Photochemistry of Ru(etp)(CO)H2 (etp = PhP(CH2CH 2PPh2)2): Fast oxidative addition and coordination following exclusive dihydrogen loss

Article abstract of DOI:10.1021/om049729x

The photochemistry of Ru(etp)(CO)H₂ (1, etp = PhP(CH ₂CH₂PPh₂)₂) has been studied by UV/vis spectroscopy following nanosecond laser flash photolysis and by NMR and IR spectroscopy following steady-state irradiation. Steady-state irradiation under CO, C₂H₄, and Et₃SiH yields Ru(etp)(CO)₂ Ru(etp)(CO)(C₂H₄), and Ru(etp)(CO)(SiEt₃)H, respectively. Laser flash photolysis (laser wavelength 308 nm) of 1 in cyclohexane generates the 16-electron transient Ru(etp)(CO). In the absence of additional ligands, Ru(etp)(CO) decays by reaction with photoejected dihydrogen, regenerating 1. When flash photolysis was performed in the presence of added ligands, the transient decays by pseudo-first-order kinetics with second-order rate constants on the order of 10⁸ dm³ mol^-1 s^-1. However, the fastest reaction rate (H₂) is only a factor of ca. 4 greater than the slowest (Et₃SiH). Activation parameters for the reaction of 1 with Et₃SiH were determined as ΔG₂₉₈? = 25.7 ± 0.1 kJ mol^-1, ΔH? = 11 ± 1 kJ mol ^-1, and ΔS? = -49 ± 4 J mol^-1 K ^-1. The evidence from the UV/vis spectrum of the transient and from the structures of the stable photoproducts indicates that Ru(etp)(CO) adopts a nonplanar geometry.

Full text of DOI:10.1021/om049729x

4034
Organometallics 2004, 23, 4034-4039
P h otoch em istr y of Ru (etp )(CO)H2 (etp )
P h P (CH2CH2P P h 2)2): F a st Oxid a tive Ad d ition a n d
Coor d in a tion F ollow in g Exclu sive Dih yd r ogen Loss
Virginia Montiel-Palma,† David I. Pattison,‡ Robin N. Perutz,*,‡ and
Claire Turner‡
Department of Chemistry, University of York, York, YO10 5DD, U.K., and Centro de
Investigaciones Qu´ımicas, Universidad Auto´noma del Estado de Morelos, Avenida
Universidad 1001, Col Chamilpa, Cuernavaca, Morelos, 62210 Me´xico
Received April 13, 2004
The photochemistry of Ru(etp)(CO)H2 (1, etp ) PhP(CH2CH2PPh2)2) has been studied by
UV/vis spectroscopy following nanosecond laser flash photolysis and by NMR and IR
spectroscopy following steady-state irradiation. Steady-state irradiation under CO, C2H4,
and Et3SiH yields Ru(etp)(CO)2, Ru(etp)(CO)(C2H4), and Ru(etp)(CO)(SiEt3)H, respectively.
Laser flash photolysis (laser wavelength 308 nm) of 1 in cyclohexane generates the 16-electron
transient Ru(etp)(CO). In the absence of additional ligands, Ru(etp)(CO) decays by reaction
with photoejected dihydrogen, regenerating 1. When flash photolysis was performed in the
presence of added ligands, the transient decays by pseudo-first-order kinetics with second-
order rate constants on the order of 108 dm3 mol-1 s-1. However, the fastest reaction rate
(H2) is only a factor of ca. 4 greater than the slowest (Et3SiH). Activation parameters for the
q
reaction of 1 with Et3SiH were determined as ∆G298 ) 25.7 ( 0.1 kJ mol-1, ∆Hq ) 11 ( 1
kJ mol-1, and ∆Sq ) -49 ( 4 J mol-1 K-1. The evidence from the UV/vis spectrum of the
transient and from the structures of the stable photoproducts indicates that Ru(etp)(CO)
adopts a nonplanar geometry.
