Temperature-dependent coordination of phosphine to five-coordinate alkenylruthenium complexes

Article abstract of DOI:10.1021/ic061389f

The red, five-coordinate complexes Ru(CO)Cl(PPh₃) ₂(CH=CHPh) and [Ru(CO)Cl(PPh₃)₂] ₂(μ-CH=CHC₆H₄CH= CH) undergo reversible coordination of PPh₃ at low temperature to p

Full text of DOI:10.1021/ic061389f

Inorg. Chem. 2007, 46, 561−567
Temperature-Dependent Coordination of Phosphine to Five-Coordinate
Alkenylruthenium Complexes
Sripriya K. Seetharaman,† Min-Chul Chung,‡ Ulrich Englich,§ Karin Ruhlandt-Senge, and
Michael B. Sponsler*
Department of Chemistry, Syracuse UniVersity, Syracuse, New York 13244-4100
Received July 25, 2006
The red, five-coordinate complexes Ru(CO)Cl(PPh3)2(CH
CH) undergo reversible coordination of PPh3 at low temperature to produce the pale yellow, six-coordinate complexes
Ru(CO)Cl(PPh3)3(CH CHPh) and [Ru(CO)Cl(PPh3)3]2( -CH CHC6H4CH CH). X-ray crystal structures of the latter
complex and of the hydride complex RuH(CO)Cl(PPh3)3 were obtained. 1H and 31P NMR spectra between 20 and
70 C exhibit large changes in both equilibrium constants and dynamic effects. Thermodynamic parameters,
) −17.5 2.0 kcal/mol and ) −57.5 7.6 eu, were obtained for PPh3 coordination to the monoruthenium
complex, and activation parameters, 20.6 0.7 kcal/mol and 41.6 2.0 eu, were obtained for the
Hq Sq
reverse decoordination. Coordination of PPh3 was not observed upon cooling of the shorter bridged complex,
[Ru(CO)Cl(PPh3)2]2( -CH CHCH CH).
dCHPh) and [Ru(CO)Cl(PPh3)2]2(µ-CHdCHC6H4CHd
d
µ
d
d
−
°
∆H
±
∆S
±
∆
)
±
∆
)
±
µ
d
d
Introduction
as are the alkenyl complexes formed from it,6 though small
ligands can still coordinate.7 With the smaller PMe3 (cone
angle 118°), six-coordinate alkenyl complexes, Ru(CO)Cl-
(PMe3)3(R1CdCHR2), are common, generally formed by
treatment of Ru(CO)Cl(PPh3)2(R1CdCHR2) with PMe3.8
As shown by many literature examples, addition of the
six-coordinate ruthenium hydride RuH(CO)Cl(PPh3)3 to
alkynes affords five-coordinate E-alkenyl complexes of the
type Ru(CO)Cl(PPh3)2(R1CdCHR2).1-3 The steric bulk of
the alkenyl ligand, despite being trans to the open coordina-
tion site, apparently leaves too little room for coordination
of the third triphenylphosphine ligand (cone angle 145°).4
A variety of smaller ligands, such as isocyanides, pyridines,
and carbon monoxide, do coordinate to the alkenyl com-
plexes to give six-coordinate, octahedral complexes.2,3,5 With
the larger phosphine P-i-Pr3 (cone angle 160°), even the
hydride complex is five-coordinate (RuH(CO)Cl(P-i-Pr3)2),
We have observed that the five-coordinate alkenyl complex
Ru(CO)Cl(PPh3)2(CHdCHPh) (1) can accept an additional
PPh3 ligand at low temperature, as shown by the equilibrium
equation in Scheme 1, to produce the saturated Ru(CO)Cl-
(PPh3)3(CHdCHPh) (1-P).9 This coordination is readily
observable, as a red CH2Cl2 solution of 1 and PPh3 becomes
pale yellow upon cooling to -78 °C. Warming the solution
again to room temperature returns the red color. The
coordination, being enthalpically favorable (negative ∆H)
and entropically unfavorable (negative ∆S), should indeed
* To whom correspondence should be addressed. E-mail: sponsler@
syr.edu.
† Current address: Lawrence University, Appleton, WI.
‡ Current address: Sunchon National University, Sunchon, Republic of
Korea.
(7) Buil, M. L.; Esteruelas, M. A.; Goni, E.; Olivan, M.; Onate, E.
Organometallics 2006, 25, 3076-3083.
