Anal. Chem. 1998, 70, 5019-5023
Luminescent Mononuclear and Dinuclear Iridium(III) Cyclometalated Complexes Immobilized in a Polymeric Matrix as Solid-State Oxygen Sensors Gaetano Di Marco,† Maurizio Lanza,*,† Antonino Mamo,‡ Ivan Stefio,‡ Cinzia Di Pietro,§ Giuseppe Romeo,§ and Sebastiano Campagna*,§
Istituto di Tecniche Spettroscopiche (ITS) del CNR, via Sperone 31, 98166 Messina, Istituto Chimico, Facolta` di Ingegneria, Universita` di Catania, 95125 Catania, and Dipartimento di Chimica Inorganica, Chimica Analitica e Chimica Fisica, Universita` di Messina, via Sperone 31, 98166 Messina, Italy
Oxygen quenching of the luminescence of mononuclear and dinuclear Ir(III) cyclometalated complexes immobilized in the pPEGMA matrixes has been studied. Linear Stern-Volmer plots, even when experiments at different emission wavelengths have been performed, were evidenced. Despite the different luminescence lifetimes of the chromophores in the absence of quencher, similar Stern-Volmer slopes have been calculated. This behavior was tentatively interpreted by taking into account the size and charge of the chromophores. Increased sizes and lower charges seem to enhance the sensitivity of the systems. Such findings could be of interest for the design of new solid-state luminescent oxygen sensors with improved performance. Chemical sensors based on luminescence are widely investigated as molecular reporters.1 Luminescence is indeed one of the most sensitive analytical techniques; it is nondestructive, relatively inexpensive, and easy to use and represents a sustainable technology. Luminescent transition metal complexes based on polypyridine ligands are extensively studied in this regard, owing to their long-lived metal-to-ligand charge-transfer (MLCT) excited states which make them quite useful also for several other applications, such as solar energy conversion2 and information storage.3 Determination of molecular oxygen concentration has important ramifications in medicinal, industrial, and environmental chemistry.1a,b As a consequence, several classes of luminescencebased solid-state oxygen sensors have been developed in the past †
ITSsCNR. Universita` di Catania. § Universita` di Messina. (1) (a) Chemical, Biochemical, and Environmental Sensors: Liebman, R. A., Wlodarczyk, M. T., Eds.; The International Society for Optical Engeneering: Bellingham, WA, 1989; Vol. 1172. (b) Fiber Optical Chemical Sensors and Biosensors; Wolfbeis, O. S., Ed.; CRC Press: Boca Raton, FL, 1991; Vol. 1. (c) Czarnik, A. W. Acc. Chem. Res. 1994, 27, 302. (d) Probe Design and Chemical Sensing; Lakowicz, J. R., Ed.; Topics in Fluorescence Spectroscopy; Plenum: New York, 1994; Vol. 4. (e) Fabbrizzi, L.; Licchelli, M.; Pallavicini, P.; Perotti, A.; Taglietti, A.; Sacchi, D. Chem. Eur. J. 1996, 2, 75. (f) de Silva, A. P.; Gunaratne, H. Q. N.; Gunnlaugsson, T.; Huxley, A. J. M.; McCoy, C. P.; Rademacher, J. T.; Rice, T. E. Chem. Rev. 1997, 97, 1515. ‡
10.1021/ac980234p CCC: $15.00 Published on Web 10/24/1998
