of Cu(II) at acidic
pH values are slightly lowered; however, this decrease is
unlikely to be due to the binding of Cu(II) to the catechola-
mide-based receptor units, since the overall shape of the pH
profiles is not changed significantly. Instead the drop can
be interpreted as intermolecular quenching based on the
observation that under similar conditions the MLCT excited
state of the parent complex [Ru(bpy)2(phen)](PF6)2 is quenched
to 88% in the presence of an equimolar amount of Cu(PF6)2.57
The presence of oxoanions, such as sulfate or phosphate,
does not significantly change the pH profiles from that
obtained for the free sensor (Supporting Information,
Figures S7 and S8). This lack of interference is especially
advantageous in biological applications, where high concen-
trations of these ions can be present. The sensor has also
been investigated in the presence of perrhenate, related
diagonally in the periodic table to molybdate, but no binding
is observed.
Composition of Complexes and Detection Limits. To
determine the composition of the Mo and V complexes
formed, solutions of [Ru(bpy)2(H2-L3)](PF6)2 and [ReBr-
(CO)3(H2-L3)] were titrated with aqueous solutions of
molybdate and vanadate (Figures 14 and 15). During the
titrations, the solutions were buffered at pH values where
the difference between the emission intensity of the free and
the bound sensors is most pronounced: pH 3 for [Ru(bpy)2-
(H2-L3)](PF6)2 and pH 4 for [ReBr(CO)3(H2-L3)]. The
optimum working pH of the latter is higher since the ortho-
OH group in [ReBr(CO)3(H2-L3)] is less acidic.
Figure 12. Emission intensity at maximum (610 nm) versus pH for 3
(gray circles), 3 + 0.5 equiv molybdate (black squares), 3 + 0.5 equiv
vanadate (black triangles), and [Ru(bpy)2(Me2-L3)]2+ + 0.5 equiv molybdate
(black circles) (0.02 mM solutions in aqueous acetonitrile).
metal ions that are able to bind to the sensor molecules is
reduced and less quenching is observed. The effect is most
pronounced for tungstate since polyoxotungstates are par-
ticularly inert.56 Pre-acidification of the analyte solutions to
pH 1 could therefore be used for the differentiation of
molybdate and tungstate.
Analogous titrations were performed with 3. As evident
from Figure 12, 3 also functions as a chemosensor for
molybdate and vanadate in slightly acidic solutions. In the
case of 3, the pre-acidification of the solutions to pH 1 did
not have a significant impact on the pH profiles obtained in
the presence of molybdate or vanadate (Supporting Informa-
tion, Figure S6). Since the 0.01 mM concentration at which
the reversibility study with [Ru(bpy)2(H2-L3)]2+ was
carried out is much lower than in the study with
[ReBr(CO)3(Me2-L3)], polyoxometalate species are less
likely to be prevalent and hysteresis is thus not observed.
In the presence of tungstate, the titration from acidic to
basic pH begins essentially as that for the reference complex
[Ru(bpy)2(H2-L3)]2+. Once the pH level of 5 is reached,
however, the decrease in emission intensity levels off and a
residual emission intensity of over 50% remains even at the
level of pH 12. No other changes to the emission spectra
are observed (Figure 13). Upon titration from basic to acidic
Upon the addition of molybdate, the emission intensity
of both sensors decreases almost linearly until a ratio of
sensor-to-molybdate of approximately 2:1 is reached. This
ratio is consistent with the predominant formation of cis-
dioxo-MoVI-dicatecholate complexes at these pH values.
Complexes of this composition are well-documented in the
(56) Tytko, K. H.; Glemser, O. AdV. Inorg. Chem. Radiochem. 1976, 19,
(57) Riklin, M.; Tran, D.; Bu, X.; Laverman, L. E.; Ford, P. C. J. Chem.
Soc., Dalton Trans. 2001, 1813.
239.
Inorganic Chemistry, Vol. 46, No. 16, 2007 6525

Batey et al.
Figure 13. Left: Emission spectra (uncorrected, λexc ) 450 nm) recorded during the titration of an acidic solution of [Ru(bpy)2(H2-L3)](PF6)2 (0.02 mM)
and sodium tungstate (0.01 mM) in aqueous acetonitrile with an aqueous tetramethylammonium hydroxide solution. Right: pH profile of [Ru(bpy)2(H2-
L3)](PF6)2 in the presence of 0.5 equiv of tungstate obtained upon titration from low-to-high pH (hollow diamonds) and high-to-low pH (black diamonds)
in comparison with the pH profile of [Ru(bpy)2(H2-L3)](PF6)2 (gray circles).