(8) Yin, J.; Yu, G.; Guan, J.; Mei, F.; Liu, S. H. J. Organomet. Chem.
2005, 690, 4265-4271.
(9) A recent report (Lam, T. C. H.; Mak, W.-L.; Wong, W.-L.; Kwong,
H.-L.; Sung, H. H. Y.; Lo, S. M. F.; Williams, I. D.; Leung, W.-H.
Organometallics 2004, 23, 1247-1252) claims to use 1-P as a room-
temperature-stable reactant, but the complex used was presumably 1.
The complex is described as orange and was prepared by the method
of ref 1, which reports 1 but not 1-P.
(10) Xia, H.; Wen, T. B.; Hu, Q. Y.; Wang, X.; Chen, X.; Shek, L. Y.;
Williams, I. D.; Wong, K. S.; Wong, G. K. L.; Jia, G. Organometallics
2005, 24, 562-569.
(11) Ahmad, N.; Levison, J. J.; Robinson, S. D.; Uttley, M. F. Inorg. Synth.
1974, 15, 45-64.
§ Current address: Cornell University, Ithaca, NY.
(1) Torres, M. R.; Vegas, A.; Santos, A. J. Organomet. Chem. 1986, 309,
169-177.
(2) Santos, A.; Lopez, J.; Gonzalez, J. J.; Tinoco, P.; Echavarren, A. M.
Organometallics 1997, 16, 3482-3488.
(3) Romero, A.; Santos, A.; Lopez, J.; Echavarren, A. M. J. Organomet.
Chem. 1990, 391, 219-223.
(4) Davila, R. M.; Staples, R. J.; Elduque, A.; Harlass, M. M.; Kyle, L.;
Fackler, J. P. J. Inorg. Chem. 1994, 33, 5940-5945.
(5) Rickard, C. F. E.; Roper, W. R.; Woodgate, S. D.; Wright, J. L. J.
Organomet. Chem. 2000, 607, 27-40.
(6) Esteruelas, M. A.; Werner, H. J. Organomet. Chem. 1986, 303, 221-
231.
10.1021/ic061389f CCC: $37.00
Published on Web 12/22/2006
© 2007 American Chemical Society
Inorganic Chemistry, Vol. 46, No. 2, 2007 561

Seetharaman et al.
Scheme 1
X-ray Crystallography of RuH(CO)Cl(PPh3)3 and 2-P2.
Crystallographic data were collected with a Bruker P4 diffractometer
equipped with a SMART CCD system20 and using Mo KR radiation
(λ ) 0.71073 Å). The data were corrected for Lorentz and
polarization effects. Absorption corrections were made using
SADABS.21 The structure solution and refinement were carried out
using the SHELX9722 crystallographic software package. The
structure was solved using direct methods. After location of all the
non-hydrogen atoms, the model was refined against F2, initially
using isotropic then anisotropic thermal displacement parameters.
Crystals of RuH(CO)Cl(PPh3)3 were grown from pentane/CH2-
Cl2. The crystals were transferred under a stream of N2 into a Petri
dish filled with a highly viscous hydrocarbon oil (Infineum). With
the aid of a microscope, a crystal was attached to the end of a
glass fiber and mounted in the cold N2 stream in the diffractometer.
RuH(CO)Cl(PPh3)3 was found to be a merohedral twin. In
complete analogy to the isostructural structure for OsH(CO)Cl-
(PPh3)3,10 systematic absences for a 31 and a 32 axis were detected.
It was found that the 2-fold axis was not a true crystallographic
one, and after interchanging h and k and reversing l (010 100 00
1), the structure was refined in the space group P31 to complete
satisfaction. An earlier structure report on RuH(CO)Cl(PPh3)3
contains the structural refinement of the compound in the space
group P32 with R1 of 9.40% with the phenyl ring positions
constrained to ideal hexagons. In our case, the R1 value drop-
ped to 3.31%, and estimated standard deviations (esd’s) for the
atomic positions and displacement parameters improved signifi-
cantly, allowing the independent refinement of all molecular
components.
be preferred at lower temperatures. The corresponding
osmium complex, Os(CO)Cl(PPh3)3(CHdCHPh), has re-
cently been reported to be stable at room temperature.10
In this paper, we report NMR studies on the coordination
of PPh3 to 1 and related complexes. Variable temperature
1
31P and H spectra have been recorded and analyzed with
respect to both thermodynamic and kinetic aspects of the
reaction. This is a rare study that provides both types of
parameters from the same datasmade more difficult by the
fact that coalescence phenomena obscured a good portion
of the equilibrium data. The quantitative results are valuable
additions to the limited literature of experimentally deter-
mined phosphine binding parameters. Our study also included
two diruthenium complexes with different bridging ligands
that affect the steric demands and coordination behavior. An
X-ray crystal structure of a PPh3 adduct [Ru(CO)Cl(PPh3)3]2-
(µ-CHdCHC6H4CHdCH) (2-P2) helps to confirm the nature
of the coordination process and the steric environment in
the complexes.
