Investigation of the Fluorescence Quenching of 1 ... - ACS Publications

Article pubs.acs.org/JPCA

Investigation of the Fluorescence Quenching of 1‑Aminoanthracene by Dissolved Oxygen in Cyclohexane Todd Pagano,*,† Nelsy Carcamo,† and Jonathan E. Kenny‡ †

Department of Science & Mathematics/Laboratory Science Technology program, Rochester Institute of Technology/National Technical Institute for the Deaf, Rochester, New York 14623, United States ‡ Department of Chemistry, Tufts University, Medford, Massachusetts 02155, United States

ABSTRACT: This study provides a detailed investigation of the fluorescence quenching mechanisms of the fluorophore, 1aminoanthracene, by dissolved oxygen in cyclohexane. Dynamic/collisional quenching dominates in the system studied, but there is also a small component of static quenching. Stern−Volmer plots revealed that the dynamic quenching constant is 0.445 ± 0.014 mM−1 and represents ∼95% of total quenching in the system. The static quenching rate constant is 0.024 ± 0.001 mM−1, and mechanisms by complex formation and “sphere of action” static quenching were examined. Compensation of steady-state fluorescence data for solvent loss during the gradual deoxygenation period of the experiment was found to be important in order to conduct a thorough evaluation of the different quenching mechanisms of the system. The enhancement factors, (Fo/F) and (τo/τ), for 1-aminoanthracene were determined to be 2.20 ± 0.01 and 2.08 ± 0.01, respectively, and the diffusion-controlled bimolecular rate constant was found to be 2.1 × 1010 ± 0.2 × 1010 M−1 s−1. The work involved the development of a novel instrumental setup that simultaneously measures several important spectroscopic parameters (steady-state fluorescence intensity, absorbance, fluorescence lifetime, and dissolved oxygen concentration) for the careful study of oxygen quenching mechanisms of 1-aminoanthracene in a cyclohexane solution.

■

INTRODUCTION Despite its advantageous versatility and sensitivity, the quantitative analysis of fluorescence spectra can be hindered by several photophysical phenomena. These phenomena can include inner filter effects, dynamic and static quenching, and resonance energy transfer. Though it is difficult to avoid these phenomena in fluorescence measurements, it is imperative to investigate and characterize the factors that cause deviations from spectra that would be expected of samples that are otherwise free from their effects. Often, experimental procedures can be modified and/or data can be manipulated to compensate for these effects. Because these photophysical phenomena are dependent upon the local environment of the fluorophore, the mechanisms can also be taken advantage of and used as tools to determine characteristics of the solution environment, provide information about physical characteristics of the sample, or monitor species in heterogeneous samples. Optical oxygen sensors based on the quenching of films1 (or even polyaromatic hydrocarbons, PAHs)2 and the detection of biomolecules using probes based on resonance energy transfer3 © 2014 American Chemical Society

are examples of analytical methods based on measurements involving these photophysical phenomena. Fluorescence quenching can be divided into dynamic and static modes. Dynamic quenching is the nonradiative deexcitation of a fluorophore through collisions with a quencher (often, molecular oxygen) during the fluorophore’s excitedstate lifetime. Fluorescence quantum yields and lifetimes decrease with increasing quencher concentration, and dynamic oxygen quenching in solutions has long been shown to be a diffusion-controlled process.4 Since dynamic quenching occurs when a fluorophore is in an excited state, the lifetime of the fluorophore plays a large role in the extent of dynamic quenching. Static quenching by complex formation occurs when a complex is formed as a result of interactions between a ground-state fluorophore and a quencher. The formed complex can be nonfluorescent and, in addition to an observed decrease Received: September 19, 2014 Revised: November 8, 2014 Published: November 26, 2014 11512

dx.doi.org/10.1021/jp5094806 | J. Phys. Chem. A 2014, 118, 11512−11520

The Journal of Physical Chemistry A

Article

Figure 1. Fiber-optic based instrumental setup for the simultaneous measurements of steady-state fluorescence, fluorescence lifetime, absorbance, and oxygen concentration. Io = incident light, IT = transmitted light (for absorbance), IF = fluorescence, (ref) = reference channel, (O2) = dissolved oxygen measurements, and (τ) = lifetime measurements.

in fluorescence intensity upon complexation, sometimes the new complex can be revealed in changes to the absorption spectrum of the sample. Perrin also proposed a “sphere of action” static quenching mechanism where, if a quencher is within a particular radius of a fluorophore at the instant that it becomes excited, quenching can occur.5−9 In order to avoid problems associated with the quantitative analysis of fluorescence data in the presence of dynamic quenching by oxygen, a method of purging solutions with an inert gas is often applied. Purging methods remove dissolved oxygen from solutions by lowering the partial pressure of oxygen above the solution in accordance with Henry’s Law. Pagano et al. of this laboratory provided useful guidelines and experimental conditions for solution deoxygenation by purging in fluorescence studies.10 Brownrigg and Kenny, also of this laboratory, reported potential evidence for the static quenching of fluorescence, in addition to strong expected dynamic quenching by dissolved oxygen, in solutions of two out of four studied PAHs in cyclohexane.11 Specifically, they showed that the Stern−Volmer plots of fluorescence intensity (Fo/F) versus oxygen concentration have slightly higher linear slopes than corresponding lifetime plots (τo/τ).11 In the study, direct oxygen concentration measurements were not made, but instead concentrations were extrapolated based on oxygen-cyclohexane contact charge-transfer bands in the deep UV portion of literature12 absorption spectra at various partial pressures of oxygen in equilibrium with cyclohexane. Also, due to instrumentation limitations, they were unable to simultaneously measure fluorescence intensity and lifetimes, potentially adding some uncertainty to reported results. They conducted an analysis of dynamic quenching in the studied systems and concluded the potential presence of a much smaller static quenching component.11 In the present study, a fluorophore solution is gradually deoxygenated while using a novel instrumental setup that simultaneously measures relevant spectroscopic data, including dissolved oxygen and solution