Table 5. Detection Limits for Molybdate
sensor
[Ru(bpy)2(H2-L3)]2+
,
[ReBr(CO)3(H2-L3)],
pH 3
pH 4
linear range/mol L-1
correlation coeff R2
slope m/mol L-1 a.u.-1
standard deviation
of the blank, sb/a.u.
detection limit
0-1.0 × 10-5
0-3.2 × 10-5
0.9982
0.9906
1.4 × 108
8.31
4.3 × 106
1.87
2-
/mol L-1 MoO4
1.8 × 10-7
1.3 × 10-6
2-
/µg L-1 MoO4
43
17
315
125
/µg L-1 Mo
Figure 14. Emission spectra of [Ru(bpy)2(H2-L3)](PF6)2 (0.02 mM) in
aqueous acetonitrile in the presence of increasing molybdate concentrations
(in aqueous acetonitrile at pH 3.0, buffer 2,6-lutidine, λexc ) 450 nm). The
Table 6. Detection Limits for Vanadate
sensor
[Ru(bpy)2(H2-L3)]2+ [ReBr(CO)3(H2-L3)],
2-
inset shows the emission intensity at 610 nm as a function of molar MoO4
,
fractions.
pH 3
pH 4
linear range/mol L-1
correlation coeff R2
slope m/mol L-1 a.u.-1
standard deviation
of the blank, sb/a.u.
detection limit
0-5.0 × 10-6
0-3.4 × 10-5
0.9977
0.9898
1.2 × 108
8.31
4.9 × 106
1.87
/mol L-1 VO4
2.0 × 10-7
1.1 × 10-6
3-
3-
/µg L-1 VO4
24
10
132
58
/µg L-1
V
tungstate. The titration curves obtained show similar trends
(Supporting Information, Figures S9 and S10).
The detection limits were determined from the linear part
of the titration curves according to eq 3,60 where sb is the
standard deviation of the blank and m is the calibration
sensitivity, which corresponds to the slope of the linear part
of the calibration curve. The standard deviation of the blank
signal was determined from a series of 10 measurements of
the emission intensity of the sensor solution in the absenceof
oxometalates. The results are summarized in Tables 5
and 6.
Figure 15. Emission spectra of [ReBr(CO)3(H2-L3)] (0.1 mM) in the
presence of increasing molybdate concentrations (in aqueous acetonitrile
at pH 3.0, buffer 2,6-lutidine, λexc ) 390 nm). The inset shows the emission
2-
intensity at 605 nm as a function of the molar MoO4 fractions.
literature.58,59 Molybdenum binding reduces the 3MLCT
emission of [ReBr(CO)3(H2-L3)] by ca. 95%, while the
emission of [Ru(bpy)2(H2-L3)](PF6)2 decreases by approxi-
mately 75%.
Analoguous titrations were carried out with 3 and 4 in
the presence of vanadate and also with 4 in the presence of
Detection limit ) 3sb/m
(3)
(59) Duhme, A.-K. J. Chem. Soc., Dalton Trans. 1997, 773.
(58) Griffith, W. P.; Nogueira, H. I. S.; Parkin, B. C.; Sheppard, R. N.;
White, A. J. P.; Williams, D. J. J. J. Chem. Soc., Dalton Trans. 1995,
1775.
(60) Skoog, D. A.; West, D. M.; Holler, F. J.; Crouch, S. R. Fundamentals
of Analytical Chemistry, 8th ed.; Thomson Brooks/Cole: Belmont,
2004; Chapters 6 and 8.
6526 Inorganic Chemistry, Vol. 46, No. 16, 2007

Two Catechol-Based Luminescent Chemosensors
Figure 16. Emission intensity of the indicated sensor (black), the sensor in the presence of a potentially competing ion (dark gray), and the sensor in the
presence of the competing ion and molybdate (light gray).
Figure 17. Emission intensity of the indicated sensor (black), the sensor in the presence of a potentially competing ion (dark gray), and the sensor in the
presence of the competing ion and vanadate (light gray).
