Tuning the excited-state properties of [M(SnR3)2(CO)2(α-diimine)] (M = Ru, Os; R = Me, Ph)

Article abstract of DOI:10.1021/ic000857x

The influences of R, the α-diimine, and the transition metal M on the excited-state properties of the complexes [M(SnR₃)₂(CO)₂(α-diimine)] (M = Ru, Os; R = Ph, Me) have been investigated. Various synthetic routes were used to prepare the complexes, which all possess an intense, sigma-bond-to-ligand charge-transfer transition in the visible region between a σ(Sn-M-Sn) and a π*(α-diimine) orbital. The resonance Raman spectra show that many bonds are only weakly affected by this transition. The room-temperature time-resolved absorption spectra of [M(SnR₃)₂(CO)₂(dmb)] (M = Ru, Os; R = Me, Ph; dmb = 4,4′-dimethyl-2,2′-bipyridine) show the absorptions of the radical anion of dmb, in line with the SBLCT character of the lowest excited state. The excited-state lifetimes at room temperature vary between 0.5 and 3.6 μs and are mainly determined by the photolability of the complexes. All complexes are photostable in a glass at 80 K, under which conditions they emit with very long lifetimes. The extremely long emission lifetimes (e.g., τ = 1.1 ms for [Ru(SnPh₃)₂(CO)₂(dmb)]) are about a thousand times longer than those of the ³MLCT states of the [Ru(Cl)(Me)(CO)₂(α-diimine)] complexes. This is due to the weak distortion of the former complexes in their ³SBLCT states as seen from the very small Stokes shifts. Remarkably, replacement of Ru by Os hardly influences the absorption and emission energies of these complexes; yet the emission lifetime is shortened because of an increase of spin - orbit coupling. The quantum yield of emission at 80 K is 1-5% for these complexes, which is lower than might be expected on the basis of their slow nonradiative decay.

Full text of DOI:10.1021/ic000857x

Inorg. Chem. 2001, 40, 277-285
277
Tuning the Excited-State Properties of [M(SnR3)2(CO)2(r-diimine)] (M ) Ru, Os; R ) Me,
Ph)
Joris van Slageren and Derk J. Stufkens*
Institute of Molecular Chemistry, Universiteit van Amsterdam, Nieuwe Achtergracht 166,
NL-1018 WV Amsterdam, The Netherlands
ReceiVed August 1, 2000
The influences of R, the R-diimine, and the transition metal M on the excited-state properties of the complexes
[M(SnR3)2(CO)2(R-diimine)] (M ) Ru, Os; R ) Ph, Me) have been investigated. Various synthetic routes were
used to prepare the complexes, which all possess an intense sigma-bond-to-ligand charge-transfer transition in
the visible region between a σ(Sn-M-Sn) and a π*(R-diimine) orbital. The resonance Raman spectra show that
many bonds are only weakly affected by this transition. The room-temperature time-resolved absorption spectra
of [M(SnR3)2(CO)2(dmb)] (M ) Ru, Os; R ) Me, Ph; dmb ) 4,4′-dimethyl-2,2′-bipyridine) show the absorptions
of the radical anion of dmb, in line with the SBLCT character of the lowest excited state. The excited-state
lifetimes at room temperature vary between 0.5 and 3.6 µs and are mainly determined by the photolability of the
complexes. All complexes are photostable in a glass at 80 K, under which conditions they emit with very long
lifetimes. The extremely long emission lifetimes (e.g., τ ) 1.1 ms for [Ru(SnPh3)2(CO)2(dmb)]) are about a
3
thousand times longer than those of the MLCT states of the [Ru(Cl)(Me)(CO)2(R-diimine)] complexes. This is
3
due to the weak distortion of the former complexes in their SBLCT states as seen from the very small Stokes
shifts. Remarkably, replacement of Ru by Os hardly influences the absorption and emission energies of these
complexes; yet the emission lifetime is shortened because of an increase of spin-orbit coupling. The quantum
yield of emission at 80 K is 1-5% for these complexes, which is lower than might be expected on the basis of
their slow nonradiative decay.
Introduction
give rise to homolysis of the metal-metal or metal-alkyl bond
with formation of radicals. For quite a few [Re(L)(CO)3(R-
diimine)]+/0 complexes these radicals and their formation have
been studied with (time-resolved) spectroscopic techniques.6,7
Interestingly, not only are most complexes photostable at low
temperature but also their 3SBLCT states are much longer-lived
Most coordination and organometallic compounds containing
a low-valent transition metal and an R-diimine ligand such as
2,2′-bipyridine (bpy) possess rather intense low-energy metal-
to-ligand charge transfer (MLCT) transitions in the visible region
of the spectrum. Best known are [Ru(bpy)3]2+ 1,2 and [Re(Cl)-
(CO)3(bpy)],3,4 which proved to be good photosensitizers for
energy- and electron-transfer processes. Although [Ru(bpy)3]2+
is more suitable as a photosensitizer because of its longer
emission lifetime and greater stability of its oxidation product,
the [Re(L)(CO)3(R-diimine)]+/0 complexes are more flexible
because the ligand L can be varied at will,4 giving rise to large
variation in excited-state properties. Thus, when L ) Cl- is
replaced by L ) I-, the HOMO obtains predominant halide
character and the low-energy transitions change character from
MLCT or dπ(Re) f π*(R-diimine) to halide-to-ligand charge
transfer (XLCT) or pπ(I-) f π*(R-diimine).5 Yet another
situation arises if L is an alkyl or metal fragment bound to Re
via a high-lying σ(Re-L) orbital. The lowest excited state then
has σ(Re-L)π*(R-diimine) or sigma-bond-to-ligand charge
transfer (SBLCT) character.6 The SBLCT states are normally
shorter-lived than the MLCT and XLCT states because they
3
3
than MLCT or XLCT states usually are. For instance, [Re-
(Br)(CO)3(dmb)] (dmb ) 4,4′-dimethyl-2,2′-bipyridine) emits
3
in a 2-MeTHF glass at 80 K from its mixed MLCT/3XLCT
state at 525 nm with a lifetime of 3.7 µs, whereas [Re(SnPh3)-
3
(CO)3(dmb)] emits from its SBLCT state at 609 nm with a
lifetime of 1.1 × 102 µs under these conditions.5,6
To increase the variation in excited-state properties further,
we have extended our photochemical studies to complexes of
the type [Ru(L1)(L2)(CO)2(R-diimine)], in which the two ligands
L1 and L2 can be varied.8 Those complexes especially appeared
to be of great interest in which both L1 and L2 are bound to Ru
by a high-lying σ orbital. The HOMO of these complexes is a
σ(L1-Ru-L2) orbital, and accordingly, the SBLCT transition
has σ(L1-Ru-L2) f π*(R-diimine) character. Depending on
the relative strengths of the Ru-L1 and Ru-L2 bonds and their
involvement in the HOMO, one of these bonds is preferably
broken on irradiation. If both Ru-L1/L2 bonds are strong, as in
the case of L1 ) L2 ) SnPh3, the complexes proved to be less
* To whom correspondence should be addressed. E-mail: stufkens@
anorg.chem.uva.nl.
(1) Meyer, T. J. Pure Appl. Chem. 1986, 58, 1193.
(2) Juris, A.; Balzani, V.; Barigeletti, F.; Campagna, S.; Belser, P.; von
Zelewski, A. Coord. Chem. ReV. 1988, 84, 85.
(3) Schanze, K. S.; MacQueen, D. B.; Perkins, T. A.; Cabana, L. A. Coord.
Chem. ReV. 1993, 63.
(4) Stufkens, D. J. Comments Inorg. Chem. 1992, 13, 359.
(5) Rossenaar, B. D.; Stufkens, D. J.; Vlcˇek, A., Jr. Inorg. Chem. 1996,
35, 2902.
(6) Rossenaar, B. D.; Lindsay, E.; Stufkens, D. J.; Vlcˇek, A., Jr. Inorg.
Chim. Acta 1996, 250, 5.
(7) Kleverlaan, C. J.; Stufkens, D. J.; Clark, I. P.; George, M. W.; Turner,
J. J.; Martino, D. M.; van Willigen, H.; Vlcˇek, A., Jr. J. Am. Chem.
Soc. 1998, 120, 10871.
(8) Aarnts, M. P.; Stufkens, D. J.; Wilms, M. P.; Baerends, E. J.; Vlcˇek,
A., Jr.; Clark, I. P.; George, M. W.; Turner, J. J. Chem.sEur. J. 1996,
2, 1556.