concentration changes due to solvent loss, so that the quenching mechanisms, including the small static quenching component, can be thoroughly investigated. Development of a Novel Instrument System. A novel fiber optic-based system that allows for the simultaneous collection of steady-state fluorescence intensity, absorbance, fluorescence lifetimes, oxygen concentration, and ancillary measurements of atmospheric pressure and temperature was designed, developed, and utilized for this study. This instrumental setup was needed in order to directly relate steady-state fluorescence and fluorescence lifetime measurements to solution oxygen concentrations in real-time as a solution is gradually deoxygenated (oxygen displacement is by nitrogen diffusion into the measured solution). Since the validity of Stern−Volmer quenching investigations based on both steady-state fluorescence and fluorescence lifetime measurements is dependent on the assumption that they are made on solutions with accurate oxygen concentrations, the simultaneous measurements allowed by this instrumental setup are vital. Also, since solvent loss is expected over the long experiment times (∼6.5 h), absorbance measurements were crucial for compensating for concentration changes of the solution. In fact, some quenching studies that require long deoxygenation times or involve vigorous bubbling with an inert gas may overestimate a quenching effect derived from the steady-state data due to concentration increases of the studied system resulting from solvent loss. With the simultaneously collected data and compensation for solvent loss, uncertainties in the data are relatively small and can be carefully examined for the different quenching mechanisms. Absorbance measurements were also used to correct fluorescence intensities for inner filter effects that impact the data over the course of the experiment and as solute concentrations slowly increase. The system allows the spectroscopic parameters to be collected through a temperature regulated cuvette holder hub by fiber optic-fitted detectors. A diagram of the instrument is shown in Figure 1. The absorbance at the excitation wavelength 11513



Article

inside the glovebag). A pressure transducer was introduced into the glovebag environment so that, during the length of experiments, very slight and continual adjustments could be made to the flow of nitrogen into the bag in order to keep the pressure consistent. Throughout the course of a trial, the pressure inside the bag was maintained at 1.002 ± 0.001 atm; at all times this was slightly higher than the ambient pressure of the laboratory. Instrumental Set-Up. The central Peltier controlled temperature hub (Qpod, Ocean Optics) holds standard 10 mm cuvettes (at a z-dimension of 8.5 mm) and has four perpendicular optical ports for SMA fiber optic connections. All internal optics consist of high-quality fused silica. The hub allows for optical slits to be manually inserted after the fiber optic connection. For this study, 2 mm slit widths were used at each of the four optical ports of the central hub. Collimating lenses were installed, after the fiber optic connection, but before the inserted slit. Small stir bars (6 mm × 1.5 mm, L × Dia., Starna Cells) were placed into the cuvette, and solutions were stirred at a low setting for the duration of each trial. The temperature of the cuvette environment was maintained at 20.00 ± 0.05 °C via the temperature control and a water bath. Lifetime measurements were taken with a portable filterbased phase shift fluorimeter (Tau Theta, Boulder, CO- MFPF1 M model). The LED modulated light source of 405 nm was operated at 1 MHz. Detection was performed by an on-board avalanche photodiode. This LED was used as the light source for absorbance measurements at the excitation wavelength, fluorescence intensity measurements, and lifetime measurements. One arm of a bifurcated fiber brought the light from the lifetime LED to the cuvette hub, while the second arm went to the reference detector. A fiber optic cable at 90° to the incident light brought the signal to the lifetime detector after going through a long-pass filter (Semrock, Rochester, NY) to monitor the fluorescence signal above 490 nm. The lifetime instrument was calibrated against Rhodamine B’s known lifetime (1.7 ns).13,14 Lifetime measurements were averaged and collected every 5 s throughout the duration of the experiment for each trial. Fluorescence measurements were conducted at a 90° angle to the incident light with a USB-4000-FL (Ocean Optics) miniature CCD array detector via a fiber optic coupling. The detector was outfitted with a Czerny−Turner monochromator having a grating with a groove density of 600 lines/mm, a 500 nm blaze wavelength, an aperture entrance slit of 200 μm, and an optical resolution of ∼7 nm fwhm. Again, the excitation light was supplied by the LED housed in the lifetime instrument. The fluorescence detector was operated at an integration time of 100 ms, and 25 scans were averaged for each measurement. Fluorescence intensity was monitored at 474 nm for the duration of each trial. Fluorescence data were later processed for concentration change effects and corrected for inner filtering. A similar miniature CCD array detector (USB-4000 UV−vis, Ocean Optics) to that used for fluorescence measurements was used for absorbance monitoring. The installed Czerny−Turner monochromator had an aperture entrance slit of 25 μm and the same grating as above, but with a 300 nm blaze wavelength and an optical resolution of ∼2 nm fwhm. The absorbance was measured directly through the central hub/cuvette via fiber optic coupling. Again, the excitation light was supplied by the LED in the lifetime instrument, so absorbance was collected only at the wavelength of excitation (405 nm). The detector was operated at an

(IT) is monitored through the cuvette and holder. Simultaneously, fluorescence measurements (IF) are collected at a 90° angle to the incident light (Io), while at the other 90° angle to the incident light, a fiber optic-based phase shift fluorescence lifetime instrument collects lifetime data (IF(τ)). A fluorescence lifetime-based oxygen sensor with an oxygen-sensitive sensing probe is used to make direct oxygen measurements (Io(O2) and IF(O2)). The setup allows for the use of a variety of lenses, filters, and optical slits as well as a reference (Io(ref)) detector to monitor the power output of the light source and correct any resultant fluctuations in the data of any of the other detectors. All measurements are taken during the process of deoxygenation to obtain important Stern−Volmer and kinetic information.

