Chloride Ion-Pairing with Ru(II) Polypyridyl Compounds in

Article pubs.acs.org/JPCA

Chloride Ion-Pairing with Ru(II) Polypyridyl Compounds in Dichloromethane William M. Ward, Byron H. Farnum, Maxime Siegler, and Gerald J. Meyer* Department of Chemistry, Johns Hopkins University, 3400 North Charles Street, Baltimore, Maryland 21218, United States S Supporting Information *

ABSTRACT: Chloride ion-pairing with a series of four dicationic Ru(II) polypyridyl compounds of the general form [Ru(bpy)3−x(deeb)x](PF6)2, where bpy is 2,2′-bipyridine and deeb is 4,4′-diethylester-2,2′-bipyridine, was observed in dichloromethane solution. The heteroleptic compounds [Ru(bpy)2(deeb)]2+ and [Ru(bpy)(deeb)2]2+ were found to be far less sensitive to ligand loss photochemistry than were the homoleptic compounds [Ru(bpy)3]2+ and [Ru(deeb)3]2+ and were thus quantified in most detail. X-ray crystal structure and 1H NMR analysis showed that, when present, the C-3/C-3′ position of bpy was the preferred site for adduct formation with chloride. Ion-pairing was manifest in UV−visible absorption spectral changes observed during titrations with TBACl, where TBA is tetrabutyl ammonium. A modified Benesi−Hildebrand analysis yielded equilibrium constants for ion-pairing that ranged from 13 700 to 64 000 M−1 and increased with the number of deeb ligands present. A Job plot indicated a 2:1 chloride-to-ruthenium complex ratio in the ion-paired state. The chloride ion was found to decrease both the excited state lifetime and the quantum yield for photoluminescence. Nonlinear Stern−Volmer plots were observed that plateaued at high chloride concentrations. The radiative rate constants decreased and the nonradiative rate constants increased with chloride concentration in a manner consistent with theory for radiative rate constants and the energy gap law. Equilibrium constants for excited state ion-pairing abstracted from such data were found to be significantly larger than that measured for the ground state. Photophysical studies of hydroxide and bromide ion-pairing with [Ru(bpy)2(deeb)]2+ are also reported.

■

INTRODUCTION Chemists have been fascinated by ion-pairs for some time, particularly when one or more of the ions is a transition metal coordination compound.1−5 Such ion-pairs have practical importance in catalysis6−8 and chromatography9 as well as in some types of batteries10 and solar cells.11,12 Their behavior is also of fundamental importance in its own right. A variety of techniques including X-ray crystallography, electrical conductivity, NMR spectroscopy, and UV−visible spectroscopy have been used to characterize a wide range of ion-pairs. Spectroscopic assays have also been utilized to determine whether ion-pairing occurs at specific sites within the cation and/or anion. In one particularly novel study, regiospecific reactivity of a cobalt coordination compound with chloride was observed.13 Herein we report studies of chloride ion-pairing with dicationic Ru(II) polypyridyl compounds in dichloromethane (DCM) solvent designed to identify whether preferred ion-pairing sites exist and the influence of such interaction on excited state decay. Contact ion-pairs between redox active ions can result in outer-sphere charge transfer absorption bands that are of great utility for fundamental electron transfer studies and can also be exploited for improved solar light harvesting. We recently reported the oxidation of iodide by the metal-to-ligand charge transfer (MLCT) excited states of Ru(II) polypyridyl compounds in dichloromethane where ion-pairing was evident.14−16 Ion-pair formation was found to greatly facilitate iodide photooxidation relative to that measured in more polar acetonitrile © 2013 American Chemical Society

solutions where ion-pairing appeared to be absent. X-ray crystallographic studies provided evidence for specific iodide adducts in the solid state that could give rise to new mechanisms for iodide oxidation and I−I bond formation.14 However, analysis of the solution UV−visible absorption spectra was complicated by what appeared to be outer-sphere charge transfer transitions that overlapped in energy with a highly perturbed MLCT absorption. It was therefore difficult to assign the absorption changes that accompanied iodide ion-pairing. The chlorine atom has a reduction potential (X•/X−) that is much more positive than that of the iodine atom and the excited state potential (2+*/+) of the ruthenium compounds. Chloride was therefore expected to be redox inactive under these conditions. Thus we undertook a study to characterize ion-pairing with chloride whose formal reduction potential precludes the appearance of outer-sphere charge transfer bands in the visible region or excited state electron transfer chemistry. A series of four Ru(II) polypyridyl compounds based on 2,2′bipyridine and/or 4,4′-(CO2CH2CH3)2-2,2′-bipyridine, Scheme 1, were synthesized and characterized. Ion-pairing with chloride was found to have a significant influence on the UV−vis absorption spectra of the compounds in dichloromethane. These ions were also found to significantly influence the MLCT excited Received: May 16, 2013 Revised: August 5, 2013 Published: August 6, 2013 8883

dx.doi.org/10.1021/jp404838z | J. Phys. Chem. A 2013, 117, 8883−8894

The Journal of Physical Chemistry A

Article

Scheme 1. Series of Four Compounds and Abbreviations Used

25.0 ± 0.1 °C were measured as various anions in argon purged DCM were titrated into the sample solution. 1 H NMR. Ruthenium compound was dissolved in approximately 0.50 mL of CD2Cl2 with added tetramethylsilane to make an approximately 3 mM solution, and spectra were recorded as aliquots of TBACl in CD2Cl2 were added. Spectra were measured on a Bruker Avance 400 MHz FT-NMR spectrometer at 298 + 0.3 K. The mole ratio of ruthenium compound to TBACl was calculated using the integration of the ester methylene peak and the terminal methyl peak from tetrabutylammonium. X-ray Crystallography. Crystals of [Ru(bpy)2(deeb)]Cl2 were grown by addition of a 10-fold molar excess of tetrabutylammonium chloride to [Ru(bpy)2(deeb)] (PF6)2 in DCM. The solution was placed in an open vial inside a larger vial with diethylether. After 7 days, red-orange platelike crystals had grown. A suitable crystal was mounted in oil on the end of a glass fiber and used for X-ray crystallographic analysis. All reflection intensities were measured at 100(2) K using a SuperNova diffractometer (equipped with Atlas detector) with Cu Kα radiation (mirror optics, λ = 1.541 78 Å) under the program CrysAlisPro (Version 1.171.36.24 Agilent Technologies, 2012). The program CrysAlisPro (Version 1.171.36.24 Agilent Technologies, 2012) was used to refine the cell dimensions. Data reduction was done using the program CrysAlisPro (Version 1.171.36.24 Agilent Technologies, 2012). The structure was solved with the program SHELXS-97 (Sheldrick, 2008) and was refined on F2 with SHELXL-97 (Sheldrick, 2008). Analytical numeric absorption corrections based on a multifaceted crystal model were applied using CrysAlisPro (Version 1.171.36.24 Agilent Technologies, 2012). The temperature of the data collection was controlled using the system Cryojet (manufactured by Oxford Instruments). The H atoms (unless specified) were placed at calculated positions using the instructions AFIX 23, AFIX 43, or AFIX 137 with isotropic displacement parameters having values 1.2 or 1.5 times Ueq of the attached C atoms.

state lifetime, quantum yield, and radiative and nonradiative rate constants. Specific sites for ion-pairing were identified that exchanged rapidly on the NMR time scale. Comparative photophysical studies of bromide and hydroxide ion-pairing are also reported.