10.1021/ic000857x CCC: $20.00 © 2001 American Chemical Society
Published on Web 12/16/2000

278 Inorganic Chemistry, Vol. 40, No. 2, 2001
van Slageren and Stufkens
ing to the procedure used for the synthesis of [Ru(I)2(CO)2(dmb)]. A
mixture of 1.0547 g (2.14 mmol) of [Ru(I)2(CO)2(MeCN)2] and 441.6
mg (2.37 mmol) of dmb was suspended in 50 mL of diethyl ether and
refluxed for 30 min. The reaction mixture was cooled to room
temperature, and the residue was filtered off (G3 glass filter), washed
with pentane, and dried in vacuo to yield the product as a yellow
powder. Yield: 94%. IR (THF): 2049, 1991 cm-1. 1H NMR (CDCl3):
δ 2.61 (s, 6H, dmb CH3), 7.38 (d, 3J ) 5.5 Hz, 2H, dmb H-5), 8.22 (s,
3
2H, dmb H-3), 8.95 (d, J ) 5.8 Hz, 2H, dmb H-6).
[Ru(I)2(CO)2(pAn-DAB)]: yield, ca. 90%. IR (THF): 2056, 2002
1
3
cm-1. H NMR (CDCl3): δ 3.88 (s, 6H, OCH3), 6.99 (d, J ) 9 Hz,
3
4H, o-C6H4OCH3), 7.68 (d, J ) 9 Hz, 4H, m-C6H4OCH3), 8.12 (s,
2H, imine H).
[Ru(I)2(CO)2(pAn-BIAN)]: yield, ca. 90%. IR (THF): 2056, 2003
1
3
cm-1. H NMR (CDCl3): δ 3.95 (s, 6H, OCH3), 7.12 (d, J ) 9 Hz,
6H, H3 + H9; see Figure 1 for numbering), 7.52 (pst, 2H, H4), 7.80
(d, J ) 8.8 Hz, 4H, H10), 8.04 (d, 2H, J ) 8.3 Hz, H5).
3
3
[Os(Cl)2(CO)2]n. The polymer [Os(Cl)2(CO)2]n was prepared
according to modified literature procedure.14 K2OsCl6 (823.3 mg, 1.7
mmol) was dissolved in a mixture of formic acid (40 mL) and
formaldehyde (aq, 40%, 15 mL). The reaction mixture was deaerated
by bubbling nitrogen through for 20 min and subsequently refluxed
for 3 days during which the color changed from dark-red to greenish
to light-yellow. The solvent was removed in vacuo, and the resulting
product was triturated with dichloromethane. The product was dissolved
in acetone and filtered to remove KCl. Evaporation of the solvent
yielded the product as an off-white powder. Yield: ca. 90%. IR
Figure 1. Schematic structures of the [M(SnR3)2(CO)2(R-diimine)]
complexes and the R-diimine ligands used.
photoreactive at room temperature and photostable and very
long-lived in their 3SBLCT state in a glass at 80 K. In the case
of [Ru(SnPh3)2(CO)2(iPr-DAB)] (iPr-DAB ) N,N′-diisopropyl-
1,4-diazabutadiene) an emission lifetime of 2.6 × 102 µs was
measured under these conditions, which is exceptional for
charge-transfer states of organometallic complexes.8 This result
prompted us to extend our investigations of these complexes
further and to develop organometallic systems that are photo-
stable at room temperature and emit in the near-infrared region
with still an appreciable lifetime to be of use as luminescent
labels in biochemical separations. For this purpose, R-diimine
ligands with low-lying π* orbitals were employed in order to
shift the emission to the NIR and to increase the barrier for the
radical formation. To further increase the photostability, Ru was
replaced by Os in [Ru(SnPh3)2(CO)2(R-diimine)] because transi-
tion metal atoms from the third row are expected to form
stronger bonds with tin than those of the first and second row.
Figure 1 shows the general structure of the R-diimine ligands
and of the complexes under study.
(THF): 2117, 2022 cm-1
.
[Os(Cl)2(CO)2(dmb)]. [Os(Cl)2(CO)2(dmb)] was prepared according
to a modified literature procedure.14 [Os(Cl)2(CO)2]n (269.0 mg, 0.85
mmol) and dmb (180.2 mg, 0.97 mmol) were dissolved in 30 mL of
n-propanol. The reaction mixture was refluxed for several hours until
IR spectral results showed complete conversion. After the solvent was
removed in vacuo, the product was purified by column chromatography
(activated silica, hexane/dichloromethane gradient elution). The product
was obtained as a light-yellow powder. Yield: ca. 90%. IR (THF):
2030, 1960 cm-1. UV-vis (THF), λmax: 296, 373 nm. 1H NMR
3
(CDCl3): δ 2.63 (s, 6H, dmb CH3), 7.45 (d, J ) 5.3 Hz, 2H, dmb
3
H-5), 8.02 (s, 2H, dmb H-3), 8.94 (d, J ) 5.7 Hz, 2H, dmb H-6).
[Os(Cl)2(CO)2(iPr-DAB)]. [Os(Cl)2(CO)2]n (201.1 mg, 0.63 mmol)
i
and Pr-DAB (182.1 mg, 1.30 mmol) were dissolved in 30 mL of
absolute ethanol. The reaction mixture was refluxed for several hours
until IR results showed complete conversion. The solvent was
evaporated, and after purification by column chromatography (activated
silica, hexane/THF ) 1:1), the product was obtained as an orange
powder. Yield: 63%. IR (THF): 2034, 1965 cm-1. UV-vis (THF),
Experimental Section
Materials. [Ru3(CO)12] (ABCR), K2OsCl6 (Alfa), I2 (Merck),
SnClPh3 (Merck, zur Synthese), SnClMe3 (Acros, 99%), 4,4′-dimethyl-
2,2′-bipyridine (dmb, Fluka), formic acid (Merck), and formaldehyde
(aq, 40%, EGA Chemie) were used as received. Solvents purchased
from Acros (THF, hexane, dichloromethane, acetonitrile, diethyl ether,
methanol, 2-MeTHF), Merck (heptane), BDH (absolute ethanol), and
Baker (n-propanol) were dried on and distilled from the appropriate
drying agent when necessary. Silica gel (kieselgel 60, Merck, 70-230
mesh) for column chromatography was dried and activated by heating
in vacuo at 160 °C overnight.
Syntheses. All syntheses were performed under a nitrogen atmo-
sphere using standard Schlenk techniques. N,N′-Diisopropyl-1,4-
diazabutadiene (iPr-DAB),9 N,N′-di(p-methoxyphenyl)-1,4-diazabuta-
diene (pAn-DAB),10 N,N′-bis(p-methoxyphenylimino)acenaphthene (pAn-
BIAN),11,12 and [Ru(SnPh3)2(CO)2(iPrDAB)]13 were prepared according
to literature procedures.
1
3
λ
max: 419 nm. H NMR (CDCl3): δ 1.50, 1.52 (d, J ) 6.6 Hz, 12H,
CH(CH3)2), 4.32 (sept, 3J ) 6.5 Hz, 2H, CH(CH3)2), 8.55 (s, 2H, imine-
CH).
[Ru(SnPh3)2(CO)2(dmb)]. [Ru(I)2(CO)2(dmb)] (206.2 mg, 0.35
mmol) was dissolved in 25 mL of THF. A solution of LiSnPh3 in THF
(prepared from SnClPh3 and freshly cut lithium metal) was added
gradually (in the dark) until IR results showed complete conversion.
Methanol (2 mL) was added to quench any unreacted LiSnPh3. The
solvent was evaporated, and after purification by column chromatog-
raphy in the dark (activated silica, hexane/dichloromethane gradient
elution) the product was obtained as a red microcrystalline powder.
Yield: ca. 50%. Anal. Calcd for C50H42N2O2RuSn2: C, 57.67; H, 4.07;
N, 2.69. Found: C, 57.33; H, 3.91; N, 2.58. FAB-MS m/z: 1042 (M+),
965 [M+ - Ph], 691 [M+ - SnPh3]. IR (THF): 1996, 1942 cm-1
.
UV-vis (THF), λmax: 327, 521 nm. 1H NMR (C6D6): δ 1.65 (s,
JSn-H ) 9 Hz, 6H, dmb CH3), 5.80 (d, 3J ) 5.6 Hz, 2H, dmb H5), 6.66
(s, dmb H3), 7.02 (m, 9H, m/p-SnC6H5), 7.41 (m, 6H, o-SnC6H5), 8.29
(d, 3J ) 5.9 Hz, 2H, dmb H-6). 13C NMR APT (C6D6): δ 20.2 (dmb-
Me), 122.8 (dmb C-5), 125.4 (dmb C-3), 127.1 (p-SnC6H5), 127.7 (m-
SnC6H5), 137.4 (JSn-C ) 36 Hz, o-SnC6H5), 145.1 (dmb C-4), 145.4
(dmb C-2), 150.7 (JSn-C ) 12 Hz, ipso-SnC6H5), 151.8 (JSn-C ) 12
Hz, dmb C-6), 208.0 (CO).
[Ru(I)2(CO)2(r-diimine)] (r-Diimine ) pAn-DAB, pAn-BIAN,
dmb). The [Ru(I)2(CO)2(R-diimine)] complexes were prepared accord-
(9) Kliegman, J. M.; Barnes, R. K. Tetrahedron 1970, 26.
(10) Kliegman, J. M.; Barnes, R. K. J. Org. Chem. 1970, 35, 3140.