■

EXPERIMENTAL SECTION Materials and Reagents. A solution was made to 7.8 × 10−5 M of 1-aminoanthracene (TCI America, Portland OR, 98% purity) [CAS# 610-49-1] by dissolving in HPLC grade cyclohexane (JT Baker, 99.995% purity). The solution was kept in a 100 mL amber glass bottle with a Teflon-lined screw cap and stored at 4 °C. The solution was used within 2 months of its preparation date. A standard 10 mm path length far-UV quartz cuvette (Starna Cells, type 3-Q-GL14-S) with a screw cap containing a septum (12.5 mm Thermolite, Restek) was used. Varying lengths of solarization-resistant silica fiber optics (Ocean Optics) with a 600 μm core diameter and SMA connections were used for light/signal transmission, according to the instrument diagramed in Figure 1. Ultrahigh purity compressed nitrogen (Airgas, Inc.) was passed through a regulator into a pyramidal-shaped glovebag through an inlet valve. The plastic glovebag (Erlab, Inc. Rowley MA) was the Captair Pyramid model of dimensions: 860 × 560 × 735 mm (L × W × H). It had two PVC medical gloves inserted for working inside the sealed bag. A seal was made via a zipper along the height of the pyramidal bag. At first, compressed nitrogen was passed through the bag, while the seal was left slightly open to ensure that all air was forced from the environment prior to sealing the bag. After filling with nitrogen, the 1-aminoanthracene solution, in the capped (and tightly sealed) cuvette, was placed in the central hub of the instrumental setup, and the glovebag was sealed. The oxygen sensing probe was previously pierced through the cuvette’s septum so that the cap could be unscrewed and removed from the cuvette vicinity at the start of each trial, while the probe remained in solution. All other optical measurements were collected through the fiber optic connections of the central hub. While capped and under the nitrogen atmosphere, baseline readings were collected using all of the connected detectors/ instruments. The cap was quickly unscrewed and dissolved oxygen began to be displaced as nitrogen diffused into the solution. This was marked as the start of each trial. An experimental hurdle involved logistics associated with the glovebag and related internal atmospheric pressure issues. The glovebag holds an atmosphere of nitrogen reasonably well but is structurally limited by not being rigid and all of the power cords, USB computer connections, and water bath tubing entering the bag produce a slightly less-than-sealed environment. While the entry point for the cords was sealed with water-resistant foam sealant, the bag never maintained a perfect seal. As such, early experimental attempts showed sporadic data that appeared to be pressure-related (and largely a function of pressure changes impacting the multiple-instrument electronics 11514



■

integration time of 625 ms, and four measurements were averaged for each absorbance measurement for each trial. Prior to each trial, the signal was background subtracted using a cyclohexane blank. During the data processing stage, absorbance data were used to supply inner filter effect corrections to fluorescence intensity measurements. An identical miniature CCD array detector to that used for absorbance measurements was used as a reference channel to monitor the incident light. The light entered the detector through one arm of the bifurcated fiber optic cable from the LED inside the lifetime instrument housing. The signal was reduced using neutral density filters before entering the reference detector. The detector was run at an integration time of 1 ms, and 100 measurements were averaged for each measurement at 405 nm for each trial. The oxygen sensor was also a phase-shift fluorimeter-based lifetime instrument (Ocean Optics) operating a 475 nm excitation source LED modulated at 100 kHz. A bifurcated fiber optic cable carried the excitation light through a HIOXY probe (Ocean Optics, especially formulated for nonpolar organic solvents) with the distal end having a coating of thin layer sol−gel matrix containing the ruthenium compound oxygen sensing material. The mechanism of operation of the oxygen sensor is based on the measurement of the fluorescence lifetime quenching of the ruthenium compound. Ruthenium compounds generally have very long fluorescence lifetimes (∼100 to 6000 ns)15 and are therefore very susceptible to quenching by oxygen. These compounds have been wellcharacterized in regard to oxygen quenching and stability, making them very useful and sensitive tools for oxygen monitoring.1,2,16−18 The sensor was housed in an 18-gauge stainless steel needle so that it could be carefully pushed through septa. The needle had an opening slightly up the shaft from the tapered point, where oxygen in solution could interact with the probe’s sensing material. The probe has a 45 s response delay, so oxygen measurements were shifted by this amount to match the timing of the other simultaneous optical measurements during each trial. The second arm of the bifurcated fiber optic cable carried the 600 nm fluorescence signal of the ruthenium compound to an avalanche photodiode detector. The probe was calibrated at the manufacturing facility using controlled concentrations of compressed oxygen gas at varying partial pressures. Experimental Details. Pre-experiment investigations of oxygen measurements of aerated solutions (following the procedure outlined in the previous manuscript by the authors)10 and air-equilibrated solutions showed that airequilibrated cyclohexane solutions were oxygen saturated. Therefore, 1-aminoanthracene solutions in cyclohexane were not aerated further with compressed air prior to the start of each trial. The fluorescence lifetime, oxygen concentration, absorbance, fluorescence intensity, reference, and temperature control were all monitored by separate laptop computers and individual data collection software. Trials were deemed to be complete after about 6.5 h of run time because lifetime plots leveled and oxygen measurements indicated essentially no remaining oxygen. The experiments were conducted in triplicate, and constants and results were calculated for each individual trial and then averages and standard deviations for the three trials were reported.

Article

RESULTS AND DISCUSSION

1-Aminoanthracene has been identified as a potential general anesthetic19 as well as a screening proxy for the elucidation of properties of other anesthetics.20 Its ability to behave similarly to other anesthetics, along with its fluorescence characteristics, enables researchers to better understand the mechanisms general anesthetics undergo in the body. One example of such an application is the study of albino tadpoles via confocal laser scanning microscopy at excitation of 488 nm and emission from 515 to 550 nm.20 According to the researchers, in addition to its increased speed and sensitivity, fluorescence provides better spatiotemporal resolution than corresponding autoradiography and NMR techniques.20 Though the data presented here on 1-aminoanthracene may be of value to researchers studying anesthetics,21 the analyte was chosen for this study due to its relatively high quantum yield (Φ = 0.61),22 sufficient fluorescence lifetime (τo = 22.8 ns, in cyclohexane),22 molar absorptivity at the excitation wavelength of this study (∼3900 M−1 cm−1 at 405 nm),22 fluorescence emission range (410−610 nm) at wavelengths that are compatible with the optics of the instruments used for the simultaneous measurements in this study and favorable solubility in cyclohexane. Essentially, 1-aminoanthracene is logistically favorable for the variety of detectors used in the complex instrumental setup as well as its photophysical properties for studying quenching mechanisms. Excitation and emission spectra for 1-aminoanthracene can be found in a recent publication.23 The simultaneously collected data obtained over time with the instrumental setup described above for the deoxygenation of 1-aminoanthracene in cyclohexane are normalized to their oxygen-saturated (time = 0) values and shown in Figure 2. Oxygen in the solution was displaced under a nitrogen atmosphere in the glovebag. The data displayed represent the average signals and error bars (±standard deviation) of three trials at a total runtime for each trial of ∼6.5 h.