■

EXPERIMENTAL SECTION

Materials. Argon gas (Airgas, 99.99%), [Ru(bpy)3]Cl2·6H2O (Aldrich, 99.95%), ammonium hexafluorophosphate (Acros, 99.5%), tetrabutylammonium chloride (TBACl; Fluka, >97%), tetrabutylammonium bromide (TBABr; Aldrich, 99%), tetrabutylammonium hexafluorophosphate (TBAH, Fluka, >98%), potassium hydroxide (Fischer, >87%), 18-crown-6 (Aldrich, 99%), triethylamine (TEA; Fluka, >99.5%), and dichloromethane (DCM; EMD, >99.8%) were used as received without further purification. [Ru(bpy)3](PF6)2 was prepared by ion exchange of [Ru(bpy) 3 ]Cl 2 ·6H 2 O with NH 4 PF 6 . [Ru(bpy)2(deeb)](PF6)2 was available from previous studies.15,16 [Ru(deeb)2(bpy)](PF6)2 and [Ru(deeb)3](PF6)2 were prepared by a modified literature method.17 Measurements. Steady State Absorption. UV−vis absorption spectra were obtained on a Varian Cary 50 UV−vis spectrophotometer at room temperature. Ten scans at approximately 1 nm resolution were averaged. The spectra were volume corrected for the addition of salt solutions. Typically 10−300 μL of salt solution was added to 5.00 mL of ruthenium compound solution. Job plot measurements were conducted with 1.00 mL of solution in a cuvette with a 0.20 cm path length. Nanosecond Transient Absorption. Argon purged DCM solutions of approximately 20 μM ruthenium compound both with and without TBACl were measured at room temperature as described previously.18 The responses to 40 laser pulses were averaged with a fluence of approximately 2 mJ/cm2. Steady State Photoluminescence. A Spex Fluorolog with a 450 W Xe lamp was utilized for steady-state photoluminescence (PL) measurements. PL spectra were acquired at 25.0 ± 0.1 °C in argon purged DCM and corrected for lamp intensity fluxuations and detector response. Samples were illuminated at the MLCT peak wavelength. Five scans at 2 nm resolution were averaged. Comparative actinometry using [Ru(bpy)3]Cl2 in water (φem = 0.042) was used to measure PL quantum yields.19 Time Resolved Photoluminescence. Excited state lifetimes were measured using a nitrogen-dye laser system described previously with a 532 nm excitation wavelength.18 Photoluminescence intensity was monitored at the PL maximum wavelength. Argon purged DCM solutions of approximately 20 μM [Ru(bpy)2(deeb)](PF6)2 or [Ru(bpy)(deeb)2](PF6)2 at

■

RESULTS The chloride salt of [Ru(bpy)2(deeb)]2+ was characterized by Xray crystallography, Table 1. The Mercury 3.1 program20 was used to determine all the chloride ions located within the sum of the van der Waals radii21 to [Ru(bpy)2(deeb)]2+. The distances from the chloride ions to the closest point on the ruthenium compound are listed in Table 2. It was found that the closest chloride counterion was situated approximately 2.5 and 2.8 Å from the bpy C-3H and C-3′H, respectively, and the next closest chloride ions were located approximately 2.8 Å from the deeb C6H. On the other hand, the deeb carbonyl carbon atoms were 8884



Article

assigned to C-6H. Since the pyridine (3,4) and (4,5) coupling constants are both 7−9 Hz, it was expected that the peak for C4H would be split by approximately the same amount by both C3H and C-5H, thereby producing a triplet pattern, whereas the (4,5) and (5,6) coupling constants were significantly different, producing a doublet of doublets pattern. Therefore, the peak at 7.474 ppm was assigned to C-5H and the peak at 8.074 ppm to C4H. For [Ru(deeb)3]2+, the peak at 8.9932 ppm had two coupling constants that were too small to be assigned to adjacent hydrogens; therefore, this peak was assigned to C-3H. The (5,6) is approximately 6 Hz and the peak with the larger second coupling constant (1.7 Hz) was assigned to the hydrogen closer to C-3H, so this peak was assigned to C-5H and the smaller second coupling constant (0.6 Hz) was assigned to C-6H. The literature pyridine values for (3,5) and (3,6) are 1−2 Hz and 0−1 Hz, in agreement with this analysis.25 As expected, the peaks in the spectrum of [Ru(bpy)2(deeb)]2+ were only shifted slightly from the homoleptic [Ru(bpy)3]2+ and [Ru(deeb)3]2+, and the same assignment process as above was used. However, with [Ru(bpy)2(deeb)]2+, symmetric hydrogens on the same bpy ligand, but different pyridyl rings, were no longer equivilent and had slightly different chemical shifts. Cross-peaks in an NOE spectrum indicated that the peak for bpy C-6H, where the bpy C6 to bpy C-6H bond points at the center of a bpy pyridyl ring, was probably downfield of the bpy C-6′H, which points at a deeb pyridyl ring. Likewise, the bpy C-5H was probably downfield of the bpy C-5′H. The NOE data did not differentiate bpy C-4H from bpy C-4′H. The 1H NMR spectra were then measured for [Ru(bpy)3]2+ and [Ru(deeb)3]2+ with 1 equiv of added TBACl. For [Ru(bpy)3]2+, the C-3H peak was significantly shifted downfield, Figure 2A; however, for [Ru(deeb)3]2+ the C-6H peak was shifted downfield, Figure 2B, while the other hydrogens displayed only minor shifts. The 1H NMR spectra were also measured for [Ru(bpy)2(deeb)]2+ as TBACl was titrated into the solution, from 0 to 42 equiv. The maximum shifts of all the hydrogens, upon addition of 1 or 42 equiv of TBACl, are shown in Figure 2C. As with the homoleptic compounds, the bpy C-3H peak shifted significantly downfield upon addition of TBACl while the deeb C-3H peak did not shift appreciably, Figure 3. Additionally, the changes in chemical shift (Δppm) for the bpy C-3H and the deeb C-6H are plotted in Figure 4A and 4B. Two equilibria were used to model the results, eq 1, where R2+ is [Ru(bpy)2(deeb)]2+, R0 is

Table 1. Crystal Parameters for [Ru(bpy)2(deeb)]Cl2 empirical formula formula weight crystal color habit temperature radiation space group. monoclinic unit cell dimensions

Z calculated density absorbance coefficient R1/wR2 [I > 2σ(I)] R1/wR2 [all refl.]

C36Cl2H32N6O4Ru 1124.35 orange-red plate 0.51 × 0.28 × 0.03 mm3 100 K Cu Kα λ = 1.541 78 Å P2/c (no. 13) a = 11.8518(3) b = 12.9483(3) c = 17.0261(6) Å β = 112.244(4)° V = 2418.39(12) Å3 2 1.544 g cm−3 μ = 8.086 mm−1 0.0578/0.1551 0.0642/0.1607

Table 2. Distances from Chlorides within Sum of van der Waals Radii atoms

distance (Å)

atoms

distance (Å)

bpy C-3H bpy C-3′H deeb C-6H bpy C-5H

2.519 2.782 2.770 2.788

bpy C-6H bpy C-6′H deeb C-6 bpy C-5′

2.867 2.881 3.364 3.439

located approximately 5.5 Å from the nearest chloride, and this chloride was also only 2.7 Å from the deeb C-6H. Chloride adducts with bpy C-3H and C-3′H are in agreement with other literature reports for tris-bipyridyl type compounds.13,22−24 The distances to the bpy C-3H and C-3′H refer to the same chloride, as seen in Figure 1, while all the other entries refer to separate chloride ions. The 1H NMR spectra of [Ru(bpy)3]2+, [Ru(bpy)2(deeb)]2+, and [Ru(deeb)3]2+ in CD2Cl2 solution were measured, and peak assignments were made as listed in Table 3. For [Ru(bpy)3]2+, the two doublets at 7.716 and 8.425 ppm were assigned to C-3H and C-6H with only one adjacent hydrogen each. The literature value for the coupling constant between hydrogens at position C3 and C-4 (notated as (3,4)) in pyridine is 7−9 Hz, while for (5,6) the literature value is 5−6 Hz.25 Therefore, the peak at 8.425 ppm was assigned to C-3H, and the peak at 7.716 ppm was

