Kinetics of substitution of weakly coordinating nitrate by chloride in (η5-Cp)Ru(CO)(ER3)ONO2 (ER3 = AsPh3, PPh3, P(p-anisyl)3, PPh2(o-anisyl), P(OPh)3) in dichloromethane

Article abstract of DOI:10.1021/om000670d

The d⁶ complexes of formula (η⁵-Cp)Ru(CO)(ER₃)ONO₂, where ER₃ = AsPh₃, PPh₃, P(p-anisyl)₃, PPh₂(o-anisyl), and P(OPh)₃, were prepared by reaction of their chloro analogues with AgNO₃ in CH₂Cl₂. They underwent moderately slow substitution of the relatively weakly coordinating nitrate by chloride in dry CH₂Cl₂ in the presence of excess [N(PPh₃)₂⁺]Cl^-. A kinetics study showed the reaction to be first order in nitrato complex and independent of chloride salt concentration under pseudo-first-order conditions. First-order rate constants for nitrate metathesis follow the trend P(OPh)₃ < PPh₃ < P(p-anisyl)₃ < AsPh₃ ? PPh₂(o-anisyl), with k₁ at 25 °C for the first four in the series from 1.4 to 4.4 × 10^-5 s^-1. Activation parameters for conversion of (η⁵-Cp)Ru(CO)(AsPh₃)ONO₂ to its chloride are ΔH? = 17(2) kcal/mol and ΔS? = -21(5) eu. Even use of a slightly moist solvent increased the rate of nitrate metathesis by 20-30% without changing the form of the rate law. The complex containing PPh₂(o-anisyl) was approximately 2 orders of magnitude more reactive (k₁ = 4.9 × 10^-3 s^-1). A likely explanation is stabilization of a coordinatively unsaturated intermediate by weak coordination of the potentially chelating o-OMe group. A mechanism entailing rate-limiting conversion of the neutral nitrato complex into a coordinatively unsaturated ion-paired species is consistent with this set of data. Semiempirical calculations (PM3(tm)), which model the structures of complexes in these systems quite well, supported such behavior. X-ray crystal structures were determined for the AsPh₃ nitrato and chloro complexes and for the PPh₃ chloro complex.

Full text of DOI:10.1021/om000670d

2270
Organometallics 2001, 20, 2270-2279
Kin etics of Su bstitu tion of Wea k ly Coor d in a tin g Nitr a te
by Ch lor id e in (η5-Cp )Ru (CO)(ER3)ONO2 (ER3 ) AsP h 3,
P P h 3, P (p-a n isyl)3, P P h 2(o-a n isyl), P (OP h )3) in
Dich lor om eth a n e
Minhton Cao, Liem V. Do, Norris W. Hoffman,* Man-Lung Kwan, J ulie K. Little,
J . Michael McGilvray, Christopher B. Morris, Bjorn C. So¨derberg, and
Andrzej Wierzbicki
Department of Chemistry, University of South Alabama, Mobile, Alabama 36688
Thomas R. Cundari
CROMIUM and the Department of Chemistry, University of Memphis,
Memphis, Tennessee 38152
Charles H. Lake
Departments of Chemistry, University of Alabama at Birmingham,
Birmingham, Alabama 35294, and Indiana University of Pennsylvania,
Indiana, Pennsylvania 15705
Edward. J . Valente
Department of Chemistry, Mississippi College, Box 4065, Clinton, Mississippi 39058
Received August 4, 2000
The d6 complexes of formula (η5-Cp)Ru(CO)(ER3)ONO2, where ER3 ) AsPh3, PPh3, P(p-
anisyl)3, PPh2(o-anisyl), and P(OPh)3, were prepared by reaction of their chloro analogues
with AgNO3 in CH2Cl2. They underwent moderately slow substitution of the relatively weakly
+
coordinating nitrate by chloride in dry CH2Cl2 in the presence of excess [N(PPh3)2 ]Cl-. A
kinetics study showed the reaction to be first order in nitrato complex and independent of
chloride salt concentration under pseudo-first-order conditions. First-order rate constants
for nitrate metathesis follow the trend P(OPh)3 < PPh3 < P(p-anisyl)3 < AsPh3 , PPh2(o-
anisyl), with k1 at 25 °C for the first four in the series from 1.4 to 4.4 × 10-5 s-1. Activation
parameters for conversion of (η5-Cp)Ru(CO)(AsPh3)ONO2 to its chloride are ∆H‡ ) 17(2)
kcal/mol and ∆S‡ ) -21(5) eu. Even use of a slightly moist solvent increased the rate of
nitrate metathesis by 20-30% without changing the form of the rate law. The complex
containing PPh2(o-anisyl) was approximately 2 orders of magnitude more reactive (k1 ) 4.9
× 10-3 s-1). A likely explanation is stabilization of a coordinatively unsaturated intermediate
by weak coordination of the potentially chelating o-OMe group. A mechanism entailing rate-
limiting conversion of the neutral nitrato complex into a coordinatively unsaturated ion-
paired species is consistent with this set of data. Semiempirical calculations (PM3(tm)), which
model the structures of complexes in these systems quite well, supported such behavior.
X-ray crystal structures were determined for the AsPh3 nitrato and chloro complexes and
for the PPh3 chloro complex.
In tr od u ction
electron density at a metal center (in addition to use of
phosphorus, carbenoid, and similar neutral ligands) via
uninegative ligands ranging in donor power from iso-
cyanate2 down to recently designed extremely weakly
coordinating anions3 such as teflate, CB11H12-, and
fluorinated derivatives of BPh4- has considerable merit.
Relatively little research has been conducted on reactiv-
ity patterns of complexes containing moderately weakly
Transition-metal complexes of weakly coordinating
anions have received extensive scrutiny for their ability
to promote carbon-carbon bond formation with special
reactivity and selectivity in the reactions of olefins.1
Such complexes often provide excellent Lewis acid sites
for catalysis of olefin polymerization and oligomeriza-
tion. Given the general desirability of controlling the
electron-richness of low-valent metal systems, tuning
(2) Araghizadeh, F.; Branan, D. M.; Hoffman, N. W.; J ones, J . H.;
McElroy, E. A.; Miller, N. C.; Ramage, D. L.; Salazar, A. B.; Young, S.
H. Inorg. Chem. 1988, 27, 3752.
(1) Tempel, D. J .; J ohnson, L. K.; Huff, R. L.; White, P. S.; Brookhart,
M. J . Am. Chem. Soc. 2000, 122, 6686.
(3) (a) Lupinetti, A. J .; Strauss, S. H. Chemtracts-Inorg. Chem. 1972,
11, 565. (b) Strauss, S. H. Chem. Rev. 1993, 93, 927.