In tr od u ction
Ru(PP3) [PP3 ) P(CH2CH2PPh2)3] reacts rapidly with
benzene, whereas Ru(dppe)2 (dppe ) Ph2CH2CH2PPh2)
shows no sign of such a C-H activation reaction. In
these systems, the switch from a tetradentate to a
bidentate phosphine alters the reactivity substantially,
although the phenyl substituent at phosphorus and the
ethane backbone are retained. The most extreme reac-
tivity toward dihydrogen or CO is observed with Ru-
(dmpe)2 (dmpe ) Me2PCH2CH2PMe2).11 The behavior
of the complex Ru(PMe3)4(H)2 contrasts with the reac-
tivity of the other complexes that we have studied in
this series. This complex is the only one to exhibit two
photochemical pathways, one involving loss of H2, the
other loss of phosphine.12
The introduction of a CO group into the ligand sphere
introduces a very useful structural reporter through the
CO-stretching vibration. It also raises the possibility of
photochemical loss of CO competing with loss of H2. It
has long been known that Ru(PPh3)3(CO)H2 undergoes
selective photodissociation of H2,13 a feature we ex-
ploited to show that H2 photodissociation occurs within
ca. 10-11 s of the initial laser flash.14 Since our studies
of Ru(PPh3)3(CO)H2 were hampered by its lack of
Multidentate phosphines are widely employed as
ligands in catalysis because of their ability to control
the coordination number and molecular geometry of the
complex. The multiple binding sites also may prevent
complete dissociation of the phosphine, which leads to
deactivation of the catalyst.1,2 Transition metal com-
plexes of polyphosphines have attracted attention as
catalysts in homogeneous reactions3 such as the hydro-
genation,4-7 hydroformylation, and isomerization of
alkenes5-7 and the functionalization, hydroamination,
and polymerization of alkynes.8,9
In previous papers, we reported the photochemistry
of a series of iron and ruthenium phosphine dihydride
complexes and showed that the reactivity of the photo-
generated 16-electron intermediate is strongly depend-
ent on the nature of the phosphine.10,11b For instance,
† Universidad Auto´noma del Estado de Morelos.
‡ University of York.
(1) Osborn, J . A.; Young, J . F.; Wilkinson, G. J . Chem. Soc., Chem.
Commun. 1965, 17.
(2) Mayer, H. A.; Kaska, W. C. Chem. Rev. 1994, 94, 1239.
(3) Bianchini, C.; Meli, A.; Peruzzini, M.; Vizza, F.; Frediani, P.;
Ram´ırez, J . A. Organometallics 1990, 9, 226.
(4) Dubois, D. L.; Meek, D. W. Inorg. Chim. Acta 1976, 19, L29.
(5) Bianchini, C.; Mealli, C.; Meli, A.; Peruzzini, M.; Zanobini, F. J .
Am. Chem. Soc. 1988, 110, 8725.
(6) Bianchini, C.; Meli, A.; Peruzzini, M.; Vizza, F.; Fujiwara, Y.;
J intoku, T.; Taniguchi, H. J . Chem. Soc., Chem. Commun. 1988, 210.
(7) Bianchini, C.; Laschi, F.; Meli, A.; Peruzzini, M.; Zanello, P.;
Frediani, P. Organometallics 1988, 7, 2575.
(8) Bianchini, C.; Frediani, P.; Meli, A.; Vizza, F.; Peruzzini, M.;
Zanobini, F. Organometallics 1989, 8, 2080.
(9) Muller, T. E.; Grosche, M.; Herdtweck, E.; Pleier, A.-K.; Walter,
E.; Yan, Y.-K. Organometallics 2000, 19, 170.
(10) Whittlesey, M. K.; Mawby, R. J .; Osman, R.; Perutz, R. N.; Field,
L. D.; Wilkinson, M. P.; George, M. W. J . Am. Chem. Soc. 1993, 115,
8627.
(11) (a) Hall, C.; J ones, W. D.; Mawby, R. J .; Osman, R.; Perutz, R.
N.; Whittlesey, M. K. J . Am. Chem. Soc. 1992, 114, 7425. (b) Cronin,
L.; Nicasio, M. C.; Perutz, R. N.; Peters, R. G.; Roddick, D. M.;
Whittlesey, M. K. J . Am. Chem. Soc. 1995, 117, 10047.
(12) Montiel-Palma, V.; Perutz, R. N.; George, M. W.; J ina, O. S.;
Sabo-Etienne, S. Chem. Commun. 2000, 1175.
(13) Geoffroy, G. L.; Bradley, M. G. Inorg. Chem. 1977, 16, 744.
10.1021/om049729x CCC: $27.50 © 2004 American Chemical Society
Publication on Web 07/16/2004

Photochemistry of Ru(etp)(CO)H2
Organometallics, Vol. 23, No. 17, 2004 4035
Ta ble 1. P r in cip a l NMR Da ta δ (m u ltip licity, J /Hz)
solubility in alkane solvents, we sought an alternative
ruthenium carbonyl phosphine dihydride complex for
further experiments.
in [2H8]-THF
complex,
temp/K
1H
31Pa
The work described in this paper concerns the photo-
chemistry of Ru(etp)(CO)H2 (1) (etp ) [PhP(CH2CH2-
PPh2)2])15 in solution as studied by time-resolved and
steady-state methods. In principle, the relatively con-
strained geometry of 1 should prevent a dimerization
pathway analogous to that proposed for Ru(PPh3)3(CO)-
H2.14 The investigation by time-resolved UV/vis spec-
troscopy demonstrates that 1 undergoes reductive elimi-
nation of dihydrogen within the instrumental response
time and establishes the rate constants for reaction of
the transient Ru(etp)(CO). Studies by NMR and IR
spectroscopy prove that only dihydrogen-loss products
are generated when reactions are performed in the
presence of ligands such as CO, C2H4, and Et3SiH.