Figure 2. Analytical measurements of 1-aminoanthracene in cyclohexane, while oxygen is displaced under a nitrogen atmosphere in a glovebag (all measurements normalized to 1 at the aerated solution starting point). Color key: blue = concentration and inner filter effect corrected fluorescence intensity, red = fluorescence lifetime, green = absorbance, purple = oxygen concentration, and orange = atmospheric pressure inside the glovebag. 11515



Article

The normalized fluorescence intensity and lifetime plots in Figure 2 follow the general expected trend for solution deoxygenation over time, where fluorescence-based signals are increased as the quencher leaves the solution. Steady-state fluorescence intensity increases at a rate faster than that of lifetime measurementsand ultimately arrives at a higher enhancement factor (the ratio of fluorescence intensity of a completely deoxygenated solution to the fluorescence intensity of an aerated solution)here, Fo/F > τo/τ. Considering that lifetime measurements are generally unaffected by concentration or static quenching, the lifetime plot versus time follows that of what might be expected under conditions of only dynamic quenching. The concentration and inner filter effectcorrected fluorescence intensity measurements, (“adjusted fluorescence”) going higher than those of lifetime measurements, are a potential indication of a static quenching component in addition to dominant dynamic quenching. From the slope of the absorbance plot, solution concentration increases at an average rate of 0.05 ± 0.01% per minute throughout the duration of the experiment. For the 3.5 mL of solution used in the cuvette, this corresponds to a measured loss of 1.6 × 10−5 ± 0.5 × 10−5 moles of cyclohexane per minute when the system is exposed to a nitrogen atmosphere. Directly bubbling nitrogen into a solution is another common method for deoxygenation but can also cause evaporation of the volatile cyclohexane solvent. By comparison, bubbling nitrogen into the solution at a flow rate of 5 mL/min resulted in a solvent loss of 2.1 × 10−5 moles of cyclohexane per minute in another study. 10 The rate of solvent loss by gradual deoxygenation under a nitrogen atmosphere is expectedly slower than in the case of direct purging by bubbling nitrogen; however, over the long experiment run times of this study, it is clear that the net effect on concentration can be significant. Corrections were made for both primary inner filter effects and solution concentration increases. Primary inner filter effect corrections were performed using Patterson’s equation24 for which its use has been confirmed in a prior publication.25 Since the light source in the instrumental setup is monochromatic (405 nm), full absorption spectra were not available (only the absorbance at the excitation wavelength is measured). Primary inner filter effects are typically the dominant factor in corrections, and if the absorbance at the wavelength of the fluorophore’s monitored emission is very small, secondary inner filter effects can be neglected. At the concentration of 1aminoanthracene used in this study, the absorbance at the monitored fluorescence emission (secondary inner filter effect) wavelength, 474 nm, is quite low (0.25 AU in raw spectra). Therefore, it is reasonable to only correct data for primary inner filter effects. Since absorbance plots showed evidence of concentration increases over the ∼6.5 h trials, the impact of inner filtering would be more pronounced as the trials proceed, making the correction for inner filter effects even more important. The primary inner filter effectcorrected fluorescence intensity was then adjusted to account for the concentration increase so that the final plot of fluorescence intensity represents that of an apparent solution whose concentration had not increased. This is a necessary adjustment when comparing to lifetime data for subtle distinctions of quenching mechanisms. Historically, for studies of dynamic quenching by oxygen in nonpolar organic solvents, direct and real-time measurements

of oxygen have not been available. The oxygen concentration measured with the fiber optic-based oxygen sensor in airequilibrated cyclohexane solutions in this study is 2.36 ± 0.01 mM. The measured oxygen concentrations of the study here are in substantial agreement with literature values (2.24−2.47 mM, average of literature values examined = 2.35 ± 0.10 mM)26−29 of oxygen solubilities in cyclohexane. Dynamic Quenching. Fluorescence can be quenched by either dynamic or static mechanisms (or a combination of the two) and is evidenced by a general decrease in fluorescence intensity at higher quencher concentrations. Steady-state fluorescence measurements are unable to distinguish between dynamic and static quenching mechanisms, but dynamic quenching is revealed by a decrease in fluorescence lifetimes with increasing quencher concentrations. Quenching by oxygen in nonpolar organic solvents is thought to be diffusion controlled, and in the case of only dynamic quenching, the Stern−Volmer equation is described as Fo/F(or τo/τ ) = 1 + KD[O2 ] = 1 + kqτo[O2 ]

(1)

where KD is the Stern−Volmer constant for dynamic quenching, Fo/F is the ratio of steady-state fluorescence intensity measured in the absence of oxygen to that measured in the presence of oxygen, τo/τ is the ratio of fluorescence lifetime measured in the absence of oxygen to that measured in the presence of oxygen, and kq is the bimolecular diffusioncontrolled rate constant and is related to the frequency of collisions between two freely diffusing molecules (the fluorophore and molecular oxygen, in the case of oxygen quenching). Fluorescence lifetime Stern−Volmer plots of eq 1 will have a slope of kq*τo (or KD), which is shown in Figure 3 for this study.

Figure 3. Stern−Volmer plots of 1-aminoanthracene in cyclohexane while oxygen is displaced under a nitrogen atmosphere in a glovebag. The blue colored data are fluorescence intensity, and the red colored data are fluorescence lifetime. The error bars represent ± standard deviation of the average signal for n = 3 trials.

Correlation coefficients show that the plots for both Fo/F and τo/τ are quite linear (R2 ∼ 0.999), but given the differences in their slopes (0.489 ± 0.015 and 0.445 ± 0.014 for Fo/F and τo/τ, respectively), there is an apparent static quenching component in addition to the prevalent dynamic quenching. Enhancement factors are 2.20 ± 0.01 and 2.08 ± 0.01 for Fo/F and τo/τ, respectively. As was the case in Pagano et al.,10 11516



Article

Table 1. Quenching Parameters of 1-Aminoanthracene by Oxygen in Cyclohexane enhancement factor [std. dev.]

slope (mM−1) of Fo/F − 1 (or τo/τ − 1) vs oxygen conc. [std. dev.]