(11) Van Asselt, R.; Gielens, E. E. C. G.; Ru¨lke, R. E.; Vrieze, K.; Elsevier,
C. J. J. Am. Chem. Soc. 1994, 116, 977.
(12) Van Asselt, R.; Elsevier, C. J.; Smeets, W. J. J.; Spek, A. L.; Benedix,
R. Recl. TraV. Chim. Pays-Bas 1994, 113, 88.
(13) Aarnts, M. P.; Stufkens, D. J.; Oskam, A.; Fraanje, J.; Goubitz, K.
Inorg. Chim. Acta 1997, 256, 93.
(14) Jandrasics, E. Z.; Keene, F. R. J. Chem. Soc., Dalton Trans. 1997,
153.

[M(SnR3)2(CO)2(R-diimine)]
Inorganic Chemistry, Vol. 40, No. 2, 2001 279
[Ru(SnPh3)2(CO)2(pAn-DAB)]. To a solution of 285 mg of
[Ru(I)2(CO)2(pAn-DAB)] in THF, 0.5 mL of NaK3 alloy was added.
Stirring at room temperature yielded a solution of a highly reactive
anionic intermediate. The remaining NaK3 alloy was filtered off using
a G3 frit, and 2 equiv of SnClPh3 were added in the dark. After column
chromatography (activated silica, dichloromethane/hexane gradient
elution) the product was obtained as a brownish-green powder. Yield:
of 292.8 mg (0.88 mmol) of SnClPh3 in 10 mL of THF. This mixture
was stirred for a few minutes, and the solvent was removed in vacuo.
After purification by column chromatography in the dark (silica, hexane/
dichloromethane gradient elution) the product was obtained as an orange
microcrystalline powder. Yield: ca. 50%. FAB-MS m/z: 1086 [M+],
1009 [M+ - Ph], 737 [M+ - SnPh3]. IR (THF): 1996, 1939 cm-1
.
1
UV-vis (THF), λmax: 287, 494 nm. H NMR (CDCl3): δ 0.95 (d,
ca. 50%. FAB-MS m/z: 1126 (M+), 1049 (M+ - Ph), 776 (M+
SnPh3). IR (THF): 2011, 1960 cm-1. UV-vis (THF), λmax: 396, 449,
-
3J ) 6.6 Hz, 12H, CH(CH3)2), 4.62 (sept, 3J ) 6.6 Hz, 2H, CH(CH3)2),
7.25 (m, 9H, m/p-SnC6H5), 7.32 (m, 6H, o-SnC6H5), 8.15 (s, JSn-H
)
1
3
570 nm. H NMR (C6D6): δ 3.18 (s, 6H, OCH3), 6.63 (d, J ) 9 Hz,
4H, o-C6H4OCH3), 6.82 (s, JSn-H ) 27 Hz, 2H, imine H), 7.09 (d,
3J ) 9 Hz, 4H, m-C6H4OCH3), 7.19 (m, 18H, m/p-SnC6H5), 7.50 (m,
12H, o-SnC6H5). 13C NMR APT (C6D6): δ 55.1 (OCH3), 114.4 (o-
C6H4OCH3), 124.5 (m-C6H4OCH3), 127.1 (p-C6H4OCH3), 128 (m/p-
SnC6H5), 138.0 (JSn-C ) 35 Hz, o-SnC6H5), 141.3 (ipso-SnC6H5), 160.5
(s, JSn-C ) 15 Hz, imine C), 204.0 (CO).
23.7 Hz, 2H, imine H). 13C NMR APT (C6D6): δ 24.8 (CH(CH3)2),
65.4 (CH(CH3)2), 128.0 (m/p-SnC6H5), 128.2 (m/p-SnC6H5), 137.7
(JSn-C ) 34 Hz, o-SnC6H5), 142.7 (ipso-SnC6H5), 148.5 (JSn-C ) 15
Hz, imine C), 187.9 (CO).
[Os(SnMe3)2(CO)2(iPr-DAB)]. [Os(SnMe3)2(CO)2(iPr-DAB)] was
prepared from [Os(Cl)2(CO)2(iPr-DAB)] and LiSnMe3 according to the
procedure used for the synthesis of [Ru(SnPh3)2(CO)2(dmb)] (vide
supra). Yield: ca. 50%. IR (THF): 1984, 1927 cm-1. UV-vis (THF),
[Ru(SnPh3)2(CO)2(pAn-BIAN)]. This complex was prepared from
[Ru(I)2(CO)2(pAn-BIAN)] and SnClPh3 according to the procedure for
[Ru(SnPh3)2(CO)2(pAn-DAB)]. Yield: ca. 50%. FAB-MS m/z: 1250
(M+), 1173 (M+ - Ph), 899 (M+ - SnPh3). IR (THF): 2011, 1960
cm-1. UV-vis (THF), λmax: 272, 321, 377sh, 400sh, 455, 607 nm. 1H
NMR (CD2Cl2): δ 3.86 (s, 6H, OCH3), 6.63 (d, 3J ) 8.7 Hz, 4H, H10;
λ
max: 257, 370, 484 nm. 1H NMR (C6D6): δ 0.28 (s, 18H, JSn-H ) 46
3
Hz, SnMe), 1.04 (d, 12H, J ) 6.6 Hz, CH(CH3)2), 4.47 (septet, 2H,
3J ) 6.6 Hz, CH(CH3)2), 7.52 (s, JSn-H ) 22 Hz, 2H, imine H). 13C
NMR APT (C6D6): δ -9.7 (JSn-C ) 228 Hz, SnCH3), 25.2 (CH-
(CH3)2)), 65.0 (CH(CH3)2), 143.0 (imine-C), 190.9 (JSn-C ) 38 Hz,
CO).
3
see Figure 1 for numbering), 6.73 (d, J ) 7.2 Hz, 2H, H3), 6.9 (m,
18H, m/p-SnC6H5), 7.03 (d, 3J ) 8.7 Hz, 4H, H9), 7.24 (m, 12H, o-Sn-
Spectroscopic Measurements. All spectroscopic measurements were
performed under a nitrogen atmosphere. Infrared spectra were recorded
on Bio-Rad FTS-7 and FTS-60A FTIR spectrophotometers (the latter
equipped with a liquid-nitrogen-cooled MCT detector) and electronic
absorption spectra on Varian Cary 4E and Hewlett-Packard 8453
spectrophotometers. NMR spectra were recorded on a Bruker AMX
300 (300.13 and 75.46 MHz for 1H and 13C, respectively) spectrometer.
Resonance Raman spectra of the complexes dispersed in KNO3 pellets
were recorded on a Dilor XY spectrometer equipped with a Wright
Instruments CCD detector, using a Spectra Physics 2040E Ar+ and
Coherent CR490 and CR590 dye lasers (with Coumarin 6 and
Rhodamine 6G dyes) as excitation sources. Steady-state emission
spectra were measured on a SPEX Fluorolog 2 (equipped with an RCA
C31034 Peltier cooled GaAs photomultiplier).
Nanosecond time-resolved electronic absorption and emission spectra
were obtained using a setup described previously.7 A Teflon mask
around the glass tube sample cell with 1 mm holes for the probe light
and a 1 cm slit for the pump light was used for the low-temperature
transient absorption spectrum. As irradiation sources, the second
harmonic (532 nm) of a Spectra Physics GCR3 Nd:YAG laser, a Quanta
Ray PDL pulsed dye laser with a Coumarin 440 solution (440 nm), or
a continuously tunable (360-700 nm) Coherent Infinity XPO laser were
used. Emission quantum yields were measured relative to a standard
solution of [Re(Cl)(CO)3(bpy)] in 2-MeTHF (Φ ) 0.028 at 77 K), using
a gate of 10 ms.
3
C6H5), 7.28 (pst, 2H, H4), 7.82 (d, 2H J ) 8.4 Hz, H5).13C NMR
APT (CD2Cl2): δ 56.2 (OCH3), 114.1 (C9; see Figure 1 for numbering),
122.8 (C3) 124.6 (C10), 127.7 (C4), 128.1 (p-SnC6H5), 128.4 (m-
SnC6H5), 128.5 (C2), 128.7 (C5), 131.1 (C6), 138.0 (JSn-C ) 34 Hz,
o-SnC6H5), 140.7 (C7), 142.9 (ipso-SnC6H5), 144.3 (C8), 159.9 (C11),
161.6 (C1), 204.08 (CO).
[Ru(SnMe3)2(CO)2(r-diimine)] (r-diimine ) dmb, iPr-DAB). The
[Ru(SnMe3)2(CO)2(R-diimine)] complexes were prepared by reaction
of [Ru(I)2(CO)2(R-diimine)] and LiSnMe3 according to the procedure
used for the synthesis of [Ru(SnPh3)2(CO)2(dmb)] (vide supra).