2
1, 300 K
-8.06 (m, see text,
105.5 (t, J PP ) 16.2, bridge)
hydride)
2
69.1 (d, J PP ) 16.2, term)
2
2, 300 K
3, 286 K
4, 257 K
109.4 (t, J PP ) 32.5, bridge)
2
76.4 (d, J PP ) 32.5, term)
2
0.71 (broad, C2H4)
1.47 (broad, C2H4)
97.5 (t, J PP ) 25, bridge)
2
74.6 (d, J PP ) 25, term)
2
2
-8.00 (ddd, J PH 61.0,
94.3 (apparent t, J PP 17.2,
21.8, 17.4, hydride)
bridge)
66.0 (d, 17.2 term)
62.1 (d, 17.2 term)
a
Bridge ) bridgehead phosphorus, term ) terminal phospho-
rus.
calculate an angle between the carbonyl groups of 92°.19
The most likely structure of 2 is shown below.
Resu lts
1. Syn th esis a n d Ch a r a cter iza tion of 1 a n d 2. The
only report of complex 1 describes its synthesis by
NaBH4 reduction of Ru(etp)(CO)Cl2.15 We preferred to
synthesize 1 by photolysis of Ru(etp)(CO)2 (2) under
dihydrogen. A saturated THF solution of 2, synthesized
according to literature methods,16 was photolyzed with
broad band UV radiation (λ > 200 nm) under a vigorous
stream of dihydrogen. The reaction was monitored
periodically by IR spectroscopy in the CO-stretching
region until negligible starting material (2) remained.
The structure of 1 has been established previously:15
the etp ligand coordinates facially on the octahedron
with CO trans to the central phosphorus atom, generat-
ing a structure with Cs symmetry (see below). Other
ruthenium complexes of etp also show a facial arrange-
ment for the phosphine ligand.17 Complex 1 is soluble
in THF and to a lesser extent in benzene and toluene;
in THF it exhibits a carbonyl band in the IR spectrum
at 1938 cm-1. Its 31P{1H} NMR spectrum is character-
istic of an AX2 spin system, with a triplet at δ 105.5
and a doublet at δ 69.1 (2J PP 16.2 Hz). The 1H NMR
spectrum shows a hydride resonance as a symmetrical
second-order multiplet centered at δ -8.02, indicative
of the presence of a mirror plane. We employed the
software gNMR to estimate the coupling constants as
cis J PH -16.5 and -22.0 and trans J PH 67.0 and J HH
3.3 Hz.18 Relevant NMR data are listed in Table 1.
Complex 1 is a pale yellow solid; its UV/vis spectrum
shows an absorption that increases into the UV with a
resolved shoulder at 300 nm.
The IR spectrum of 2 shows bands of almost equal
relative intensity at 1948 and 1884 cm-1 in THF. The
31P{1H} NMR spectrum in [2H8]-THF displays a doublet
at δ 76.4 and a triplet at δ 190 with coupling constants
(2J PP) of 32 Hz. The IR relative intensities were used to
2. La ser F la sh P h otolysis. We studied the photo-
chemistry of 1 by laser flash photolysis in cyclohexane
solution. At the laser wavelength of 308 nm, it has an
absorption coefficient of 1.04 × 103 dm3 mol-1 cm-1
.
a . Un d er Ar gon . A transient species was generated
within the instrumental response time on laser flash
photolysis (λexc ) 308 nm) of 1 in cyclohexane at 295 K.
A plot of ∆A-1 versus time was linear, consistent with
second-order kinetics. The slope of the line yielded k2/
ꢀl ) 7.28 × 105 s-1 where ∆A is the change in absor-
bance, k2 is the second-order rate constant, ꢀ is the
molar absorption coefficient, and l is the path length (1
cm). Such behavior is typical of a photodissociation
reaction that generates equal concentrations of tran-
sient and dissolved H2 and is followed by thermal
recombination.11 At 390 nm, the original absorbance was
not restored even after a period of hundreds of micro-
seconds, indicating that the back-reaction was incom-
plete in that period and that permanent photoproducts
may also have been formed. Nevertheless, the UV/vis
spectrum recorded after flash photolysis (ca. 50 laser
shots) was found to be essentially unchanged.
b. Tr a n sien t Sp ectr u m u n d er H2. The UV/vis
transient spectrum was recorded under a partial pres-
sure of dihydrogen (700 Torr) in order to ensure the
reversibility of the system and prevent the buildup of
any stable photoproducts. The spectrum measured
point-by-point 200 ns after the flash from 800 to 360
nm shows increasing absorbance toward the UV with a
maximum at 390 nm (Figure 1). This spectrum re-
sembles those of species such as Ru(PPh3)3(CO)14 and
Fe(dmpe)2,10 which also show broad maxima in the near
UV region.
