Flash Kinetic Spectroscopic Studies of Dinuclear Rhodium( I

complex, it is of interest to determine if the MF6-Xe system forms such complexes. The CT transition in IrF6-Xe is the closest to the ground state; th...
0 downloads 0 Views 489KB Size
Gray et al.

/

Spectroscopic Studies of Dinuclear Rhodium(I) Complexes

complex, it is of interest to determine if the MF6-Xe system forms such complexes. The CT transition in IrF6-Xe is the closest to the ground state; thus, it is expected that the IrF6-Xe complex would be the most stable. The usual approach for determining stability of such complexes employs the method of Benesi and Hildebrand4 to find the equilibrium constant for complex formation. However, the near-IR data obtained for solid IrF6-Xe (Tables I1 and 111) and the near-IR-visible data for gas-phase IrF6-Xe allow a simpler, though perhaps more approximate, alternative method to be utilized. The fact that there is no evidence of the IrFb-Xe complex in the room temperature vapor indicates that the 1:l complex is not strongly bound. The observed frequency shifts, increased line widths (see Table HI), and intensity enhancements in the 0.1% IrF6-Xe solid samples indicate that there is significant CT interaction between IrF6 and Xe in the ground state. However, the solid state data pertain to an IrF6.12Xe complex, whereas information on the 1:1 complex is of more intrinsic interest. Equation 1 and the assumption that the complex is weak enough to allow n' to be set equal to 12 provide an approximate relationship between the 1:12 and the 1:l complexes. The stabilization energy of the ground state [12X0(rg(~A2))] of the IrF6.12Xe complex can be estimated by assuming that eq 2 can be applied to the frequency shift data for the I's(2T1) state to find @ ~ ( r g ( ~ T land ) ) , that @o(r~(4A2)) is roughly the same since both states stem from the (t2,)3 configuration. These considerations lead to a value for 12x0 (rS(4A2)) of -300 cm-l. The stabilization energy of the ground ~ t a t e X ~ ( F ~ ( of ~ Athe ~ )1:l ) complex is then -25 cm-I. One would certainly expect this to be a lower limit since saturation effects5 have not been taken into account. The original

485

assumption that the ground state complex is weak thus appears well justified. The general picture that emerges for these weak charge transfer complexes between Xe and MF6 molecules is then as follows. The ground state is neutral probably with a shallow broad potential minimum somewhere near r D 4 %.,the Xe-Xe approximate distance in a crystal lattice or a liquid. It is possible that there are a few vibrational quanta in this well but this is not a necessary condition imposed by our data. The excited state potential well is much deeper and more narrow, and the potential minimum is such that r*(MFs-Xe+) < rO(MF6Xe). These considerations also account nicely for the very broad ( > l o 000 cm-l) Franck-Condon envelope observed for the CT transitions in all systems. VI. Conclusion The new electronic transitions which appear when certain transition metal hexafluorides are dissolved in liquid xenon can be assigned as intermolecular charge-transfer transitions. The concomitant charge-transfer complexes are weakly bound.

-

References and Notes (1) E. R. Bernstein and G. R. Meredith, J. Cbem. Phys., 64, 375 (1976). (2) E. R . Bernstein and J. D.Webb, Mol. Pbys. in press: E. R. Bernstein, G. R . Meredith, and J. D.Webb, ibid., in press. (3) P. R . Hammond, J. Cbem. SOC.A, 3826 (1971), and references to earlier work therein. (4) R . S.Mulliken and W. 8. Person, "Molecular Complexes, a Lecture and Reprint Volume", Wiiey, New York, N.Y., 1969. (5) R. S. Mulliken, J. Am. Cbem. SOC.,74, 811 (1952). (6) N. Bartlett, Angew. Cbem., int. Ed. Engl., 7, 433 (1968). (7) N. Bartlett as quoted by R. N. Compton, J. Cbem. Pbys., 66, 4478 (1977). (8) C. D.Cooper, R. N. Compton, and P. W. Reinhardt, Abstracts, lXth international Conference on the Physics of Electronic and Atomic Collisions, Vol. 2, J. S.Risley and R. Geballe, Ed., University of Washington Press, Seattle, Wash., 1975, p 922.