[Ru(SnMe3)2(CO)2(iPr-DAB)]: yield, ca. 50%. IR (THF): 1993,
1936 cm-1. UV-vis (THF), λmax
:
277, 404, 511 nm. 1H NMR
3
(CDCl3): δ 0.03 (s, JSn-H ) 43 Hz, 18H, SnMe), 1.38 (d, 12H, J )
3
6.6 Hz, CH(CH3)2), 4.47 (septet, 2H, J ) 6.6 Hz, CH(CH3)2), 7.82
(s, JSn-H ) 26 Hz, 2H, imine H). 13C NMR APT (C6D6): δ -8.5
(JSn-C ) 198 Hz, SnCH3), 25.0 (CH(CH3)2)), 63.8 (CH(CH3)2), 141.6
(imine-C), 181.5 (JSn-C ) 48 Hz, CO).
[Ru(SnMe3)2(CO)2(dmb)]: yield, ca. 50%. IR (THF): 1984, 1928
1
cm-1. UV-vis (THF), λmax: 257sh, 409, 598 nm. H NMR (C6D6) δ:
0.10 (s, JSn-H ) 38 Hz, 18H, SnMe 2.63 (s, JSn-H ) 11 Hz, 6H, dmb
CH3), 6.07 (d, 3J ) 5.9 Hz, 2H, dmb H-5), 7.08 (s, 2H, dmb H3), 8.70
(d, 3J ) 5.9 Hz, 2H, dmb H6). 13C NMR APT (C6D6): δ -10.1 (SnMe),
20.5 (dmb CH3), 122.8 (JSn-C ) 13 Hz, dmb C5), 124.4 (JSn-C ) 8
Hz, dmb C3), 144.3 (JSn-C ) 16 Hz, dmb C4), 149.7 (JSn-C ) 16 Hz,
dmb C2), 151.5 (JSn-C ) 12 Hz, dmb C6), 210.5 (CO).
[Os(SnPh3)2(CO)2(dmb)]. [Os(Cl)2(CO)2(dmb)] (332.4 mg, 0.66
mmol) was dissolved in 25 mL of THF. A solution of LiSnPh3 in THF
(prepared from SnClPh3 and freshly cut lithium metal) was added
gradually (in the dark) until IR results showed complete conversion.
Methanol (2 mL) was added to quench any unreacted LiSnPh3. The
solvent was evaporated, and after purification by column chromatog-
raphy in the dark (silica, hexane/dichloromethane gradient elution) the
product was obtained as a red microcrystalline powder. Yield: ca. 50%.
FAB-MS m/z: 1130 (M+), 1053 [M+ - Ph], 781 [M+ - SnPh3]. IR
(THF): 1989, 1930 cm-1. UV-vis (THF), λmax: 305, 358sh, 514 nm.
1H NMR (C6D6): δ 1.67 (s, JSn-H ) 9 Hz, 6H, dmb CH3), 5.71 (d,
3J ) 5.9 Hz, 2H, dmb H5), 6.62 (s, dmb H3), 7.02 (m, 9H,
m/p-SnC6H5), 7.41 (m, 6H, o-SnC6H5), 8.45 (d, 3J ) 6.0 Hz, 2H, dmb
H6), 13C NMR APT (C6D6): δ 20.1 (dmb CH3), 123.1 (dmb C5), 126.2
Photochemical quantum yields were determined by observation of
the decay of the first absorption band of solutions of the complexes in
dichloromethane at 21.0 °C by in situ irradiation in a Varian Cary 4E
spectrophotometer using previously described procedures.7
Results
(I) Syntheses. The procedure used for the synthesis of
[Ru(SnPh3)2(CO)2(iPr-DAB)]15 is not suitable for the preparation
of the [Ru(SnR3)2(CO)2(R-diimine] complexes containing R-di-
imine ligands other than R-DAB. Instead, these complexes were
prepared from [Ru(I)2(CO)2(R-diimine)], which was obtained
by reaction of the appropriate R-diimine ligand with [Ru(I)2-
(CO)2(MeCN)2],16 and LiSnR3.
However, during the preparation of [Ru(SnPh3)2(CO)2(pAn-
BIAN)] the last step resulted in decomposition. [Ru(I)2(CO)2-
(pAn-BIAN)] was therefore prepared first and subsequently
reduced using a sodium-potassium alloy to give a highly
(dmb C3), 127.2 (m/p-SnC6H5), 127.7 (m/p-SnC6H5), 137.5 (JSn-C
)
35 Hz,, o-SnC6H5), 144.0 (JSn-C ) 16 Hz, dmb C4), 145.2 (JSn-C ) 13
Hz, dmb C2), 151.7 (dmb C6), 150.7 (JSn-C ) 11 Hz, ipso-SnC6H5),
190.8 (CO).
[Os(SnPh3)2(CO)2(iPr-DAB)]. Os(Cl)2(CO)2(iPr-DAB) (182.5 mg,
0.40 mmol) was dissolved in 30 mL of THF. After adiition of NaK2.8
alloy (0.5 mL), the color changed from orange to green to brown-yellow.
The reaction mixture was filtered and added in the dark to a solution
(15) Aarnts, M. P.; Wilms, M. P.; Peelen, K.; Fraanje, J.; Goubitz, K.;
Hartl, F.; Stufkens, D. J.; Baerends, E. J.; Vlcˇek, A., Jr. Inorg. Chem.
1996, 35, 5468.
(16) Irving, R. J. J. Chem. Soc. 1956, 2879.

280 Inorganic Chemistry, Vol. 40, No. 2, 2001
van Slageren and Stufkens
Table 1. Electronic Absorption Spectral Data and Transient Absorption Lifetimes of the M(SnR3)2(CO)2(R-diimine) Complexes at Room
Temperature
electronic absorption (nm)
transient
entry
metal
R
R-diimine
toluene
CH2Cl2 (ꢀ)a
MeCN
∆b (nm)
lifetimec (µs)
1d
2
3
4
5
6
7
8
9
Ru
Ru
Ru
Ru
Ru
Ru
Os
Os
Os
Ph
Ph
Ph
Ph
Me
Me
Ph
Ph
Me
iPr-DAB
pAn-DAB
pAn-BIAN
dmb
523
577
614
542
517
630
497
537
485
519 (6.8)
572 (2.7)
605 (16)
529 (3.8)
515 (7.2)
612 (5.3)
495 (6.2)
519 (4.1)
484 (6.9)
508
567
601
503
511
581
485
496
482
0.57
0.31
0.35
1.4
0.23
1.3
0.50
1.5
0.13
1.0
1.9
3.6
1.0
2.6e
0.50
1.5
2.5
1.4
iPr-DAB
dmb
iPr-DAB
dmb
iPr-DAB
a ꢀ in 103 M-1 cm-1
of overlap of transient and ground-state absorptions.
. .
b ∆ ) νmax(MeCN) - νmax(toluene) in 103 cm-1 c In THF at room temperature. d From ref 13. e Uncertainty is large because
Table 2. Resonance Raman Data of the Complexes [M(SnR3)2(CO)2(iPr-DAB)] (M ) Ru, Os; R ) Ph, Me) and [Ru(Cl)(Me)(CO)2(iPr-DAB)]
in KNO3
resonance Raman data
metal
Rua
Os
Ru
Os
R
νs(CO)
νs(CN)
δs(CH)
δ(DAB)
Ph
Ph
Me
Me
1473s
1467s
1473s
1470m
1283s
1272s
1286s
1279s
1166w
1168w
1178s
1172s
953s
958s
971s
968s
836s
844s
850s
850s
651w
657w
651m
646w
610m
614m
615m
614m
419w
424m
425w
420m
247m
256m
264w
251w
197m
211w
229m
189w
496w
494w
RuClMeb
2033w
1568s
487m
a From ref 8. b [Ru(Cl)(Me)(CO)2(iPr-DAB)], from ref 17.
reactive intermediate. This intermediate was then allowed to
react with 2 equiv of SnClPh3, yielding the desired product.
This method, which has also been used for the synthesis of [Ru-
(Me)(I)(CO)2(bpy)],17,18 was also successfully used for the
synthesis of [Ru(SnPh3)2(CO)2(pAn-DAB)].
less solvatochromic (∆ν ) 0.2 × 103 to 0.6 × 103 cm-1) than
those of the aromatic dmb complexes (∆ν ) 1.2 × 103 to
1.5 × 103 cm-1) (Table 1, entries 4, 6, and 8).
On going to a 2-MeTHF glass at 80 K, the absorption bands
of the complexes become narrower and shift by ca. 20 nm to
shorter wavelengths. They become asymmetric for the R-DAB
complexes, while those of the aromatic R-diimine compounds
show a pronounced shoulder on their short-wavelength side.
Because the DFT calculations on the model complex [Ru-
(SnH3)2(CO)2(H-DAB)] do not show the presence of any close-
lying electronic transition,15 this shoulder is attributed to a
vibrational sideband.