2
2
2
(14) Colombo, M.; George, M. W.; Moore, J . N.; Pattison, D. I.;
Perutz, R. N.; Virrels, I. G.; Ye, T.-Q. J . Chem. Soc., Dalton Trans.
1997, 2857.
(15) Michos, D.; Luo, X.-L. Crabtree, R. H. Inorg. Chem. 1992, 31,
4245.
(16) J ia, G.; Meek, D. W. Inorg. Chim. Acta 1990, 178, 195.
(17) (a) Delgado, G.; Rivera, A. V.; Sua´rez, T.; Fontal, B. Inorg. Chim.
Acta 1995, 233, 145. (b) Sheldrick, W. S.; Brandt, K. Inorg. Chim. Acta
1994, 217, 51. (c) Albinati, A.; J iang, Q. Z.; Ruegger, H.; Venanzi, L.
M. Inorg. Chem. 1993, 32, 4940.
c. Ad d ition of Qu en ch in g Liga n d s. (i) Dih yd r o-
gen . Addition of a small pressure of dihydrogen caused
an increase in the rate of decay of the transient species,
which decayed obeying pseudo-first-order kinetics. When
using partial pressures of H2 > 600 Torr, no residual
(18) Budzelaar, P. H. M. g-NMR, version 4; Cherwell Scientific
Publishing Limited: Oxford, 1995-1997.
(19) Braterman, P. S. Metal Carbonyl Spectra; Academic Press:
London, 1975.

4036 Organometallics, Vol. 23, No. 17, 2004
Montiel-Palma et al.
F igu r e 1. Point-by-point transient spectrum measured
100 ns after laser flash photolysis of Ru(etp)(CO)H2 (1) in
cyclohexane under 700 Torr of dihydrogen. The line is a
fit to the points with a polynomial of order 6.
F igu r e 3. (a) Kinetic trace obtained upon flash photolysis
of 1 in cyclohexane in the presence of CO (0.96 × 10-3 mol
dm-3). The inset plot shows the fitting to pseudo-first-order
behavior. (b) Plot of kobs versus [CO] (λexc ) 308 nm, λmon
390 nm).
)
Ta ble 2. Secon d -Or d er Ra te Con sta n ts for
Rea ction of Ru (etp )CO w ith Ad d ed Qu en ch er s
quencher
k2 (dm3 mol-1 s-1
)
H2
C2H4
CO
(9.0 ( 0.1) × 108
(8.4 ( 0.5) × 108
(6.8 ( 0.8) × 108
(2.1 ( 0.5) × 108
Et3SiH
was found when monitoring at different wavelengths.
Second-order rate constants are collected in Table 2.
(ii) Ca r bon Mon oxid e. When flash photolysis of a
cyclohexane solution of 1 was carried out in the presence
of dissolved CO and monitored at 390 nm, the kinetic
data fitted pseudo-first-order behavior. The formation
of a long-lived photoproduct was indicated by high
residual absorbance, which increased to shorter wave-
lengths (Figure 3a). A plot of kobs versus [CO] was linear
(Figure 3b) and gave a second-order rate constant of (6.8
( 0.8) × 108 dm3 mol-1 s-1 at 390 nm; no significant
variation was found when monitoring at other wave-
lengths.
(iii) Eth en e a n d Tr ieth ylsila n e. The reactions of
the transient with ethene and triethylsilane were
investigated at several wavelengths and with different
concentrations of the added reagents. Addition of C2H4
or Et3SiH to cyclohexane solutions of 1 resulted in
quenching of the transient, which decayed over a period
of microseconds. Plots of kobs versus [quencher] showed
linear dependence on the quencher concentration (Fig-
ure 4a). No significant variation in the rate constants
was found in either case when monitoring at different
wavelengths.
F igu r e 2. (a) Decay of the transient (monitoring wave-
length, λ ) 390 nm) generated upon laser flash photolysis
(excitation wavelength, λ ) 308 nm) of 1 in cyclohexane at
295 K under a partial pressure of 30 Torr of H2. The inset
plot shows the fitting to pseudo-first-order behavior. (b) Plot
of kobs versus concentration of dissolved H2, yielding, from
the slope, the second-order rate constant for reaction of the
transient with added H2.
change in absorbance remained at the end of the decay
(Figure 2a). The rate of the reaction of the transient
with dihydrogen was measured over a range of partial
pressures of H2 that corresponded to concentrations of
dissolved dihydrogen from 1.3 to 19.4 × 10-4 mol dm-3 20
.