© 1998 American Chemical Society
few years.4-7 Most of them are based on Ru(II) complexes immobilized in polymeric matrixes. The polymer has to exhibit particular properties such as oxygen permeability and physical and chemical stability. In the last years, investigation has been extended to chromophores made of different metals such as Os(II),5e Ir(III),6d,7b and Pt(II).8 The driving force for this extension is the search for chromophores that (i) exhibit higher luminescence quantum yields and longer lifetimes and (ii) are available to be excited at very low energies. We recently reported a solid-state oxygen sensor based on the luminescent Ir(III) cyclometalated complex [Ir(ppy)2(dpt-NH2)](PF6) (4; ppy ) 2-phenylpyridine anion; dpt-NH2 ) 4-amino-3,5di-2-pyridyl-4H-1,2,4-triazole) immobilized in a polymerized poly(ethyleneglycol) ethyl ether methacrylate (pPEGMA).7b The system exhibited excellent properties as a quenchometric oxygen sensor, such as linear Stern-Volmer plot, photo- and thermal stability, and reproducibility. Here we report an extension of this (2) (a) Hagfelt, A.; Graetzel, M. Chem. Rev. 1995, 95, 49. (b) Bignozzi, C. A.; Schoonover, J. R.; Scandola, F. Prog. Inorg. Chem.. 1997, 44, 1. (c) Balzani, V.; Campagna, S.; Denti, G.; Juris, A.; Serroni, S.; Venturi, M. Acc. Chem. Res. 1998, 31, 26. (3) (a) Balzani, V.; Credi, A.; Scandola, F. In Transition Metals in Supramolecular Chemistry; Fabbrizzi, L., Poggi, A., Eds.; Kluwer: Dordrecht, The Netherlands, 1994; p 1. (b) Lehn, J.-M. Supramolecular Chemistry; VCH: Weilheim, 1995. (c) Collin, J.-P.; Gavin ˜a, P.; Heitz, V.; Sauvage, J.-P. Eur. J. Inorg. Chem. 1998, 1. (4) (a) Del Bianco, A.; Baldini, F.; Bacci, M.; Klimant, I.; Wolfbeis, O. S. Sens. Actuators 1993, B11, 347. (b) Preininger, C.; Klimant, I.; Wolfbeis, O. S. Anal. Chem. 1994, 66, 1841. (c) Klimant, I.; Wolfbeis, O. S. Anal. Chem. 1995, 67, 3160, and refs. therein. (5) (a) Bacon, J. R.; Demas, J. N. Anal. Chem. 1987, 59, 2780. (b) Demas, J. N.; De Graff, B. A. Anal. Chem. 1991, 63, 829. (c) Hartmann, P.; Leiner, M. J. P.; Lippitsch, M. E. Anal. Chem., 1995, 67, 88. (d) Xu, W.; Schmidt, R.; Whaley, M.; Demas, J. N.; De Graff, B. A.; Karikari, E. K.; Farmer B. L. Anal. Chem. 1995, 67, 3172. (e) Xu, W.; Kneas, K. A.; Demas, J. N.; De Graff, B. A. Anal. Chem. 1996, 68, 2605. (6) (a) Kaneko, M.; Hayakawa, S. J. Macromol. Sci.-Chem. 1988, A25, 1255. (b) Li, X. M.; Ruan, F. C.; Wong, K. Y. Analyst 1993, 118, 289. (c) Kavandi, J.; Callis, J.; Gouterman, M.; Khalil, G.; Wright, D.; Green, E.; Burns, D.; McLachlan, B. Rev. Sci. Instrum. 1990, 61, 3341. (d) Vander Donckt, E.; Camerman, B.; Hendrick, F.; Herne, R.; Vandeloise, R. Bull. Soc. Chim. Belg. 1994, 103, 207. (7) (a) Di Marco, G.; Lanza, M.; Campagna, S. Adv. Mater. 1995, 7, 468. (b) Di Marco, G.; Lanza, M.; Pieruccini, M.; Campagna, S. Adv. Mater. 1996, 8, 576. (8) Pang, Z.; Gu, X.; Yekta, A.; Masoumi, Z.; Foucher, D.; Coll, J.; Winnik, M. A.; Manners, I. Adv. Mater. 1996, 8, 768.