The recently synthesized complexes [Os(Cl)2(CO)2(R-di-
imine)] 14 proved to be excellent starting compounds for the
preparation of [Os(SnPh3)2(CO)2(R-diimine)]. The synthesis
started with the formation of the polymer [Os(Cl)2(CO)2]n, which
was allowed to react with the R-diimine ligand. Subsequent
addition of LiSnPh3 to a solution of this complex afforded [Os-
(SnPh3)2(CO)2(R-diimine)] in the case of dmb. However, in the
i
case of the Pr-DAB complex, the last step of this reaction
To further characterize the SBLCT transition, we studied the
resonance Raman (rR) spectra of the complexes. Upon excitation
into an allowed electronic transition, such rR spectra normally
show resonance enhancement of Raman intensity for those
vibrations that are most strongly coupled to this transition.19 In
other words, this technique allows us to characterize the
electronic transition by revealing which bonds of the complex
are affected most. RR spectra were recorded for all [M(SnR3)2-
(CO)2(iPr-DAB)] (M ) Ru, Os; R ) Me, Ph) complexes under
study and for comparison also for [Ru(Cl)(Me)(CO)2(iPr-DAB)]
in order to find out in which way the rR spectra are affected by
the type of electronic transition (SBLCT vs MLCT). The
wavenumbers of the most strongly enhanced Raman bands are
collected in Table 2. The spectrum of [Ru(Cl)(Me)(CO)2(iPr-
DAB)] is rather simple (Figure 2a); it shows a strong rR effect
for νs(CN) of the iPr-DAB ligand at 1568 cm-1 and for a band
at 487 cm-1, belonging to either ν(Ru-CH3) or νs(Ru-CO).
A weaker effect is observed for νs(CO) at 2033 cm-1. This
spectrum is characteristic of excitation into an MLCT transition
because such a transition is accompanied by reduction of the
iPr-DAB ligand (rR effect for νs(CN)) and oxidation of the
central metal atom (rR effect for νs(CO)).
sequence resulted in decomposition, and [Os(SnPh3)2(CO)2-
(iPr-DAB)] was therefore synthesized from [Os(Cl)2(CO)2-
(iPr-DAB)] using the procedure used for [Ru(SnPh3)2(CO)2(pAn-
BIAN)].
All compounds are strongly colored microcrystalline powders.
They have a trans-(SnR3, SnR3), cis-(CO, CO) configuration,
as can be seen from their IR and NMR spectra.13 The complexes
are photostable in the solid state but photolabile in solution to
varying degrees.
(II) Electronic Absorption and Resonance Raman Spectra.
All complexes under study show an intense absorption band at
500-600 nm (Table 1), which has been assigned to a σ(Sn-
Ru-Sn) f π*(iPr-DAB) sigma-bond-to-ligand charge transfer
transition in the case of [Ru(SnPh3)2(CO)2(iPr-DAB)].13 The
absorption bands are only weakly solvatochromic. For instance,
∆ν ) νmax(MeCN) - νmax(toluene) ) 0.57 × 103 cm-1 for [Ru-
(SnPh3)2(CO)2(iPr-DAB)], which is much less than the solva-
tochromism of the MLCT band of, for example, the isostructural
complex [Ru(Cl)(Me)(CO)2(iPr-DAB)] (∆ν ) 1.9 × 103
cm-1).17 Furthermore, the absorption bands of the R-DAB and
pAn-BIAN complexes (Table 1, entries 1, 2, 5, 7, and 9) are
(17) Nieuwenhuis, H. A.; Stufkens, D. J.; Oskam, A. Inorg. Chem. 1994,
33, 3212.
(18) Rohde, W.; tom Dieck, H. J. Organomet. Chem. 1987, 328, 209.
(19) Clark, R. J. H.; Dines, T. J. Angew. Chem., Int. Ed. Engl. 1986, 25,
131.

[M(SnR3)2(CO)2(R-diimine)]
Inorganic Chemistry, Vol. 40, No. 2, 2001 281
Figure 2. Resonance Raman spectra of (a) [Ru(Cl)(Me)(CO)2(iPr-
Figure 3. Transient absorption difference spectra (solid lines) and
ground-state absorption spectra (dotted lines) of [Os(SnPh3)2(CO)2-
(dmb)] in THF at room temperature (A) and in a 2-MeTHF glass at 90
K (B). The delays between the transient absorption spectra are 200 ns
(A) and 20 µs (B).
DAB)] (λexc ) 457.9 nm), (b) [Ru(SnPh3)2(CO)2(iPr-DAB)] (λexc
)
457.9 nm), (c) [Os(SnPh3)2(CO)2(iPr-DAB)] (λexc ) 488.0 nm), (d) [Ru-
(SnMe3)2(CO)2(iPr-DAB)] (λexc ) 476.5 nm), and (e) [Os(SnMe3)2-
-
(CO)2(iPrDAB)] (λexc ) 476.5 nm). Asterisks denote NO3 peaks.
The bands observed in the rR spectra of the [M(SnPh3)2(CO)2-
(iPr-DAB)] complexes are weaker than those of [Ru(Cl)(Me)-
(CO)2(iPr-DAB)], while more bands are resonantly enhanced.
The spectra of, for example, [Ru(SnPh3)2(CO)2(iPr-DAB)] show
rR effects for stretching and deformation modes of the iPr-DAB
ligand (1473, 1283, 953, and 836 cm-1), while νs(CO) is not
observed at all. Some of these resonance effects are exceptional,
and their occurrence will be explained in the Discussion. In
addition, several lower-frequency rR bands (at 610, 419, 247,
and 197 cm-1 for [Ru(SnPh3)2(CO)2(iPr-DAB)]) are observed,
which belong to (combined) metal-ligand stretching and
deformation modes.20,21 The band at about 250 cm-1 is
tentatively assigned to a vibration having predominant ν(M-
Sn) or ν(Sn-M-Sn) (M ) Ru, Os) character because it is
observed for all SnR3 complexes having a lowest SBLCT
transition, but not for any other complex. The corresponding
osmium complex [Os(SnPh3)2(CO)2(iPr-DAB)] has virtually the
same rR spectrum, while the spectra of the corresponding
[M(SnMe3)2(CO)2(iPr-DAB)] complexes additionally show a
strong rR effect for a band at ca. 1170 cm-1, which is assigned
to a CH3 deformation mode of the SnMe3 ligand.22
band between 350 and 400 nm, have also been observed in the
TA spectra of [Re(Br)(CO)3(dmb)],23 [Re(SnPh3)(CO)3(dmb)],6
and [Re(CH3)(CO)3(dmb)]7 and in the spectrum of reduced
[Re(Br)(CO)3(dmb)].23 They closely resemble the bands found
24
in the absorption spectrum of the [dmb]•- radical anion and
are therefore assigned to the intraligand transitions of the
[dmb]•- radical anion in the SBLCT states of the complexes.
At room temperature the excited states, which have lifetimes
varying between 0.5 and 3.6 µs, are quenched by oxygen, which
confirms their triplet character. For instance, the transient
lifetime of [Os(SnPh3)2(CO)2(dmb)] is reduced by a factor of
10 in the presence of oxygen.
Nanosecond time-resolved emission spectra were recorded
for the compounds in a 2-MeTHF glass at 90 K, under which
conditions the complexes are completely photostable. The
emission data, collected in Table 3, show that the emitting states
of all [M(SnR3)2(CO)2(R-diimine)] complexes are very long-
3
lived, much longer than the MLCT state of the structurally
related compound [Ru(Cl)(Me)(CO)2(iPr-DAB)]. The longest
lifetime (τ ) 1.1 ms) is observed for [Ru(SnPh3)2(CO)2(dmb)],
in which complex the dmb is a rigid aromatic ligand. The
SnMe3-substituted complexes have slightly shorter emission
lifetimes than the SnPh3 ones, which were previously observed
for [Re(SnR3)(CO)3(phen)] (R ) Me, Ph).25
(III) Time-Resolved Electronic Absorption and Emission
Spectra. Nanosecond time-resolved absorption (TA) spectra of
the complexes were measured in THF at room temperature and
for [Os(SnPh3)2(CO)2(dmb)] also in a 2-MeTHF glass at 90 K.
Both spectra of the latter complex are shown in Figure 3. In
most spectra the bleaching of the ground-state absorption is not
observed because of the much stronger excited-state absorption.
The TA spectra of all complexes are very similar and consist
of strong absorptions with maxima at ca. 350 and 530 nm and
a weak, broad band above 600 nm. The low-temperature
spectrum of [Os(SnPh3)2(CO)2(dmb)] shows that this latter
absorption is a separate band and not the tail of the 530 nm
band. The 90 K spectrum of [Os(SnPh3)2(CO)2(dmb)] shows
that the 530 nm absorption band consists of two components.
Similar features, i.e., a broad, composite band around 500 nm,
a very broad and weak absorption above 600 nm, and an intense
Figure 4 shows the absorption and excitation spectra of
[Ru(SnPh3)2(CO)2(dmb)], together with its continuous wave
emission spectrum excited at 500 nm. The excitation spectrum
does not deviate from the absorption spectrum, which means
that the population of the emissive state has the same efficiency
throughout the first absorption band.