Figure 1. ORTEP diagram for [Ru(bpy)2(deeb)]2+ showing the closest four chloride ions. The interactions of the chlorides (light green dashed lines) with the bpy C-3H hydrogens are more clearly seen in the viewing angle on the left, and the interactions with the deeb C-6H are more clearly seen in the view on the right. Ellipsoids drawn at the 50% probability level. Note: the electrically neutral compound only has two chlorides. 8885



Article

Table 3. 1H NMR Spectra and Assignments complex

chemical shift (ppm)

splitting pattern

coupling constant (Hz)

integration ratio

assignment

[Ru(bpy)3]2+

7.474 7.716 8.074 8.425 1.4244 4.4877 7.9267 7.9966 8.9932 1.4290 4.4927 7.4754 7.5115 7.6689 7.7186 7.9401 7.9838 8.0872 8.1007 8.4387 8.9825

dd d t d t q dd dd dd t q ddd ddd ddd ddd dd dd dt dt ddd dd

7.6, 5.2 4.2 7.3 8.1 7.1 7.1 6.2, 0.7 5.8, 1.7 1.7, 0.6 7.0 7.1 7.6, 5.8, 1.4 7.6, 5.8, 1.4 5.6, 1.4, 0.6 5.6, 1.4, 0.6 5.8, 0.6 5.8, 1.7 7.9, 1.5 7.9, 1.5 8.2, 1.2 1.8, 0.7

6.0a 6.0 6.0 6.0 18.0a 12.2 5.7 5.9 5.8 6.1 4.0a 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 4.0 1.9

C-5H, 5′H C-6H, 6′H C-4H, 4′H C-3H, 3′H ester CH3 ester CH2 C-6H, 6′H C-5H, 5′H C-3H, 3′H Ester CH3 Ester CH2 bpy-C-5′H bpy-C-5H bpy-C-6′H bpy-C-6H deeb-C-6H deeb-C-5H bpy-C-4′H bpy-C-4H bpy-C-3H deeb-C-3H

[Ru(deeb)3]2+

[Ru(bpy)2(deeb)]2+

a

Used as reference for integration ratios.

M−1. The NMR shifts at low concentrations, Figure 4 bottom right, did not satisfactorily fit any of the attempted models. K1

R2 + + X− ⇌ [R2 +, X−]+ K2

[R2 +, X−]+ + X− ⇌ [R2 +, 2X−] Δppm =

Δδmax [R2 +, X−]+ [R2 +, 2X−] + Δδmax 2 R0 R0

(1)

(2)

The electronic absorption spectra (UV−vis) of the ruthenium compounds in dichloromethane were altered when TBACl was added, as seen in Figure 5A. The metal-to-ligand charge transfer (MLCT) absorption maximum of [Ru(bpy)2(deeb)]2+ at 478 nm was red-shifted and decreased in intensity, while absorbances at approximately 425 nm displayed a slight growth and those at 362 nm were red-shifted with small changes in intensity. The deeb π → π* and the bpy π → π* absorption bands in the ultraviolet region both decreased in intensity without a shift in energy (data not shown). The MLCT absorption spectra changed monotonically with added chloride, and the isosbestic points were lost at the highest concentrations used. These spectral changes plateaued at higher Cl− concentrations such that the spectra measured with 7 and 10 equiv of Cl− were very similar, Figure 5A inset. The spectral changes observed with TBACl were reversed when a 10-fold excess of TBAPF6 was added. The presence of 10 equiv of TBAPF6 in a solution of just [Ru(bpy)2(deeb)]2+ produced only negligibly small changes in intensity without any measurable wavelength shifts. There were qualitatively similar changes for [Ru(bpy)(deeb)2]2+ and [Ru(deeb)3]2+. However, [Ru(bpy)3]2+ displayed smaller changes and a net blue shift in the MLCT maximum with increased TBACl. Specific details of these spectral changes, as well as related spectral data observed with added TBABr and KOH, are listed in the Supporting Information. Benisi and Hildebrand have previously described how spectroscopic titration data can be analyzed to abstract equilibrium constants.26 In its original form, the Benesi−

Figure 2. 1H NMR shifts upon addition of 1 equiv of TBACl to (A) [Ru(bpy)3] (PF6)2 or (B) [Ru(deeb)3] (PF6)2 in CD2Cl2. A positive shift is downfield, and a negative shift is upfield. (C) 1H NMR shifts upon addition of 1 equiv (top numbers) or 42 equiv (bottom numbers) of TBACl to [Ru(bpy)2(deeb)]2+(PF6)2 in CD2Cl2.

the initial concentration of [Ru(bpy)2(deeb)]2+ before ionpairing, and X− is Cl−. The system of four equations (two equilibria equations, mass balance of ruthenium compound and mass balance of chloride) was solved for the four concentration variables [R2+], [X], [[R2+,X−]+], and [[R2+,2X−]] using assumed values for K1 and K2. The NMR shifts were fit to eq 2 where Δδmax is the asymptotic maximum shift observed by varying K1 and K2. In order to reduce the number of free-floating variables, it was assumed that the singly ion-paired species produced half the change in chemical shift as the doubly ion-paired species. The equilibrium constants extracted were K1 = 480 M−1 and K2 = 70 8886



Article

Figure 3. The 1H NMR of [Ru(bpy)2(deeb)] (PF6)2 in CD2Cl2 with the indicated equivalents of TBACl. (A) The bpy C-3 proton resonance. (B) The deeb C-3 proton resonance.

Figure 4. Changes in the 1H NMR spectral peaks for bpy C-3H (black squares) and deeb C-6H (red triangles) of approximately 3 mM [Ru(bpy)2(deeb)] (PF6)2 in CD2Cl2 as TBACl is titrated into the solution. Calculated fits are depicted as green lines. (A) Full concentration range. (B) Close-up of the first 1.0 mM of added TBACl.

Figure 5. (A) UV−vis absorption spectrum of a 26 μM [Ru(bpy)2(deeb)](PF6)2 solution in CH2Cl2 titrated with TBACl. The red arrows indicate the direction of change with added TBACl. The inset shows the absorbance change measured at 448 nm with an overlaid fit to a modified Benisi− Hildebrand analysis from which an equilibrium constant of Kobs = 17 600 ± 700 M−1 was abstracted (left). (B) Job plot for [Ru(bpy)2(deeb)](PF6)2 with TBACl. The red line is a calculated plot based on two equilibria with K1 = 130 000 M−1 and K2 =60 000 M−1.

R2 + + X− ⇌ [R2 +, X−]+

Hildebrand equation assumed a large excess of one of the constituents which was not a valid assumption for this study. Therefore, the relation was reformed as described below. An analysis of the chloride titration data with [Ru(bpy)2(deeb)]2+ is shown as the inset of Figure 5A. Due to the lack of structure in the data, a single equilibrium was assumed, eq 3. The terms in the equilibrium equation can be reformed and then solved for the concentration of the ion-paired species, eq 4, where R0 is the initial concentration of [Ru(LL)3]2+, X0 is the initial concentration of the added anion, and Kobs is the equilibrium constant for ion-pairing; only one of the roots of the quadratic makes physical sense.

