A Fluorescent Temperature Probe Based on the Association between

Spectroscopic studies of a fluorescent fluoresceinophane formed via a practical synthetic route. Sergio Andr?s P?rez Guar?n , Derek Tsang , W. G. Sken...
0 downloads 0 Views 58KB Size
Anal. Chem. 1998, 70, 3974-3977

Technical Notes

A Fluorescent Temperature Probe Based on the Association between the Excited States of 4-(N,N-Dimethylamino)benzonitrile and β-Cyclodextrin Iddys D. Figueroa,† Mohamed El Baraka,‡ Edwin Quin˜ones,* and Osvaldo Rosario

Department of Chemistry, University of Puerto Rico, P.O. Box 23346, San Juan, Puerto Rico 00931-3346 Michel Deumie´

LBPC, Universite´ de Perpignan, 52 Avenue de Villeneuve, 66860 Perpignan Cedex, France

A temperature probe based on the fluorescence properties of the two excited states of 4-(N,N-dimethylamino)benzonitrile (DMABN) in equilibrium with β-cyclodextrin (CD) in aqueous solution is presented. The fluorescence intensity of the Franck-Condon excited state (FB) as a function of temperature shows a straight line with a correlation better than 0.99 in the 283-308 K temperature interval. On the other hand, the fluorescence intensity of the twisted internal charge-transfer state (FA) remains constant in the same temperature interval because the binding of DMABN in the A* state to CD is isoenthalpic and entropy driven. It is found that the FA/ FB ratio is independent of the excitation intensity at a specified temperature, shows a linear relationship with temperature, and allows temperature measurements with a resolution of (2.5 K. There has been a long-standing interest in developing fluorescent molecular probes to study microheterogeneous solutions, interfaces, porous materials, and colloids.1-4 Using molecular probes, it is possible to perform measurements of micropolarity, microviscosity, concentrations of ions such as Ca2+ and Mg2+, and pH in such complex systems.5 The use of reactive probes has been particularly important to investigate biological systems.5 On

the other hand, the information that one can extract from microheterogeneous systems employing molecular probes may be limited by a number of physical properties. In particular, the molecular size of the probe,6 the magnitude of its dipole moment, and its capacity to interact with a given microenvironment (e.g., formation of hydrogen bonds) dictate the location of the probe with respect to the bulk solution. Other aspects to be considered in developing a molecular probe are its chemical reactivity and partitioning into different phases.7 The temperature is an important parameter to characterize a wide variety of systems (e.g., sprays, living cells), which is sometimes challenging to measure with the appropriate position resolution, and the development of fluorescence probes is a promising direction to address this need. Fluorescent probes based on the ratio of the fluorescence intensity of the excimer to the monomer bands of added aromatic organic molecules have been developed to measure the temperature of fuel sprays.8 The temperature of organic solutions as well as solvent relaxation may be determined from the shift of the maxima of fluorescence bands of organic as well as inorganic molecular probes.9-12 The electronic transitions of inorganic crystals such as dysprosium yttrium-aluminum-garnet (YAG) have been used to measure the temperature of reactive surfaces.13 The development of a fiber-

Present address: Department of Chemistry, Texas A & M University, College Station, TX 77843-3255. ‡ Permanent address: LBPC, Universite ´ de Perpignan, 52 Avenue de Villeneuve, 66860 Perpignan Cedex, France. (1) Kalyanasundaram, K. Photochemistry in Microheterogeneous Systems; Academic Press: New York, 1987. (2) Lakowicz, J. R. Principle of Fluorescence Spectroscopy; Plenum Press: New York, 1983. (3) (a) Kalyanasundaram, K.; Thomas, J. K. J. Am. Chem. Soc. 1977, 99, 20392044. (b) Fendler, J. H. Membrane Mimetic Chemistry; Wiley-Interscience: New York, 1982. (4) Klafter, J.; Drake, J. M. Molecular Dynamics in Restricted Geometries; John Wiley: New York, 1989. (5) Haugland, R. P. Handbook of Fluorescent Probes and Research Chemicals; Molecular Probes: Eugene, OR, 1992.

