undecanoate Dissolved in Organic Liquids and Supercritical Carbon

Department of Chemistry, New Mexico Institute of Mining and Technology, ... at 470-480 nm (E1) forms within 2 ns of optical excitation; however, it is...
4 downloads 0 Views 203KB Size
Download PDF
1820

J. Phys. Chem. B 2002, 106, 1820-1832

The Photophysics of 6-(1-Pyrenyl)hexyl-11(1-pyrenyl)undecanoate Dissolved in Organic Liquids and Supercritical Carbon Dioxide: Impact on Olefin Metathesis Siddharth Pandey,*,† Maureen A. Kane,‡ Gary A. Baker,‡ Frank V. Bright,*,‡ Alois Fu1 rstner,*,§ Gu1 nter Seidel,§ and Walter Leitner*,§ Department of Chemistry, New Mexico Institute of Mining and Technology, Socorro, New Mexico 87801, Department of Chemistry, Natural Sciences Complex, UniVersity at Buffalo, The State UniVersity of New York, Buffalo, New York 14260-3000, and Max-Planck-Institut fur Kohlenforschung, Kaiser-Wilhelm-Platz 1, D-45470 Mulheim an der Ruhr, Germany ReceiVed: April 19, 2001; In Final Form: July 26, 2001

The Leitner and Fu¨rstner groups reported (Fu¨rstner, A.; Koch, D.; Langemann, K.; Leitner, W.; Six, C. Angew Chem., Int. Ed. Engl. 1997, 36, 2466) on the ring closing metathesis (RCM) of a 16-membered diene dissolved in supercritical CO2 (scCO2). The authors found that the cyclic product, indicative of an intramolecular RCM event, was formed in excellent yield when the CO2 density was high, but oligomers were formed by an acyclic diene metathesis (ADMET) pathway at lower CO2 densities. These results suggest that changes in the CO2 density lead to changes in the intra- vs intermolecular interactions between the 16-membered diene dissolved in scCO2. To assess this issue in more detail, we have prepared 6-(1-pyrenyl)hexyl-11-(1-pyrenyl)undecanoate [1-Py(CH2)10COO(CH2)61-Py] in which we replaced the terminal alkenes of Letiner and Fu¨rstner’s original diene with the fluorophore pyrene. We have studied the pyrene excimer formation of 1-Py(CH2)10COO(CH2)61-Py when it is dissolved in five organic solvents (cyclohexane, dichloromethane, ethanol, acetonitrile, and dimethyl sulfoxide) and supercritical carbon dioxide (scCO2) to determine how the tail segments interact with each other. The result show that the excimer formation mechanism is completely different when 1-Py(CH2)10COO(CH2)61-Py is dissolved in scCO2 or organic liquids. In liquids, excimer formation is purely dynamic in nature, there are two formation pathways to the excimer, and all the rates can be understood with the help of Kamlet-Taft linear solvent energy relationships. In scCO2, we found that the 1-Py(CH2)10COO(CH2)61-Py excimer-to-monomer intensity ratio (E/M) correlates directly with (1) the observed RCM yield for Leitner and Fu¨rstner’s original 16-membered diene and (2) the solvent refractive index function. The steady-state and time-resolved fluorescence of 1-Py(CH2)10COO(CH2)61-Py dissolved in scCO2 show that there are two excimers that form in scCO2 and their relative contributions change in a systematic way with changes in the CO2 pressure/density. Interestingly, the typical dynamically formed excimer species that emits at 470-480 nm (E1) forms within 2 ns of optical excitation; however, it is not the dominant species at low CO2 densities. E1 is equivalent to the species that goes on to form the RCM product in Leitner and Fu¨rstner’s original reaction. The second excimer (E2) emits in the 410-440 nm region. E2 is associated with intermolecular preassociated forms of the pyrene residues within a collection of 1-Py(CH2)10COO(CH2)61-Py molecules, and this species dominates at low CO2 densities. E2 is equivalent to the species that goes on to form the oligomeric product in the original Leitner and Fu¨rstner reaction. As the CO2 density increases, the E1 excimer contribution increases relative to the E2 excimer contribution. The combination of the fluorescence and reaction outcome results are used to explain Leitner and Fu¨rstner’s previous density-dependent RCM yields.