Fo/F: 2.20 [0.01] τo/τ: 2.08 [0.01]

Fo/F: 0.489 [0.015] τo/τ: 0.445 [0.014]

τo (ns) [std. dev.]

diffusion coefficient, D1‑AA (cm2/s) in cyclohexane

diffusion coefficient, DO2 (cm2/s) in cyclohexane

kdc (M−1 s−1) calculated using eq 2

kq (M−1 s−1) [std. dev.]

21.8 [1.1]

0.7 × 10−5 a

5.3 × 10−5 b

2.2 × 1010 a,b

2.1 × 1010 c [0.2]

a Using eq 2a from the radius of similar molecule, anthracene. bUsing value for oxygen diffusion in cyclohexane from ref 38. cfrom Stern−Volmer lifetime plots of eq 1.

calculated). However, this equation assumes that diffusing molecules are larger than solvent molecules.5 Part of the system studied here (molecular oxygen in cyclohexane) does not satisfy this assumption. Therefore, eq 2a can be used to calculate the diffusion coefficient value for 1-aminoanthracene in cyclohexane, but an experimental literature value is used for the oxygen diffusion coefficient in cyclohexane. Using eq 2a and 0.893 mPa·s for cyclohexane’s viscosity at 20 °C,37 3.27 Å for the approximated radius (actually cited for anthracene),36 a diffusion coefficient of 0.7 × 10−5 cm2/s is calculated for 1-aminoanthracene in cyclohexane. This value is in good agreement with a literature diffusion coefficient value (0.6 × 10−5 cm2/s) for a similar molecule, 9,10-diphenylanthracene, in cyclohexane.37 The diffusion coefficient for oxygen in cyclohexane is taken from the literature38 to be 5.3 × 10−5 cm2/s. The separate diffusion coefficients for 1-aminoanthracene (approximated using eq 2a and the literature value for the molecular radius of anthracene) and oxygen in cyclohexane, along with separate molecular radii of 1-aminoanthracence and molecular oxygen for the system can be plugged into eq 2 in order to calculate kdc. Here, a kdc of 2.2 × 1010 M−1 s−1 is calculated and is in very good agreement with the experimentally determined kq by this study (2.1 × 1010 ± 0.2 × 1010 M−1 s−1). Parameters of the diffusion-controlled rate constant are summarized in Table 1. Static Quenching. Due to the large dynamic quenching component in systems similar to the one studied here, small static quenching contributions can be obscured. However, the presented data suggest that, though much smaller in magnitude, there is a static quenching component present. Again, static quenching may be separated into two potential forms: (1) complex formation and (2) “sphere of action” quenching.5 For only static quenching by formation of complexes with oxygen, the Stern−Volmer equation is expressed as the linear function:

enhancement factors found in this study are higher than those of Berlman (2.07)22 by about 6%, and Berlman might have had remaining oxygen in the solutions that he was using for Fo measurements. If we take our values here for Fo/F to be correct (τo/τ would be a better comparison, but Berlman does not provide such a value), we calculate that Berlman had about 0.3 mM of oxygen remaining in the solution that he identified as deoxygenated. The Stern−Volmer slope of Fo/F in Figure 3 is greater than that of τo/τ by 9−10%. The deoxygenated lifetime measurement of 1-aminoanthracene (21.8 ± 1.1 ns in the present study) is about 4% lower than Berlman’s value (22.8 ns, also measured in cyclohexane)22 but is within error bars. The discrepancies are likely due to different instrumentation used to obtain lifetime measurements, and other PAH lifetime data from literature have also been found to be within a similar range below Berlman’s lifetime measurements.11,30,31 Quenching parameters and results from Stern−Volmer plots of this study are summarized in Table 1. The slopes of the Stern−Volmer plots derived from lifetime data yield KD values. For the system studied here, KD is found to be 0.445 ± 0.014 mM−1. Related 1/KD values (2.24 ± 0.07 mM) for trials are significant in that they are representative of the oxygen concentrations at which half of the fluorescence is quenched.5 Also using eq 1 and Stern−Volmer plots derived from lifetime data, along with the measured fluorescence lifetime of 1-aminoanthracene in the absence of oxygen, the bimolecular rate constant, kq, can be determined. From the trials of this study, kq is calculated to be 2.1 × 1010 ± 0.2 × 1010 M−1 s−1, which is in very good agreement with those obtained in similar oxygen quenching system studies.4,11,32−34 Furthermore, this value is in excellent agreement with that expected of a diffusion-controlled rate constant (≈1−3 × 1010 M−1 s−1).5,32,35 Diffusion-controlled bimolecular rate constants for quenching systems can be calculated using the Smoluchowski equation:5 kdc = 4π (N /1000)(D1 ‐ AA + DO2)(r1 ‐ AA + rO2)

Fo/F = 1 + KS[O2 ]

where KS is the general Stern−Volmer constant for static quenching. In the case of static quenching by formation of complexes, KS can be expressed as11

(2)

where kdc is the diffusion-controlled rate constant, D1‑AA and DO2 are the diffusion coefficients for 1-aminoanthracene and molecular oxygen in cyclohexane, respectively; r1‑AA and rO2 are the molecular radii for 1-aminoanthracene and molecular oxygen, respectively; N is Avogadro’s number; and 1/1000 is a conversion factor for units of cm3 to liters. In quenching studies, the separate diffusion coefficients (D) for the fluorophore and quencher in the solvent can be calculated using the Stokes−Einstein equation5,36 D = kT /6πηr

(3)

KS = [M−O2 ]/[M][O2 ]

(4)

where fluorophore, M, forms a ground-state complex, M−O2, with oxygen. KS is the ground-state equilibrium constant for the formation of a complex and the concentration of the products can be written as11 [M−O2 ] = KS[M][O2 ]

(4a)

Researchers have also proposed a “sphere of action” static quenching mechanism that happens over a distance when a quencher is within the requisite volume of the fluorophore at the instant it becomes excited.6−9,39,40 This type of static quenching, when combined with dynamic quenching, is discussed in the next section.