[R2 +, X−] =

R2 + = [Ru(LL)3 ]2 +

(3)

1⎛ 1 ⎞ ⎜R 0 + X 0 + ⎟ 2⎝ Kobs ⎠ ±

2 1⎛ 1 ⎞ ⎜R 0 + X 0 + ⎟ − R 0X 0 4⎝ Kobs ⎠

(4)

Using eq 4, the Benesi−Hildebrand equation then becomes eq 5, where ΔA is the change in absorbance at 448 nm and Δε is the difference in extinction coefficients between [Ru(bpy)2(deeb)]2+ 8887



Article

Table 4. Photophysical Results UV−vis

a

lifetime

compd

anion

λmax(Abs) (nm)

Δε (M−1 cm−1)

Kobs (M−1)

[Ru(bpy)3](PF6)2 [Ru(bpy)2(deeb)](PF6)2 [Ru(bpy)(deeb)2](PF6)2 [Ru(deeb)3](PF6)2 [Ru(bpy)2(deeb)](PF6)2 [Ru(bpy)2(deeb)](PF6)2

Cl− Cl− Cl− Cl− Br− OH−

453 448 479 448 448 448

1500 ± 20 2340 ± 20 1500 ± 30 1800 ± 60 1430 ± 30 1650 ± 50

13700 ± 700 17600 ± 600 19700 ± 1800 64000 ± 9000 17000 ± 2000 2100 ± 400

Kobs,ES (M−1) a

81000 ± 9000 47000 ± 3000 a

118000 ± 6000 51000 ± 4000

Not determined due to irreversible photochemistry.

to [Ru(bpy)2(deeb)]2+ are shown in the inset of Figure 6. Stern− Volmer plots28 were constructed from the calculated lifetimes, Figure 7A. With excess TBAPF6 the lifetime decreased by up to 9%, while the other salts resulted in decreases of 25−30%. The Stern−Volmer plots were nonlinear as the lifetimes became independent of the salt concentration at high salt concentrations. There was no clear evidence of static quenching as the PL amplitudes immediately following the laser pulse displayed no discernible dependence on salt concentration. The equation for the average lifetime of a mixture of two photoluminescent species was reformed into eq 6.28 Using eq 4 for the concentration of the ion-paired species, the changes in lifetime were fit to eq 6 and an equilibrium constant for ion-pairing with the excited state, Kobs,ES, was extracted. Table 4 compares the equilibrium constants obtained from UV−vis and lifetime measurements.

and the ion-paired species and l is the path length. The experimental data were fit using a least-squares analysis to find the best values of Kobs and Δε. The equilibrium constants for ruthenium compounds and anions are listed in Table 4. ΔA = lΔε[R2 +, X−]

(5)

In order to determine the stoichiometry of ion-pairing, a Job plot27 was constructed for the addition of TBACl to [Ru(bpy)2(deeb)]2+, Figure 5B, where the sum of the concentrations of [Ru(bpy)2(deeb)]2+ and TBACl was kept constant at 207 ± 11 μM. The maximum change in absorbance at 448 nm was at a [Ru(bpy)2(deeb)]2+ mole fraction of approximately 0.3 which corresponds to a 1:2 ratio for [Ru(bpy)2(deeb)]2+:TBACl. For comparison, the solid red line is a calculated plot based on two equilibria with K1= 130 000 M−1 and K2 = 60 000 M−1. Light excitation into the MLCT absorption band resulted in room temperature photoluminescence, PL, Figure 6. The PL

⟨τ ⟩ =

α1τ12 + α2τ22 α1τ1 + α2τ2 [R2 +] R0

α1 =

α2 =

[R2 +, X−] R0

τ1 is the lifetime without added anion, and τ2 is the asympotic lifetime at infinite anion concentration. ⟨τ ⟩ =

R 0τ12 + (τ22 − τ12)[R2 +, X−] R 0τ1 + (τ2 − τ1)[R2 +, X−]

(6)

The relative PL quantum yield was calculated using eq 7 below where ϕ is the PL quantum yield, A is the absorbance of the sample compound at the excitation wavelength, η is the refractive index of the solvent, and D is the integrated PL.29,30 The subscripts R and x refer to the reference compound and the unknown, respectively. The literature value for [Ru(bpy)3]Cl2 in water was used as the reference19 (ϕr = 0.042), and the refractive indices of water and CH2Cl2 used were 1.3387 and 1.4242.31

Figure 6. Steady state photoluminescence spectra of argon purged 6 μM [Ru(bpy)2(deeb)]2+ with the indicated TBACl concentrations (left). Time resolved PLI decay measured after 532 nm pulsed light excitation of an argon purged 31 μM [Ru(bpy)2(deeb)]2+ dichloromethane solution with added TBACl. The red arrows indicate the direction of change as TBACl is added.

2 ⎛ A ⎞⎛ η ⎞ ⎛ D ⎞ ϕx = ϕR ⎜ R ⎟⎜⎜ x ⎟⎟ ⎜ x ⎟ ⎝ Ax ⎠⎝ ηR ⎠ ⎝ DR ⎠

intensity decreased, red-shifted, and became sharper with increased chloride concentration. For example, the PL maximum red-shifted by 36 mV when 10 equiv of TBACl was added. Similar behavior was observed for [Ru(bpy)(deeb)2]2+; however, light excitation of [Ru(bpy)3]2+ and [Ru(deeb)3]2+ resulted in net photochemistry, as evidenced by the appearance of long wavelength absorption bands in solutions that had undergone prolonged irradiance and were therefore not studied further. Pulsed light excitation of [Ru(bpy)2(deeb)]2+ and [Ru(bpy)(deeb)2]2+ resulted in PL decays that were well described by a first-order kinetic model. The addition of excess salts resulted in a decrease in the lifetime. Typical data for the addition of chloride

(7)

The PL quantum yield (ϕ) values and the excited state lifetimes (τ) were used to calculate radiative (kr) and nonradiative (knr) rate constants, eqs 8 and 9.28 The values of ϕ0/ϕ, kr,0/kr, and knr/ knr,0 were plotted in Figure 7B. Note that both ϕ and kr decreased with added salt while knr increased. The ratio shown in Figure 7B for knr was inverted. kr = k nr = 8888

ϕem τ

(8)

1 − kr τ

(9) dx.doi.org/10.1021/jp404838z | J. Phys. Chem. A 2013, 117, 8883−8894


Article

Figure 7. (A) Stern−Volmer plots of lifetimes of approximately 20 μM [Ru(bpy)2(deeb)](PF6)2 solutions in CH2Cl2 with added TBAPF6 (black squares), TBACl (red circles), TBABr (green triangles), KOH (dark blue upside down triangles), and [Ru(bpy)(deeb)2](PF6)2 with added TBACl (light blue diamonds). (B) Ratios of the quantum yield (ϕ), kr, and knr for a 6 μM [Ru(bpy)2(deeb)](PF6)2 solution in CH2Cl2 measured with the indicated TBACl concentration.

Figure 8. [Ru(bpy)2(deeb)](PF6)2 with added TBACl. (A) Natural logarithm of the nonradiative rate constant as a function of the energy gap. (B) Radiative rate constant as a function of the cube of energy gap.

Figure 9. Transient absorption data for argon purged [Ru(bpy)2(deeb)](PF6)2 in (A) neat CH2Cl2 and (B) 10 mM TBACl CH2Cl2. The pink colored absorption spectra at the bottom are of the ground state. The red arrows depict the direction of change with time.

a medium frequency vibration, and C is a collection of other terms.2,33−40

As TBACl was added, the energy gap decreased as evidenced by the red shift of the PL spectra. The radiative rate constant is often related to this energy gap (E) and the transition moment integral (μ),32 eq 10, where ε0 is the permittivity of free space and c is the speed of light. 3

kr =

E μ2 3πε0ℏ4c 3

μ2 = |⟨ψgs|μ|̂ ψes⟩|2

ln k nr = C −

γ0E ℏωm

⎛ E ⎞ γ0 = ln⎜ ⎟−1 ⎝ SMℏωm ⎠

(11)

In Figure 8A, the natural logarithm of the nonradiative rate constant was plotted against the energy gap and fit to a straight line with a slope of −15.4 eV−1 that gave a Huang−Rhys factor of 0.4 based on an average medium frequency vibration of 1300 cm−1. This Huang−Rhys factor is considerably less than that for [Ru(bpy)3]2+ and indicates an excited state geometry that is less

(10)

The nonradiative rate constant is often related to the energy gap law which can be given in the simplified form of eq 11, where E is taken to be the PL maximum, SM is the Huang−Rhys factor, ωm is 8889