10.1021/om000670d CCC: $20.00 © 2001 American Chemical Society
Publication on Web 05/03/2001

Kinetic Studies of (η5-Cp)Ru(CO)(ER3)ONO2 in CH2Cl2
Organometallics, Vol. 20, No. 11, 2001 2271
coordinating anions such as ONO2 and O2PF2-.4 Ru-
-
tane, and then dried overnight in vacuo. Yield 233 mg (86.9%).
Anal. Calcd for C24H20AsClORu: C 53.80; H 3.76. Found: C
53.61; H 3.78. AsPh3-ligand NMR: 1H δ 7.48 (m); 13C δ 134.4,
132.9, 130.1, 128.8. Single crystals for X-ray analysis were
grown by vapor diffusion of pentane into a toluene solution
(ca. 5 mg in 12 mL).
The chloro compounds 2-5 were prepared in analogous
fashion on a similar scale, with yields in the 79-91% range.
(η5-Cp )Ru (CO)(P P h 3)Cl (2). Yellow-orange microcrystals.
Its spectroscopic data matched those previously reported.5
Single crystals for X-ray analysis were grown by vapor
diffusion of pentane into a toluene solution (ca. 5 mg in 12
mL).
thenium(II) is a d6 system generally slow to undergo
ligand substitution reactions unless weakly coordinating
ligands are present. We report here a study of the
kinetics of nitrate-anion replacement in the pseudo-
octahedral system (η5-Cp)Ru(CO)(ER3)ONO2 by chloride
+
from [N(PPh3)2 ]Cl- (abbreviated hereafter as (PPN+)Cl-)
in the weakly polar solvent CH2Cl2.
(η5-Cp )Ru (CO)(P (p-a n isyl)3)Cl (3). Lemon-yellow micro-
crystals. Anal. Calcd for C27H26ClO4PRu: C 55.72; H 4.50.
Found: C 55.58; H 4.49. P(p-anisyl)3-ligand NMR: 1H δ 7.48
(d, 8.8 Hz, 6H), 6.89 (d, 8.8 Hz, 6H), 3.82 (s, 9H); 13C δ 161.0,
135.0, 126.7, 113.8.
(η5-Cp )Ru (CO)(P P h 2(o-a n isyl))Cl (4). Lemon-yellow mi-
crocrystals. Anal. Calcd for C25H22ClO2PRu: C 57.53; H 4.25;
Found C 57.38; H 4.24. PPh2(o-anisyl)3-ligand NMR: 1H δ 7.8-
7.3 (m, 9H), 6.90 (m, 3H), 3.60 (s, 3H); 13C δ 160.3, 135.3, 134.7,
134.0, 133.6, 132.8, 132.2, 130.2, 129.8, 128.4, 128.2, 127.9,
127.7, 120.5, 110.9, 54.8.
Exp er im en ta l Section
(η5-Cp )Ru (CO)(P (OP h )3)Cl (5). Pale-yellow microcrystals.
Its spectroscopic data matched those previously reported.6b
(η5-Cp )Ru (CO)(AsP h 3)ONO2 (6). Complex 1 (102 mg,
0.198 mmol) was stirred with AgNO3 (43.7 mg, 0.248 mmol)
in 20 mL of CH2Cl2 for 18 h protected from light. Although
the reaction mixture was prepared aerobically, the flask in
which it was stirred was blanketed by nitrogen to avoid
exposure to excess moisture. Then the mixture was filtered
through Celite and concentrated to about 3 mL under reduced
pressure. Into the resulting yellow-orange solution were
dripped 3 mL of diethyl ether and then 40 mL of heptane. The
microcrystalline yellow-orange precipitate afforded was fil-
tered, washed with four 5-mL portions of heptane, and then
dried overnight in vacuo. Yield 88 mg (82%). Anal. Calcd for
Sta r tin g Ma ter ia ls. Tertiary phosphines, P(OPh)3, and
AsPh3 were purchased from Strem Chemicals and used as
received. (η5-Cp)Ru(CO)2Cl was prepared according to the
procedure of Humphries and Knox from Ru3(CO)12 (Strem).5
Reagent toluene, heptane, and diethyl ether (Mallinckrodt)
were used as received. Reagent dichloromethane (Mallinck-
rodt) for synthetic purposes was stored over 4 Å molecular
sieves in air. That for kinetics was dried 4 days and then stored
under N2 over 4 Å molecular sieves which had been dried by
heating 2 days at 130 °C/20 mTorr). (PPN+)Cl- (Strem) was
recrystallized from dichloromethane/diethyl ether, heated at
90 °C/20 mTorr, and stored under N2.
Gen er a l Sp ectr om etr ic P r oced u r es. FTIR spectra were
recorded in 2.5-mm and 5-mm path length CaF2 cells in
dichloromethane solution. NMR spectra of CDCl3 solutions
were recorded at 90 MHz. 1H and 13C chemical shifts are
relative to respective internal TMS, and 31P chemical shifts
are relative to external 85% aqueous H3PO4. 13C signals for
the carbonyl ligands were not recorded due to the time
required to afford useful signal-to-noise ratios. The more
important IR and NMR spectroscopic parameters for com-
pounds 1-10 are presented in Table 1; others are listed with
their respective compounds below in this section. UV-visible
spectra were run in dichloromethane solution in quartz 1-cm
cuvettes on a Shimadzu 2100 spectrometer fitted with multiple-
cell changer and thermoelectric temperature control.
C
24H20AsNO4Ru: C 51.26; H 3.58. Found: C 51.13; H 3.56.
AsPh3-ligand NMR: 1H δ 7.41 (m); 13C δ 134.3, 132.8, 130.5,
129.1. Single crystals for X-ray analysis were grown by vapor
diffusion of hexanes into a toluene solution (ca. 5 mg in 10
mL).
The nitrato complexes 7-10 were prepared in analogous
fashion on a similar scale, with yields in the 77-89% range.
(η5-Cp )Ru (CO)(P P h 3)ONO2 (7). Yellow-orange microcrys-
tals. Anal. Calcd for C24H20NO4PRu: C 55.60; H 3.89. Found:
C 55.77; H 3.92. PPh3-ligand NMR: 1H δ 7.44 (m); 13C δ 133.4,
130.8, 128.7.
(η5-Cp )Ru (CO)(P (p-a n isyl)3)ONO2 (8). Yellow-orange mi-
crocrystals. Anal. Calcd for C27H26NO7PRu: C 55.72; H 4.50.
Found: C 55.58; H 4.49. P(p-anisyl)3-ligand NMR: 1H δ 7.41
(m, 9H), 6.92 (m, 6H), 3.62 (s, 9H); 13C δ 161.3, 134.8, 125.1,
114.1, 55.2.
(η5-Cp )Ru (CO)(P P h 2(o-a n isyl))ONO2 (9). Yellow-orange
microcrystals. Anal. Calcd for C25H22NO5PRu: C 57.74; H 4.04.