(6) Cox, G. S.; Hauptman, P. J.; Turro, N. J. Photochem. Photobiol. 1984, 39, 597-601. Chandar, P.; Somasundaran, P.; Turro, N. J. J. Colloids Interface Sci. 1987, 117, 31-46. (7) El Baraka, M.; Deumie´, M.; Viallet, P.; Lampidis, T. J. J. Photochem. Photobiol. 1991, 62, 195-216. (8) Murray, A. M.; Melton, L. A. Appl. Opt. 1985, 24, 2783-2787. (9) Seilmeier, A.; Scherer, P. O. J.; Kaiser, W. Chem. Phys. Lett. 1984, 105, 140-146. (10) Castner, E. W.; Maroncelli, M.; Fleming, G. R. J. Chem. Phys. 1987, 86, 1090-1097. (11) Schrum, K. F.; Williams, A. M.; Haerther, S. A.; Ben-Amotz, D. Anal. Chem. 1994, 66, 2788-2790 and references therein. (12) Kitamura. N.; Kim, H. B.; Kawanishi, Y.; Obata, R.; Tazuke, S. J. Phys. Chem. 1986, 90, 1488-1491. (13) Goss, L. P.; Smith, A. A.; Post, M. E. Rev. Sci. Instrum. 1989, 60, 37023706.

3974 Analytical Chemistry, Vol. 70, No. 18, September 15, 1998

S0003-2700(98)00300-X CCC: $15.00



© 1998 American Chemical Society Published on Web 08/08/1998

optic temperature sensor based on the temperature dependence of the pKa of tris(hydroxymethyl)aminometane was reported.14 In this work, we report a fluorescent molecular probe to measure the temperature of aqueous media. To this end, we use the photophysical equilibria between two excited states of DMABN in the presence of CD. It is found that, at a fixed concentration of DMABN and CD, the fluorescence intensity emanating from the B* state (FB) decreases upon increasing the temperature in the 283-308 K temperature range, while the intensity of the fluorescence band of the A* state (FA) remains invariant. It is demonstrated that the FA/FB ratio exhibits two important properties: (a) it is independent of the excitation intensity at a fixed temperature and, therefore, insensitive to instrumental fluctuations, and (b) it exhibits a perfect linear relationship with respect to the temperature of the solution. EXPERIMENTAL SECTION Apparatus. UV absorption spectra were determined with a Cary 3 UV-visible spectrophotometer (Varian). The corrected emission spectra were taken with a SPF-4800 (SLM Aminco) spectrofluorometer. Slit widths of 4 nm were used in both the emission and absorption monochromators. The samples were excited at 295 nm. To keep constant the temperature of the solutions, a water circulator (Brinkman, model RM6) was adapted to the spectrofluorometer. This device was also used in temperature variation experiments. Since the cavity of the fluorometer can hold two fluorescence cells, the temperature of the system was measured directly from the sample not excited before and after collecting each fluorescence spectrum. The excitation intensity was varied using constant optical density filters. Those filters were placed outside the fluorometer sample cavity to avoid light scattering. Reagents. β-Cyclodextrin (Aldrich) was used as received, and absorption and fluorescence spectra were recorded in water to confirm its purity. 4-(N,N-Dimethylamino)benzonitrile (Aldrich) was recrystallized from a 75% ethanol/water solution. Fresh solutions were prepared using deionized water. RESULTS The present work is concerned with the development of a fluorescent temperature probe based on the association of two excited states of DMABN to CD:

DMABN + CD T DMABN-CD

(1)

where DMABN-CD stands for an association complex, which could be either an inclusion complex or a surface complex. The maximum (299 nm) of the absorption spectrum of DMABN neither shifts as function of concentration of CD nor overlaps with the maximum of the first fluorescence band of DMABN. Therefore, no inner filter effect could affect the fluorescence intensity measurements, regardless of whether CD has been added. In aqueous media, the DMABN molecule emits from two energetically and physically different electronic excited states, namely the Franck-Condon excited state (B*) and the twisted

(14) Straub, A. E.; Seitz, R. W. Anal. Chem. 1993, 65, 1491-1492.