A plot of kobs versus [H2] was linear (Figure 2b) and gave
a second-order rate constant k2 ) (9.0 ( 0.1) × 108 dm3
mol-1 s-1. No significant variation in the rate constant
d . Act iva t ion P a r a m et er s for R ea ct ion w it h
Et3SiH. Pulsed laser flash photolysis of 1 in cyclohexane
in the presence of Et3SiH (2.5 × 10-3 mol dm-3) was
carried out at seven different temperatures in the range
(20) Wilhelm, E.; Battino, R. Chem. Rev. 1973, 73, 1. Gas solubilities
in cyclohexane were taken as follows: H2 (3.8 × 10-3 mol dm-3 atm-1);
CO (9.3 × 10-3 mol dm-3 atm-1); ethene (1.4 × 10-1 mol dm-3 atm-1).

Photochemistry of Ru(etp)(CO)H2
Organometallics, Vol. 23, No. 17, 2004 4037
served at δ 0.71 and 1.47, which sharpened at lower
temperature (220 K, δ 0.59 and 1.43). The IR spectrum
recorded in [2H8]-THF displays a strong carbonyl band
at 1910 cm-1. In marked contrast with Ru(dmpe)2H2
and Ru(depe)2H2, no hydride resonances attributable to
1
vinyl hydride products were observed in the H NMR
spectrum.10,11
Insights into the geometry of 3 can be obtained from
the ab initio calculations carried out by Ogasawara et
al. on the model compound Ru(PH3)4(C2H4), which show-
ed a preference for the ethene ligand to lie in the equa-
torial position due to stronger back-donation of the
metal into its π* orbital.21,22 On the other hand, there
is little energy difference between isomers of Ru(PH3)4-
(CO) having CO in axial or equatorial sites. The etp
ligand is almost certain to adopt a facial geometry like
the isoelectronic complexes [Co(etp)(CO)L]X (L ) CO,
P(OMe)3, PPh2H, PEt3, PPh2Me; X ) BF4, PF6).23 We
therefore conclude that 3 probably adopts the structure
shown below.
c. Rea ction w ith Et3SiH. Photolysis of 1 with λ >
290 nm in the presence of 20 equiv of Et3SiH yielded a
new hydride multiplet centered at δ -8.00. Alterna-
tively, the same product could be generated by photoly-
sis of 3 in the presence of excess Et3SiH in [2H8]-THF.
The integrations of the resonances in the 1H NMR
spectrum are consistent with the formation of a product
of the formula Ru(etp)(CO)(H)(SiEt3) (4). The room-
temperature 29Si{1H} NMR spectrum of 4 shows a
doublet of triplets resonance at δ 18.1 with coupling
F igu r e 4. (a) Plot of kobs versus [Et3SiH] (λexc ) 308 nm,
λmon ) 390 nm). (b) Eyring plot showing the temperature
variation of the second-order rate constant for reaction with
Et3SiH.
Ta ble 3. Tem p er a tu r e Dep en d en ce of
Secon d -Or d er Ra te Con sta n t for Rea ction of
Ru (etp )CO w ith Et3SiH
2
constants J SiP 79.2 and 16.0 Hz. In the 31P{1H} NMR
spectrum at ambient temperature, complex 4 shows a
2
T (K) k2 (108 dm3 mol-1 s-1
)
T (K) k2 (108 dm3 mol-1 s-1
)
triplet resonance at δ 94.3 with J PP 17.2 Hz and two
broad signals at δ 66.0 and 62.1. At 257 K, the hydride
resonance of the 1H NMR spectrum resolves into a
doublet of doublets of doublets (2J PH ) 61.0, 21.8, and
17.4 Hz), and the two highest field resonances (δ 66.0
and 62.1) in the 31P{1H} NMR spectrum resolve into
doublets (2J PP ) 17.2 Hz). A selective heteronuclear
decoupling experiment was also performed at 257 K.
Decoupling of the 31P signal at δ 49.3 removed a doublet
coupling from the hydride multiplet in the 1H NMR
spectrum, leaving a doublet of doublets resonance with
coupling constants 61.0 and 17.4 Hz. The retention of
the large trans coupling indicated that the central
phosphorus atom is not trans to the hydride. Decoupling
of the other two phosphorus signals provided less precise
information due to their proximity. However, the simi-
larity of the chemical shift of the triplet resonance of
the product in the 31P{1H} NMR spectrum when com-
pared to that of 1 prompts us to postulate that the
bridgehead phosphorus atom lies trans to CO. We also
note that the fluxional behavior may be associated with
an η2-H-SiEt3 intermediate that allows exchange of the
two terminal phosphorus ligands. We therefore assign
the structure of 4 as shown below.