Article

distorted from the ground state than for [Ru(bpy)3]2+.41 In Figure 8B the radiative rate constant is plotted against the cube of the energy gap for PL and fit to a straight line using a leastsquares analysis with a slope of 1.2 × 105 s−1 eV−3. Nanosecond transient absorbance measurements of [Ru(bpy)2(deeb)]2+ with and without 10 mM TBACl were measured, Figure 9B and 9A, respectively. Difference spectra measured under both conditions were typical of MLCT excited states with a bleach of the visible absorption band and intense transitions in the ultraviolet region. With excess TBACl, isosbestic points were maintained at 400 and 528 nm; the subtle spectral changes measured in comparison to that measured without chloride were correlated to the changes in the ground state spectra. The spectra measured at different delay times after pulsed light excitation were found to be the same when normalized indicating that one excited state was generated within the 10 ns instrument response time. There was no evidence for permanent photochemistry or for production of the reduced ruthenium compound.

to [Ru(deeb)3]2+, Table 5. Additionally, the deeb C-6H are located directly above the aromatic ring of the adjacent bpy Table 5. Selected 1H NMR Changes change in chemical shift bpy C-3H

deeb C-6H

0.135 − 0.761

− 0.179 0.150

ligand, and in the case of [Ru(bpy)2(deeb)]2+ the shift in the deeb C-6H appeared to be a secondary effect from chloride association with the bpy C-3H. However, one would then expect the C-6H in [Ru(bpy)3]2+ to be influenced by this secondary effect, and they were not. In any case, the primary site of interaction with [Ru(bpy)2 (deeb)]2+ is the bpy C-3H. Interestingly, the addition of only 0.1 equiv of chloride to [Ru(bpy)2(deeb)]2+ resulted in a significant shift in these resonances without broadening or the appearance of new peaks. Since a geometry in which a single chloride would simultaneously interact with 40 hydrogens on 20 bpy ligands is not possible on steric grounds alone, the chloride ion must be exchanging sites between the ligands on the NMR time scale. The titration data is also consistent with an equilibrium with a rapid exchange. Although the 1H NMR spectra and the crystal structure describe the [Ru(bpy)2(deeb)]2+ complex in two distinct phases, a comparison provides several interesting points which illuminate the atomic level interactions between the ruthenium compound and halide ions. (1) In the crystal structure, the closest chlorides to the ruthenium compound are situated near the two bpy C-3H and the deeb C-6H, and in solution phase these are also the hydrogens that undergo the largest shift in the 1H NMR spectrum. In addition, a previously reported14 crystal structure of [Ru(bpy)2(deeb)]2+ I2 was reanalyzed, and it was found that the closest iodides were located 3.0− 3.2 Å from the two bpy C-3H in that structure as well. (2) The chloride near the bpy C-3H is coplanar with the bipyridine ligand, both of the C--H--Cl angles are 171°, and the H−Cl distance is approximately 2.5 and 2.8 Å, which is less than the sum of the van der Waals radii (2.95 Å).21 These criteria support what has previously been described as a carbon−hydrogen−chloride hydrogen bond.44 In the 1H NMR spectrum, the bpy C-3H are the most acidic hydrogens (most downfield) in the complex except for the deeb C-3H which do not display any indication of interaction with the chlorides. On this basis the ion-pair interaction can be described as an acid−base adduct. (3) In the crystal structure the ethyl ester groups are rotated away from the deeb C-3H, and it may be that the oxygen lone pair electrons coulombically repel any chloride ions. Alternatively, space filling models show that in solution phase, it is possible for the ethyl ester groups to rotate and sterically block the deeb C-3H from interacting with chloride. Either one of these or a combination of both may explain why there is no interaction with the deeb C-3H. (4) Even though it is reasonable to expect that the chlorides would interact with the ethyl ester carbonyl carbon atoms, there is a conspicuous lack of evidence in both the crystal

■

DISCUSSION Chloride ion-pairing with the series of four ruthenium compounds in dichloromethane was found to have a significant influence on the metal-to-ligand charge transfer (MLCT) excited states. Studies of the homoleptic compounds, [Ru(deeb)3]2+ and [Ru(bpy)3]2+, were complicated by the well documented appearance of ligand loss photochemistry, eq 12.34,35 [Ru(bpy)3 ]C12 + hv → Ru(bpy)2 C12 + bpy

compd [Ru(bpy)3]2+ plus 1 equiv TBACl [Ru(deeb)3]2+ plus 1 equiv TBACl [Ru(bpy)2(deeb)]2+ plus 42 equiv TBACl

(12)

Interestingly, this photochemistry was far less efficient for the heteroleptic compounds. While the origins of this very disparate photochemistry for homo- versus heteroleptic compounds are not well understood, it was not explored in more detail herein.19,36,42,43 The inefficient photochemistry was exploited as it enabled a more detailed characterization of the excited state properties of the heteroleptic compounds. The site(s) of adduct formation between chloride and the ruthenium compounds was quantified by 1H NMR studies and in the solid state with a crystal structure of [Ru(bpy)2(deeb)]Cl2. Ion-pairing was shown for the first time to influence both the radiative and nonradiative rate constants of MLCT excited states. Below we discuss these and relevant literature results in more detail. Ion-Pair Interactions. Chloride titration studies with 1H NMR spectroscopy has provided new insights into the site(s) of ion-pairing in this important class of ruthenium polypyridyl compounds. As chloride was added to the solution, several NMR resonances were shifted considerably downfield while others remained essentially at the same position. The downfield shift can be rationalized as a coulombic attraction between the hydrogen and the negatively charged chloride that elongates the C−H bond and thereby lowers the electron density around the hydrogen atom. The chloride is much farther away than the carbon atom, and any electron density provided by the chloride’s more diffuse orbitals does not compensate for the loss. When 1 equiv of TBACl was added to the homoleptic complexes [Ru(bpy)3]2+ and [Ru(deeb)3]2+, the greatest peak shifts occurred for resonances associated with C-3H and C-6H, respectively, indicating the preferred site of interaction for each of those ligands. When 42 equiv of TBACl were added to the heterolepetic [Ru(bpy)2(deeb)]2+, the bpy C-3H shifted almost 6 times as far as when 1 equiv of TBACl was added to [Ru(bpy)3]2+. In contrast, the deeb C-6H shifted less than when 1 equiv was added 8890



Article

Equilibrium Constants. In CH2Cl2 solutions it is known that TBA+ and Cl− will ion-pair46, and the data described herein show that these ruthenium compounds also ion-pair with chloride, bromide, and hydroxide. Therefore, in a solution of [Ru(bpy)2(deeb)](PF6)2 and TBACl, it is reasonable to describe the ground state equilibria with the equations in Scheme 2.