Introduction Supercritical fluids provide a solvent that exhibits liquidlike densities and gaslike mass transfer. These features make supercritical fluids attractive for extractions, separations, and reactions.1-9 In addition, subtle changes in the system pressure allow one to continuously tune the physicochemical properties (e.g., density, dielectric constant, and refractive index) of a supercritical fluid. Carbon dioxide is the most widely used * To whom all correspondence should be directed. Voice: 716-645-6800 ext. 2162. FAX: 716-645-6963. E-mail: [email protected]. † Department of Chemistry, New Mexico Institute of Mining and Technology. ‡ Department of Chemistry, Natural Sciences Complex. § Max-Planck-Institut fur Kohlenforschung.

supercritical fluid because it exhibits mild critical parameters (critical temperature, Tc ) 31.1 °C; critical pressure, Pc ) 73.8 bar), it is inexpensive, it is nonflammable and nontoxic, it can be easily separated from other system components, it can be recycled and/or reused, and it is environmentally benign.1-9 The Leitner and Fu¨rstner groups recently investigated the ring closing metathesis (RCM) of a 16-membered diene in supercritical CO2 (scCO2).10 The authors found that the cyclic product, indicative of an intramolecular RCM event, was formed in excellent yield when the CO2 density was greater than 0.65 g/mL; however, below 0.65 g/mL, oligomers were formed by an acyclic diene metathesis (ADMET) pathway. These results suggest that changes in the CO2 density lead to changes in the intra- vs intermolecular interactions between the 16-membered

10.1021/jp011497h CCC: $22.00 © 2002 American Chemical Society Published on Web 01/23/2002

Photophysics of 1-Py(CH2)10COO(CH2)61-Py

J. Phys. Chem. B, Vol. 106, No. 7, 2002 1821 SCHEME 2

Figure 1. Molecular structure of 1-Py(CH2)10COO(CH2)61-Py.

SCHEME 1

cyclic diene dissolved in scCO2. The purpose of this paper is to determine the origin of this behavior. Toward this end, we have prepared an analogue of Leitner and Fu¨rstner’s original diene in which the terminal alkene moieties are both replaced with the fluorescent probe pyrene. We use the steady-state and time-resolved pyrene emission to determine how the tail segments behave at low concentration in several liquids and in scCO2. One of the more attractive ways to investigate the behavior of chain termini is to selectively label the “tails” with a fluorophore like pyrene that can form an excimer.11-43 In this way one can address questions ranging from tail-tail cyclization/unfolding dynamics to inter- and intramolecular associations of chain termini/tails themselves. Previous work has shown that the excimer emission depends on chain length and chain structure, the actual fluorophore chemistry, the exact labeling site on the fluorophore and the chain, the tether length and its chemistry, and the solvent quality/composition, temperature, and pressure.11-43 As an example, in organic liquids at room temperature, the time-resolved monomer and excimer intensity decay profiles for 1,3-di(2-pyrenyl)propane are well-described by a double-exponential decay law (i.e., a Birks-like model44 suffices); however, a triple-exponential decay model is required to describe the monomer and excimer decay profiles for 1,3di(1-pyrenyl)propane.18,20-22 Zachariasse and co-workers have argued that there are two kinetically distinct excimer configurations for 1,3-di(1-pyrenyl)propane where one of the excimers has an asymmetric configuration and the other possesses a symmetric structure.18,20-22 In this paper, we report on the excimer formation behavior of 6-(1-pyrenyl)hexyl-11(1-pyrenyl)undecanoate [1-Py(CH2)10COO(CH2)61-Py] (Figure 1) dissolved at low concentration in several organic solvents and in scCO2 as a function of density. Theory In its simplest embodiment, excimer formation is described within a Birks framework (Scheme 1).11-44 For an intramolecular excimer, the rate coefficients shown in Scheme 1 are all unimolecular. Following electronic excitation by a short pulse of electromagnetic radiation, Scheme 1predicts that the monomer time-resolved fluorescence intensity (IM(t)) will decay as the sum of two exponentials, the excimer time-resolved emission

intensity (IE(t)) will decay as the difference of two exponentials, and certain of the decay constants (λi) must be identical (i.e., λ1 ) λ3 and λ2 ) λ4).