Experimental Section
Crystals of 2-P2 were grown by slow diffusion over 2 weeks at
-78 °C from a CH2Cl2 solution of 2 and 2 equiv of PPh3 over
which was layered 4 volumes of a 20:1 mixture of pentane and
THF. The nearly colorless crystals quickly turned orange upon
warming to room temperature, making it necessary to keep the
crystals below -30 °C at all times. Moreover, removal of the
crystals from the mother liquor resulted in immediate solvent loss
and destruction of the crystals. The use of the hydrocarbon oil was
hampered by its low-temperature viscosity. With a microscope
positioned next to the diffractometer, a few crystals in mother liquor
were quickly transferred into a Petri dish containing a few milliliters
of diethyl ether cooled with dry ice. A crystal was then mounted
at the end of a grease-tipped fiber and quickly moved into the stream
of cold N2 gas (transfer time ca. 2 s) on the diffractometer.
Compound 2-P2 was found to contain a large number of lattice
solvent molecules, explaining the immediate desolvation upon
removal of the mother liquor. Mounting from cooled diethyl ether
(-78 °C) prevented desolvation, while avoiding the decomposition
of the crystals observed upon decoordination of PPh3. Some solvents
of crystallization were found to be disordered. The positions were
refined using split parameters as well as restraints.22 Two solvent
molecules, one CH2Cl2 and one THF, could not be refined
satisfactorily and were removed from the refinement using the
squeeze function in PLATON.23
All sample manipulations were carried out under a dinitrogen
atmosphere in a glove box. Solvents were either degassed (diethyl
ether) or purified by distillation or vacuum transfer from CaH2
(pentane, CH2Cl2, and CD2Cl2). RuH(CO)Cl(PPh3)3,11 Ru(CO)Cl-
(PPh3)2(CHdCHPh)1 (1), and [Ru(CO)Cl(PPh3)2]2(µ-CHdCHC6H4-
CHdCH)12,13 (2) were prepared according to the reported proce-
dures. Commercial triphenylphosphine was used without purification.
1H and 31P NMR spectra were recorded on a Bruker DPX-600
at seven different temperatures: 20, 5, -10, -25, -40, -55, and
-70 °C for two samples containing 1 and a slight excess of PPh3
in CD2Cl2 (sample 1 under air and sample 2 under vacuum). For
samples 1 and 2, the peaks obtained at temperatures ranging from
20 to -40 °C were taken as averaged peaks representing mixtures
of 1, PPh3, and 1-P. After verification that the weighted-average
method would be accurate, the relative amounts of 1, PPh3, and
1-P at each temperature were calculated by weighted-average
interpolation relative to the chemical shifts and coupling constants
obtained for the pure components.14 The equilibrium constants for
the coordination process at each temperature were then computed
and used to create van’t Hoff plots. The chemical shifts and coupling
constants from which the thermodynamic data were calculated are
given in the Supporting Information, along with more detailed
descriptions of the methods used. NMR line-shape fitting was
carried out using the programs SwaN-MR (version 3.6.1)15,16 and
MEXICO (version 3.0).17-19
(18) Bain, A. D.; Rex, D. M.; Smith, R. N. Magn. Reson. Chem. 2001, 39,
122-126.
(19) Bain, A. D.; Duns, G. J. Can. J. Chem. 1996, 74, 819-824.
(20) SMART, Data Collection Software, version 4.050; Siemens Analytical
Instruments, Inc.: Madison, WI, 1966.
(21) Sheldrick, G. M. SADABS: University of Go¨ttingen: Gottingen,
Germany, 1996.
(22) Sheldrick, G. M. SHELXTL-Plus: Program Package for Structure
Solution and Refinement; Siemens Analytical X-Ray Instruments,
Inc.: Madison, WI, 1999.