As is evident from Tables 5 and 6, the Ru-based sensor is
generally more sensitive than the Re-based sensor. The lower
detection limits for molybdate and vanadate in the case of
[Ru(bpy)2(H2-L3)]2+ are a consequence of the steeper
gradients of the calibration curves, which are caused by the
much higher initial emission intensity of the Ru-based
luminophore. Although the Ru-based sensor shows a higher
residual emission intensity in the presence of molybdate and
vanadate than the Re-based sensor, its far stronger emission
in the absence of molybdate or vanadate results in a higher
calibration sensitivity and lower detection limits.
The fact that the colorimetric determination of molybdate
concentrations with catechol has a detection limit of 11 mg
L-1 61 shows that the attachment of the catechol receptor unit
to Ru- or Re-based luminophores has improved the sensitivity
of the method dramatically. The detection limits achieved
with [Ru(bpy)2(H2-L3)]2+ and [ReBr(CO)3(H2-L3)] reach
those reported for organic fluorescent reagents. Reported
detection limits range from 100 µg L-1 of Mo reported for
Alizarin red S62 or 10 µg L-1 of Mo at pH 3.0-3.7 for
bathophenanthrolinedisulfonate63 to 0.08 µg L-1 of Mo at
pH 9.2 for 2-hydroxy-1-naphthaldehydene-8-aminoquino-
line.64 For vanadate, detection limits range from 190 µg L-1
of V for the extractive colorimetric detection with 6-chloro-
3-hydroxy-7-methyl-2-(2-thienyl)-4H-chromen-4-one65 or the
use of 3,3′-dimethylnaphtidine in a stopped-flow injection
system66 to 20 µg L-1 for the direct spectrophotometric
determination of V between pH 4.0 and 5.5 with 1,5-
diphenylcarbohydrazide.67
Compared with these organic fluorophores, the Ru- and
Re-based luminophores in [Ru(bpy)2(H2-L3)]2+ and [ReBr-
(CO)3(H2-L3)] have longer luminescence lifetimes and larger
differences between the absorption and emission maxima.
These photophysical properties are of advantage in the
investigation of biological samples, since they can be used
to discriminate the signal against background fluorescence.
Competition Studies. To assess the practicability of [Ru-
(bpy)2(H2-L3)]2+ and [ReBr(CO)3(H2-L3)] as chemosensors,
their ability to detect molybdate in the presence of potentially
competing ions was investigated at pH levels of 3 and 4,
respectively. For this purpose, a stoichiometric quantity of
the competing ions was added to the sensor solutions prior
to the addition of molybdate. The emission intensity at 610
or 605 nm was then recorded 3 min after each addition.
Figure 16 shows the observed decrease of emission intensity
in comparison to the emission intensity of the free sensors.
(61) Seifter, S.; Novic, B. Anal. Chem. 1951, 23, 188.
(62) Blanco, C. C.; Campana, A. G.; Barrero, F. A.; Ceba, M. R. Anal.
Chim. Acta 1993, 283, 213.
(63) Pal, B. K.; Singh, K. A.; Dutta, K. Talanta 1992, 971.
(64) Jiang, C.; Wang, J.; He, F. Anal. Chim. Acta 2001, 439, 307.
(65) Agnihotri, N.; Dass, R.; Mehtra, J. R. Anal. Sci. 1999, 15, 1261.
(66) Palomeque, M. E.; Lista, A. G.; Band, B. S. F. Anal. Chim. Acta 1998,
366, 287.
(67) Ahmed, N. J.; Banoo, S. Talanta 1999, 48, 1085.
Inorganic Chemistry, Vol. 46, No. 16, 2007 6527

Batey et al.
While the addition of most ions causes only small
variations of the emission intensity of the sensors, the
addition of Cu(II) causes a drop to ca. 80% of the original
intensity in both cases. This drop is consistent with the effect
that was observed for copper in the corresponding pH
profiles.
significantly on moving from a [ReBr(CO)3(phen)]- to a [Ru-
(bpy)2(phen)]2+-type luminophore. Spectrophotometric ti-
trations, however, show that the pKa value of the 2-hydroxy-
group is significantly lower in [Ru(bpy)2(H2-L3)]2+. Conse-
quently, the optimum working pH for this sensor is lower
that that for [ReBr(CO)3(H2-L3)].