Found: C 57.45; H 4.07. PPh2(o-anisyl)3-ligand NMR: 1H δ
7.41 (m, 9H), 6.92 (m, 3H), 3.29 (s, 3H); 13C δ 160.6, 134.2,
133.8, 133.0, 132.6, 130.7, 130.6, 130.2, 128.7, 128.4, 128.1,
55.0.
(η5-Cp )Ru (CO)(P (OP h )3)ONO2 (10). Pale-yellow microc-
rystals. Anal. Calcd for C24H20NO7PRu: C 50.89; H 3.56.
Found: C 50.61; H 3.52. P(OPh)3-ligand NMR: 1H δ 7.32 (m);
13C δ 151.0, 129.8, 125.6, 121.4.
(η5-Cp )Ru (CO)(AsP h 3)Cl (1). Synthesis of this chloride
and its phosphorus-ligand analogues was similar to that
described in the literature,6 except that toluene instead of
xylenes was used as solvent. A solution initially containing
CpRu(CO)2Cl (129 mg, 0.500 mmol) and AsPh3 (168 mg, 0.549
mmol) in 50 mL of toluene was heated under nitrogen at reflux
for 30 h. The resulting clear yellow solution was cooled and
then handled aerobically. It was concentrated to about 5 mL
under reduced pressure. To the resulting yellow-orange liquid
was added dropwise 3 mL of diethyl ether and then 40 mL of
heptane. The microcrystalline yellow-orange precipitate af-
forded was filtered, washed with four 5-mL portions of hep-
(4) Branan, D. M.; Hoffman, N. W.; McElroy, A. E.; Prokopuk, N.
C.; Robbins, M. J .; Hill, W. E.; Webb, T. R. Inorg. Chem. 1991, 30,
1200 and references therein.
(5) Humphries, A. P.; Knox, S. A. R. J . Chem. Soc., Dalton Trans.
1975, 1710.
(6) Earlier studies have used xylene at reflux. (a) Brown, D. A.;
Lyons, H. J .; Manning, A. R. Inorg. Chim. Acta 1970, 4, 428. (b) Brown,
D. A.; Lyons, H. J .; Sane, R. T. Inorg. Chim. Acta 1970, 4, 621. (c)
J ungbauer, A.; Behrens, H. Z. Naturforsch., Teil B 1978, 33, 1083.
(η5-Cp )Ru (CO)(P P h 2(o-a n isyl))Br (11). Nitrate 9 (109.7
mg, 0.2000 mmol) was dissolved in 10 mL methanol. To the
resulting yellow-orange solution was added solid NaBr (52.7
mg, 0.512 mmol). Stirring of the mixture for 15 min produced
a microcrystalline yellow solid in a yellow liquid. Slow dropwise
addition of 30 mL of deionized water afforded more solid and
an essentially colorless liquid. The lemon-yellow solid was

2272 Organometallics, Vol. 20, No. 11, 2001
Cao et al.
filtered, washed with 3:1 (v/v) water/methanol, and then dried
first in air and next at 50 °C/50 mTorr. Yield 94.2 mg (83.2%).
Anal. Calcd for C25H22BrO2PRu: C 53.02; H 3.92. Found: C
53.08; H 4.04. PPh2(o-anisyl)3-ligand NMR: 1H δ 7.8-7.3 (m,
9H), 6.90 (m, 3H), 3.60 (s, 3H); 13C δ 160.0, 135.1, 133.8, 132.8,
132.5, 132.8, 130.3, 129.9, 128.3, 127.9, 127.0, 117.2, 111.0,
54.9.
Excel and evaluated in standard pseudo-first-order fashion
(linear regression of ln(A - A∞) versus time, discarding data
past 3 half-lives when instrumental noise became significant)
using the estimated A∞ values. Values for kobs were identical
within 3% using the three methods. No pattern related to
[(PPN+)Cl-]o was observed; values were scattered randomly
(standard deviation 0.20 × 10-5) around a mean of 4.37 × 10-5
s-1
.
UV-Visible Kin etics Exp er im en ts. Although the reac-
tion mixtures were not sensitive to O2, they were prepared
and handled under N2 to rigorously minimize exposure to
water. All glassware and the quartz cuvettes were heated for
at least 6 h in an oven at 150 °C. Upon their removal hot from
the oven, they were exposed to an N2 atmosphere so that any
surface making contact with the reaction mixtures could
transfer essentially no water. Dichloromethane dried as previ-
ously described over molecular sieves afforded results identical
to those when it was dried by prolonged reflux over and
distillation from CaH2. The cuvettes were placed into large-
neck Schlenk tubes under N2, reaction mixtures were trans-
ferred in by cannulae, and the cuvettes were tightly sealed
with Teflon stoppers and then inserted into the spectrometer’s
thermostated cell-holder. For each nitrato complex, the reac-
tion was first monitored by repetitive scans at appropriate
intervals over 550-350 nm to ensure clean isosbestic behavior
and to determine the optimum wavelength at which to monitor
absorbance over time. All experiments were run under pseudo-
first-order conditions, with [(PPN+)X-]o/[CpRu(CO)(ER3)-
ONO2]o ratio g 10 (X ) Cl for all five nitrates; X ) Br for only
nitrate 9). Fixed-wavelength absorbance versus time runs
employed three mixtures of different initial-concentration
ratios run simultaneously using the cell-changer. An example
for complex 6 below provides sufficient detail to describe our
techniques.
IR Con fir m a tion of Clea n Con ver sion of Nitr a te 6 in to
Ch lor id e 1. A mixture of 6 (8.4 mg, 0.015 mmol) and
(PPN+)Cl- (25.3 mg, 0.045 mmol) in a Schlenk tube under N2
was dissolved in 10.0 mL of dichloromethane. [A lower
[(PPN+)Cl-]o/[Ru-ONO2]o ratio was used here to preclude
interference from benzenoid overtone/combination bands of
PPN+.] A sample of the yellow-orange solution was transferred
into a 5-mm CaF2 cell, which was tightly sealed with Teflon
stoppers. The solution was scanned automatically via macro
control every 19 min. Spectral overlay (Figure 2) of selected
scans showed clean isosbestic behavior. This technique was
also used for the other nitrato complexes to confirm their clean
conversion into their respective chloro analogues and the
bromo analogue 11.
Activa tion P a r a m eter s for Con ver sion of Nitr a te 6
in to Ch lor id e 1. Multiple runs at 20.0 and 30.0 °C similar to
those described above at 25.0 °C were carried out and analyzed
to generate pseudo-first-order rate constants. Those experi-
ments afforded values which also were apparently randomly
scattered around their respective means of 2.82(11) × 10-5 s-1
and 7.69(16) × 10-5 s-1. Values for ∆H‡ and ∆S‡ were
determined from the regression analysis of an Eyring plot of
ln(k/T) versus 1/T using the equations ∆H‡/R ) - slope and
∆S‡/R ) y-intercept.8a Uncertainties were calculated using
MathCad.