(12) Santos, A.; Lopez, J.; Montoya, J.; Noheda, P.; Romero, A.; Echa-
varren, A. M. Organometallics 1994, 13, 3605-3615.
(13) Jia, G.; Wu, W. F.; Yeung, R. C. Y.; Xia, H. P. J. Organomet. Chem.
1997, 539, 53-59.
(14) London, R. E. J. Magn. Reson., Ser. A. 1993, 104, 190-196.
(15) Balacco, G. J. Chem. Inf. Comput. Sci. 1994, 34, 1235-1241.
(16) A newer program by the same author is available: Balacco, G. http://
www.inmr.net.
(23) Spek, A. L. In PLATON for Windows; Utrecht University: The
Netherlands, 2000.
(17) Bain, A. D. http://www.chemistry.mcmaster.ca/bain/exchange.html.
562 Inorganic Chemistry, Vol. 46, No. 2, 2007

Coordination to Alkenylruthenium Complexes
Figure 2. Temperature-dependent 1H NMR spectra of 1 and PPh3 in
CD2Cl2.
Figure 1. Temperature-dependent 31P NMR spectra of 1 and a slight excess
of PPh3 in CD2Cl2.
The 31P NMR spectrum at -70 °C showed a mixture of
1-P and excess PPh3 with no observable peak for 1,
indicating that the coordination equilibrium constant was very
high at this temperature. As the temperature was raised, the
peaks broadened, and the coordinated and free PPh3 peaks
coalesced between -40 and -25 °C to an averaged peak.
The gradual upfield shift of the averaged peak between -25
and 20 °C (from -0.7 to -4.1 ppm) is consistent with an
equilibrium shift away from 1-P and toward 1 plus PPh3.
Likewise, the gradual downfield shift of the averaged peak
representing the trans PPh3 ligands of 1 and 1-P (from 25.4
to 31.5 ppm) over the same temperature range is also
consistent with this equilibrium shift. The observation that
the 20 °C shifts (31.5 and -4.1 ppm) do not quite match
those observed for separate samples of 1 and PPh3 (31.6 and
-4.6 ppm) is consistent with an equilibrium that favors 1
and PPh3, though not to the complete exclusion of 1-P.
Crystallographic data (excluding structure factors) for the two
structures have been deposited with the Cambridge Crystallo-
graphic Data Center as supplementary publications 615733 (RuH-
(CO)Cl(PPh3)3) and 615734 (2-P2). Copies of these data can be
obtained free of charge on application to CCDC, 12 Union Road,
Cambridge CB21EZ, U.K. (fax: (+44) 1223-336-033; e-mail:
deposit@ccdc.cam.ac.uk).
Results and Discussion
Coordination of Triphenylphosphine. Ru(CO)Cl(PPh3)2-
(CHdCHPh) (1) was prepared from the reaction of RuH-
(CO)Cl(PPh3)3 with phenylacetylene.1 The addition occurs
with loss of PPh3 to give the five-coordinate complex. When
the crude mixture, already containing 1 equiv of free
phosphine, was cooled, a gradual and reversible color change
was observed from red at ambient temperature to pale yellow
1
at -78 °C. 31P and H NMR spectra recorded for clean
1
mixtures of 1 and PPh3 between 20 and -70 °C (Figures 1
and 2) were consistent with the low-temperature coordination
of PPh3 to give Ru(CO)Cl(PPh3)3(CHdCHPh) (1-P). In
particular, the -70 °C 31P spectrum showed a doublet at 23.9
ppm and a triplet at 3.4 ppm (JPP ) 15.3 Hz) with an
integration ratio of 2:1. Further evidence was obtained in
the form of an X-ray crystal structure of a complex closely
related to 1-P (see below).
The series of H NMR spectra recorded at the same
temperatures (Figure 2) is also interpretable in terms of the
same coordination changes. The 20 °C doublets at 8.36 and
5.60 ppm represent fast-exchange, averaged signals for the
R- and â-alkenyl hydrogens, respectively, coming close to
the 20 °C values for pure 1 (8.41 and 5.58 ppm). As the
temperature was lowered, these signals shifted toward each
other, with the R-signal becoming lost in the PPh3 signals
Inorganic Chemistry, Vol. 46, No. 2, 2007 563

Seetharaman et al.