Figure 1. Fluorescence spectra of an aqueous solution of 10 µM DMABN (A), 1 mM CD, and (B) 3 mM CD at the following temperatures: (a) 283, (b) 288, (c) 293, (d), (e) 303, and (f) 308 K.

internal charge-transfer state (A*).15-17 A photophysical description of the present system can be found elsewhere.18 The properties of the DMABN-CD system that make it a candidate for development of a temperature probe are given below. First, the fluorescence intensity from the B* state, with maximum at 350 nm, decreases with temperature, whereas that arising from the A* state, with maximum around 530 nm, is insensitive to temperature changes, as shown in Figure 1. For the experiments presented in Figure 1, the DMABN concentration was kept constant, and two concentrations of CD were examined, 1 and 3 mM. The relative fluorescence intensities of the A* and B* bands are different for the two concentrations studied at a given temperature because they represent different positions of the equilibrium. El Baraka et al. reported that (a) the DMABN-CD equilibrium involving the B* state is exothermic while the association of the A* state with the CD is isoenthalpic, and (b) the emission yield of both excited states is larger for the fraction of molecules associated with the CD.18 That is to say, the B*state fluorescence intensity decreases with temperature because the equilibrium shifts to the left (a larger fraction of DMABN is transferred to the aqueous phase where the emission yield is lower). In contrast, the A*-state fluorescence intensity remains (15) Rotkiewicz, K.; Grellmann, K. H.; Grabowski, Z. R. Chem. Phys. Lett. 1973, 19, 315-318. (16) Lippert, E.; Rettig, W.; Bonacic-Koutecky, V.; Heisel, F.; Miehe, J. A. Advances in Chemical Physics; Prigogine, I., Rice, S. A., Eds.; Wiley: New York, 1987; Vol. 68, pp 1-173. (17) Rettig, W. Angew. Chem., Int. Ed. Engl. 1986, 25, 971-988. (18) El Baraka, M.; Garcı´a, R.; Quin ˜ones, E. J. Photochem. Photobiol. 1994, 79, 181-187.

Analytical Chemistry, Vol. 70, No. 18, September 15, 1998

3975

Figure 2. Graph of FA/FB versus temperature for different CD concentrations: 1 (0), 3 (O), and 5 mM (4). The concentration of DMABN was 20 µM.

constant with temperature because the complexation of this state is isoenthalpic.18 In the experiments previously described, the excitation intensity was kept constant. Second, a plot of the ratio of the A*- and B*-state intensities, FA/FB, against temperature exhibits a perfectly linear behavior in the 1-5 mM CD concentration range, and this is displayed in Figure 2 for three CD concentrations. The magnitude of the slope that corresponds to the 1 mM CD solution is largest and offers the greatest sensitivity for measuring temperature among the three curves presented. Because the absolute fluorescence intensity of the spectra decreases when the CD concentration is reduced, it is not practical to work at CD concentrations below 1 mM. Accordingly, in the present work, we studied the effect of the DMABN concentration keeping the CD concentration fixed at 1 mM. We carried out experiments changing the DMABN concentration in the 5.2-25 µM concentration range. Since not much change in the slope was observed as a function of DMABN concentration, we decided to work at 25 µM DMABN (the absorbance is approximately 0.5) because the solutions exhibit the strongest fluorescence signal. At these concentration conditions, there is no overlap between the absorption and the fluorescence spectra at 350 nm. Third, the equilibrium represented by eq 1 is reversible in the temperature range studied. This was confirmed by collecting spectra, increasing the temperature of the solutions, and then cooling the samples to the original temperature and collecting the spectra again. Fourth, since the fluorescence intensity emanating from the A* state remains constant with temperature, its magnitude is directly proportional to the excitation intensity, Io, and one can write FA ) KIo, where K is a constant independent of temperature. The FA/FB ratio was examined against the excitation intensity at 288, 298, and 308 K (data not shown) using constant optical density filters to attenuate the fluorometer excitation beam up to 10% of its original fluence. Taking into account the uncertainty in the measurements, it was found that the FA/FB ratio is constant with respect to the excitation intensity. 3976 Analytical Chemistry, Vol. 70, No. 18, September 15, 1998

Figure 3. Graph of FA/FB versus temperature changing the temperature in intervals of 1 K to illustrate that the present system allows temperature measurements with a precision better than (2.5 K. A solution 25 µM DMABN and 1 mM CD was examined.