structure and the 1H NMR data for any significant interaction. Overall, the crystal structure and NMR data for [Ru(bpy)2(deeb)]2+ indicate that upon ion-pairing, there is a significant interaction between the chlorides and the bpy C3H, as well as possibly the deeb C-6H, which perturbs the electronic structure of the complex and could well be described as an acid−base adduct. Further indication of the perturbed electronic structure can be seen in the photophysical data. In the UV−vis spectra, both peak positions in energy and oscillator strengths were reversibly perturbed, indicating an interaction that alters energy levels and/ or transition integrals without permanent chemistry. The excited state absorption spectrum with added chloride was only altered in a manner consistent with the ground state UV−vis changes. The excited state spectrum measured with transient absorption returned to baseline with approximately the same lifetime as that measured by PL, indicating that no photochemistry or redox chemistry was occurring. The spectral shifts in the peak maximum for steady state photoluminescence were correlated with similar shifts in the absorbance. Both absorption and photoluminescence involve the same electronic ground state; however, for PL the upper level was the 3MLCT thermally equilibrated excited (thexi) state, and for absorbance the upper level was the 1MLCT Franck−Condon state. Upon addition of 10 equiv of chloride the PL energy was perturbed less (36 mV) than the absorbance (48 mV). Therefore, the 3MLCT thexi state was probably located farther from the chloride than the 1MLCT. This presumably results from the coulombic repulsion of chloride ions with the ligand in which the excited state was localized upon in the 3MLCT thexi state. Interestingly, both the absorbance and PL energies were decreased by 1.9% on addition of 10 equiv of chloride. These minor changes to the steady state photoluminescence spectrum, excited state lifetimes, and quantum yields for PL as TBACl was titrated into the solution were all consistent with a perturbed electronic structure in the excited state. A chloride adduct with the bpy C-3H’s of [Ru(bpy)3]2+ would inductively raise the energy levels of delocalized bpy π* orbitals to a greater extent than the metal based d-orbitals, consistent with the observed blue shift in the MLCT absorption for [Ru(bpy)3]2+. On the other hand, a chloride adduct with the bpy C-3H’s of [Ru(bpy)2(deeb)]2+ would inductively raise the energy levels of the metal based orbitals more than the deeb π* orbitals. This would cause the observed red shift in the PL and absorbance spectra from or to a ligand localized orbital, consistent with previous reports.45 Excited State Relaxation. The decrease in integrated steady state photoluminescence and excited state lifetime were fundamentally related to changes in the radiative and nonradiative rate constants for excited state decay.2 The radiative rate constant decreased with the cube of the energy gap, indicating qualitative agreement with the theoretical treatment of Strickler and Berg.32 On the other hand, as the energy gap decreased, the number of available vibration modes increased and the nonradiative rate constant (knr) increased in accordance with the energy gap law.2,33−35,37,39,40 The increase in knr far exceeded the decrease in kr, causing a decrease in steady state photoluminescence, excited state lifetime, and quantum yield of PL with the addition of chloride. In summary, the significant changes in kr and knr with chloride resulted from the energy level changes due to ion-pairing.

Scheme 2. Equilibria in CH2Cl2 Solution with [Ru(bpy)2(deeb)](PF6)2 and TBACl

Additionally, there is no reason to assume that the excited state equilibrium constants would be the same as the ground state equilibrium constants. Experimentally, the UV−vis spectrum and 1 H NMR reported on the ground state equilibrium while the photoluminescence and transient absorption reported on the excited state equilibrium. The fact that the normalized transient absorption spectra were independent of the observation time and returned cleanly to the ground state with the same lifetime measured by photoluminescence indicated that the excited state equilibrium was fully established within approximately 10 ns. Interestingly, the equilibrium constants for the excited state obtained from lifetime measurements were much greater than the corresponding equilibrium constants obtained from ground state UV−vis measurements. In the MLCT excited state of ruthenium polypyridyl compounds with a coordinated deeb ligand, the excited state is localized on the deeb ligand. For example, with [RuII(bpy)2(deeb)]2+, the more Lewis acidic metal center in the excited state, [RuIII(bpy)2(deeb−)]2+*, will inductively attract electron density from the bpy ligands. This in turn increases the acidity of the bpy C-3H, forming a stronger adduct with chloride resulting in a larger equilibrium constant. Whereas the actual solution dynamics are an interplay of all the equilibria, the first two are expected to be the most relevant for the experimental effects described herein. However, even though two equilibria were expected, the presence of all three species, the dicationic ruthenium complex, the singly ion-paired monocation, and the doubly ion-paired neutral species, could not be detected spectroscopically. Only a single set of isosbestic points were observed in the UV−vis spectra until reaching the highest chloride concentrations where no isosbestic points were seen. Additionally, the Benesi−Hildebrand analysis of the absorption changes associated with ion-pairing were found to be well described by a single equilibrium constant whose magnitude increased with the number of deeb ligands present on the ruthenium compound. The changes in the excited state photophysical properties (excited state lifetime and quantum yield of photoluminescence) also did not contain sufficient structure to model two equilibrium processes, and therefore each experiment yielded only a single, “observed” equilibrium constant that presumably represent a weighted combination of 8891



Article

the actual equilibrium constants. Based on coulombic considerations, the first equilibrium constant is expected to be significantly greater than the second. The experimental Kobs varied from 480 to 118 000 M−1, probably due to the complex equilibria mentioned above and the varied effects each species has on the measured quantities. Using Bjerrum ion-pairing theory,1 the first equilibrium constant was calculated to be K1 = 290 000 M−1 while using the Eigen−Fuoss equation,47,48 the calculations resulted in K1 = 117 000 and K2 = 500. The experimental and theoretical equilibrium constants are at least of a similar order of magnitude. The Job plot provided the clearest evidence for the expected 1:2 cation-to-anion ratio. Additionally, the 1H NMR data point to chloride−hydrogen−carbon hydrogen bonds in the ion-paired state which would require two chlorides to form adducts with the two bpy ligands, and in this case a two-equilibria model was used. Deviations from the modeled fit at low TBACl concentrations can be attributed to water in the solution that was evident in the NMR spectra that would preferentially pair with the chlorides. Chloride, bromide, and hydroxide can be thought of as redox inactive analogues for iodide that is used extensively as a redox mediator in dye-sensitized solar cells (DSSCs).49,50 The data reported herein measured in the low dielectric constant solvent dichloromethane may have some relevance to DSSCs. While the spectral changes associated with ion-pairing were significant, they were small. For example, the subtle changes to the visible absorption spectra reported herein could easily be missed in an operational solar cell where interfacial heterogeneity is thought to play a key role. The influence of ion-pairing on excited state injection should be negligible, unless the excited state was very short-lived or was remote to the semiconductor surface or both.51 The ∼30% change in excited state lifetime would not be expected to influence injection rates that are often found to occur on picosecond and shorter time scales. Ion-pairing between the oxidized sensitizer and iodide has in fact been invoked to rationalize rapid regeneration under some conditions.12,52 Unfortunately the oxidized sensitizer formed after excited state injection is often assumed to be a “cation” even though the charge of the ground state dye molecule is generally unknown and is rarely neutral or cationic. Nevertheless, the electric field generated by injected electrons53 certainly influences the interfacial dielectric constant, and the data described herein show that careful absorption measurements may yield direct evidence for ion-pairing.

From the equilibrium expression: R + X ⇌ [R, X]R = [Ru(LL)3 ]2 + Keq =

[RX] [R][X]

Keq =

[RX] (R 0 − [RX])(X 0 − [RX])

R 0 = [R] + [RX] X 0 = [X] + [RX]

(R 0 − [RX])(X 0 − [RX]) =

[RX] Keq

⎛ 1 ⎞⎟ [RX]2 − ⎜⎜R 0 + X 0 + [RX] + R 0X 0 = 0 Keq ⎟⎠ ⎝

Using the quadratic formula: [RX] =

ΔA 1⎛ 1 ⎞⎟ = ⎜⎜R 0 + X 0 + Δε 2⎝ Keq ⎟⎠ −

2 ⎡ ⎛ ⎞⎤ ⎢ 1 ⎜R 0 + X 0 + 1 ⎟⎥ − R 0X 0 ⎢⎣ 2 ⎜⎝ Keq ⎟⎠⎥⎦

From Beer’s law: AR = [R]εR l

ARX = [RX]εRX l

A 0 = R 0εR l

From mass balance: R 0 = [R] + [RX] A soln = AR + ARX

ΔA = AR + ARX − A 0 = [R]εR l + [RX]εRX l − R 0εR l − ([R] − R 0)εR l + [RX]εRX l

Substituting [R] − R 0 = −[RX] ΔA = −[RX]εR l + [RX]εRX l = [RX](εRX − εR )l ΔA = [RX]Δεl

■

Derivation of Equation for Lifetime Analysis From ref 22, pages 99 and 141−142, the average lifetime of a two-component system is as follows.