Flash Kinetic Spectroscopic Studies of Dinuclear Rhodium( I) Complexes Vincent M. Miskowski,la Glenn L. Nobinger,IaDavid S. Kliger,la George S. Hammond,la Nathan S. Lewis,lb Kent R. Mann,lb and Harry B. Graylb* Contributionfrom the Division of Natural Sciences, University of California, Santa Cruz, California 95060, and Contribution No. 5561 from the Arthur Amos Noyes Laboratory of Chemical Physics, California Institute of Technology, Pasadena, California 91 12.5. Received April 11 I977 ~

Abstract: Excitation of concentrated acetonitrile solutions of [Rh(tol)4]PF6 (to1 is p-methylphenyl isocyanide) at 562 nm, where strong absorption attributable to [Rh2(tol)8l2+ occurs, produces emission (A, 697 nm) with a quantum yield of 0.0065 and a lifetime of 6 2 ns. Excitation of [Rh2(bridge)d](BPh& (bridge is 1,3-diisocyanopropane) in acetonitrile solution at 553 nm gives emission a t 656 nm with a quantum yield of 0.056 and a lifetime of 6 2 ns. The emission is assigned to 'A*,, 'AI, (2a1, lal,,) in both dinuclear Rh(1) complexes. Excitation also gives rise to long-lived transient absorptions attributable to the following dinuclear and trinuclear species: -8 ps, Rh2(bridge)42+; 0.09 ps, [Rh2(tol)8l2+;and 0.14 ps, [Rh,(tol)1213+. In the dinuclear complexes, this transient is most likely 3A2u,which is the triplet excited-state partner of 'A2,. The results suggest that the spin-orbit components of 3A2upossess very little singlet character.

-

-

Introduction In recent work it has been e ~ t a b l i s h e d that ~ - ~ planar Rh(1) isocyanide complexes oligomerize in solution, through formation of weak metal-metal bonds. The oligomerization can be followed conveniently by monitoring electronic absorption spectra, as transitions attributable to dimers, trimers, and 0002-7863/78/1500-0485$01.00/0

higher oligomers occur at progressively lower energies. As part of our continuing study of the physical and chemical properties of Rh(1) oligomers, we have performed flash kinetic spectroscopic experiments on solutions of [Rh(tol)4]PF6, where to1 is p-methylphenyl isocyanide, and on the dinuclear complex, [Rh2(bridge)4] (BPhq)2,3 where bridge is 1,3-diisocyanopropane.

0 1978 American Chemical Society

486

Journal of the American Chemical Society A("0.5

450

500

I

I

100:2

/ January 18,1978

A(nm)--

550 600 I

/

I

700 800 900 I

I

3 50 I

1

I

4 50

550

650

800

I 1

1

1

' I

151

t

A

0.4

t

I

5

0

0) -

a

0.3

E E -

Q(

5

0.2

L

::

n

a 0.I

0

+Y X I 0-3 (crn-1)

Figure 1. Absorption (--) and corrected emission (- - -) spectra of [Rh(tol)4]PFs in acetonitrile solution at room temperature; [RhItot= 4.4 x 10-3 M.