Analytical Chemistry, Vol. 70, No. 23, December 1, 1998 5019
Chart 1
study, in which three new Ir(III) complexes are used. The new complexes are [Ir(L)(L1)](PF6)2 (1; L ) 2,6-bis(7′-methyl-4′phenyl-2′-quinolyl)pyridine; L1 is a monoanion of L), [Ir(L1)2](PF6) (2), and [(ppy)Ir(dpt-cy-dpt)Ir(ppy)](PF6)2 (3). The structural formulas of the compounds are shown in Chart 1 (in this chart the metal-N and metal-C bond distances are not in scale and the tridimensional arrangement is not shown). EXPERIMENTAL SECTION Synthesis and characterization of the polymer used as the matrix have been reported in a former work.7b The complexes 1-4 were available from earlier studies in our laboratories.9 All the investigated systems were prepared by solubilizing pPEGMA in ethanol and adding the complex in a suitable ratio (see below). The solution was left under magnetic stirring at room temperature for 1 day. The crude mixture was inserted drop by drop into a stainless steel net. The solvent evaporation produces a film (∼0.1 mm) in which the concentration of the Ir chromophore in the solid solution is ∼5 × 10-4 M. The luminescence equipment and the apparatus for gas mixing have been reported previously.7b RESULTS All the complexes immobilized in the polymers shown intense luminescence spectra with maximums in the range 500-650 nm. Luminescence lifetimes were fitted by monoexponential decay (9) (a) Calogero, G.; Giuffrida, G.; Serroni, S.; Ricevuto, V.; Campagna, S. Inorg. Chem. 1995, 34, 541. (b) Mamo, A.; Stefio, I.; Parisi, M. F.; Credi, A.; Venturi, M.; Di Pietro, C.; Campagna, S. Inorg. Chem. 1997, 36, 5947. (c) Romeo, G.; Serroni, S.; Campagna, S., manuscript in preparation.
5020 Analytical Chemistry, Vol. 70, No. 23, December 1, 1998
Figure 1. Emission spectra of polymeric matrixes doped with 1 (full line), 2 (dashed line), and 3 (dotted line). The spectra shown are not corrected for photomultiplier response.
functions and were in the range 10-7-10-5 s. Luminescence intensities and lifetimes were significantly reduced on increasing oxygen concentration. Stern-Volmer fits of luminescence quenching were calculated at different emission wavelengths for the various complexes and were found to be linear in all the cases and independent from emission wavelengths. Figure 1 shows the luminescence spectra of 1-3; Figure 2 shows emission spectra of 3 on changing oxygen concentration; typical Stern-Volmer fits for the luminescence quenching by oxygen of 1-3 are displayed in Figure 3. Figure 4 shows response time, relative luminescence output, and reversibility of the complex/polymer systems on cycling from argon-saturated to oxygen-saturated atmosphere. Table 1 collects the luminescence lifetimes in argon- and oxygen-
Figure 2. Emission spectrum of 3-doped/pPEGMA at different oxygen exposures: argon-saturated (full line), air-equilibrated (dashed line), and oxygen-saturated (dotted line).
saturated atmosphere and quenching rates for all the four systems investigated. DISCUSSION The interest in the luminescence properties of metal-polypyridine complexes, as far as their use as quenchometric oxygen sensors is concerned, is related to the nature of the excited state. Actually, the luminescence of such complexes usually originates from triplet MLCT levels exhibiting long-lived lifetimes (typically in the microsecond time scale).10 The triplet-triplet energytransfer process (eq 1; in this equation, M stands for a generic
*M + O2 f M + *O2
(1)
M-containing chromophore) involving oxygen is therefore quite efficient. Stern-Volmer plots have been reported for oxygen quenching of several immobilized complexes, and nonlinearity of the plots has usually been found. Such a nonlinearity was attributed to a microheterogeneity of the binding sites.5 In contrast to the results of many research groups, we found in previous investigations7 linear Stern-Volmer plots and assumed that the nature of the immobilizing matrixes used allowed the metal chromophores to experience similar environments. We would like to point out, anyway, this does not mean the matrixes are perfectly homogeneous, because a narrow distribution of lifetimes (and therefore of different sites) could be not evidenced by our experimental methods. The luminescence spectra of the chromophore/matrix systems studied here (Figure 1, Table 1) are similar to the spectra exhibited by the complexes 1-4 in fluid solution at room temperature,9 where the emission has been attributed to MLCT levels. Furthermore, the luminescence lifetimes of the systems reported here have the same order of magnitude of the fluid solution luminescence lifetimes of 1-4 (compare lifetimes given in Table 1 with those of 1-4 in deoxygenated acetonitrile fluid solution at room temperature, which are9 (1) 0.33 µs; (2) 2.30 (10) (a) Meyer, T. J. Pure Appl. Chem. 1986, 58, 1193. (b) Juris, A.; Balzani, V.; Barigelletti, F.; Campagna, S.; Belser, P.; von Zelewsky, A. Coord. Chem. Rev. 1988, 84, 85. (c) Meyer, T.-J. Acc. Chem. Res. 1989, 22, 163.