280
285
291.5
297
1.61
1.72
2.04
2.24
302
307
313
2.42
2.57
2.90
280-313 K (Table 3). Plots of ln(k2/T) against 1/T and
ln k2 versus 1/T were used to calculate the activation
parameters for the formation of product Ru(etp)(CO)-
(H)(SiEt3) from the intermediate (Figure 4b). The
resulting parameters were determined as ∆G298q ) 25.7
( 0.1 kJ mol-1, ∆Hq ) 11 ( 1 kJ mol-1, and ∆Sq ) -49
( 4 J mol-1 K-1
.
3. Stea d y-Sta te P h otolysis. The solution photo-
chemistry of 1 was probed in order to interpret the
results of the time-resolved experiments and identify
the ultimate photoproducts.
a . Rea ction w ith CO. A sample of 1 dissolved in
[2H8]-THF was rigorously degassed and placed under
an atmosphere of CO. The sample was irradiated (λ >
290 nm) for 2 h, and IR and NMR spectra were acquired.
The spectra allowed unambiguous assignment of the
photoproduct as 2.16
b. Rea ction w ith C2H4. Complex 1 was found to
react with ethene over a period of hours to yield Ru-
(etp)(CO)(C2H4) (3) quantitatively on irradiation (λ >
290 nm) in THF. The reaction was followed conveniently
by 31P{1H} NMR spectroscopy. Depletion of the reso-
nances for 1 was accompanied by growth of a new triplet
at δ 97.5 and a doublet at δ 74.6 (2J PP 25 Hz), indicative
of an AX2 spin system. In the 1H NMR spectrum
measured at 286 K, new broad resonances were ob-

4038 Organometallics, Vol. 23, No. 17, 2004
Montiel-Palma et al.
Sch em e 1. P h otoch em ica l Rea ction s of
Ru (etp )(CO)(H)2 (1)
Sch em e 2. P ossible Str u ctu r es for Ru (etp )(CO)
three structural possibilities (A, B, C) of Cs symmetry
and the agostic structure (D) shown in Scheme 2.
Tr a n sien t Rea ctivity a n d Str u ctu r a l Im p lica -
tion s. In the absence of quenchers, Ru(etp)(CO) reacted
back with dihydrogen, decaying over a period of hun-
dreds of milliseconds (390 nm) and obeying second-order
kinetic behavior. This can be understood by considering
the recombination of the Ru(etp)(CO) and H2, which are
generated in equal concentrations, giving a second-order
rate law.11 This reaction probably competes with other
pathways when the reaction is carried out under an
argon atmosphere. The transient Ru(etp)(CO) reacted
with quenchers H2, CO, Et3SiH, and C2H4 following
pseudo-first-order kinetics with rate constants that
followed the trend H2 > C2H4 > CO > Et3SiH (Table
2). The plots of observed rate constant versus quencher
concentration were linear for a wide range of quencher
concentrations. On the basis of its very high rate
constants for reaction with a variety of quenchers, we
deduce that Ru(etp)(CO) is a “naked” fragment as
opposed to Ru(etp)(CO)‚‚‚S (S ) solvent). Solvation
would be expected to reduce the rate constants signifi-
cantly.25,26
Discu ssion
The results from our investigations on the solution
photochemistry of 1 are summarized in Scheme 1.
F la sh P h otolysis. Tr a n sien t Id en tifica tion a n d
Str u ctu r al Im plication s. The primary transient formed
upon flash photolysis of 1 in cyclohexane solution is
demonstrated from the following arguments to be 16-
electron Ru(etp)(CO). First, reaction with H2 (>600
Torr) results in complete regeneration of the dihydride
precursor as indicated by the absence of residual ab-
sorbance. Second, the stable products formed upon
steady-state photolysis all result from dihydrogen loss.
Third, the transient reacts with other quenchers with
second-order kinetics. The occurrence of a cyclometala-
tion reaction with a phosphine phenyl group can be
ruled out since it would result in first-order kinetics and
the kinetics would be independent of added quenchers.
The very high value of k2 for the reaction of Ru(etp)-
(CO) with CO can also be used to argue against the
formation of a triplet ground-state intermediate. It is
well documented that spin-forbidden reactions of triplet-
state complexes with singlet CO to form singlet photo-
products proceed at a lower rate than for singlet-state
complexes as has been demonstrated for 3Fe(CO)4.27
Additionally, high-spin ground states are highly un-
usual for ruthenium and osmium compounds. Recently
it has been shown through parahydrogen-induced po-
larization of NMR spectra that Ru(dppe)(CO)2 has a
singlet ground state.28
The UV/vis spectrum of the transient generated upon
flash photolysis of 1 in cyclohexane shows a steadily
increasing absorbance toward the UV with a maximum
at 390 nm. The spectrum contrasts with the multiband
UV/vis spectra of complexes Ru(dmpe)2 and related
square-planar complexes with two bidentate ligands.11
However, it resembles the UV/vis spectra of fragments
M(PP3) (M ) Ru, Os), which show broad maxima at 395
and 390 nm, respectively. In these cases, the constraint
imposed by the PP3 phosphine prevents the fragments
from adopting a square planar geometry. Three struc-
tures have been proposed for the M(PP3) intermedi-
ates: C3v, Cs, and one with an agostic interaction of the
phenyl ligand to the metal center. Of these, the agostic
structure offers the most satisfactory explanation of the
observed behavior of such species.24 In the case of Ru-
(etp)(CO), the UV/vis spectral evidence prompts us to
disregard a square planar structure and consider the
The second-order rate constants of Ru(etp)(CO) vary
only by a factor of ca. 4 (k2(H2)/k2(Et3SiH)) and are ca.