Scheme 2
Table 1. Crystal Data for RuH(CO)Cl(PPh3)3 and 2-P2
RuH(CO)Cl(PPh3)3‚
2-P2‚3CH2Cl2‚
CH2Cl2
3THF
formula
fw
C56H48Cl3OP3Ru
1036.27
P31
12.5676(6)
12.5676(6)
26.2720(17)
90
90
120
3593.6(3)
3
C67.5H64Cl4O2.5P3Ru
1250.46
C2/c
26.803(2)
21.2043(15)
24.917(2)
90
94.079(2)
90
14125(2)
8
space group
a (Å)
b (Å)
c (Å)
R (˚)
â (˚)
γ (˚)
V (Å3)
Z
below -25 °C. The â-signal, observable at all temperatures,
broadened until -40 °C and then sharpened again into a
presumably slow-exchange doublet representing 1-P. No
doublet was observable at -70 or -55 °C for 1, to be
expected at 5.30 or 5.35 ppm, respectively, on the basis of
spectra of pure samples of 1.
The styryl phenyl signals in Figure 2 show smaller
temperature-dependent shifts without broadening. The phos-
phine phenyl signals, in contrast, broadened as the temper-
ature dropped to -25 °C and then split into many signals
over a large chemical shift range at -70 °C. The large
number of signals indicates that 1-P must experience one or
more types of restricted rotations, placing phenyl hydro-
gens in different and slowly exchanging environments. A
very broad peak integrating for 1H was observed at 8.5 ppm
in the -55 and -70 °C spectra, apparently representing a
PPh3 hydrogen that resides in a particularly deshielded
environment.
In a completely analogous fashion, coordination of PPh3
to the dinuclear complex 2 was also observed, ultimately
producing 2-P2 (Scheme 2). Temperature-dependent solution
colorations were essentially identical to those observed for
Scheme 1. Spectral interpretations were somewhat compli-
cated by the unsymmetrical nature of the intermediate 2-P,
but 31P and 1H spectral features were completely consistent
with a stepwise coordination process.
Very interestingly, the dinuclear complex 3,24 with a
shorter, C4H4 bridging ligand, showed no evidence of PPh3
coordination, even at low temperatures. The color of a
solution of 3 and PPh3 remained red when cooled to -78
°C, and no qualitative changes were observed in the 31P or
1H NMR spectra between 20 and -70 °C. The shorter bridge
apparently brings the coordinated PPh3 ligands on the two
Ru centers into close enough proximity that they are unable
to bend away to accommodate an incoming PPh3 ligand.
T (K)
95(2)
1.437
0.0331
0.0781
91(2)
1.176
0.0853
0.2032
D
calc (g cm-3
)
R(F)a
Rw(F)C
a R(F) ) Σ||Fo| - |Fc||/Σ|Fo|; Rw(F)C ) [Σ w(Fo2 - Fc )2/Σ w(Fo )2]1/2
.
2
2
Table 2. Selected Bond Lengths (Å) and Angles (deg) for
RuH(CO)Cl(PPh3)3 and 2-P2
RuH(CO)Cl(PPh3)3
2-P2
Ru-Cl
2.499(1)
2.396(1)
2.358(1)
2.505(1)
1.837(6)
1.141(6)
Ru-Cl
Ru-P2
Ru-P3
Ru-P1
Ru-C6
C-O
2.452(2)
2.418(2)
2.413(3)
2.552(2)
1.822(10)
1.172(10)
2.073(8)
Ru-P1
Ru-P2
Ru-P3
Ru-C55
C-O
Ru-C1
P1-Ru-P2
P1-Ru-P3
P2-Ru-P3
Cl-Ru-P3
C55-Ru-P3
Ru-C55-O
154.02(4)
105.78(4)
99.27(4)
99.11(4)
88.4(2)
P2-Ru-P3
P2-Ru-P1
P3-Ru-P1
Cl-Ru-P1
C6-Ru-P1
Ru-C6-O
C1-Ru-P2
C1-Ru-P3
C1-Ru-P1
C1-Ru-Cl
C1-Ru-C6
159.99(8)
98.61(8)
100.67(8)
98.80(8)
85.0(3)
178.5(7)
81.4(2)
80.2(2)
178.9(5)
172.4(2)
88.8(2)
87.4(3)
alkenyl structure Ru(CO)Cl(PPh3)3(PhCdCH2).1 The former
was prepared in CH2Cl2 and the latter in CH2Cl2/CH3OH.
In our hands, only 1 was obtained in either solvent.