CONCLUSION The data suggests that chloride preferentially formed adducts with the hydrogen atoms in the 3 and 3′ positions of bpy in the low dielectric solvent dichloromethane. This preference suggests that the sites of ion-pairing can be controlled in heteroleptic Ru2+ compounds. The ion-pairing produced a perturbation in the electronic structure such that various photophysical properties were altered in a measurable and predictable manner. Changes to the radiative and nonradiative rate constants were in accordance with known theoretical equations for the radiative rate constant and the energy gap law and suggest that these photophysical effects are due to changes in energy levels. Observed equilibrium constants were extracted that were in general agreement with Bjerrum and Eigen−Fuoss equations.

⟨τ ̅ ⟩ = α1 =

⟨τ ⟩ =

=

■

APPENDIX 1 Derivation of Equation for Benesi−Hildebrand Analysis26

= 8892

α1τ12 + α2τ22 α1τ1 + α2τ2

[R] R0

α2 =

[RX] R0

R 0 = [R] + [RX]

α1τ12 + α2τ22 α1τ1 + α2τ2 [R] 2 τ R0 1

+

[RX] 2 τ R0 2

(R) τ R0 1

+

[RX] τ R0 2

[R]τ12 + [RX]τ22 [R]τ1 + [RX]τ2 dx.doi.org/10.1021/jp404838z | J. Phys. Chem. A 2013, 117, 8883−8894

The Journal of Physical Chemistry A ⟨τ ⟩ = =

Article

(R 0 − [RX])τ12 + [RX]τ22 (R 0 − [RX])τ1 + [RX]τ2

Semi-Random Donor/Acceptor Polymer: Fullerene Solar Cells. J. Phys. Chem. C 2013, 117, 6940−6948. (12) Clifford, J. N.; Palomares, E.; Nazeeruddin, M. K.; Grätzel, M.; Durrant, J. R. Dye Dependent Regeneration Dynamics in DyeSensitized Nanocrystalline Solar Cells: Evidence for the Formation of a Ruthenium Bipyridyl Cation/Iodide Intermediate. J. Phys. Chem. C 2007, 111, 6561−6567. (13) Brasch, N. E.; Buckingham, D. A.; Clark, C. R.; Simpson, J. Reactivity of a Regiospecific Ion Pair. Comparisons of Cl- and H2O Entry into the Five-Coordinate Intermediate Generated from p[Co(tren)(NH3)OH2]3+Cl. Inorg. Chem. 1996, 35, 7728−7734. (14) Marton, A.; Clark, C. C.; Srinivasan, R.; Freundlich, R. E.; Narducci Sarjeant, A. A.; Meyer, G. J. Static and Dynamic Quenching of Ru(II) Polypyridyl Excited States by Iodide. Inorg. Chem. 2006, 45, 362−369. (15) Farnum, B. H.; Gardner, J. M.; Marton, A.; Narducci-Sarjeant, A. A.; Meyer, G. J. Influence of Ion Pairing on the Oxidation of Iodide by MLCT Excited States. Dalton Trans. 2011, 40, 3830−3838. (16) Farnum, B. H.; Jou, J. J.; Meyer, G. J. Visible Light Generation of I−I Bonds by Ru-tris(diimine) Excited States. Proc. Natl. Acad. Sci. U.S.A. 2012, 109, 15628−15633. (17) Kelly, C. A.; Farzad, F.; Thompson, D. W.; Meyer, G. J. ExcitedState Deactivation of Ruthenium(II) Polypyridyl Chromophores Bound to Nanocrystalline TiO2 Mesoporous Thin Films. Langmuir 1999, 15, 731−737. (18) Argazzi, R.; Bignozzi, C. A.; Heimer, T. A.; Castellano, F. N.; Meyer, G. J. Enhanced Spectral Sensitivity from Ruthenium(II) Polypyridyl Based Photovoltaic Devices. Inorg. Chem. 1994, 33, 5741−5749. (19) Van Houten, J.; Watts, R. J. Temperature Dependence of the Photophysical and Photochemical Properties of the Tris(2,2′bipyridyl)ruthenium(II) Ion in Aqueous Solution. J. Am. Chem. Soc. 1976, 98, 4853−4858. (20) Macrae, C. F.; Bruno, I. J.; Chisholm, J. A.; Edgington, P. R.; McCabe, P.; Pidcock, E.; Rodriguez-Monge, L.; Taylor, R.; van de Streek, J.; Wood, P. A. Mercury CSD 2.0New Features for the Visualization and Investigation of Crystal Structures. J. Appl. Crystallogr. 2008, 41, 466−470. (21) Bondi, A. van der Waals Volumes and Radii. J. Phys. Chem. 1964, 68, 441−451. (22) Fortin, S.; Beauchamp, A. L. Preparation and Characterization of Oxorhenium(V) Complexes with 2,2′-Biimidazole: The Strong Affinity of Coordinated Biimidazole for Chloride Ions via N−H...Cl− Hydrogen Bonding. Inorg. Chem. 2000, 39, 4886−4893. (23) Fortin, S.; Beauchamp, A. L. Preparations, Characterizations, and Structures of (Biimidazole)dihalobis(triphenylphosphine)rhenium(III) Salts: A Strong Ion-Pairing and Acid−Base Properties. Inorg. Chem. 2001, 40, 105−112. (24) Ayme, J.-F.; Beves, J. E.; Leigh, D. A.; McBurney, R. T.; Rissanen, K.; Schultz, D. A Synthetic Molecular Pentafoil Knot. Nat. Chem. 2011, 4, 15−20. (25) Silverstein, R., Bassler, G., Morrill, T. Spectroscopic Identification of Organic Compounds, 4th ed.; John Wiley and Sons: New York, 1981. (26) Benesi, H. A.; Hildebrand, J. H. A Spectrophotometric Investigation of the Interaction of Iodine with Aromatic Hydrocarbons. J. Am. Chem. Soc. 1949, 71, 2703−2707. (27) Job, P. Job Plot. Ann. Chim. Appl. 1928, 9, 113−203. (28) Lakowicz, J. R. Principles of Fluorescence Spectroscopy, 2nd ed.; Kluwer Academic/Plenum Publishers: New York, 1999. (29) Parker, C. A.; Rees, W. T. Correction of Fluorescence Spectra and Measurement of Fluorescence Quantum Efficiency. Analyst 1960, 85, 587−600. (30) Crosby, G. A.; Demas, J. N. Measurement of Photoluminescence Quantum Yields. Review. J. Phys. Chem. 1971, 75, 991−1024. (31) Castellano, F. N.; Heimer, T. A.; Tandhasetti, M. T.; Meyer, G. J. Photophysical Properties of Ruthenium Polypyridyl Photonic SiO2 Gels. Chem. Mater. 1994, 6, 1041−1048.

R 0τ12 − (τ12 − τ22)[RX] R 0τ1 − (τ1 − τ2)[RX]

By forming the equation in terms of [RX], eq 4 can be substituted into the equation above.

■

ASSOCIATED CONTENT

S Supporting Information *

Table containing detailed changes in UV−vis spectra of the ruthenium compounds upon addition of ionic salts. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Telephone: (410) 516-7319. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was funded by the Division of Chemical Sciences, Geosciences, and Biosciences, Office of Basic Energy Sciences of the U.S. Department of Energy through Grant DE-FG0296ER14662 (G.J.M.). The authors thank Dr. Cathy Moore for her assistance and guidance with the 1H NMR experiments.