Just as for [Ru(SnPh3)2(CO)2(iPr-DAB)]8 a weak emission
is observed at the low-energy side of the broad emission band,
which is only produced by excitation at the low-energy side of
the first absorption band. Its lifetime is somewhat shorter than
that of the much stronger high-energy component. For instance,
the emission of [Ru(SnPh3)2(CO)2(dmb)] has a lifetime of 1.1
ms at 440 nm excitation (Table 3) and 0.76 ms at 532 nm
excitation. Because the lifetimes do not differ much, taking into
(20) Kokkes, M. W.; Snoeck, T. L.; Stufkens, D. J.; Oskam, A.; Chris-
tophersen, M.; Stam, C. H. J. Mol. Struct. 1985, 131, 11.
(21) Andre´a, R. R.; de Lange, W. G. J.; Stufkens, D. J.; Oskam, A. Inorg.
Chim. Acta 1988, 149, 77.
(22) Kleverlaan, C. J.; Stufkens, D. J.; Fraanje, J.; Goubitz, K. Eur. J. Inorg.
Chem. 1998, 1243.
(23) Rossenaar, B. D.; Stufkens, D. J.; Vlcˇek, A., Jr. Inorg. Chim. Acta
1996, 247, 247.
(24) Krejcˇ´ık, M.; Vlcˇek, A. A. J. Electroanal. Chem. 1991, 313, 243.
(25) Luong, J. C.; Faltynek, R. A.; Wrighton, M. S. J. Am. Chem. Soc.
1980, 102, 7892.

282 Inorganic Chemistry, Vol. 40, No. 2, 2001
van Slageren and Stufkens
Table 3. Emission Data of [M(SnR3)2(CO)2(R-diimine)] and [Ru(Cl)(Me)(CO)2(iPr-DAB)] in a 2-MeTHF Glass at 90 K
λabs
λem
∆Eabs-em
τ
Φem
Φ
ISCkr
knr
entry
metal
R
R-diimine
(nm)
(nm)
(103 cm-1
)
(102 µs)
(10-2
)
(102 s-1
0.55
)
(104 s-1
)
1
2
3
4
5
6
7
8
9
Rua
Ru
Ru
Ru
Ru
Ru
Os
Os
Os
Ph
Ph
Ph
Ph
Me
Me
Ph
Ph
Me
iPr-DAB
pAn-DAB
pAn-BIAN
dmb
495
552
595
495
507
567
478
485
478
387
633
767
821
604
733
736
655
589
714
650
5.3
5.0
4.6
3.6
6.1
4.0
5.7
3.6
6.9
10
2.6
0.72
0.68
11
0.62
0.60
0.32
2.3
1.5
0.37
5.7
0.62
0.10
iPr-DAB
dmb
iPr-DAB
dmb
0.58
3.4
1.8
1.5
3.13
0.42
iPr-DAB
0.16
0.003
[Ru(Cl)(Me)(CO)2(iPr-DAB)]b
a From ref 8. b From ref 47.
0.034
11
384
from the σ(Ru-R) or σ(Ru-Sn) orbitals to π*(R-diimine) are
normally not observed, and the lowest-energy transitions of these
latter complexes have dπ(Ru) f π*(R-diimine) (MLCT) char-
acter. This situation changes completely when two metal
fragments are coordinated in an axial position to Ru (or Os), as
in the case of [M(SnR3)2(CO)2(R-diimine)], according to density
functional (DFT) MO calculations on the model complex [Ru-
(SnH3)2(CO)2(H-DAB)].15 The HOMO, denoted as σ(Sn-Ru-
Sn), consists of contributions from the antisymmetric combi-
nation of the Sn fragment σ orbitals Sn(sp3-sp3) (42%) and
from the Ru(5p) (15%) and H-DAB(π*) (27%) orbitals. This
implies a strong σ-π* interaction, i.e., a large delocalization
of electron density from the Sn-Ru-Sn σ bond over the
H-DAB ligand. According to the calculations, the LUMO of
the model complex is also delocalized because it has contribu-
tions from H-DAB(π*) (61%), Ru(4dyz) (11%), and Sn(sp3-
sp3) (27%). The σ(Sn-M-Sn) f π*(R-diimine) transition
between the HOMO and LUMO is strongly allowed. In view
of the nature of the orbitals involved, this transition is called
sigma-bond-to-ligand charge transfer.26 Because of the strong
σ-π* interaction, the lowest-energy (SBLCT) transitions of the
complexes under study are less solvatochromic than the MLCT
transition of the isostructural complex [Ru(Cl)(Me)(CO)2(iPr-
DAB)] (see Table 1). The population of π*(iPr-DAB) in the
ground state, due to the strong σ-π* interaction, causes a
lengthening of the CN bond and a shortening of the CC bond
because the lowest π* orbital of an R-diimine such as iPr-DAB
is antibonding between the N and C atoms of the NdC-CdN
skeleton and is bonding between the central C atoms. This is
evident from the crystal structure of [Ru(SnPh3)2(CO)2(iPr-
DAB)] in which CN and CC bond lengths of 1.34 and 1.39 Å,
respectively, were found. This indicates much more σ-π*
interaction than for [Ru(I)(Me)(CO)2(iPr-DAB)] having CN and
CC bond lengths of 1.26 and 1.48 Å, respectively. As can be
seen from Table 1, the absorption bands of the R-DAB and
pAn-BIAN complexes are less solvatochromic than those of the
aromatic dmb compounds. This means that the σ and π* orbitals
of the dmb complexes have less interaction and, accordingly,
their σ(Sn-M-Sn) f π*(R-diimine) (M ) Ru, Os) transitions
have more charge-transfer character. The strong σ-π* interac-
tion of the R-DAB complexes causes the SBLCT transition to
occur at higher energy than expected on the basis of its π*-
orbital energy. This effect becomes evident when SnPh3 is
replaced by the more electron-donating SnMe3 in the complexes
[Ru(SnR3)2(CO)2(R-diimine)] (R ) Ph, Me; R-diimine )
iPr-DAB, dmb) (Table 1). A red shift of the absorption band is
then observed in the case of the dmb complexes but not for the
iPr-DAB compounds.
Figure 4. Emission spectrum (solid line, λexc ) 500 nm), excitation
spectrum (dashed line, λem ) 620 nm), and ground-state absorption
spectrum (dotted line) of [Ru(SnPh3)2(CO)2(dmb)] in a 2-MeTHF glass
at 90 K.
account a difference in emission energy, both emissions most
likely belong to the same excited state of the complex in a
different environment or isomeric form. The former explanation
was given in the case of [Re(SnPh3)(CO)3(bpy)] because the
effect was not observed for this complex in its solid state.25
Variable excitation wavelength time-resolved emission measure-
ments, using a continuously tuneable Coherent Infinity XPO
laser, showed that both the emission maximum and lifetime are
constant for excitation wavelengths covering most of the
absorption band and only start to change at the extreme long-
wavelength side of the absorption band.
Replacement of ruthenium by osmium has only a small effect
on the absorption and emission energies of the [M(SnPh3)2(CO)2-
(R-diimine)] complexes. Despite this, the emission lifetimes of
the Os complexes are much shorter because of the increase of
spin-orbit coupling (SOC) going from Ru to Os. For all
complexes the quantum yields of emission (Table 3) from the
3SBLCT states are rather small in view of their very long
emission lifetimes. In the next section we will discuss this
observation and its consequences in more detail.
Discussion
The complexes under study belong to a group of R-diimine
compounds in which two coligands are bound to the central
metal atom via high-lying σ orbitals. These coligands may be
alkyl groups or metal fragments. The lowest-energy transitions
of these complexes are fundamentally different from those of
complexes with only one such coligand., e.g., [Ru(Cl)(R)(CO)2-
(R-diimine)]17 or [Ru(Cl)(SnPh3)(CO)2(R-diimine)].8 Transitions
(26) Djurovich, P. I.; Watts, R. J. Inorg. Chem. 1993, 32, 4681.

[M(SnR3)2(CO)2(R-diimine)]
Inorganic Chemistry, Vol. 40, No. 2, 2001 283
The differences between the [M(SnPh3)2(CO)2(R-diimine)]
complexes possessing a low-energy SBLCT transition and the
isostructural complex [Ru(Cl)(Me)(CO)2(iPr-DAB)] having a
lowest MLCT transition are not reflected in the absorption
spectra but become evident when their resonance Raman spectra
and especially their photophysical and photochemical behavior
are compared. The main difference is the observation of rather
strong rR effects for a few vibrations in the case of [Ru(Cl)-
(Me)(CO)2(iPr-DAB)] and weak rR effects for many vibrations
in the case of the [M(SnR3)2(CO)2(iPr-DAB)] complexes. The
latter observation confirms the delocalized character of the
SBLCT transition during which many bonds are only weakly
distorted in the excited state. This weakness of distortion is also
demonstrated by the emssion spectra (vide infra). The rR spectra
of [Ru(SnPh3)2(CO)2(iPr-DAB)] (which is taken as a representa-
tive for all the [M(SnR3)2(CO)2(iPr-DAB)] complexes) show
strong rR effects for bands at 1473, 1283, 953, and 836 cm-1
,
Figure 5. Qualitative potential energy curves of ground and excited
states of [Ru(SnPh3)2(CO)2(iPr-DAB)] (solid line) and [Ru(SnPh3)2-
(CO)2(pAn-DAB)] (dotted line).
while νs(CO) is not observed at all. The absence of νs(CO)
implies that the charge density at the central metal atom is hardly
affected by the electronic transition. This result agrees with the
main conclusion from the DFT calculations on the model
complex [Ru(SnH3)2(CO)2(H-DAB)] that the central metal atom
and the carbonyls are hardly involved in the σ f π* (SBLCT)
transition.15 The rR band at 1473 cm-1 is assigned to νs(CN),
which is lower in frequency than for [Ru(Cl)(Me)(CO)2(iPr-
DAB)] (1568 cm-1) because of the strong σ-π* interaction.