Moistu r e Effect u p on th e Kin etics of Con ver sion of
Nitr a te 6 in to Ch lor id e 1. A reaction mixture of 6 (5.0 mg,
0.0090 mmol) and (PPN+)Cl- (57.4 mg, 0.090 mmol) in 10.0
mL of dry dichloromethane was prepared using standard
anhydrous techniques. A sample was transferred as usual into
a dry cuvette, which was stoppered and inserted into the
thermostated cell changer at 25.0 °C. The remainder of the
solution was transferred via cannula under N2 into an un-
heated 50-mL flask open to the atmosphere, a Teflon stirbar
was added, the flask was fitted with rubber septum, and the
solution was stirred 2 min. A 3-mL aliquot was transferred
by unheated airtight syringe (Teflon plunger - exposure to
relatively small glass surface area) into an unheated quartz
cuvette, which was stoppered and inserted into the thermo-
stated cell changer. The remaining solution was stirred
another 5 min and then transferred, this time using a 10-mL
all-glass syringe (exposure to large glass surface area) into
another unheated quartz cuvette. It was also inserted into the
cell changer, and absorbance at 438 nm was recorded auto-
matically every 24 s for 17.5 h for all three samples. Pseudo-
UV-Visible Mon itor in g of Kin etics of Con ver sion of
Nitr a te 6 in to Ch lor id e 1. Into an oven-dried Schlenk tube
under N2 containing a Teflon-coated magnetic stirbar was
placed 52.2 mg (0.0910 mmol) (PPN+)Cl- weighed under N2
(if weighed in air, its mass slowly increased as it picked up
water from lab air of ca. 70% relative humidity). Next 5.1 mg
(0.0091 mmol) 6 (weighed in air) was poured into the tube and
10.0 mL of dry dichloromethane was added by volumetric pipet
(taken hot from oven, with all liquid-contact surfaces shielded
from air). As soon as the reactants had dissolved, a portion of
the reaction mixture was transferred under N2 by cannula into
a dry cuvette. The cuvette was then stoppered and placed into
the thermostated cell-holder set for 25.0 °C; after allowing 4
min for the solution to reach constant temperature, scans over
550-350 nm were run automatically every 10.8 min for 18 h.
Spectral overlay (see Figure 1) of selected scans showed clean
isosbestic behavior. For determination of pseudo-first-order
rate constants and dependence upon [(PPN+)Cl-], 50.0 mL of
an 0.904 mM stock solution of 6 (25.4 mg, 0.0452 mmol, in
dichloromethane) was prepared via volumetric pipet in a
Schlenk tube. Then 10.00-mL aliquots of the stock solution
were transferred by volumetric pipet into four Schlenk tubes
containing different respective masses of (PPN+)Cl-: (a) 60.8
mg (0.106 mmol; [(PPN+)Cl-]o/[Ru-ONO2]o ) 11.7), (b) 82.7
mg (0.144 mmol; [(PPN+)Cl-]o/[Ru-ONO2]o ) 15.9), (c) 116.0
mg (0.2021 mmol; [(PPN+)Cl-]o/[Ru-ONO2]o ) 22.4), (d) 125.7
mg (0.2190 mmol; [(PPN+)Cl-]o/[Ru-ONO2]o ) 24.2). As soon
as the reactants had dissolved, a portion of each mixture was
transferred under N2 by cannula into respective dry cuvettes;
each cuvette was stoppered and placed into the thermostated
cell-changer array set for 25.0 °C. After allowing 4 min for
the solution to reach constant temperature, absorbance at 438
nm (decreasing) was recorded automatically every 24 s for 17.5
h. Since the reactions had not yet reached completion, the
absorbance-time data were converted into ASCII format and
imported into programs designed to estimate A∞ values and
then compute kobs (FITEX for Mac7a and FINAL7b for PC). As
a check on these values, the data were also imported into MS
first-order rate constants were determined in
a fashion
identical to that described for anhydrous runs above, with
values of R2 > 0.999 found for all three.
Gen er a t ion of (η5-Cp )R u (CO)(η2-P h 2P -C6H 4-o-OMe)+.
To a solution of 4 (12.5 mg, 0.0240 mmol) in 15 mL of
dichloromethane was added excess AgBF4 (8.3 mg, 0.043
mmol). The mixture was stirred for 1.0 h in the dark and
filtered through Celite. The resulting solution was somewhat
darker than the initial solution of 4. A 2-mL aliquot was taken
for IR analysis. After the spectrum had been run, (PPN+)Cl-
(2.5 mg, 0.0047 mmol) was added to this aliquot, and the IR
spectrum was recorded within 5 min of mixing. The remaining
solution was concentrated to ca. 1 mL under reduced pressure.
(7) (a) Program from the J ack Norton group (then at Colorado State
University) written by S. Kristjansdottir. (b) MathCad program by ,
A. Wierzbicki.
(8) Atwood, J . D. Inorganic and Organometallic Reaction Mecha-
nisms, 2nd ed.; VCH: New York, 1997: (a) pp 13-14; (b) pp 153-155.

Kinetic Studies of (η5-Cp)Ru(CO)(ER3)ONO2 in CH2Cl2
Organometallics, Vol. 20, No. 11, 2001 2273
Ta ble 1. P er tin en t Sp ectr oscop ic Da ta for
Attempts to isolate a solid by slowly adding 2 mLof diethyl
ether followed by 10 mL of cyclohexane afforded an intractable
tar.
(η5-Cp )Ru (CO)(ER3)X
δ 1H
δ
13C
(Cp)d
a
δ
31Pb (Cp)c
Sem iem p ir ica l Ca lcu la tion s. PM3(tm) calculations were
conducted using the Spartan platform from Wavefunction, Inc.
(either as PC-Spartan Pro, Spartan, or J aguar).9 The mini-
mization routine was prohibited from using the pseudodiago-
nalization algorithm during the SCF procedure. The default
fixed planar η5-cyclopentadienyl group ligand from the menu
was used unless otherwise specified. Where necessary, the
Ru-ONO2 distance was constrained to the value measured
by X-ray crystallography for 6. A variety of isomers of
pseudooctahedral (η3-Cp)Ru(CO)(η2-Ph2P(o-anisyl))ONO2 were
constructed by changing the relative locations of the P and O
ligating atoms of η2-Ph2P(o-anisyl). From this effort, the lowest-
energy isomer obtained upon PM3(tm) optimization was used
for the analyses presented later.