■

REFERENCES

(1) Marcus, Y.; Hefter, G. Ion Pairing. Chem. Rev. 2006, 106, 4585− 4621. (2) Vining, W. J.; Caspar, J. V.; Meyer, T. J. The Influence of Environmental Effects on Excited-State Lifetimes. The Effect of Ion Pairing on Metal-to-Ligand Charge Transfer Excited States. J. Phys. Chem. 1985, 89, 1095−1099. (3) Macchioni, A. Ion Pairing in Transition-Metal Organometallic Chemistry. Chem. Rev. 2005, 105, 2039−2074. (4) Sutton, L. R.; Scheloske, M.; Pirner, K. S.; Hirsch, A.; Guldi, D. M.; Gisselbrecht, J.-P. Unexpected Change in Charge Transfer Behavior in a Cobalt(II) Porphyrin−Fullerene Conjugate That Stabilizes Radical Ion Pair States. J. Am. Chem. Soc. 2004, 126, 10370−10381. (5) McCosar, B. H.; Schanze, K. S. Ion-Pair Charge Transfer Photochemistry in Rhenium(I) Borate Salts. Inorg. Chem. 1996, 35, 6800−6808. (6) Llewellyn, D. B.; Adamson, D.; Arndtsen, B. A. A Novel Example of Chiral Counteranion Induced Enantioselective Metal Catalysis: The Importance of Ion-Pairing in Copper-Catalyzed Olefin Aziridination and Cyclopropanation. Org. Lett. 2000, 2, 4165−4168. (7) Abramo, G. P.; Li, L.; Marks, T. J. Polynuclear Catalysis: Enhancement of Enchainment Cooperativity between Different SingleSite Olefin Polymerization Catalysts by Ion Pairing with a Binuclear Cocatalyst. J. Am. Chem. Soc. 2002, 124, 13966−13967. (8) Brak, K.; Jacobsen, E. N. Asymmetric Ion-Pairing Catalysis. Angew. Chem., Int. Ed. 2013, 52, 534−561. (9) Choi, M. M. F.; Douglas, A. D.; Murray, R. W. Ion-Pair Chromatographic Separation of Water-Soluble Gold MonolayerProtected Clusters. Anal. Chem. 2006, 78, 2779−2785. (10) Yamakawa, N.; Jiang, M.; Key, B.; Grey, C. P. Identifying the Local Structures Formed During Lithiation of the Conversion Material, Iron Fluoride, in a Li Ion Battery: A Solid-State NMR, X-ray Diffraction, and Pair Distribution Function Analysis Study. J. Am. Chem. Soc. 2009, 131, 10525−10536. (11) Li, K.; Khlyabich, P. P.; Li, L.; Burkhart, B.; Thompson, B. C.; Campbell, J. C. Influence of Exciton Diffusion and Charge-Transfer State Dissociation Efficiency on the Short-Circuit Current Densities in 8893



Article

Os(dcb)2(CN)2/TiO2 Photoelectrochemical Cells. J. Phys. Chem. B 1998, 102, 7577−7581. (53) Ardo, S.; Sun, Y.; Castellano, F. N.; Meyer, G. J. Excited-State Electron Transfer from Ruthenium−Polypyridyl Compounds to Anatase TiO2 Nanocrystallites: Evidence for a Stark Effect. J. Phys. Chem. B 2010, 114, 14596−14604.

(32) Strickler, S. J.; Berg, R. A. Relationship between Absorption Intensity and Fluorescence Lifetime of Molecules. J. Chem. Phys. 1962, 37, 814−822. (33) Englman, R.; Jortner, J. The Energy Gap Law for Nonradiative Decay in Large Molecules. J. Lumin. 1970, 1−2, 134−142. (34) Caspar, J. V.; Meyer, T. J. Photochemistry of Tris(2,2′bipyridine)ruthenium(2+) Ion (Ru(bpy)32+). Solvent Effects. J. Am. Chem. Soc. 1983, 105, 5583−5590. (35) Caspar, J. V.; Meyer, T. J. Photochemistry of MLCT Excited States. Effect of Nonchromophoric Ligand Variations on Photophysical Properties in the Series cis-Ru(bpy)2L22+. Inorg. Chem. 1983, 22, 2444− 2453. (36) Allen, G. H.; White, R. P.; Rillema, D. P.; Meyer, T. J. Synthetic Control of Excited-State Properties. Tris-Chelate Complexes Containing the Ligands 2,2′-Bipyrazine, 2,2′-Bipyridine, and 2,2′-Bipyrimidine. J. Am. Chem. Soc. 1984, 106, 2613−2620. (37) Kober, E. M.; Caspar, J. V.; Lumpkin, R. S.; Meyer, T. J. Application of the Energy Gap Law to Excited-State Decay of Osmium(II)−Polypyridine Complexes: Calculation of Relative Nonradiative Decay Rates from Emission Spectral Profiles. J. Phys. Chem. 1986, 90, 3722−3734. (38) Photochemistry and Photophysics of Coordination Compounds I; Balzani, V., Campagna, S., Eds.; Springer: Berlin, Germany, 2007; Vol. 280. (39) Caspar, J. V.; Sullivan, B. P.; Kober, E. M.; Meyer, T. J. Application of the Energy Gap Law to the Decay of Charge Transfer Excited States, Solvent Effects. Chem. Phys. Lett. 1982, 91, 91−95. (40) Durham, B.; Caspar, J. V.; Nagle, J. K.; Meyer, T. J. Photochemistry of Tris(2,2′-bipyridine)ruthenium(2+) Ion. J. Am. Chem. Soc. 1982, 104, 4803−4810. (41) McCusker, J.; McCusker, C. Synthesis and Spectroscopic Characterization of CN-Substituted Bipyridyl Complexes of Ru(II). Inorg. Chem. 2011, 50, 1656−1669. (42) Bargawi, K. R.; Llobet, A.; Meyer, T. J. Synthetic Design of MLCT Excited States. Ligand-Substituted, Mono-2,2′-bipyridine Complexes of Ruthenium(II). J. Am. Chem. Soc. 1988, 110, 7751−7759. (43) Van Houten, J.; Watts, R. J. Photochemistry of Tris(2,2′bipyridyl)ruthenium(II) in Aqueous Solutions. Inorg. Chem. 1978, 17, 3381−3385. (44) Taylor, R.; Kennard, O. Crystallographic Evidence for the Existence of C−H...O, C−H...N, and C−H...Cl Hydrogen Bonds. J. Am. Chem. Soc. 1982, 104, 5063−5070. (45) Ferguson, J.; Mau, A. W. H.; Sasse, W. H. F. Localization of Eectronic Excitation Energy in Ru(2,2′-bipyridine)2(2,2′-bipyridine4,4′-dicarboxylic acid)2+ and Related Complexes. Chem. Phys. Lett. 1979, 68, 21−24. (46) Mo, H.; Wang, A.; Wilkinson, P. S.; Pochapsky, T. C. Closed-Shell Ion Pairs: Cation and Aggregate Dynamics of Tetraalkylammonium Salts in an Ion-Pairing Solvent. J. Am. Chem. Soc. 1997, 119, 11666− 11673. (47) Fuoss, R. M. Ionic Association. III. The Equilibrium between Ion Pairs and Free Ions. J. Am. Chem. Soc. 1958, 80, 5059−5061. (48) Powers, M. J.; Meyer, T. J. Medium and Distance Effects in Optical and Thermal Electron Transfer. J. Am. Chem. Soc. 1980, 102, 1289−1297. (49) Boschloo, G.; Hagfeldt, A. Characteristics of the Iodide/Triiodide Redox Mediator in Dye-Sensitized Solar Cells. Acc. Chem. Res. 2009, 42, 1819−1826. (50) Rowley, J. G.; Farnum, B. H.; Ardo, S.; Meyer, G. J. Iodide Chemistry in Dye-Sensitized Solar Cells: Making and Breaking I−I Bonds for Solar Energy Conversion. J. Phys. Chem. Lett. 2010, 1, 3132− 3140. (51) Johansson, P. G.; Kopecky, A.; Galoppini, E.; Meyer, G. J. Distance Dependent Electron Transfer at TiO2 Interfaces Sensitized with Phenylene Ethynylene Bridged RuII−Isothiocyanate Compounds. J. Am. Chem. Soc. 2013, 135, 8331−8341. (52) Alebbi, M.; Bignozzi, C. A.; Heimer, T. A.; Hasselmann, G. M.; Meyer, G. J. The Limiting Role of Iodide Oxidation in cis8894


Chloride Ion-Pairing with Ru(II) Polypyridyl Compounds in

Recommend Documents