X-ray Crystal Structures of RuH(CO)Cl(PPh3)3 and
2-P2. Although RuH(CO)Cl(PPh3)3 has been known since
1964 and has appeared in over 450 papers, its crystal
structure has only been reported recently.25 As mentioned
above, the former data analysis did not recognize the
merohedral twinning, resulting in nonsatisfactory data analy-
sis and geometrical values. We here report an improved
structure analysis (Tables 1 and 2).
The structure of RuH(CO)Cl(PPh3)3 (Figure 3) is very
similar to that of the isostructural Os analogue; indeed, the
twinning observed for this analogue was also observed for
the Os congener. The isostructural relationship was found
despite differences in solvent incorporation: one molecule
of CH2Cl2 in the Ru crystal vs one molecule of ethanol in
the Os crystal. The unit cell parameters are slightly contracted
for the Ru case, by 0.44% (a and b) and 0.20% (c), but the
temperature for data collection was also lower (95 vs 295
K). The bond lengths to the metal atom (not including the
An early report on the chemistry of RuH(CO)Cl(PPh3)3
claimed that the reaction with phenylacetylene produced
either 1 or a pale yellow complex assigned the six-coordinate
(24) Xia, H.; Yeung, R. C. Y.; Jia, G. Organometallics 1998, 17, 4762-
(25) Snelgrove, J. L.; Conrad, J. C.; Yap, G. P. A.; Fogg, D. E. Inorg.
Chim. Acta 2003, 345, 268-278.
4768.
564 Inorganic Chemistry, Vol. 46, No. 2, 2007

Coordination to Alkenylruthenium Complexes
Figure 4. Molecular structure of 2-P2 with 30% thermal ellipsoids for
non-carbon atoms. Hydrogen atoms have been omitted for clarity.
distances of 2.413 and 2.418 Å are shorter than the Ru-P
(equatorial) distance of 2.552 Å.
The X-ray structure of the mononuclear osmium analogue,
Os(CO)Cl(PPh3)3(CHdCH-p-tolyl), has been reported.10
Compared to this complex, the M-L distances in 2-P2 are
generally slightly shorter (by 1.5% or less), but the Ru-P
(equatorial) distance is longer by 2.1% (0.05 Å). The Ru
complex has the trans phosphine groups angled slightly more
toward the alkenyl ligand (81.4° and 80.2°) relative to the
Os complex (84.2° and 83.3°).
The tilting of the trans PPh3 ligands toward the alkenyl
ligand in 2-P2 is the opposite of what is seen in five-
coordinate alkenyl complexes. For example, in the five-
coordinate complex Ru(CO)Cl(PhCdCHPh)(PPh3)2, the
P-Ru-CR (alkenyl) angles are greater than 90° (99.4° and
97.6°), averaging 17.7° greater than those in 2-P2 (81.4° and
80.2°). The adjustment of the trans PPh3 ligands upon
coordination of a third PPh3 ligand is presumably responsible
for the radically different coordination behavior of 2 and 3.
The longer bridge in 2 allows sufficient room for this
adjustment, whereas the shorter bridge of 3 does not. A DFT
geometry optimized structure of 326 supports this view,
showing the PPh3 groups on the two Ru centers to be in
close proximity.
Thermodynamic and Kinetic Analysis of the Variable
Temperature NMR Spectra. The NMR spectra of Figures
1 and 2 contain both thermodynamic information, showing
a striking, temperature-dependent change in the equilibrium
constant, and kinetic information, showing line-broadening
and coalescence effects. Equilibrium constants might be
extracted from the spectra in three ways: (1) by integration
of peaks involved in slow exchange with respect to the NMR
time scale (below the coalescence temperature); (2) by using
the chemical shift of a fast-exchanging, averaged peak (above
coalescence) to calculate the relative amounts of the con-
tributing species; and (3) by simulation and line-shape fitting
of the peaks, a method that can be applied for spectra at any
temperature. Line-shape analysis also gives rate constants
for the exchange processes.
Figure 3. Molecular structure of RuH(CO)Cl(PPh3)3 with 30% thermal
ellipsoids for non-carbon atoms. Hydrogen atoms have been omitted for
clarity.
hydride, which was not found in either structure) vary by
1.5% or less, but the C-O distance is 6.3% longer in the
Ru structure (1.141 vs 1.073 Å). Bond angles at the metal
center differ by less than 2.6°, with a 4.7° difference for the
M-C-O angle (178.9° vs 174.2°).