The observation of a rR effect for a band at 1283 cm-1 is
exceptional. It has only been observed for complexes such as
[W(CO)4(R-DAB)] (R ) p-tolyl, mesityl),27 [Re{Re(CO)5}-
(CO)3(iPr-DAB)],20 and [Ru(L1)(L2)(CO)2(iPr-DAB)] (L1, L2 )
metal fragment)8,28,29 in which there is a very strong dπ-π* or
σ-π* interaction (π-backbonding). According to preliminary
calculations, it is a coupled δs(CH) + νs(CN) vibration in which
δs(CH) is a symmetric in-plane deformation of the imine
hydrogen atoms. This coupling, which is responsible for the
resonance enhancement of this vibration, most probably occurs
because of the small energy difference between these two local
modes. In the case of [Ru(SnPh3)2(CO)2(iPr-DAB)] strong rR
effects are also observed for deformation modes of iPr-DAB at
953 and 836 cm-1. Again, these vibrations are always observed
when there is a strong dπ-π* or σ-π* interaction.8,27-29 The
observation of ν(M-Sn) and δ(CH3) (for the SnMe3 complexes)
indicates that the transition indeed occurs from a σ(Sn-M-
Sn) orbital rather than from a dπ(M) orbital.
lifetimes (Table 1) although the dmb ligand is much more rigid
3
and the SBLCT state of its complex is at somewhat higher
energy (Table 3). The large influence of the photoreactivity on
i
the excited-state lifetime also becomes evident when the Pr-
DAB ligand is replaced by an R-diimine with a lower-lying π*
orbital such as pAn-DAB or pAn-BIAN. The SBLCT states are
then lower in energy, and although the energy gap law (EGL)
predicts a decrease of excited-state lifetime, this lifetime
becomes much longer because of the larger photostability (Table
1). For instance, the quantum yield for the photoreaction of the
[Ru(SnPh3)2(CO)2(R-DAB)] complexes in CH2Cl2 at room
i
temperature is 0.10 for R ) Pr but only 0.006 for R ) pAn.
This increase of photostability is most probably due to an
increase of the barrier for this reaction, as clarified in Figure 5
by the qualitative potential curves for ground and 3SBLCT states
of the [Ru(SnPh3)2(CO)2(R-DAB)] (R ) iPr, pAn) complexes.
Because the Os-Sn bonds are stronger than the Ru-Sn bonds,
the [Os(SnPh3)2(CO)2(R-diimine)] complexes are more photo-
stable than the Ru ones (e.g., Φ ) 0.038 for [Os(SnPh3)2(CO)2-
(iPr-DAB)] and Φ ) 0.10 for its Ru analogue). As a result,
their SBLCT states are also longer-lived than those of the Ru
compounds, despite the larger spin-orbit coupling constant of
the Os atom. Because of their photostability, the lifetimes of
the Os complexes at room temperature also increase when iPr-
DAB is replaced by dmb, i.e., when the R-diimine becomes
more rigid (Table 1).
According to the TA spectra, the complexes [M(SnPh3)2-
(CO)2(R-diimine)] have much longer excited-state lifetimes at
room temperature (τ ) 0.5-3.6 µs; Table 1) than [Ru(Cl)(Me)-
(CO)2(iPr-DAB)] (τ ) 63 ns)30 even though the former
complexes are photolabile. This photolability, not observed for
[Ru(Cl)(Me)(CO)2(iPr-DAB)], is a specific property of com-
plexes having a lowest SBLCT state and involves a homolytic
splitting of a M-Sn bond from this state.31 Because of this
The differences between the 3SBLCT states of the [M(SnPh3)2-
(CO)2(R-diimine)] complexes and the 3MLCT state of [Ru(Cl)-
(Me)(CO)2(iPr-DAB)] become even more pronounced at low
temperature, under which conditions all complexes are photo-
stable. The emitting 3SBLCT states of the [M(SnR3)2(CO)2(R-
3
3
diimine)] complexes are much longer-lived than the MLCT
photolability, the SBLCT states of [Ru(SnPh3)2(CO)2(dmb)]
state of [Ru(Cl)(Me)(CO)2(iPr-DAB)], an effect that was already
noted for the complex [Ru(SnPh3)2(CO)2(iPr-DAB)].8 This
increase of lifetime is caused by a decrease of distortion of the
complexes in their lowest excited state, which is reflected in a
decrease of the apparent Stokes shift (i.e., the energy difference
between the absorption and emission maxima, ∆Eabs-em; see
Table 3) from 10 × 103 cm-1 to (3.5-6.0) × 103 cm-1 going
from [Ru(Cl)(Me)(CO)2(iPr-DAB)] to [M(SnR3)2(CO)2(R-di-
and [Ru(SnPh3)2(CO)2(iPr-DAB)] have coincidentally the same
(27) Balk, R. W.; Stufkens, D. J.; Oskam, A. J. Chem. Soc., Dalton Trans.
1982, 275.
(28) Aarnts, M. P.; Wilms, M. P.; Stufkens, D. J.; Baerends, E. J.; Vlcˇek,
A., Jr., Organometallics 1997, 16, 2055.
(29) Van Slageren, J.; Hartl, F.; Stufkens, D. J. Eur. J. Inorg. Chem. 2000,
847.
(30) Nieuwenhuis, H. A.; Stufkens, D. J.; McNicholl, R. A.; Al-Obaidi,
A. H. R.; Coates, C. G.; Bell, S. E. J.; McGarvey, J. J.; Westwell, J.;
George, M. W.; Turner, J. J. J. Am. Chem. Soc. 1995, 117, 5579.
(31) Aarnts, M. P.; Stufkens, D. J.; Vlcˇek, A., Jr. Inorg. Chim. Acta 1997,
266, 37.
3
imine)]. This implies that for the SBLCT state the potential
energy curve is shifted less with respect to that of the ground
state and that the vibrational overlap between these curves is

284 Inorganic Chemistry, Vol. 40, No. 2, 2001
van Slageren and Stufkens
smaller for [M(SnPh3)2(CO)2(iPr-DAB)] than for [Ru(Cl)-
(Me)(CO)2(iPr-DAB)],32-34 causing a decrease of the rate
constant for nonradiative decay knr (EGL effect). In fact, knr
decreases by a factor of 1000 going from [Ru(Cl)(Me)(CO)2-
(iPr-DAB)] to [Ru(SnPh3)2(CO)2(iPr-DAB)] (Table 3) even
though the emission energy hardly changes. Correspondingly,
the emission lifetime, which is mainly determined by knr,
increases by a factor of ca. 1000 from 0.30 to 2.6 × 102 µs.
This increase of vibrational overlap with an increase of distortion
is clear from Figure 5. The overlap is minimal if the equilibrium
distance is the same in the ground and excited states but
increases dramatically if there is an appreciable lengthening of
this distance in the excited state.
is 8 ms at 6.5 K.42 For analogous transition metal complexes
the excited-state lifetime is shorter, e.g., τ ) 6.1 µs for [Pt-
(bpy)(mnt)] (mnt ) maleonitriledithiolate) in the solid state at
77 K.
Replacement of ruthenium by osmium has only a small effect
on the absorption and emission energies of the [M(SnPh3)2(CO)2-
(R-diimine)] complexes. Thus, the energy of the 3SBLCT state
hardly varies with M. Despite this, the excited-state lifetime is
much shorter because of an increase in spin-orbit coupling. In
contrast to this, MLCT states show a decrease of both the
emission energy and lifetime when Ru is replaced by Os. For
instance, the complex [Ru(bpy)3]Cl2 emits in a 77 K ethanol/
methanol glass at 584 nm with a lifetime of 5.3 µs, while [Os-
(bpy)3]Cl2 emits at 710/773 nm with a lifetime of 0.83 µs under
these circumstances.43 Complexes with a lowest IL state also
show a decrease of excited-state lifetime going from a second-
to a third-row transition metal. For instance, the excited-state
lifetime of [M(bpy)3]3+ is 2.2 ms for M ) Rh 39 but is only
0.080 ms for M ) Ir.40 Likewise, the excited-state lifetime of
[M(TPP)] (TPP ) tetraphenylporphyrine) is 2800 µs for M )
Pd and 291 µs for M ) Pt.38
If an R-DAB ligand is replaced by pAn-BIAN and finally
by a fully aromatic ligand, the R-diimine becomes more rigid.