IM(t) ) R1 exp (-λ1t) + R2 exp (-λ2t)

(1)

IE(t) ) R3 exp (-λ3t) + R4 exp (-λ4t)

(2)

There is one additional constraint associated with Scheme 1: the recovered preexponential factors for the excimer intensity decay (i.e., R3 and R4) must be equal to one another but opposite in sign. If the time-resolved fluorescence intensity decay profiles are not well-described by eqs 1 and 2 and the constraint R3/R4 ) -1, then a more sophisticated model is required.11-43 Scheme 2 is an example of a more complex excimer formation model. Here, we consider the situation wherein each end group of the bichromophoric species can be excited separately and each of these excitation events follows its own kinetic pathway (i.e., exhibits distinguishable rates) leading to the eventual formation of an intramolecular excimer that possesses the same molecular configuration irrespective of which end group chromophore is excited initially. Upon intramolecular dissociation, as Scheme 2 illustrates, the formation of only the more thermodynamically stable of the two excited-state monomers (dashed arrows) occurs. Scheme 2 predicts that the excited monomer and excimer will decay as the sum of three exponentials:35

IM(t) ) R1 exp (-λ1t) + R2 exp (-λ2t) + R3 exp (-λ3t) (3) IE(t) ) R4 exp (-λ4t) + R5 exp (-λ5t) + R6 exp (-λ6t) (4) with the constraints that

R 4 + R5 + R6 ) 0

(5a)

λ1 ) λ4; λ2 ) λ5; and λ3 ) λ6

(5b)

If Scheme 2 is operational, the decay constants and important linear combinations (λi) are given by35

λ1 ) ka2 + kM

(6)

2λ2,3 ) (Ax + Ay) - [(Ax + Ay)2 + 4(ka1kd - AxAy)]1/2 (7) Ax ) kM + ka1; Ay ) kE + kd

(8)

R5/R4 ) [(λ2 - λ1)(Ax - λ3)]/[(λ3 - λ2)(Ax - λ1)] (9) Notice that Scheme 2 reverts to the classical Birks model (Scheme 1) if ka1 ) ka2.35 Thistlethwaite et al.42 described the intramolecular excimer formation of 1,3-bis(1-pyrenyl)propane in solution by using a

1822 J. Phys. Chem. B, Vol. 106, No. 7, 2002 SCHEME 3

kinetic model (Scheme 3) that involved two simultaneously operating but noninteracting sub-schemes identical to that describing intermolecular excimer formation. Expressions for IM(t) and IE(t) are then obtained by simple addition of the equations describing the fluorescence response functions in each of the two subschemes 3A and 3B:

IM(t) ) R1A exp(-λ1At) + R1B exp(-λ1Bt) + R2A exp(-λ2At) + R2B exp(-λ2Bt) (10) IE(t) ) R3[exp(-λ1At) - exp(-λ2At)] + R4[exp(-λ1Bt) - exp(-λ2Bt)] (11) As with the classical Birks model (Scheme 1), Scheme 3 requires that IE(t) be described by two pairs of preexponentials that are equal in magnitude but opposite in sign. Experimental Section Synthesis of 1-Py(CH2)10COO(CH2)61-Py.45 The chemical transformations are summarized in reaction 1:

In brief, 9-BBN (300 mg, 2.46 mmol) is added to a solution of 5-hexenyl-10-undecenoate (325 mg, 1.23 mmol) in THF (50 mL) and the resulting mixture is stirred for 30 min at ambient temperature. KOMe (185 mg, 2.64 mmol) is then introduced and stirring is continued until a clear solution of the corresponding borate complex (11B NMR: δ ) -1.9 ppm) has formed. After addition of 1-bromopyrene (633 mg, 2.25 mmol) and (dppf)PdCl2 (57 mg, 3 mol %) the mixture is refluxed for 30 min. For workup, the solvent is evaporated, the residue is dissolved in tert-butyl-methyl ether and water, the organic phase is dried (Na2SO4) and evaporated, and the crude product is purified by flash chromatography on silica using a gradient of hexane/ethyl acetate (4:1 f 2:1 f 1:1, each containing 1% v/v CH2Cl2) as the eluent. This affords the title compound as a beige solid (453 mg, 55%). 1H NMR (300 MHz, CD2Cl2): δ ) 8.28 (d, J ) 9.3, 1H), 8.19-8.17 (m, 2H), 8.15-8.08 (m, 2H), 8.138.01 (m, 2H), 7.98-7.96 (m, 1H), 7.90-7.85 (m, 1H), 4.05 (t, J ) 6.6, 2H), [3.34 (t, J ) 6.9), 3.31 (t, J ) 6.9); 4H], 2.26 (t,

Pandey et al. J ) 7.3, 2H), 1.92-1.78 (m, 4H), 1.7-1.4 (m, 9H), 1.4-1.23 (m, 9H). 13C NMR (75 MHz, CD2Cl2): δ ) 174.0, 138.0, 137.7, 131.9, 131.4, 130.1, 130.05, 129.0, 127.9, 127.7, 127.4, 127.3, 126.82, 126.78, 126.2, 125.4, 125.4, 125.15, 125.11, 124.99, 124.97, 123.97, 123.90, 64.5, 34.7, 33.9, 33.8, 32.3, 32.2, 30.2, 29.94, 29.85, 29.78, 29.65, 29.5, 29.1, 26.3, 25.4. MS: m/z (rel intensity) ) 670 (50) [M+], 335 (4), 228 (5), 215 (100). Materials and Methods for Photophysical Studies. 1-Ethylpyrene (1-EP) was purchased from Molecular Probes, Inc. (Eugene, OR) and it was used without further purification. The following solvents were used as received: cyclohexane (CH, spectrophotometric grade, 99+%, Acros), dichloromethane (DCM, certified ACS, stabilized with amylene, 99.9%, Fisher), ethanol (EtOH, dehydrated 200 proof, Pharmaco), acetonitrile (ACN, HPLC grade, 99.93+%, Sigma-Aldrich), dimethyl sulfoxide (DMSO, certified ACS, 99.9%, Fisher). Supercritical fluid chromatography grade CO2 was obtained from Scott Speciality Gases (Plumsteadville, PA). All stock solutions were refrigerated in the dark at 4 °C. Instrumentation. All absorbance measurements were carried out by using a Spectronic Model 1201 UV-vis spectrophotometer (Milton Roy) with a 1 nm spectral band-pass. The scan rate was set at 200 nm/min, and all spectra were carefully blank corrected. Steady-state excitation and emission experiments were performed with an SLM-AMINCO model 48000 MHF fluorometer (Spectronic Instruments) using a 450 W xenon arc lamp as the excitation source and single-grating monochromators served as wavelength selection devices. All emission and excitation spectra were background corrected by using appropriate blanks. Most of the time-resolved fluorescence measurements were carried out by using an IBH model 5000W SAFE timecorrelated single photon counting fluorescence lifetime instrument with a N2-filled flashlamp, operating at a 40 kHz repetition rate, as the excitation source. The excitation wavelength was set to 337 nm and the excited-state fluorescence intensity decay data were acquired at the desired emission wavelengths. Singlegrating monochromators were used for wavelength selection. All time-resolved intensity decay profiles were recorded under magic angle polarization conditions. The typical time resolution for an experiment was 0.47 ns/channel with 1024 total channels within the multichannel analyzer (MCA). To avoid pulse pile up, the count rate at the reference and emission detectors was maintained at less than 2% of the flash lamp repetition rate. Data were acquired until there were at least 104 counts in the peak MCA channel. The instrument response and the monomer and excimer decay traces were always acquired simultaneously. Several experiments were also performed on a few of the scCO2 samples by using phase-modulation fluorescence. For these experiments we used the SLM-AMINCO model 48000 MHF fluorometer and a He-Cd laser (325 nm) for excitation. The high-pressure system consists of an Isco model 260-D microprocessor-controlled syringe pump that is purged and charged with CO2, stainless steel tubing and plumbing, a highpressure optical cell46 with fused silica windows and a stainless steel body, a home-built temperature controller ((0.1 °C), and a pressure monitoring system ((0.2 bar). The high-pressure cell internal volume is 3.5 mL. Data Analysis. Although numerous methods are available for recovering the kinetic terms from IM(t) and IE(t), we have opted to use a global analysis strategy.47 In this approach, the monomer and excimer decay traces are simultaneously fit to a given model (e.g., Scheme 1) such that a self-consistent set of kinetic terms are fit for and recovered. We have used the global analysis strategy here because it provides a more rigorous set