Differences with the previously reported structure for RuH-
25
(CO)Cl(PPh3)3 can be attributed to incorporated solvent
(none noted in the previous structure vs one molecule of
CH2Cl2 in the current structure) and to the refinement
differences noted in the experimental section. Unit cell
parameters are slightly smaller in our structure (by 0.76%
for a and b and 0.11% for c), but this is consistent with the
lower temperature used (95 vs 203 K). Bond length esd’s
are 4 to 8 times smaller in our structure.
The two trans PPh3 ligands lean away from the third PPh3
ligand, with cis P-Ru-P angles of 105.8° and 99.3° and a
trans P-Ru-P angle of 154.0°. The unique Ru-P bond
distance of 2.505 Å is significantly longer than the trans
Ru-P distances of 2.358 and 2.396 Å.
Crystals of 1-P and 2-P2 were grown from a ternary
solvent system (pentane, CH2Cl2, and THF) over 2-3 weeks
at -78 °C. The initially white crystals turned orange upon
warming, well before reaching ambient temperature. The
process was not reversible, as the color remained upon
recooling. Consequently, crystals of both 1-P and 2-P2 were
handled and mounted for collection of diffraction data at low
temperature, but multiple problems were encountered. A
successful collection was achieved for 2-P2 (Tables 1 and
2), despite strong sensitivity of the crystals not only to
temperature but also to solvent loss as well as mechanical
pressure.
The integration method was not usable for these spectra,
since no peaks were observable for 1 below coalescence. In
extracting equilibrium constants from chemical shifts of
averaged peaks, the assumption is commonly made that the
The structure of 2-P2, shown in Figure 4, is similar to that
of RuH(CO)Cl(PPh3)3 in the arrangement of phosphine
ligands. The trans PPh3 ligands again lean away from the
remaining PPh3, with cis P-Ru-P angles of 100.7° and
98.6° and a trans P-Ru-P angle of 160.0°. The trans Ru-P
(26) Freedman, T. B.; Sponsler, M. B. Unpublished work.
Inorganic Chemistry, Vol. 46, No. 2, 2007 565

Seetharaman et al.
observed chemical shift is a direct weighted average of the
chemical shifts of the exchanging species. However, London
has shown that very large deviations from the weighted
average are possible for ligand binding equilibria and that
accurate calculations sometimes require inclusion of kinetic
data.14 Making use of rate constants from line-shape fitting
(see below), we have found that the weighted-average
approximation should be accurate for calculation of all
Figure 5. Enthalpy diagram for the coordination of PPh3 to 1.
1
averaged signals in the 31P and H spectra of Figures 1
Line-shape analysis was done for both the 31P spectra and
1
and 2.
the five styryl hydrogens in the H spectra. Details of the
The weighted-average approximation was used to calculate
equilibrium constants from the two averaged 31P signals and
from four of the five averaged styryl 1H signals between 20
and -25 or -40 °C. The HR signal could not be used, since
it was obscured in 1-P. These calculations took into account
the slight excess of PPh3 present, and the details are given
as Supporting Information. The equilibrium constants ob-
tained from each signal were used to construct a van’t Hoff
plot, providing six independent determinations of ∆H and
∆S.
analysis are presented in Supporting Information.
Since line-shape analysis involves optimization of both
thermodynamic and kinetic parameters, both van’t Hoff and
Eyring plots can potentially be produced. However, with both
thermodynamic and kinetic parameters to be optimized and
with the need for assumptions, such as temperature-
independence of some of the chemical shifts (see Supporting
Information for more detail), we were unable to uniquely
and accurately obtain the desired parameters. Therefore, we
used ∆H and ∆S values fixed at the average values presented
above and used the simulation as a means only to obtain
The alkenyl coupling constant, JH-H, for 1, at 13.3 Hz,
differs very significantly from the value for 1-P at 16.9 Hz.27
Like the chemical shifts, the coupling constants are also
averaged in the fast-exchange signals, and the large difference
in this case provides another means to obtain equilibrium
constants. Thus, equilibrium constants were calculated from
the observed coupling constants for both alkenyl 1H signals
between 20 and -25 or -40 °C. From these values, two
additional determinations of ∆H and ∆S were achieved
through van’t Hoff plots.
1
rate constants. With the duplicate 31P and H data sets, we
obtained four independent determinations of ∆Hq and ∆Sq
for both coordination and dissociation through Eyring plots.