The subsequent addition of molybdate to the solutions that
already contain the competing ions results in the emission
intensity falling to ca. 40% of the original intensity level of
[Ru(bpy)2(H2-L3)]2+ and to ca. 5% of the intensity level of
[ReBr(CO)3(H2-L3)]. This decrease is in accordance with the
residual emission intensities expected from the corresponding
pH profiles in the presence of molybdate at pH levels 3 and
4, respectively. It can therefore be concluded that none of
the competing ions tested affects the molybdate-binding
ability of [Ru(bpy)2(H2-L3)]2+ and [ReBr(CO)3(H2-L3)] if
present in concentrations equal to those of molybdate.
However, Cu(II) and, in the case of [ReBr(CO)3(H2-L3)],
Fe(III) may give rise to false positives in the absence of
molybdate if the pH dependence is not further investigated.
The practicability of [Ru(bpy)2(H2-L3)]2+ and [ReBr(CO)3-
(H2-L3)] in the analysis of vanadate in the presence of
potentially competing ions was also investigated. As evident
from Figure 17, vanadate still quenches the emission of both
sensors in the presence of competing ions. While the
approximately 35% decrease in the emission intensity of
[ReBr(CO)3(H2-L3)] upon the addition of vanadate is con-
sistent with the residual intensity at pH 4 in the pH profile,
the emission quenching of [Ru(bpy)2(H2-L3)]2+ appears to
be less complete. From the pH profile, a decrease of ca. 60%
of the intensity of free [Ru(bpy)2(H2-L3)]2+ is expected. In
contrast, the intensity decrease observed in the presence of
the competing ions varies considerably and averages only
ca. 30%. This observation suggests that the solutions have
not reached equilibrium after 3 min, when the emission
intensity was recorded. Similar delays were not observed in
the competition study with [ReBr(CO)3(H2-L3)] since the
solutions were far more concentrated and the reaction rates
were higher. The conditions used in the competition study
with the Re-based sensor are thus more suitable for high-
throughput analyses.
[Ru(bpy)2(H2-L3)]2+ and [ReBr(CO)3(H2-L3)] both signal
the presence and concentration of molybdate and vanadate
through a decrease in emission intensity. In addition, [ReBr-
(CO)3(H2-L3)] also detects tungstate. The sensitivity for
molybdate achieved with [Ru(bpy)2(H2-L3)]2+ is almost an
order of magnitude higher than the one reached with [ReBr-
(CO)3(H2-L3)] and 3 orders of magnitude higher than the
one published for the colorimetric determination of molyb-
date with catechol. Linking the catechol-based receptor unit
to the metal-based luminophore brought the detection limits
for molybdate and vanadate down to a level that falls within
the range reported for highly fluorescent organic chro-
mophores. In contrast to most organic fluorescent reagents,
the [ReBr(CO)3(phen)]- and [Ru(bpy)2(phen)]2+-type lumi-
nophores have long luminescence lifetimes and emission
maxima at >600 nm. These properties could be used to
discriminate the signal against biological background fluo-
rescence.
Our approach allowed us to combine the photophysical
advantages of the luminophores with the excellent selectivity
of the catecholamide-based receptor unit for molybdate,
tungstate, and vanadate. Due to the profound influence of
the luminophore on the sensitivity, acid-base, and metal-
binding properties of the chemosensors, Ru and Re com-
plexes of this type have the potential to be tuned to suit
particular applications by the selection of appropriate pho-
tophysical properties and charge of the luminophore.
Acknowledgment. We thank Prof. R. N. Perutz and Dr.
N. Reddig for helpful discussions, the EPSRc National Mass
Spectrometry Service Centre in Swansea for the HR mass
spectra and the EPSRC [GR/R16464/01] for financial sup-
port.
Note Added after ASAP Publication. This article was
released ASAP on July 6, 2007 with errors in Figures 16
and 17 where the Re complex should not have shown a
positive charge. The correct version was posted on July 13,
2007.
Summary and Conclusions
Two new luminescent chemosensors for biologically
relevant oxometalates have been synthesized and character-
ized, and their spectroscopic properties have been studied.
Since the sensors contain the same catecholamide receptor
unit but different luminophores, the impact of the lumino-
phore on the sensor properties could be investigated. The
crystal structures of the sensors demonstrate that the con-
formation of H2-L3 in the solid state does not change
Supporting Information Available: X-ray crystallographic data
in CIF format and Figures S1-S11 are available free of charge
via the Internet at http://pubs.acs.org. In addition, CCDC files
648122 and 648123 contain the crystallographic data. These data
can be obtained free of charge from The Cambridge Crystal-
lographic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
IC700554N
6528 Inorganic Chemistry, Vol. 46, No. 16, 2007