Photophysics of 1-Py(CH2)10COO(CH2)61-Py of fitting criteria, it yields all the kinetic parameters directly, and the accuracy of the recovered kinetic terms is improved significantly because orthogonal data sets are being modeled simultaneously.47 All experiments were performed at least three times on different days by using fresh preparations. The average of all results is reported and uncertainties reflect ( one standard deviation. Sample Preparation in Organic Solvents. All 1-Py(CH2)10COO(CH2)61-Py and 1-EP samples were made as 1 µM solutions and they were transferred to standard 1 cm2 quartz cuvettes that are fitted with a freeze-pump-thaw assembly. Before acquiring any spectroscopic data, all liquid solutions were freeze-pump-thaw degassed four times to remove dissolved O2 (O2 is a ubiquitous quencher of molecular fluorescence and has differential quenching efficiencies toward excited monomer and excimer species).48-51 Sample Preparation in scCO2. A 10-50 µL aliquot of 1-EP or 1-Py(CH2)10COO(CH2)61-Py dissolved in ethanol was pipetted directly into the high-pressure optical cell such that the final concentration within the high-pressure cell was ∼10 µM. The ethanol solvent was driven from the cell by using a gentle stream of argon gas. A Teflon-coated stir bar was placed inside the optical cell. Prior to charging the optical cell with CO2, the cell and all its plumbing were carefully flushed with CO2 (2-4 bar) to remove any residual O2 occupying the lines, connections, and the cell. After removing residual O2, the cell was heated to 35 °C. CO2 was then pumped into the cell until the desired starting pressure was reached. The cell contents were stirred continuously and equilibrium was established within 20-30 min of a pressure change. Equilibrium was evidenced by the pump flow rate going to zero ((3 µL/min) and by steady-state emission spectra that were stable ((4% RSD) over time. Typical experiments were performed from low to high pressure. Additional experiments were performed by loading the 1-Py(CH2)10COO(CH2)61-Py samples into the cell, heating the cell to the experimental temperature, pressurizing the cell initially to 250-300 bar, and then decompressing the cell along an isotherm. In these experiments the sample was sometimes maintained at 250-300 bar for many days prior to the decompression step. The results from “high to low” and “low to high” runs yielded statistically equivalent spectra and timeresolved intensity decay traces. Results and Discussion Steady-State Spectroscopy of 1-Py(CH2)10COO(CH2)61Py Dissolved in Organic Liquids. Figure 2 presents the normalized steady-state emission spectra of 1-Py(CH2)10COO(CH2)61-Py dissolved in each liquid solvent. All spectra have been normalized to the highest energy peak in the monomer portion of the spectrum. The monomer emission with its vibronic features is observed between 370 and 430 nm in the liquids. The excimer emission is clearly seen in the 440-550 nm region and the excimer-to-monomer intensity ratio (E/M) is a strong function of the solvent. At this concentration level there is no detectable excimer emission from 1-EP in any of these liquids (results not shown). The latter result argues that the observed excimer for 1-Py(CH2)10COO(CH2)61-Py arises only from an intramolecular process. Also note the relative magnitude of the peak at ∼376 nm relative to the peak at ∼394 nm (i.e., I376/I394 > 1). One must be very cautious when interpreting “excimer” emission and carefully distinguish between classic excimers that form in and exist only in the excited state and related species

J. Phys. Chem. B, Vol. 106, No. 7, 2002 1823

Figure 2. Normalized steady-state emission spectra (λex ) 337 nm) of 1-Py(CH2)10COO(CH2)61-Py dissolved in organic solvents. Spectra have been normalized to the highest energy emission peak.