The complex is then even less distorted in its excited state, and
this results in a smaller apparent Stokes shift, a smaller value
of knr, and a longer emission lifetime. Of course, part of the
decrease of knr is caused by the fact that the dmb complex emits
at somewhat higher energy. The rigidity of the dmb ligand,
3
combined with the specific properties of the SBLCT state,
For all complexes under study the quantum yields of emission
causes the complex [Ru(SnPh3)2(CO)2(dmb)] to have an ex-
tremely long emission lifetime of 1.1 ms in a glass at 90 K.
Emission lifetimes this long are virtually unknown for charge-
transfer states of organometallic compounds. Various other types
of long-lived excited states are known, and they all feature a
dimished involvement of the transition metal atom in the excited
state. Thus, for Ru(II) R-diimine complexes in which a low-
lying intraligand (IL) state interacts with the MLCT state, the
excited-state lifetime may increase by 2 orders of magnitude.
An example is [Ru(bpy-pyr)(bpy)2]2+; a [Ru(bpy)3]2+ type
complex in which one bpy ligand has been functionalized with
a pyrene group. In this case the MLCT emission decay is
biexponential, the longer-lived component (τ ) 50 µs at room
temperature) being due to internal conversion from the higher-
3
Φem from the SBLCT state are rather small in view of their
very long emission lifetimes. By use of the equation Φem/τ )
Φisckr, values are obtained for Φisckr that do not exceed 1.8 ×
102 s-1 (Table 3). In fact, these values are much lower than
that found for the isostructural complex [Ru(Cl)(Me)(CO)2(iPr-
DAB)] (Φisckr ) 11 × 102 s-1) having a lowest 3MLCT excited
state. In general, 3MLCT states have radiative decay constants
kr that are even higher and range between 104 and 105 s-1 35
.
3
These low values of Φisckr for the emission from the SBLCT
state are rather unexpected because the electronic transition to
1
the corresponding SBLCT state is strongly allowed (ꢀ ) (3-
15) × 103 M-1 cm-1; Table 1). The values of Φisckr and of the
emission quantum yields are therefore not lower than expected
because kr is small but because Φisc is much smaller than unity.
Lower values of Φisc are obtained if crossing between 1SBLCT
3
3
lying pyrene IL state to the MLCT state.35 When an ethynyl
group is inserted between the bpy and pyrene units, the IL and
MLCT states are in thermal equilibrium, leading to a single-
exponential decay with a lifetime of 46 µs.36 An even longer
excited-state lifetime was found for [Ru(CN2-np)(bpy)2]2+ (CN2-
np ) naphtho[2,3-f][1,ω]phenanthroline-9,14-dicarbonitrile);
τ ) 464 µs at 77 K.37 Excited states that are virtually purely IL
in character can have even longer lifetimes. Examples include
various metalloporphyrins38 and the complexes [M(bpy)3]3+
(M ) Rh, Ir).39,40 Ligand-to-ligand charge transfer (L′LCT)
states, among which the SBLCT states may be reckoned in view
of the limited involvement of the transition metal, can be long-
lived as well.41 In the L′LCT state of [Zn(4-Cl-PhS)2(phen)]
negative charge has been transferred from the thiolate donors
to the phenanthroline acceptor. The lifetime of this 3L′LCT state
3
and SBLCT is slow or if there is a competing intersystem
1
crossing from the SBLCT state to another nonemitting state
3
of MLCT character. According to recent CASSCF/CASPT2
calculations of the ground state and some singlet and triplet-
excited states of the model complex [Ru(SnH3)2(CO)2(Me-
DAB)],44 both factors may play an important role here. Thus,
the calculated energy difference between the 1SBLCT and
3SBLCT states of 5400 cm-1 is rather high compared with that
1
3
between the lowest MLCT and MLCT states (1500 cm-1).
This high-energy difference may cause a slowing down of the
intersystem crossing to such an extent that fluorescence from
the 1SBLCT state can compete with this process. In fact a short-
lived and only slightly Stokes shifted emission was observed,
1
most probably belonging to fluorescence from the SBLCT
state.6,8 This luminescence, which is not due to fluorescence
from solvent impurities or any other artifact, has a lifetime of
800 ps for [Ru(SnPh3)2(CO)2(iPr-DAB)] in a 2-MeTHF glass
at 90 K. This lifetime was determined with a Hamamatsu streak
camera setup,45 using a nitrogen laser (λexc ) 337 nm) as the
excitation source.
(32) Damrauer, N. H.; Boussie, T. R.; Devenney, M.; McCusker, J. K. J.
Am. Chem. Soc. 1997, 119, 8253.
(33) Treadway, J. A.; Loeb, B.; Lopez, R.; Anderson, P. A.; Keene, F. R.;
Meyer, T. J. Inorg. Chem. 1996, 35, 2242.
(34) Strouse, G. F.; Schoonover, J. R.; Duesing, R.; Boyde, S.; Jones, W.
E., Jr.; Meyer, T. J. Inorg. Chem. 1995, 34, 473.
(35) Baba, A. I.; Shaw, J. R.; Simon, J. A.; Thummel, R. P.; Schmehl, R.
H. Coord. Chem. ReV. 1998, 171, 43.
(36) Harriman, A.; Hissler, M.; Khatyr, A.; Ziessel, R. Chem. Commun.
1999, 735.
(37) Albano, G.; Belser, P.; Cola, L. D.; Gandolfi, M. T. J. Chem. Soc.,
Chem. Commun. 1999, 1171.
(38) Eastwood, D.; Gouterman, M. J. Mol. Spectrosc. 1970, 35, 359.
(39) DeArmond, M. K.; Hillis, J. E. J. Chem. Phys. 1971, 54, 2247.
(40) Flynn, C. M., Jr.; Demas, D. M. J. Am. Chem. Soc. 1974, 96, 1959.
(41) Vogler, A.; Kunkely, H. Comments Inorg. Chem. 1990, 9, 201.
(42) Kutal, C. Coord. Chem. ReV. 1990, 99, 213.
(43) Fabian, R. H.; Klassen, D. M.; Sonntag, R. W. Inorg. Chem. 1980,
19, 1977.
(44) Turki, M.; Daniel, C. 34th International Conference on Coordination
Compounds, Edinburgh July 8-14, 2000. Coord. Chem. ReV., to be
published.
(45) Lauteslager, X. Y.; van Stokkum, I. H. M.; van Ramesdonk, H. J.;
Brouwer, A. M.; Verhoeven, J. W. J. Phys. Chem. A 1999, 103, 653.

[M(SnR3)2(CO)2(R-diimine)]
Inorganic Chemistry, Vol. 40, No. 2, 2001 285
A second noteworthy result of the CASSCF/CASPT2 calcu-
Conclusions
3
lations is the presence of a MLCT state very close in energy
1
The results of this study show that the photostability of these
complexes at room temperature can be increased appreciably
by using an R-diimine with a low-lying π* orbital and osmium
instead of ruthenium. In this way virtually photostable com-
plexes were prepared with lifetimes of ca. 4 µs. In a glass at 80
K all complexes are photostable, and because the complexes
are only weakly distorted in their emitting 3SBLCT states
according to the Stokes shift and resonance Raman spectra,
extremely long emission lifetimes of up to 1 ms were obtained.
Ab initio calculations suggest that this may be due to an
inefficient decay to the emitting 3SBLCT state, but this will be
investigated further by ultrafast time-resolved absorption studies.
to the absorbing SBLCT state. Intersystem crossing to this
3MLCT state will compete with decay to the 3SBLCT state, the
more because crossing between 1SBLCT and 3MLCT has been
1
found to be much more efficient than between SBLCT and
3SBLCT states.46 Moreover, occupation of the 3MLCT state in
question is not expected to give rise to any strong additional
emission because the electronic transition to the corresponding
1MLCT state, observed as a weak band at ca. 400 nm in the
case of [Ru(SnPh3)2(CO)2(iPr-DAB)], is overlap-forbidden.
We therefore propose that the low emission quantum yields
of the complexes under study are the result of two effects, i.e.,
1
3
the large energy gap between the SBLCT and SBLCT states
and the presence of a nonemissive 3MLCT state close in energy
Acknowledgment. Many thanks are due to Dr. C. Daniel
(CNRS, Strasbourg) for very fruitful discussions and for
providing us with the theoretical data of [Ru(SnH3)2(CO)2(Me-
DAB)] before publication.
1
to the SBLCT state.
(46) Daniel, C.; Guillaumont, D.; Ribbing, C.; Minaev, B. J. Phys. Chem.
A 1999, 103, 5766.
(47) Nieuwenhuis, H. A.; Stufkens, D. J.; Vlcˇek, A., Jr. Inorg. Chem. 1995,
34, 3879.
IC000857X