(2a)

where k is the Boltzmann constant, T is temperature, η is the temperature dependent viscosity of the solvent (cyclohexane), and r is again the molecular radius (of either the fluorophore or quencher, depending on which diffusion coefficient is being 11517



Article

Table 2. Constants for the Oxygen Quenching of 1-Aminoanthracene with Oxygen in Cyclohexane

a

KD (mM−1) [std. dev.]

1/KD (mM) [std. dev.]

correlation coefficient (R2) of Stern−Volmer Plots

KS (mM−1) [std. dev.]

% static quenching (KS/(KD + KS)*100) [std. dev.]

0.462a [0.010]

2.16a [0.05]

0.014a [0.003]

3.0a [0.6]

0.445b [0.014]

2.24b [0.07]

0.9993 (Fo/F) 0.9992 (τo/τ) 0.9994a (Fo/F)

0.024c [0.001]

5.1c [0.3]

b

c

Calculated using eq 5a. From Stern−Volmer lifetime plots of eq 1. Calculated using eq 6.

Combined Dynamic and Static Quenching. For systems with combined dynamic quenching and static quenching by complex formation, the Stern−Volmer equation can be expressed as5 Fo/F = (1 + KD[O2 ])(1 + KS[O2 ])

eqs 7b and 6 were used, respectively. When lifetime data are substituted for the dynamic term, eq 7a can become linear (as is the case for the system in our data set) as described by11 (Fo/F )/(τo/τ ) = 1 + V [O2 ])

(5)

For situations with combined dynamic and static quenching, eqs 6 (derived from static quenching due to complex formation) and 7c (derived from static quenching due to “sphere of action”) are the same in form. The linearity of both is in agreement with the Stern−Volmer plots presented for this data set. Brownrigg and Kenny note that the validity of eq 6 is dependent on the assumption that oxygen concentrations in fluorescence intensity and lifetime measurements must be identical;11 a condition that their study was not directly able to meet but is satisfied by the use of the instrumental setup for simultaneous measurements in this study. KS is calculated from the slope of a plot of eq 6 and assessed as the static quenching component of this system. Here, KS is found to be 0.024 ± 0.001 mM−1 and represents only 5.1 ± 0.3% of the total quenching (when the dynamic component is taken from the fluorescence lifetime slopes of the Stern−Volmer plot and quenching is defined as the sum of KD and KS). Results of the plots of eq 6 are summarized and compared to those of eq 5a in Table 2. The quenching parameters determined from this data set via eqs 5a and 6 are in relatively good agreement. In a system like the one studied here, where KS ≪ KD, dynamic quenching is the dominant quenching mechanism, even the relatively small KS values can indicate a static quenching component. As mentioned, Brownrigg and Kenny also postulated a potential static quenching portion to two of the four PAHs that they studied, and perhaps due to experimental limitations at the time, the relative standard deviation in the static quenching component was about 30%.11 The relative standard deviation in the static quenching component for the present study is about twice as favorable, allowing the static quenching component to be elucidated with more certainty. Brownrigg and Kenny found the quenching in the studied system to be about 2% static for naphthalene, the PAH which displayed the greatest static quenching component, and attributed it to a potential formation of PAH-oxygen charge-transfer complexes.11 Likewise, Lakowicz observed 3− 5% static quenching for fluorophores that he studied in aqueous and organic solutions and attributed the static portion to a “sphere of action” quenching mechanism.32 However, he states that for a similar PAH−O2 quenching system, a formed complex would have a very weak association (∼1 kcal/mol),32 which is similar in strength to van der Waals interactions. In his study, using a special high-pressure fluorescence cell, he equilibrated solutions with high pressures of oxygen and examined absorption spectra for signs of formed complexes but could not identify anyeven at oxygen concentrations ∼375 times greater than that of the aerated solutions of the current study.32 Therefore, a “sphere of action” mechanism seems more reasonable for explaining static quenching in the system

and expanded to Fo/F − 1 = (KD + KS)[O2 ] + KDKS[O2 ]2

(5a)

This relationship becomes second order in quencher concentration, [O2]. A plot of eq 5a can yield a second-order polynomial fitted trendline, where two solutions of two unknowns (KD and KS) can be calculated. The solution that is comparable in rate to a diffusion-controlled constant process is assigned to KD and the other to KS.5 The values of KS and KD calculated from eq 5a can be compared in order to assess the relative magnitude of dynamic and static quenching in this system. As expected using this equation, KD (0.462 ± 0.010 mM−1) is much larger than KS (0.014 ± 0.003 mM−1), meaning that the system is dominated by dynamic quenching and static quenching only represents 3.0 ± 0.6% of the total quenching in the system, where quenching is defined as the sum of KD and KS. Values of KD and KS are summarized in Table 2. For situations where the product in the right-hand factor of eq 5a is much less than 1, the plot should be close to linear, as is the case with the data set presented here. The comparison of steady-state fluorescence intensity and lifetime measurements can be used to differentiate dynamic and static quenching. When lifetime data is substituted into eq 5, the relationship with static quenching by complex formation becomes (Fo/F )/(τo/τ ) = 1 + KS[O2 ]

(6)

According to Lakowicz, combined dynamic quenching and “sphere of action” static quenching can be described as5 Fo/F = (1 + KD[O2 ]) exp(V [O2 ])

(7)

where V is the volume of the “sphere of action” (volume per molecule) of radius r, which can be calculated from the volume of the sphere (V = 4/3πr3). “Sphere of action” static quenching can make a plot of Fo/F exponential. However, the exponential factor can be expanded when the oxygen concentration is much less than 1, and when truncated at the linear term, eq 7 becomes Fo/F = (1 + KD[O2 ])(1 + V [O2 ])

(7a)

and expanded to Fo/F − 1 = (KD + V )[O2 ] + KDV [O2 ]2

(7c)

(7b)

As seen by comparing eqs 7b and 5a, the V in “sphere of action” static quenching plays the same role as KS in static quenching by complex formation. In the system studied here, the radii were found to be 17.7 ± 1.3 and 21.1 ± 0.4 Å when 11518



■

presented here, but since the Stern−Volmer plots cannot discriminate between complex formation from weak association constants and charge-transfer complexes, it is possible that oxygen charge-transfer complexes could also be forming and contributing to KS.