TABLE 1: Excimer-to-Monomer Intensity Ratio (E/M ) I480/I376) for 1-Py(CH2)10COO(CH2)61-Py Dissolved in Organic Liquids, with Solvent Dielectric Constant (E), Viscosity (η), and Kamlet-Taft Parameters r, β, and π* at 20 °C solvent ACN CH DCM DMSO EtOH a

E/M

a,b

2.03 ( 0.05 37.5 2.66 ( 0.05 2.02 1.59 ( 0.04 9.08 0.70 ( 0.04 46.66 1.83 ( 0.05 24.55

η (mPa s)a,b

‚η

Rc

βc

π* c

0.375 0.980 0.449 2.20 1.078

14.1 1.98 4.08 102.7 26.5

0.19 0.00 0.30 0.00 0.83

0.31 0.00 0.00 0.76 0.77

0.75 0.00 0.82 1.00 0.54

Reference 56. b Reference 57. c Reference 58.

that look like an excimer but are formed from preassociated inter- or intramolecular pyrene species prior to optical excitation.32 Several methods exist to discriminate between these socalled static dimers and true dynamic excimers, including acquisition of the electronic absorbance spectra32 and acquisition of the emission wavelength-dependent excitation scans at several emission wavelengths.52 If the 1-Py(CH2)10COO(CH2)61-Py system is described by a purely intramolecular excimer, the normalized emission wavelength-dependent excitation scans will be independent of the emission wavelength.52 If these excitation spectra are shifted relative to one another, this is a direct indication of inter- or intramolecular preassociation between the pyrene residues in the ground state. Figure 3 presents a series of normalized emission wavelengthdependent excitation scans for 1-Py(CH2)10COO(CH2)61-Py dissolved in each liquid solvent at the monomer (s) and excimer (- -) emission maxima. Within our measurement precision, these spectra demonstrate that there is no evidence for inter- or intramolecular pyrene residue preassociation in any of these liquids. Identical results (not shown) were seen when we changed the 1-Py(CH2)10COO(CH2)61-Py concentration between 0.1 and 50 µM. These results confirm that the observed excimer emission is purely an excited-state phenomenon in each of the liquid solvents and there is no detectable ground-state dimer formation prior to optical excitation. There are several correlations that have been used to assess the effects of solvent on the extent of excimer formation (i.e., E/M).13-18,22,42,44,53-55 In Table 1 we report the 1-Py(CH2)10COO(CH2)61-Py E/M in each liquid solvent along with solvent

1824 J. Phys. Chem. B, Vol. 106, No. 7, 2002

Pandey et al.

Figure 3. Emission wavelength-dependent excitation spectra of 1-Py(CH2)10COO(CH2)61-Py dissolved in organic solvents when emission is monitored at 376 nm (s) or 476 nm (- -).

dielectric constant ()56,57 and viscosity (η)56,57 as well as the Kamlet-Taft58 solvent acidity (R), basicity (β), and polarity/ polarizability (π*) indices. A plot of E/M vs ‚η is linear, but the corresponding r2 value is poor (0.73). Although removing a solvent is not a particularly attractive strategy, there is a significant improvement in r2 (0.94), if we omit the DCM data. In light of the fact that DCM is chlorinated, leading to the possibility of differential excimer-monomer quenching efficiency, the data point for DCM might not be expected to correlate with the other solvents. By using the Kamlet-Taft parameters, we sought to treat all five solvents together and use multiple linear regression to fit our experimental E/M data to a Kamlet-Taft function: E/M )