Cr ysta llogr a p h y. Crystals of 1‚tolu en e, 2, and 6 were
sealed into thin-walled glass capillaries. Cell information and
data collection were carried out on an Enraf Nonius CAD4 (1
and 6) and a Siemens R3m/v (2) diffractometer with mono-
chromatized Mo KR radiation (λ ) 0.71073A). Intensity data
were corrected for Lorentz and polarization effects. Structures
were discovered with SHELXTL-PC10 and SHELXS-97.11
Information on the crystallographic experiments is given in
Table 4
. Models included positions and anisotropic displacement
factors for all non-H atoms and calculated H-atom positions
and fixed isotropic displacement factors for H-atoms in ap-
propriate geometries. Structures were refined by full-matrix
least-squares methods against the intensities to convergence,
with form factors taken from the standard source.12 Data were
also corrected for absorption effects using SHELXTL-PC and
XEMP.10 In 2, a disorder in the locations of the Cl and carbonyl
ligands was included in the model refinement with the major
conformer having occupancy 0.893(8).
compd
ER3
X
νCO
1958
ONO2 1971
Cl 1959
ONO2 1972
Cl 1953
ONO2 1967
1962
PPh2(o-anisyl) ONO2 1974
PPh2(o-anisyl) Br 1961
1
6
2
7
3
8
4
9
11
5
AsPh3
AsPh3
Cl
4.93 83.2
4.98 82.5
PPh3
PPh3
48.92 4.89 85.6 (2.9)
50.21 4.94 85.0 (2.9)
44.19 4.88 85.5
45.50 4.93 84.9
41.37 4.80 85.3
43.75 4.87 84.5
41.15 4.79 85.5
P(p-anisyl)3
P(p-anisyl)3
PPh2(o-anisyl) Cl
P(OPh)3
P(OPh)3
Cl
1988 -24.44 4.53 85.1 (2.9)
10
ONO2 1988 -22.33 4.66 84.5 (2.9)
a
b
In CH2Cl2, cm-1
.
In CDCl3, ppm above external 85% H3PO4.
c In CDCl3, ppm above TMS. In CDCl3, ppm above TMS; coupling
d
constant to 31P in parentheses.
for donation into the CO π* orbitals.4,14 Similar patterns
1
were seen for the H and 13C Cp resonances (about δ
0.05 and 0.6 higher, respectively for the nitrates than
1
their chloride analogues).15
ν
C≡O, δ H (Cp), and δ 31P
values for the PPh2(o-anisyl) complexes 4 and 9 appear
anomalously low given the donor strength of PPh2(o-
anisyl) relative to the other pnictogen ligands, perhaps
consistent with the dramatically greater reactivity of
9.
Kin etics Stu d ies of th e Su bstitu tion of Ch lor id e
for Nitr a te in Com p lexes 6-10. Under pseudo-first-
order conditions (Cl-/Ru g 10), the nitrato complexes
reacted smoothly with (PPN+)Cl- to quantitatively
afford their respective chloro analogues (eq 1).
(η5-Cp)Ru(CO)(ER3)ONO2 + (PPN+)Cl- f
-
(η5-Cp)Ru(CO)(ER3)Cl + (PPN+)ONO2 (1)
Resu lts a n d Discu ssion
Repetitive scans of reaction mixtures using UV-
visible (550-350 nm) and IR spectroscopy (2020-1900
cm-1) displayed clean isosbestic behavior. Respective
examples are provided for reaction of complex 6 in
Figures 1 and 2. Isosbestic points were observed at 465,
384, and 359 nm and at 1968 cm-1, respectively. Similar
results were obtained for reactions of the other nitrates,
including those of 9 with (PPN+)Br- to form the bromide
11.
The complexes of formula (η5-Cp)Ru(CO)(ER3)ONO2
were prepared in high yield by treating their chloro
analogues with a slight excess of AgNO3. The chloro
To determine the values of kobs for each anion-
metathesis reaction, the absorbance at ca. 440 nm of
the anhydrous reaction mixture was recorded at short
intervals at constant temperature. For example, the
conversion of 6 to 1 was monitored at 438 nm, tracking
complexes were synthesized in high yield by heating (η5-
Cp)Ru(CO)2Cl with a slight excess of phosphorus or
arsenic ligand in refluxing toluene.6 Those containing
PPh3 and P(OPh)3 were reported earlier by other
groups. IR and NMR data pertinent to structure and
bonding considerations are reported in Table 1. As
expected, νC≡O were higher (roughly 13 cm-1) for the
nitrato complexes than for their respective chloro ana-
logues; such a pattern is typically observed for species
with less electron density at the metal center available
5
6a
the replacement of a more intense band for 6 (λmax
)
430 nm) with one less intense for 1 (λmax ) 448 nm).
Only for the conversion of complex 9 was the reaction
sufficiently fast to practically afford measured values
of A∞. For the others, absorbance-time data were
imported into programs7 using exponential fitting rou-
tines to generate estimates of A∞ and then calculate
values of kobs. Combination of A∞ values from these
(9) (a) PCSpartan Pro, version 1.01 and Spartan, version 5.1.
Wavefunction, Inc., 18401 Von Karman Ave., Suite 370, Irvine, CA
92612. (b) Titan, version 1. Wavefunction, Inc., 18401 Von Karman
Ave., Suite 370, Irvine, CA 92612; Schrodinger, Inc., 1500 SW First
Avenue, Suite 1180, Portland, OR 97201.
(10) Siemens SHELXTL-PC Manual, V4.2; Siemens Analytical
Instruments; Madison, WI, 1992.
(11) XEMP, 1989. Empirical Absorption Correction. Siemens Ana-
lytical X-ray Inc., 1989.
(12) Churchill, M. R. Inorg. Chem. 1973, 12, 1213.
(13) SHELXL-97/SHELXS-97. Sheldrick, G. Programs for solution
and refinement of crystal structures from X-ray diffraction data,
University of Go¨ttingen, 1997.
(14) A similar pattern is observed for ν(NtN) in trans-Ir(PPh3)2-
(N2)X complexes of the weakly coordinating trifluoracetate anion and
chloride ion. Kubota, M.; Chappell, T. G.; Powers, S. P Inorg. Chem.
1979, 18, 3615.
(15) Kulawiec, R. J .; Faller, J . W.; Crabtree, R. H. Organometallics
1990, 9, 745 and references therein.

2274 Organometallics, Vol. 20, No. 11, 2001
Cao et al.
Ta ble 2. F ir st-Or d er Ra te Con sta n ts for
An h yd r ou s Con ver sion of Nitr a te to Ha lid e
Com p lex
complex
ligand
halide temp (°C)
k (s-1
)
6
6
6
7
8
9
9
9
10
AsPh3
AsPh3
AsPh3
PPh3
P(p-anisyl)3
PPh2(o-anisyl)
PPh2(o-anisyl)
PPh2(o-anisyl)
P(OPh)3
Cl-
Cl-
Cl-
Cl-
Cl-
Cl-
Cl-
Br-
Cl-
20.0
25.0
30.0
25.0
25.0
25.0
20.0
20.0
25.0
2.82(11) × 10-5
4.37(20) × 10-5
7.69(16) × 10-5
2.10(18) × 10-5
3.78(20) × 10-5
4.90(10) × 10-3
2.71(13) × 10-3
2.64(15) × 10-3
1.40(21) × 10-5
similar in reactivity to its PPh3 analogue. Donor strength
for AsPh3 in binding to soft, low-valent metal moieties
has been estimated in thermochemical studies16 as just
slightly less than that of PPh3. νC≡O in CpRu(CO)-
(EPh3)X for both X ) ONO2 and Cl follow the trend
EPh3 ) P(OPh)3 < PPh3 < AsPh3 < P(p-anisyl)3. P(p-
anisyl)3 appears to be a significantly stronger donor
than AsPh3 in this system, with νC≡O 4-5 cm-1 lower
for complexes of the former (Table 1). Tertiary arsines
are also slightly less sterically demanding than their
phosphine analogues.17 Thus, the presence of AsPh3 in
the coordination sphere should be less effective in a
dissociative pathway than that of P(p-anisyl)3.