The average values and standard deviations were ∆Hq ) 3.1
( 0.7 kcal/mol and ∆Sq ) -16.0 ( 2.0 for coordination
and ∆Hq ) 20.6 ( 0.7 kcal/mol and ∆Sq ) 41.6 ( 2.0 eu
for dissociation. The ∆Hq and ∆H values are combined in a
potential enthalpy diagram for the reaction in Figure 5.
The activation parameters obtained are fully consistent
with expectations for a coordination reaction: a low enthalpic
but significant entropic barrier to coordination. In the
decoordination direction, the enthalpic barrier is comparable
to the ligand binding energy, with a large gain of entropy
noted even at the presumably late transition state.
1
One set of 31P and H NMR spectra were recorded for
each of two samples, and our analysis of chemical shifts and
coupling constants therefore provided 16 independent de-
terminations of ∆H and ∆S. The averages and standard
deviations of these values were ∆H ) -17.5 ( 2.0 kcal/
mol and ∆S ) -56.5 ( 7.6 eu. As expected for a
coordination reaction, both values are negative. The equi-
librium constants implied by the ∆H and ∆S values are 2.9
at 20 °C and 1.7 × 106 at -70 °C. Given concentrations of
1 and PPh3 of 14 and 17 mM, respectively, in the sample
corresponding to Figures 1 and 2, the percent of 1 in the
bound form (1-P) should be 4.5% at 20 °C and 99.98% at
-70 °C.
As expected, the PPh3 binding energy of 17.5 kcal/mol is
low relative to experimental values reported for stable PPh3
complexes. Ligand dissociation energies obtained from
exchange reactions, requiring knowledge of another ligand
dissociation energy, have been reported for PPh3 complexes
of Cr, Mo, and Ni, ranging from 33 to 36 kcal/mol.28 A lower
PPh3 binding energy of 11 kcal/mol was obtained for an Ag
complex by using NMR line-shape analysis.29
Conclusions
Complexes of the type Ru(CO)Cl(PPh3)2(alkenyl) have
been widely reported as stable, five-coordinate complexes.
Coordination of a third PPh3 ligand, though unfavorable at
ambient temperature, was made to proceed for two such
complexes (1 and 2) with very high equilibrium constants
by lowering the temperature by 75-90 °C. From dynamic
NMR data that simultaneously showed a strong equilibrium
shift and coalescence, we were able to obtain both thermo-
dynamic and kinetic parameters for the ligand binding to 1,
giving a complete energy profile for the coordination/
decoordination process. Whereas coordination of PPh3 to
the dinuclear complex 2, with a CHdCHC6H4CHdCH
bridge, was observed at low temperature, the shorter complex
3, with a CHdCHCHdCH bridge, showed no signs of such
coordination.
(27) The value for 1 was the same at 20 and -70 °C, while the value for
1-P could only be measured at -70 °C. The latter was assumed to be
temperature-independent.
(28) Dias, P. B.; Minas de Piedade, M. E.; Simoes, J. A. M. Coord. Chem.
ReV. 1994, 135/136, 737-807.
(29) Carmona, D.; Ferrer, J.; Lamata, M. P.; Oro, L. A.; Limbach, H. H.;
Scherer, G.; Elguero, J.; Jimena, M. L. J. Organomet. Chem. 1994,
470, 271-274.
Acknowledgment. We thank the donors of The Petro-
leum Research Fund, administered by the American Chemi-
cal Society, for support of this research. This work was also
supported by the National Science Foundation under Grant
No. CHE-0320583. We also gratefully acknowledge funds
566 Inorganic Chemistry, Vol. 46, No. 2, 2007

Coordination to Alkenylruthenium Complexes
Supporting Information Available: NMR data and analysis
methods, thermodynamic and kinetic results for the coordination
of PPh3 to 1, and X-ray crystallographic data for RuH(CO)Cl(PPh3)3
and 2-P2. This material is available free of charge via the Internet
at http://pubs.acs.org.
from the NSF (CHE-9527898), the W. H. Keck Foundation,
and Syracuse University, which made possible the purchase
of the X-ray diffractometer. We thank Jacob Alexander and
Weijie Teng for data-handling assistance with the X-ray
results, David J. Kiemle for collection of the NMR data, and
Professor Alex D. Bain, McMaster University, for helpful
correspondence.
IC061389F
Inorganic Chemistry, Vol. 46, No. 2, 2007 567