Experimental Section The ligands p-CNPhCH3 (tol) and 1,3-diisocyanopropane (bridge)3 were prepared from the corresponding amines by standard proced u r e ~ Both . ~ ligands were purified by vacuum distillation. (Note! Several attempted distillations of bridge in which the crude product was orange resulted in violent explosions. Extreme caution is advised.) [Rh(tol).#F6/ This complex was prepared by adding an excess of to1 to a warm benzene solution of [Rh(COD)CI]2 (the latter compound was prepared by a standard procedure6). A precipitate formed immediately, and was filtered and washed with benzene. This precipitate was dissolved in water and a solution of KPF6 was added. The precipitate was filtered, dried, and recrystallized from CH2C12/ CHC13: yellow crystals; T(CN) 2160 cm-l (CH2C12 solution). Anal. Calcd: C , 53.65; H, 3.94; N, 7.82; Rh, 14.36. Found: C, 53.79; H, 3.80; N , 8.03; Rh, 14.06. [Rh2(bridge)4](BPh4)2*2CH3CN.A stoichiometric amount of NaBPh4 in methanol was added to a methanol solution of [Rh2(bridge)4]Clz (the latter was obtained as a blue powder by adding a stoichiometric amount of bridge to a chloroform solution of [Rh(COD)C1]2). The purple solid was then recrystallized from acetonitrile, T(CN) 2172 cm-l (KBr disk). Anal. Calcd: C, 66.37; H, 5.41; N, 10.75. Found: C, 65.59; H, 5.49; N, 10.24. Emission Spectra. A Perkin-Elmer MPF-3A fluorescence spectrometer was employed for measurements at 25 OC in solution. All spectra were corrected for phototube and monochromator response by using a computer program that corrects raw emission data point by point. The data are then replotted either on a wavelength or an energy scale. Spectra were obtained using the standard 90' to incidence technique. Quantum Yields. The quantum yields in undegassed fluid acetonitrile solutions at 25 OC were obtained by matching the absorbance of the sample and a reference (Ru(bpy)3C12 in aqueous solution (4 = 0.042)7 for [Rh2(bridge)4I2+; [Rh2(bridge)4I2+ in acetonitrile solution for [Rh2(t01)8]~+)at 505 nm for [Rh2(bridge)4l2+ and 510 nm for [ R h ~ ( t o l ) s ] ~Spectra +. were then obtained as outlined above for both sample and reference under exactly the same instrumental conditions. The spectra were corrected and plotted on an energy scale, and the relative areas were calculated with the aid of a computer program, using the trapezoidal approximation to the area under the curve. Approximately 50 trapezoids were used for each spectrum. The ratios of the areas were taken a,s the ratios of the quantum yields without further correction. The error limits of f 2 0 % are appropriate owing

-

v x iO3(cm-')

Figure 2. Absorption (--) and corrected emission (- - -) spectra of [Rh~(bridge)4](BPh4)*in acetronitrile solution ( lod4 M) at room temperature.

to the fact that the absorbance changes with wavelength are different for the sample and reference near the exciting wavelength. Flash Kinetic Spectroscopic Experiments. The flash kinetic spectroscopic studies employed an apparatus that has been described previously.8 For the present studies, in addition to the Nz laser (337 nm excitation), we employed the following dye lasers: Rhodamine 6G (580 nm excitation); Coumarin 495 (550 nm excitation); Nile Blue Perchlorate (705 nm excitation). The laser pulse widths were in all cases approximately 10 ns. This establishes a lower limit for measurable transient lifetimes of -2 ns, as shorter lifetimes could not be deconvoluted from the laser pulse. Samples for flash studies were degassed either by three freeze-thaw cycles or, for much of the work with [Rh(tol)4]PF6 solutions, by thorough argon bubbling. Air-saturated solutions of the latter compound displayed unquenched emission, whereas transient absorption lifetimes (see Results and Discussion) were reduced approximately 40%, suggesting diffusion-controlled quenching by 0 2 .

Results and Discussion Dimer Emission. Excitation of concentrated (10-2-10-3 M) CHsCN solutions of [Rh(tol)4]PFs into the dimer absorption band (Amax 562 nm) results in emission (A, 697 nm) at room t e m p e r a t ~ r eIn . ~ Figure 1, we show representative absorption and emission spectra. Note that the weak absorption feature near 700 nm is attributable to trimers, whereas absorption at X < 500 nm is primarily due to monomers. The emission quantum yield and lifetime were found to be 0.0065 and immeasurably short (G2 ns), respectively. Next we examined [Rh2(bridge)4I2+in acetonitrile solution. Here we could employ low concentrations (10-5- 10-4 M), as there is no evidence indicating that this dinuclear complex dissociates to monomeric specie^.^ Excitation into the lowest energy absorption band (A, 553 nm) again gives emission (Amax 656 nm), as illustrated in Figure 2. The emission quantum yield was found to be 0.056. For this complex, careful scrutiny of laser excitation and emission pulse shapes suggests that the emission lifetime is near that of the laser pulse, but it is still too small to measure ( T G 2 ns).

Gray et al.

/ Spectroscopic Studies of Dinuclear Rhodium(I) Complexes

487

Table I. Values of Measured and Calculated Emission Parameters for Dinuclear Rhodium(I) Complexes in Acetonitrile Solution at

-

0

1.5

Room Temperature Complex

4~~

TF,

[Rh2(tol)8I2+ 0.0065 [Rhz0.056 (bridge)4I2+

ns