F igu r e 1. Selected successive UV-visible scans for con-
version of nitrate 6 into chloride 1.
An Eyring plot of the data for conversion of the AsPh3
complex 6 into 1 over the temperature range 20-30 °C
afforded values of ∆H‡ ) 17(2) kcal/mol and ∆S‡
)
-21(5) eu, with R2 ) 0.99. [The slowness of the reaction
and the volatility of solvent dichloromethane limited the
temperature range available.] The moderately large
negative entropy of activation argues strongly against
a mechanism involving rate-determining dissociation of
CO. It is, however, consistent with one in which nitrate
leaves the coordination sphere to form an ion-pair in
which nitrate is both tightly electrostatically attracted
to the resulting CpRu cation and highly solvated by CH2-
Cl2 molecules. [Negative entropies of activation have
been observed for anion metathesis using (PPN+)Cl- in
dichloromethane for FeBr42- and CpMoI2(PMe3)2; how-
ever, unlike our situation, both reactions exhibit first-
order dependence upon chloride concentration and
would be expected to create transition states more
F igu r e 2. Selected successive FTIR scans for conversion
of nitrate 6 into chloride 1.
programs and the individual spectroscopic data in a
standard spreadsheet pseudo-first-order linear-regres-
sion routine afforded similar numbers. Values for kobs
for a given run were always identical within 3% using
the three methods. Each set of nitrate (6-10) into
chloride metathesis experiments was run at 25 °C. That
for 6 was also conducted at 20 and 30 °C to allow
determination of activation parameters. Data appear in
Table 2. Also included are results at 20 °C for the
conversion of nitrate 9 into both its respective chloride
6 and bromide 11. Those rates are identical within
experimental error (i.e., no dependence upon either
ordered than the reactants alone.18,19
]
The complex containing the o-OMe-substituted phos-
phine PPh2(o-anisyl) was much more reactive than the
rest of the set (k1 over 2 orders of magnitudes greater).
nature or concentration of entering halide, and kobs
)
k1). These results afford the rate expression rate )
k1[(η5-Cp)Ru(CO)(ER3)ONO2].
ν
C≡O appeared higher than expected for both its nitrato
and chloro complexes based upon its substituents (Table
2), and its values for δ 1H (Cp) and δ 31P were similarly
inconsistent. Unfortunately, multiple attempts to grow
single crystals of 4 and 9 suitable for X-ray analysis
failed. A likely explanation for the major difference in
reactivity of 9 is stabilization of a coordinatively unsat-
urated intermediate by weak coordination of the poten-
tially chelating o-OMe group. Shaw20 publicized such
chemistry extensively, and many similar types of bi-
dentate hemilabile phosphorus ligands have been stud-
First-order rate constants for nitrate metathesis in
CpRu(CO)(ER3)ONO2 based upon ancillary pnictogen
ligand ER3 follow the order P(OPh)3 < PPh3 < P(p-
anisyl)3 < AsPh3 , PPh2(o-anisyl). For the less reactive
nitrate complexes containing phosphorus ligands, me-
tathesis rate qualitatively follows ligand donor strength,
with only a relatively small range exhibited. Such a
trend can be rationalized in terms of a mechanism
whose rate-determining step involves dissociation of
nitrate from the coordination sphere to form a tight ion-
pair. A stronger donor ligand can be pictured as (i)
weakening the Ru-ONO2 bond modestly by putting
more electron density upon the Ru atom and (ii)
somewhat better stabilizing the coordinatively unsatur-
ated positively charged intermediate. The AsPh3 com-
plex seems somewhat anomalous, for it should be quite
(16) Rottink, M. K.; Angelici, R. J . Inorg. Chem. 1993, 32, 3282 and
references therein.
(17) Tolman, C. A. Chem. Rev. 1977, 77, 313
(18) Algra, G. P.; Balt, S. Inorg. Chem. 1981, 20, 1102.
(19) Poli, R.; Owens, B. E.; Linck, R. G. Inorg. Chem. 1992, 31, 662.
(20) Empsall, H. D.; Hyde, E. M.; J ones, C. E.; Shaw, B. L. J . Chem.
Soc., Dalton Trans. 1974, 1980.