aR + bβ + cπ + d. The values of the recovered a, b, c, and d coefficients are as follows: 0.49, -0.73, -1.32, 2.72, respectively, and the fit is reasonably good (r2 ) 0.88). These results argue that an increase in solvent acidity causes an increase in the 1-Py(CH2)10COO(CH2)61-Py E/M, an increase in solvent basicity decreases E/M, and an increase in solvent polarity/ polarizability causes E/M to decrease. Time-Resolved Fluorescence of 1-Py(CH2)10COO(CH2)61Py Dissolved in Organic Liquids. On the basis of literature precedent,11-43,53-55 we use the 1-EP excited-state fluorescence lifetime (τ) to determine the 1-Py(CH2)10COO(CH2)61-Py monomer decay rate (kM ) 1/τ). The excited-state intensity decay traces for 1-EP dissolved in each liquid are best described

Photophysics of 1-Py(CH2)10COO(CH2)61-Py

J. Phys. Chem. B, Vol. 106, No. 7, 2002 1825

Figure 4. Time-resolved fluorescence intensity decay data for the 1-Py(CH2)10COO(CH2)61-Py dissolved in CH at 376 nm (monomer, part A) and 476 nm (excimer, part B), instrument response function (IRF), and the best fit to a double exponential decay model. Residuals are shown in parts C and D. The model parameters are collected in Table 3.

Figure 5. Time-resolved intensity decay data for the 1-Py(CH2)10COO(CH2)61-Py dissolved in CH at 376 nm (monomer, part A) and 476 nm (excimer, part B), instrument response function (IRF), and the best fit triple exponential decay model. Residuals are shown in parts C and D. The model parameters are collected in Table 3.

TABLE 2: Excited-fluorescence Lifetimes (τ) and Decay Rates (kM ) 1/τ) for 1-Ethylpyrene Dissolved in Organic Liquids at 20 °C solvent

τ (ns)

kM (10-6, s-1)

acetonitrile (ACN) cyclohexane (CH) dichloromethane (DCM) dimethyl sulfoxide (DMSO) ethanol (EtOH)

187.6 ( 7.7 228.3 ( 10.4 102.0 ( 2.9 131.6 ( 3.8 215.5 ( 8.8

5.33 ( 0.22 4.38 ( 0.20 9.80 ( 0.28 7.60 ( 0.22 4.64 ( 0.19

by a single-exponential decay model (χ2 e 1.1). Table 2 presents the 1-EP τ and kM values. These kM values are used within a

global analysis strategy47 to recover the 1-Py(CH2)10COO(CH2)61-Py excimer formation and dissociation rates (vide infra). The time-resolved fluorescence intensity decay traces for 1-Py(CH2)10COO(CH2)61-Py in each liquid solvent were acquired at 376 nm (monomer) and 480 nm (excimer). Typical data for 1-Py(CH2)10COO(CH2)61-Py dissolved in CH at 20 °C, the lamp instrument response function (IRF), and best fits to double and triple exponential models are shown in Figures 4 and 5, respectively. Based on previous studies of intramolecular excimer formation in organic liquids,11-43 one would not expect the 1-Py(CH2)10COO(CH2)61-Py intensity decay traces to follow

1826 J. Phys. Chem. B, Vol. 106, No. 7, 2002

Pandey et al.

TABLE 3: Recovered Intensity Decay Parameters for 1-Py(CH2)10COO(CH2)61-Py Dissolved in Organic Liquids at 20 °Ca λem (nm)

τ1b (ns)

τ2b (ns)

376 480 376 480 376 480 376 480

1.7 1.7 2.4 2.4 2.2 2.2 1.5 1.5

58.0 58.0 55.2 55.2 54.5 54.5 2.6 2.6

376 480 376 480 376 480 376 480

4.4 4.4 5.7 5.7 5.7 5.7 5.4 5.4

65.3 65.3 55.1 55.1 54.7 54.7 9.1 9.1

376 480 376 480 376 480 376 480

6.3 6.3 7.3 7.3 7.4 7.4 7.0 7.0

58.4 58.4 50.4 50.4 49.4 49.4 8.3 8.3

376 480 376 480 376 480 376 480

11.6 11.6 13.3 13.3 13.1 13.1