100 times faster than those of Ru(PP3);24 we have found
(25) Hall, C.; Perutz, R. N. Chem. Rev. 1996, 96, 3125.
(26) (a) Kelly, J . M.; Bonneau, R. J . Am. Chem. Soc. 1980, 102, 1220.
(b) Kelly, J . M.; Long, C.; Bonneau, R. J . Phys. Chem. 1983, 87, 3344.
(27) (a) Weitz, E. J . Phys. Chem. 1987, 91, 3945. (b) Barton, T. J .;
Grinter, R.; Thomson, A. J .; Poliakoff, M. J . Chem. Soc., Chem.
Commun. 1977, 841. (c) Poliakoff, M.; Weitz, E. Acc. Chem. Res. 1987,
20, 408. (d) Harvey, J . N.; Aschi, M. Faraday Discuss. 2003, 124, 129.
(e) Poliakoff, M.; Turner, J . J . Angew. Chem. Int. Ed. 2001, 40, 2809.
(f) Snee, P. T. Payne, C. K.; Mebane, S. D.; Kotz, K. T.; Harris, C. B.
J . Am. Chem. Soc. 2001, 123, 6909. (g) Ihee, H.; Cao, J .; Zewail, A. H.
Angew. Chem. Int. Ed. 2001, 40, 1532.
(21) Ogasawara, M.; Maseras, F.; Gallego-Planas, N.; Streib, W. E.;
Eisenstein, O.; Caulton, K. G. Inorg. Chem. 1996, 35, 7468.
(22) Ogasawara, M.; Maseras, F.; Gallego-Planas, N.; Toyota, K.;
Streib, W. E.; Komiya, S.; Eisenstein, O.; Caulton, K. G. Organome-
tallics 1997, 16, 1979.
(23) DuBois, D. L.; Meek, D. W. Inorg. Chem. 1976, 15, 3076.
(24) Osman, R.; Pattison, D. I.; Perutz, R. N.; Bianchini, C.; Casares,
J . A.; Peruzzini, M. J . Am. Chem. Soc. 1997, 119, 8459.
(28) Schott, D.; Sleigh, C. J .; Lowe, J . P.; Duckett, S. B.; Mawby, R.
J .; Partridge, M. G. Inorg. Chem. 2002, 41, 2960.

Photochemistry of Ru(etp)(CO)H2
Organometallics, Vol. 23, No. 17, 2004 4039
Ta ble 4. Com p a r ison of log k2/d m 3 m ol-1 s-1 for
Rea ction of Ru th en iu m P h osp h in e Tr a n sien ts
w ith Qu en ch er s, Q
(Wilmad 528-PP) were fitted with Young’s taps. Ru3(CO)12
(Aldrich) and bis(2-diphenylphosphinoethyl)phenylphosphine
(etp, Strem) were used without further purification. NMR
spectra were recorded either on a Bruker MSL300 operating
at 300.13 MHz (1H), 121.49 MHz (31P), or 75.47 MHz (13C) or
on a Bruker DRX400 operating at 400.13 MHz (1H), 161.97
MHz (31P), or 79.49 MHz (29Si).
transient, reference
Ru(dmpe)2, Ru(depe)2, Ru(PP3), Ru(PPh3)3- Ru(etp)(CO),
Q
10
11
24
(CO), 14
this work
H2
CO
Et3SiH
C2H4
9.79
9.66
8.32
8.34
8.60
7.96
5.04
5.30
6.30
6.00
5.73
5.88
7.92
8.49
7.95
8.96
8.83
8.32
8.93
Syn th esis of 2. Complex 2 was synthesized by a modifica-
tion of a literature route.16 Ru3(CO)12 (0.18 g, 0.3 mmol) and
etp (0.49 g, 0.9 mmol) in toluene (50 mL) were refluxed in a
three-necked round-bottomed flask for 4.5 h, forming a yellow-
orange solution. The solvent was reduced until a precipitate
started to appear; hexane (5 mL) was added and the mixture
left to crystallize overnight at 273 K. Filtration of the mixture
yielded yellow crystals of 2 that were washed three times with
hexane (10 mL) and dried in vacuo (yield 0.62 g, 70%).