REFERENCES

(1) Feng, Y.; Cheng, J.; Zhou, L.; Zhou, X.; Xiang, H. Ratiometric Optical Oxygen Sensing: A Review in Respect of Material Design. Analyst 2012, 137, 4885−4901. (2) Wang, X. D.; Wolfbeis, O. S. Optical Methods for Sensing and Imaging Oxygen: Materials, Spectroscopies and Applications. Chem. Soc. Rev. 2014, 43, 3666−3761. (3) Mehta, V. N.; Kailasa, S. K.; Wu, H. F. Surface Modified Quantum Dots as Fluorescent Probes for Biomolecule Recognition. J. Nanosci. Nanotechnol. 2014, 14, 447−459. (4) Ware, W. R. Oxygen Quenching of Fluorescence in Solution: An Experimental Study of the Diffusion Process. J. Phys. Chem. 1962, 66, 455−458. (5) Lakowicz, J. R. Principles of Fluorescence Spectroscopy, 3rd ed.; Springer: New York, 2006; Vol. XXVI, p 954. (6) Perrin, F. Law of Decrease in the Fluorescent Power According to Concentration. C. R. Hebd. Seances Acad. Sci. 1924, 178, 1978. (7) Melavanki, R. M.; Kusanur, R. A.; Kadadevaramath, J. S.; Kulakarni, M. V. Quenching Mechanisms of 5BAMC by Aniline in Different Solvents Using Stern-Volmer Plots. J. Lumin. 2009, 129, 1298−1303. (8) Patil, N. R.; Melavanki, R. M.; Nagaraja, D.; Patil, H. D.; Sanningannavar, F. M.; Kapatakar, S. B. Quenching of the Fluorescence of ENCDTTC by Aniline and Carbon Tetrachloride in Different Organic Solvents. J. Mol. Liq. 2013, 180, 112−120. (9) Moriya, T. Excited-State Reactions of Coumarins in AqueousSolutions. II. The Fluorescence Quenching of 7-Ethoxycoumarins by Halide Ions. Bull. Chem. Soc. Jpn. 1984, 57, 1723−1730. (10) Pagano, T.; Biacchi, A. J.; Kenny, J. E. Nitrogen Gas Purging for the Deoxygenation of Polyaromatic Hydrocarbon Solutions in Cyclohexane for Routine Fluorescence Analysis. Appl. Spectrosc. 2008, 62, 333−336. (11) Brownrigg, J. T.; Kenny, J. E. Fluorescence Intensities and Lifetimes of Aromatic Hydrocarbons in Cyclohexane Solution: Evidence of Contact Charge-Transfer Interactions with Oxygen. J. Phys. Chem. A 2009, 113, 1049−1059. (12) Munck, A. U.; Scott, J. F. Ultra-Violet Absorption of Oxygen in Organic Solvents. Nature 1956, 177, 587. (13) Boens, N.; Qin, W. W.; Basaric, N.; Hofkens, J.; Ameloot, M.; Pouget, J.; Lefevre, J. P.; Valeur, B.; Gratton, E.; Vandeven, M.; et al. Fluorescence Lifetime Standards for Time and Frequency Domain Fluorescence Spectroscopy. Anal. Chem. 2007, 79, 2137−2149. (14) Ryder, A. G.; Power, S.; Glynn, T. J.; Morrison, J. J. TimeDomain Measurement of Fluorescence Lifetime Variation with pH. In Biomarkers and Biological Spectral Imaging; Bearman, G. H., Bornhop, D. J., Levenson, R. M., Eds.; SPIE: San Jose, CA, 2001; Vol. 4259, pp 102−109. (15) Morris, K. J.; Roach, M. S.; Xu, W. Y.; Demas, J. N.; DeGraff, B. A. Luminescence Lifetime Standards for the Nanosecond to Microsecond Range and Oxygen Quenching of Ruthenium(II) Complexes. Anal. Chem. 2007, 79, 9310−9314. (16) Meyer, T. J. Photochemistry of Metal Coordination Complexes - Metal to Ligand Charge-Transfer Excited States. Pure Appl. Chem. 1986, 58 (9), 1193−1206. (17) Hartmann, P.; Leiner, M. J. P.; Lippitsch, M. E. Response Characteristics of Luminsecent Oxygen Sensors. Sens. Actuators, B 1995, 29, 251−257. (18) Juris, A.; Balzani, V.; Barigelletti, F.; Campagna, S.; Belser, P.; Vonzelewsky, A. Ru(II) Polypyridine Complexes - Photophysics, Photochemistry, Electrochemistry, and Chemi-Luminescence. Coord. Chem. Rev. 1988, 84, 85−277. (19) Emerson, D. J.; Weiser, B. P.; Psonis, J.; Liao, Z. Z.; Taratula, O.; Fiamengo, A.; Wang, X. Z.; Sugasawa, K.; Smith, A. B.; Eckenhoff, R. G.; et al. Direct Modulation of Microtubule Stability Contributes to Anthracene General Anesthesia. J. Am. Chem. Soc. 2013, 135, 5389− 5398. (20) Butts, C. A.; Xi, J.; Brannigan, G.; Saad, A. A.; Venkatachalan, S. P.; Pearce, R. A.; Klein, M. L.; Eckenhoff, R. G.; Dmochowski, I. J.