Fifth, we determined experimentally that the fluorescence yield of DMABN in water (Φw) is temperature independent. Figure 3 displays the linear relationship that exists between FA/FB and the temperature in the 282-294 K range. In these experiments, the concentrations of DMABN and CD were 25 µM and 1 mM, respectively. Figure 3 also shows that we can measure the temperature of this solution with a precision of (2.5 K. We carried out temperature variation experiments such as the one described in Figure 3 at pH 2.0, 6.2, and 8.1, and the FA/FB versus temperature curves were also linear with correlations larger than 0.99, although the sensitivity at the different pH’s varies slightly. Even though fluorescent molecules that are not sensitive to pH, such as pyrene, are preferred for this application, their fluorescence properties are not very sensitive to temperature changes.19 DISCUSSION The optical excitation of DMABN in aqueous media produces two physically different electronic states. The B* state is neutral, while the A* state is a charge-transfer state.15-17 It is clear from Figure 1 that the two association equilibria are not coupled, since the reaction involving the B* state can be shifted by changing the temperature, while this is not the case for the reaction involving the A* state. The fluorescence intensity emanating from the B* state in a solution containing CD may be written as

FB ) bIo(Φw[B]w + Φc[B]c)

(2)

where  and b stand for the optical path and the molar absorptivity of DMABN, respectively. The symbols Φw (Φc) and [B]w ([B]c) denote the fluorescence quantum yield of the B* state in water (associated to CD) and the molar concentration of molecules in the B* state in the aqueous phase (associated to CD), respectively. Dividing FA ) KIo by eq 2, one finds that the FA/FB ratio is independent of the excitation intensity, (19) Waris, R.; Acree, W. E.; Street, K. W. Analyst 1988, 113, 1465-1467.

γ FA ) FB (Φw[B]w + Φc[B]c)

(3)

where γ is a constant. The FA/FB ratio can be related to the association constant of the B* state corresponding to eq 1,

Keq )

[B]c [B]w[CD]o

(4)

where, in the present case, the following condition holds: [CD] ) [CD]o - [B]c ≈ [CD]o. Substituting eq 4 into eq 3 gives

FA γ ) FB [B]w(Φw + ΦcK [CD] ) eq o

(5)

El Baraka et al. found that the magnitude of the equilibrium constant for the complexation of the B* state decreases from 1050 to 50 M-1 upon increasing the temperature from 296 to 303 K.18 While Φw is constant with respect to temperature changes, it is not possible to determine the temperature dependence of Φc because the equilibrium shifts with temperature. From a linear least-squares analysis of the data displayed in Figure 3, we obtained the intercept and the slope, b ) -22.45 and m ) 0.809, respectively, with a correlation coefficient of 0.996.

The lack of selectivity of CD toward binding different substrates restricts the possible applications of the present system. In fact, Warner and co-workers and Hamai have shown that the addition of alcohols affects CD/pyrene equilibria because the CD/ pyrene/alcohol ternary inclusion complex is more stable.20,21 The possible binding of DMABN to metal ions needs to be experimentally assessed for each particular application. For practical applications of the present system, one must construct calibration plots of the FA/FB ratio as a function of temperature at the CD and DMABN concentrations specified in Figure 3, exciting around 295 nm and using band path filters at 350 and 530 nm. Specific applications include the construction of fiber-optic temperature sensors or an instrument to measure the temperature of aqueous sprays. ACKNOWLEDGMENT Financial support from Fondos Institucionales para la Investigacio´n de la Universidad de Puerto Rico, and the DOE-EPSCoR Program, is gratefully acknowledged. I.D.F. was supported by the NIH-SUBE Program (Grant No. S06RR08102-17) and M.E.B. by the Industry University Research Center (Grant No. 624035). Received for review March 16, 1998. Accepted July 7, 1998. AC980300N (20) Nelson, G.; Patonay, G.; Warner, I. M. Anal. Chem. 1988, 60, 274-279. (21) Hamai, S. J. Phys. Chem. 1989, 93, 2074-2078. Tee, O. S.; Gadosy, T. A.; Giorgi, J. B. J. Chem. Soc., Perkin Trans. 2 1993, 1705-1706.

Analytical Chemistry, Vol. 70, No. 18, September 15, 1998

3977