Syn t h esis of 1. A saturated solution of 2 (100 mg, 0.14
mmol) in THF was placed in a cylindrical Pyrex cell (20 mm
path length, 50 mm diameter) fitted with a greaseless cone
and a tubing connection. The solution was irradiated (λ > 290
nm, 150 W mercury arc) while stirring continuously and
bubbling dihydrogen vigorously via a needle introduced through
a silicone septum. The reaction was monitored by IR spectro-
scopy. After 12 h, little starting material remained and the
solution was carefully decanted from the cell and evaporated
under vacuum. The residue was extracted with benzene and
the volume reduced until some solid started to precipitate.
Hexane (2 mL) was added and the solution left to crystallize
at 273 K. The resulting yellow solid was pure according to IR
and NMR spectroscopy (yield 70 mg, 73%).
no evidence for reaction of Ru(etp)(CO) with benzene.
The rate constants are, however, similar to those of Ru-
(PPh3)3(CO) (see Table 4). The transient Ru(etp)(CO)
is less discriminating in its reactivity than Ru(dmpe)2
or Ru(depe)2. The values of ∆Hq and ∆Sq for the reaction
of Ru(etp)(CO) with Et3SiH are similar to those of Ru-
(dmpe)2 with a significant entropic component to the
barrier. In the case of Ru(PP3) the small variation in
rate constants and the small value of ∆Sq were associ-
ated with an agostic interaction similar to D.24 The
higher value of the rate constants and the more negative
value of ∆Sq in Ru(etp)(CO) indicate that any agostic
stabilization of the transient is kinetically insignificant.
Although we favor structure A on the basis of the
arguments presented above and the geometry of the
final products, we cannot rigorously exclude the other
structures (Scheme 2).
La ser F la sh P h otolysis. The laser flash photolysis equip-
ment for excitation with 308 nm29,30 has been described in
detail elsewhere. In brief, a XeCl excimer laser (energy 35 mJ
pulse-1, 30 ns pulse width, 32.5 kV supply voltage) was used
as excitation source for photolysis with 308 nm, and a pulsed
Xe arc lamp acted as the monitoring source. All measurements
were made at 295 ( 1 K unless otherwise stated. Samples for
temperature measurements above ambient were equilibrated
in a thermostated bath for a few minutes prior to flash
photolysis. To avoid thermal reactions and inaccuracies caused
by sample degradation in the determination of activation
parameters, the temperature was not raised above 313 K and
the samples were changed every two to three readings.
Stea d y-Sta te P h otolysis. Steady-state photolysis was
performed with samples contained in Pyrex ampules or Pyrex
NMR tubes (cutoff ca. 290 nm) fitted with a Young’s stopcock.
The photolysis source was either a Xe arc (ILC, 300 W) or a
medium-pressure Hg arc (Philips HPK, 125 W). In both cases,
a water filter (10 cm path) was placed between the arc and
the sample to remove heat.
Con clu sion s
Reductive elimination of dihydrogen from Ru(etp)-
(CO)H2 occurs within the instrument response time
(<100 ns) upon laser flash irradiation. The resulting
transient species, assigned as 16-electron Ru(etp)(CO),
is rapidly quenched by coordination or oxidative addi-
tion of a variety of reagents. The second-order rate
constants decrease in the order k2(H2) > k2(C2H4) > k2-
(CO) > k2(Et3SiH), but there is little selectivity toward
these reagents. Product studies confirmed that dihy-
drogen loss is the only photochemical pathway.
Exp er im en ta l Meth od s
Gen er a l P r oced u r es. All syntheses and manipulations
were carried out under argon using standard Schlenk, high-
vacuum line, or glovebox techniques. Solvents for general use
were dried by refluxing for several hours over sodium/ben-
zophenone transferred by cannula and stored under argon in
ampules fitted with Young’s taps. [2H8]-THF was dried over
potassium and distilled under high vacuum directly into the
NMR tubes before use. Cyclohexane for laser flash photolysis
experiments was obtained from Sigma-Aldrich (99.9+% HPLC
grade) and dried by refluxing over calcium hydride under
argon, distilled, and collected. Gases used for flash photolysis
were BOC research grade 99.999% purity. All NMR tubes
Ackn owledgm en t. V.M.P.thanks CONACYT(Me´xico)
for a Ph.D. grant.
OM049729X
(29) Haddleton, D. M.; McCamley, A.; Perutz, R. N. J . Am. Chem.
Soc. 1988, 110, 1810.
(30) Hall, C.; J ones, W. D.; Mawby, R. J .; Osman, R.; Perutz, R. N.;
Whittlesey, M. K. J . Am. Chem. Soc. 1992, 114, 7425.