µs; (3) 1.80 µs; and (4) 0.87 µs). The same MLCT attribution therefore also holds in the immobilized systems. However, the difference in the luminescence lifetime of each chromophore on passing from fluid solution to the polymeric matrix suggests the presence of polymer environmental effects. As one can see from Figures 2 and 3 and from Table 1, both luminescence intensities and lifetimes of the systems decrease in the presence of oxygen. The constancy of the shape of the emission profile (Figure 2) under different oxygen concentrations guarantees that no ground-state interaction between molecular oxygen and the chromophore occurs, so that the decreased luminescence can be safely attributed only to a quenching process. As already reported previously, even for the complexes studied here for the first time, linear plots were obtained. It was pointed out that an elegant way to determine the heterogeneity of the binding sites is the examination of the variation in quenching rate constant on changing emission wavelength.5d We therefore performed the experiments at different (at least three) emission wavelengths; in each case, quite similar rates were found (differences within 5%).
I0/I ) 1 + kSV[O2]
(2a)
kSV ) kqτ0
(2b)
On considering the Stern-Volmer (SV) equation (eq 2),11 and looking at the luminescence lifetime in absence of oxygen (Table 1), one would expect quite different slopes for the SV plots. Surprisingly, we find similar SV rates for 1-4, which lead to calculation of different values of kq for the various systems (Table 1). Because kq is expected12 to depend on the diffusion constant kd and on the rate constant of the energy transfer in the encounter complex ken (eq 3), it is at a first sight surprising to found different
kq ) kd ken
(3)
values for this constant. In fact, kd essentially depends on the permeability of oxygen within the bulk matrix (expected to be identical for all the systems) and ken should be constant in the series of the complexes because of the similar nature of the excited states involved and the adiabatic regime usually found for bimolecular energy transfer. However, according to Smoluchowski,13 kd depends (eq 4)14 on the van der Waals radii of the
kd ) 3/4πN′ArMQ3
(4)
chromophore and quencher species and the mutual diffusion coefficients. Oxygen radius and mutual diffusion coefficient can be considered constant within the series of systems here studied, so that only the radius of the chromophore can play a role. By (11) I0 and I are the luminescence intensities in the absence, and in the presence of quencher, τ0 is the lifetime in the absence of quencher and kq is the quenching constant. (12) Gilbert, A.; Baggott, J. Essentials of Molecular Photochemistry; Blackwell Scientific Publications: Oxford, U.K., 1991. (13) Smouluchowski, M. Z. Phys. Chem. 1917, 92, 129. (14) In eq 4, N′A is the Avogadro number divided by 103, and rMQ is the encounter distance between reactants, usually calculated as the sum of the van der Waals radii of the reactants.