Kinetic Studies of (η5-Cp)Ru(CO)(ER3)ONO2 in CH2Cl2
Organometallics, Vol. 20, No. 11, 2001 2275
Ta ble 3. Selected Bon d Len gth s a n d An gles for Com p lexes 1‚Tolu en e, 2, a n d 6
complex
1‚toluene
2
6
formula
Ru-E (Å)
Ru-CCp (Å)
CpRu(CO)(AsPh3)Cl‚toluene
Ru(1)-As(1) ) 2.412(1)
Ru(1)-C(1) ) 2.200(11)
Ru(1)-C(2) ) 2.221(12)
Ru(1)-C(3) ) 2.226(18)
Ru(1)-C(4) ) 2.169(14)
Ru(1)-C(5) ) 2.175(12)
Ru(1)-C(0) ) 1.862
As(1)-C(11) ) 1.935(9)
As(1)-C(21) ) 1.931(8)
As(1)-C(31) ) 1.931(8)
C(0)-O(0) ) 1.15
C(1)-C(2) ) 1.363(23)
C(1)-C(5) ) 1.344(17)
C(2)-C(3) ) 1.408(22)
C(3)-C(4) ) 1.380(22)
C(4)-C(5) ) 1.403(25)
Ru(1)-Cl(1) ) 2.455(2)
CpRu(CO)(PPh3)Cl
CpRu(CO)(AsPh3)ONO2
Ru(1)-As(1) ) 2.435(1)
Ru(1)-C(1) ) 2.163(13)
Ru(1)-C(2) ) 2.185(12)
Ru(1)-C(3) ) 2.230(8)
Ru(1)-C(4) ) 2.243(7)
Ru(1)-C(5) ) 2.222(8)
Ru(1)-C(0) ) 1.852(8)
As(1)-C(11) ) 1.939(6)
As(1)-C(21) ) 1.934(5)
As(1)-C(31) ) 1.951(8)
C(0)-O(0) ) 1.139(10)
C(1)-C(2) ) 1.426(15)
C(1)-C(5) ) 1.386(20)
C(2)-C(3) ) 1.472(20)
C(3)-C(4) ) 1.351(15)
C(4)-C(5) ) 1.373(15)
Ru(1)-O(1) ) 2.113(6)
N(1)-O(1) ) 1.294(7)
Ru(1)-P(1) ) 2.3120(7)
Ru(1)-C(1) ) 2.231(4)
Ru(1)-C(2) ) 2.15(3)
Ru(1)-C(3) ) 2.207(4)
Ru(1)-C(4) ) 2.185(4)
Ru(1)-C(5) ) 2.218(4)
Ru(1)-C(0) ) 1.872(6)
P(1)-C(11) ) 1.824(2)
P(1)-C(21) ) 1.822(2)
P(1)-C(31) ) 1.824(3)
C(0)-O(0) ) 1.132(8)
C(1)-C(2) ) 1.410(8)
C(1)-C(5) ) 1.374(8)
C(2)-C(3) ) 1.349(8)
C(3)-C(4) ) 1.363(8)
C(4)-C(5) ) 1.397(8)
Ru(1)-Cl(1) ) 2.332(18)
Ru-Cco (Å)
E-C (Å)
CtO (Å)
C-C (Å)
other bond distances (Å)
N(1)-O(2) ) 1.209(12)
N(1)-O(3) ) 1.212(11)
As(1)-Ru(1)-O(1) ) 86.9(1)
As(1)-Ru(1)-C(0) ) 90.0(2)
O(1)-Ru(1)-C(0) ) 97.8(3)
Ru(1)-C(0)-O(0) ) 173.8(9)
angles about Ru (deg)
As(1)-Ru(1)-Cl(1) ) 90.7(1)
As(1)-Ru(1)-C(0) ) 88.7(3)
Cl(1)-Ru(1)-C(0) ) 93.0(3)
Ru(1)-C(0)-O(0) ) 176.3(8)
P(1)-Ru(1)-Cl(1) ) 90.45(3)
P(1)-Ru(1)-C(0) ) 88.19(12)
Cl(1)-Ru(1)-C(0) ) 92.78(15)
Ru(1)-C(0)-O(0) ) 178.3(8)
angle in carbonyl ligand (deg)
ied, including the Shell Higher Olefin Process industrial
application.21 Kulawiec et al.14 have reported complexes
of formula (η5-Cp)Ru(CO)(η2-PPh2(o-XC6H4))+, where X
) Cl or Br but not F, analogous to our postulated
intermediate (η5-Cp)Ru(CO)(η2-PPh2(o-MeOC6H4))+. Evi-
dence for our intermediate species was obtained by
stirring chloro complex 4 with an excess of AgBF4 in
CH2Cl2. A white precipitate formed, and the νC≡O band
shifted from 1962 to 1985 cm-1. Such a frequency shift
(+23 cm-1) is appropriate for the OMe-coordinated
cation, given that the o-OMe ligand should be a signifi-
cantly better donor than the o-Br and o-Cl ligands.
[Larger frequency shifts, 1971 to 2014 cm-1, were
observed by Kulawiec et al. upon stirring (η5-Cp)-
Ru(CO)(η1-PPh2(2-XC6H4))Cl with excess AgPF6.]
The effect of small amounts of water upon the reaction
rate was demonstrated by running the kinetics experi-
ments with some exposure to atmospherically derived
moisture. Allowing progressively more contact of the
reaction mixture with unheated glass surfaces as de-
scribed in the Experimental Section caused no change
in rate law but did increase the value of kobs (in
multiples of 10-5 s-1) from 4.33 to 5.11 to 5.51 at 25 °C.
F igu r e 3. Thermal ellipsoid plot of complex 6 with 50%
probablility enclosures.
molecular units, which are separated by normal van der
Waals distances. There are no abnormally short inter-
molecular contacts. Molecular geometry of this complex
is depicted in Figure 3; selected distances and angles
are presented in Table 3 and crystallographic data in
Table 4. The geometry around the ruthenium(II) is best
described as a “three-legged piano-stool” conformation.
The individual Ru-C(η5-C5H5) distances ranged from
Ru(1)-C(1) ) 2.163(13) Å to Ru(1)-C(4) ) 2.243(7) Å.
The C-C distances associated with the cyclopentadienyl
ring ranged from 1.351(15) to 1.472(20) Å. The three
“legs” of this molecule were composed of a terminal
carbonyl, a nitrate, and a triphenylarsine ligand, with
the angles about these substituents relatively close to
90°. The environment around the arsenic atom was a
The value of kobs never increased beyond 5.8 × 10-5 s-1
,
no matter how wet the solvent was allowed to become.
Similar results were observed in enhancement of me-
tathesis for all the nitrates. This effect can be rational-
ized easily in terms of the proposed rate-determining
step as conversion of coordinated nitrate to ion-paired
nitrate. Water would likely hydrogen-bond to coordi-
nated nitrate, weakening the Ru-ONO2 bond and
stabilizing the ion-pair intermediate in the weakly polar
solvent dichloromethane.22
Cr yst a l St r u ct u r e of Com p lex 6. The crystal
structure consists of an ordered arrangement of discrete
somewhat distorted tetrahedron, with
Ru-As-C
(21) Parshall, G. W.; Ittel, S. D. Homogeneous Catalysis, 2nd. ed.;
Wiley: New York, 1992; pp 70-71.
(22) We earlier observed strong perturbation of an equilibrium
involving Rh-F. Branan, D. M.; Hoffman, N. W.; McElroy, E. A.; Miller,
N. C.; Ramage, D. L.; Schott, A. F.; Young, S. H. Inorg. Chem. 1987,
26, 2915.
116(1)° and C-As-C 102(1)°, with distortion likely
due to the bulkiness of the Ru moiety. Complete
crystallographic data are available in the Supporting
Information.

2276 Organometallics, Vol. 20, No. 11, 2001
Ta ble 4. Cr ysta llogr a p h ic Deta ils for 1‚Tolu en e, 2, a n d 6
Cao et al.