■

CONCLUSION The method detailed above, including the use of the novel instrument system for the simultaneous optical measurements of several parameters, shows significant potential for the detailed analysis of fluorescence quenching mechanisms. When measurements are carefully taken and uncertainties in data are small, in addition to dynamic quenching components, the data show a small static quenching component due to either complex formation or “sphere of action”. Using several forms of the Stern−Volmer relationship, it was found that for the system involving 1-aminoanthracene fluorescence quenching by oxygen in cyclohexane, ∼95% of the total quenching can be attributed to the dynamic process. KSV from the slope of Fo/F of the Stern−Volmer plot (Figure 3) equals 0.489 ± 0.015 mM−1, which is within error bars of the sums of KD and KS determined from eqs 5a and 6. It has been shown that attention to, and compensation for, solvent loss during deoxygenation studies, and the need for inner filter effect corrections, is crucial for detailed investigations of quenching mechanisms. With the custom-designed instrument used in this study, valuable real-time relationships are drawn between fluorescence intensity and fluorescence lifetime data at directly measured oxygen concentrations. Meanwhile, the absorbance measurements monitor solution concentration changes during the deoxygenation process. As a result, uncertainties in data are smaller, and the static quenching component is able to be better detected, providing opportunity for more rigorous study of the combination of fluorescence quenching mechanisms of fluorophore-quencher systems in solution. The success of the developed instrumental system, employed for the simultaneous measurements of parameters crucial to the study of kinetic processes occurring as a function of solution deoxygenation, warrants its use in further studies. Dynamic and static quenching rate constants for larger data sets of different analytes, solvents, concentration ranges, and inert gases responsible for displacing dissolved oxygen could provide additional and detailed information into these mechanisms. Continued investigations into the differentiation between the two static quenching mechanisms, and in which conditions the two flourish, could prove useful.

■

Article

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]; Phone: 585-475-4539. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported in part by an NTID/RIT Innovation Grant. The authors would like to thank Annemarie Ross, Kyle Edenzon, and Grace Kennedy for assistance in collecting data; Hai Tang for helping to make the image of the fiber-optic based instrumental setup; and Susan B. Smith and Morgan Bida for editing and helpful comments on the manuscript. 11519



Article

Identification of a Fluorescent General Anesthetic, 1-Aminoanthracene. Proc. Natl. Acad. Sci. U.S.A. 2009, 106, 6501−6506. (21) Eckenhoff, R. Promiscuous Ligands and Attractive Cavities. Mol. Interventions 2001, 1, 258−268. (22) Berlman, I. B. Handbook of Fluorescence Spectra of Aromatic Molecules. Academic Press: New York, 1971. (23) Prichystalova, H.; Almonasy, N.; Nepras, M.; Bures, F.; Dvorak, M.; Michl, M.; Cermak, J.; Burgert, L. N-Triazinyl Derivatives of 1-and 9-Aminoanthracene: Synthesis and Photo-Physical Properties. J. Fluoresc. 2013, 23, 425−437. (24) MacDonald, B. C.; Lvin, S. J.; Patterson, H. Correction of Fluorescence Inner Filter Effects and the Partitioning of Pyrene to Dissolved Organic Carbon. Anal. Chim. Acta 1997, 33, 155−162. (25) Pagano, T. E.; Kenny, J. E. Assessment of Inner Filter Effects in Fluorescence Spectroscopy Using the Dual Pathlength Method: A Study of the Jet Fuel JP-4. In Internal Standardization and Calibration Architectures for Chemical Sensors; Shaffer, R. E., Potyrailo, R. A., Eds.; SPIE: Boston, MA, 1999; Vol. 3856, pp 289−297. (26) Novak, K. Chem. Abstr. 1963, 58, 69833 cited in ref 11. (27) Luhring, P.; Schumpe, A. Gas Solubilities (H2, He, N2, CO, O2, Ar, CO2) in Organic Liquids at 293.2 K. J. Chem. Eng. Data 1989, 34, 250−252. (28) Wild, J. D.; Sridhar, T.; Potter, O. E. Solubility of Nitrogen and Oxygen in Cyclohexane. Chem. Eng. J. 1978, 15, 209−214. (29) Battino, R.; Rettich, T. R.; Tominaga, T. The Solubility of Oxygen and Ozone in Liquids. J. Phys. Chem. Ref. Data 1983, 12, 163− 178. (30) Amata, C. D.; Burton, M.; Helman, W. P.; Ludwig, P. K.; Rodemeye, S. A. Luminescence Decay Times of Ultraviolet-Excited Scintillator Solutions. J. Chem. Phys. 1968, 48, 2374−2376. (31) Hobbs, S. E.; Hieftje, G. M. Scintillator-Based Nanosecond Light-Sources for Time-Resolved Fluorometry. Appl. Spectrosc. 1995, 49, 15−19. (32) Lakowicz, J. R.; Weber, G. Quenching of Fluorescence by Oxygen-Probe for Structural Fluctuations in Macromolecules. Biochemistry 1973, 12, 4161−4170. (33) Arik, M.; Celebi, N.; Onganer, Y. Fluorescence Quenching of Fluorescein with Molecular Oxygen in Solution. J. Photochem. Photobiol., A 2005, 170, 105−111. (34) Abdel-Shafi, A. A.; Worrall, D. R.; Wilkinson, F. Singlet Oxygen Formation Efficiencies Following Quenching of Excited Singlet and Triplet States of Aromatic Hydrocarbons by Molecular Oxygen. J. Photochem. Photobiol., A 2001, 142, 133−143. (35) Patterson, L. K.; Porter, G.; Topp, M. R. Oxygen Quenching of Singlet and Triplet States. Chem. Phys. Lett. 1970, 7, 612−614. (36) Kowert, B. A.; Dang, N. C.; Sobush, K. T.; Seele, L. G. Diffusion of Aromatic Hydrocarbons in N-Alkanes and Cyclohexanes. J. Phys. Chem. A 2001, 105, 1232−1237. (37) Kowert, B. A.; Jones, J. B.; Zahm, J. A.; Turner, R. M. SizeDependent Diffusion in Cycloalkanes. Mol. Phys. 2004, 102, 1489− 1497. (38) Krieger, I. M.; Mulholla, G. W.; Dickey, C. S. Diffusion Coefficients for Gases in Liquids from Rates Solution of Small Gas Bubbles. J. Phys. Chem. 1967, 71, 1123−1129. (39) Frank, J. M.; Wawilow, S. J. Effective Spheres in the Extinction Process in Fluorescent Liquids. Z. Phys. 1931, 69, 100−110. (40) Sarkar, A.; Ali, M.; Baker, G. A.; Tetin, S. Y.; Ruan, Q. Q.; Pandey, S. Multiprobe Spectroscopic Investigation of Molecular-Level Behavior within Aqueous 1-Butyl-3-Methylimidazolium Tetrafluoroborate. J. Phys. Chem. B 2009, 113, 3088−3098.

11520


Investigation of the Fluorescence Quenching of 1 ... - ACS Publications

Recommend Documents