Analytical Chemistry, Vol. 70, No. 23, December 1, 1998
5021
Figure 3. Stern-Volmer plots for oxygen quenching of the studied systems. I0 and I are the luminescence intensities in the absence and in the presence of oxygen, respectively, and pO2 stands for the partial pressure of the oxygen. Table 1. Lifetimes in Argon-Saturated (τ0) and Oxygen-Saturated Atmosphere (τq), Stern-Volmer Slopes (kSV), Quenching Rates (kq), and Quenching Rates Normalized to the Chromophore Size (k′q) for the Various Systems Investigated samples
τ0 (µs)
τq (µs)
KSV (atm-1)
kqa (atm-1 s-1)
k′q (atm-1 s-1)
1 2 3 4
2.10 1.40 1.13 1.30
0.76 0.87 0.39 0.52
1.47 1.56 1.60 1.60
0.70 1.11 1.42 1.23
0.70 1.11 0.71 1.23
a
Figure 4. Response time, relative luminescence changes, and reversibility of the polymeric matrixes doped with 1-3 complexes on cycling from argon-saturated to oxygen-saturated atmosphere.
simple considerations, similar sizes can be assumed for the mononuclear chromophores 1, 2, and 4, and an approximately double size can be assumed for the dinuclear chromophore 3. With such an approximation, a normalized (to the dimension) k′q is obtained (Table 1), which evidences as the systems 1 and 3 have a similar constant, whereas another constant is exhibited by 2 and 4. This analysis suggests that at least another factor should be held into account to justify the apparently different k′q for the two series of systems. In an attempt to rationalize our results, we propose that a factor to be considered is the charge of the chromophores, which for the systems 1 and 3 is +2, whereas for 2 and 4 it is +1. Apparently, an enhanced charge of the chromophore seems to reduce k′q. In light of the above discussion, we hypothize that the microenvironment of the chromophore could depend on its charge and size. This seems to be reasonable on considering that the polymeric chains can be differently 5022 Analytical Chemistry, Vol. 70, No. 23, December 1, 1998
kq ) kSV/τ0 × 103.
organized around the chromophore on changing charge and size. Actually, it is foreseen that chromophores with enhanced charge could interact more strongly with the polymer chains, with a consequent stronger packing of the chains around the chromophores. The chromophores could therefore experience different local shieldings and as a consequence, modification could occur in k′q. Such modifications could be attributed to the following: (i) a different ken, which could be due to restricted orientation/distance between the reactants within the encounter complex imposed by a different matrix conformation/package around the chromophores; (ii) a different apparent kd, which could depend on slow diffusion of molecular oxygen within the inner solvation spheres of the chromophores. There is no simple way to discriminate between (i) and (ii), which should have the same consequences for the quenching efficiency. Finally, we checked the time response, relative luminescence output, and reversibility of the new systems on cycling from argonsaturated to oxygen-saturated atmosphere (Figure 4). The experiments were repeated several times for each system within three months and no degradation of the properties of the sensors was noted. The different luminescence intensities of the various
systems qualitatively agree with the relative quantum yields in solution of the complexes.9 CONCLUSIONS We reported the oxygen sensor ability of a series of mononuclear and dinuclear Ir(III) cyclometalated complexes immobilized in the pPEGMA matrixes. From the slopes of the SV plots, such systems have properties comparable with the more used Ru(II) polypyridine complexes as far as the oxygen sensitivity (which is related to the SV slope) is concerned. This confirms that Ir(III) cyclometalated complexes are a valid alternative to the more extensively used Ru(II) complexes as the active component of luminescent solid-state oxygen sensors. Furthermore, our results seem to suggest that, beside the lifetime of the chro-
mophore and the permeability of matrixes, the size and the charge of the chromophore could also affect the sensitivity and the response of the sensor system. Such findings could be of interest for the design of new solid-state luminescent sensors with improved performance. We are engaged in a systematic study aimed to further check the role of charge and size of chromophores in this regard. ACKNOWLEDGMENT This work was supported by MURST and CNR. Received for review March 4, 1998. Accepted July 24, 1998. AC980234P
Analytical Chemistry, Vol. 70, No. 23, December 1, 1998
5023