complex
1‚toluene
2
6
formula
fw
C
628.0
31H28AsOClRu
C24H20POClRu
491.9
C24H20AsNO4Ru
562.4
size (mm)
system
space group
a
b
c
0.4, 0.25, 0.25
monoclinic
P2(1)/c (No. 14)
17.228(4)
7.831(1)
19.718(2)
90
110.84(2)
90
2486(1)
0.5, 0.2, 0.15
triclinic
0.3, 0.25, 0.25
triclinic
P-1 (No. 2)
10.253(3)
10.760(1)
11.959(3)
69.25(1)
68.00(2)
69.51(2)
1107.4(6)
2
P-1 (No. 2)
9.381(2)
10.589(2)
11.534(2)
76.64(2)
79.20(2)
78.33(2)
1079.7(4)
2
R
â
γ
V
Z (molecules/cell)
density
4
1.678
1.513
1.687
trans. coeff. (max, min)
F(000)
0.83, 0.67
1264
0.87, 0.65
496
0.64, 0.25
560
θ range, hkl range
2.5-22.5 18; 7; (21
1.8-32.5 14;(16;(17
2.5-22.5 11;(11;(12
scan type
ω
ω
ω-2θ
intensities measd
3299
9241
3071
unique refl. (Rmerge
obsd reflns (cutoff)
parameters
)
3179 (3.9%)
2302 (I > 2σI)
269
7824 (4.0%)
5655 (I > 2σI)
267
2880 (4.5%)
2519 (I > 2σI)
281
R, wR (observed)
R, wR (all data)
goodness-of-fit
0.0470, 0.0572
0.0716, 0.0779
0.95
0.0446, 0.0973
0.0708, 0.1085
1.03
0.0600, 0.0599
0.0654, 0.0698
0.82
data/parameter ratio
diff map features
8.6
29.3
+0.86, -0.87
10.2
+2.09, -1.49
+0.96, -1.40
Cr ysta l Str u ctu r e of Com p lex 1. A toluene mol-
ecule of solvation was present in the crystal structure.
This solvent molecule exhibited static disorder. The
disorder pattern can be easily rationalized as the
averaging of two distinct toluene molecules which are
related by a 180° rotation. The resultant averaged
electron density possessed a center of symmetry. The
six-membered toluene ring was treated as fixed hexagon
during refinement. There was also static disorder as-
sociated with the arrangement of the CpRu(CO)(AsPh3)-
Cl molecules in the crystal structure. Minor scrambling
between the carbonyl and chloro ligands of each mol-
ecule was observed. Unfortunately, this disorder mani-
fested itself as an impossibly short C-O bond length of
1.012(16) Å. All attempts to model the disorder were
unsuccessful due to the resolution of the study. The final
structure is reported with the distances associated with
the carbonyl fixed at “reasonable” distances. The C0-
O0 and Ru1-C0 bond distances were constrained
respectively to interatomic distances of 1.15 and 1.862
Å. The resulting structure modeled the actual electron
density and the refinement converged (largest ∆/σ )
0.023, which was related to the disordered toluene
solvent molecule) with R ) 0.0470 and wR ) 0.0572 for
those 2302 reflections with Fo > 4σ(Fo). The final
difference Fourier map showed no anomalous features
after refinement, with the residual electron density in
the range -1.40 to 0.96 e Å-3. The ORTEP diagram
appears as Figure 4, and pertinent structural data are
presented in Tables 3 and 4. Complete crystallographic
data are available in the Supporting Information.
Cr ysta l Str u ctu r e of Com p lex 2. We decided to
obtain a new crystal structure of 2 because the original23
displayed an R factor above 0.10 and we wished to
compare the PM3(tm)-computed structure with an
experimentally determined structure of higher quality.
F igu r e 4. Thermal ellipsoid plot of complex 1‚toluene with
50% probablility enclosures.
In the later stages of model refinement, it became clear
that the carbonyl and chlorine groups were disordered.
[The carbonyl bond length was very short; in fact, much
shorter than the 1.145(20) Å typical of most carbonyl
complexes.] Although the Fourier difference map showed
residual peaks in the vicinity of the metal atom and the
carbonyl group, an empirical absorption correction failed
to eliminate the features near the carbonyl. Since the
Ru-Cl and Ru-(CO) distances were similar, a model
in which a disorder in the Cl and CO groups occurred
was examined, with immediate improvement in the
model. The model with contributions from this disorder
converged with a major conformer having a site oc-
cupancy of 0.893(8). The C-O bond length (of the major
isomer) became 1.132 Å. The alternate Cl, C, and O
atoms were modeled with isotropic displacement factors,
and the metrics associated with them were close to their
isomeric counterparts with an appropriate increase in
errors since their density was small. Final R [I > 2σ(I)]
) 0.0446, wR2 ) 0.0973, and for all data R1 ) 0.0708,
wR2 ) 0.1085. The ORTEP diagram appears as Figure
(23) Wilczewski, T.; Dauter, Z. J . Organomet. Chem. 1986, 312, 349.

Kinetic Studies of (η5-Cp)Ru(CO)(ER3)ONO2 in CH2Cl2
Organometallics, Vol. 20, No. 11, 2001 2277
Ta ble 5. Com p a r ison of (η5-Cp )Ru (CO)(ER3)X Cr ysta l a n d Com p u ted Str u ctu r es
ER3
X
Ru-CO
CO
Ru-ER3
Ru-X
X-Ru-E
X-Ru-C
C-Ru-E
AsPh3 (crystal)
NO3
NO3
Cl
Cl
Cl
1.852(8)
1.872
1.139(10)
1.177
2.485(1)
2.459
2.412(1)
2.448
2.312
2.253
2.113(6)
2.113/fixed
2.455(2)
2.323
2.401
2.331
86.9(1)
89.0
90.8(1)
89.94
90.45
91.9
97.8(3)
96.5
92.6(3)
89.31
92.77
90.24
90.0(2)
91.1
88.5(3)
91.78
88.18
92.57
AsPh3 (PM3(tm)b)
AsPh3‚toluene (crystal)c
AsPh3 (PM3(tm))
PPh3 (crystal)d
1.862c
1.856
1.15c
1.182
1.132
1.182
1.872
1.856
PPh3 (PM3(tm))
Cl
a
Experimental and calculated bond lengths and angles are reported in angstroms and degrees, respectively. b Ru-ONO2 bond constrained
at 2.11 Å. c Carbonyl/chloro disorder afforded unrealistically short CO bond length of 1.012(16) Å, so Ru-C and CO arbitrarily were
assigned more reasonable values on the basis of Cambridge Structural Databased carbonyl/chloro disorder observed here, but structure
solution parameters allowed theoretically reasonable values for Ru-C and CO bond lengths.
Ta ble 6. Com p u ted Hea ts of F or m a tion of Nitr a te
Com p lexes a n d P r op osed Ca tion In ter m ed ia tes
∆Hf of
nitrate
complex
(kcal/mol)
∆Hf of proposed
cation intermediate
(kcal/mol)
difference
(kcal/mol)
ER3
AsPh3
PPh3
P(p-anisyl)3
-94.272 41.844
-145.782 -9.588
-260.656 -127.211
136.116
136.194
133.445
132.127
127.429
146.842
PPh2(o-anisyl) -175.056 (η1-PPh2(o-anisyl)) -42.929
PPh2(o-anisyl) -175.056 (η2-PPh2(o-anisyl)) -47.627
P(OPh)3
-318.573 -171.731
a
Calculated heats of formation determined using PM3(tm)-
optimized geometries.
for our proposed mechanism for the nitrate-to-chloride
metatheses can be estimated by comparing the differ-
ences in computed energies (standard heats of forma-
tion) for the nitrato complexes and the (η5-Cp)Ru(CO)-
(ER3)+ intermediates. Although the absolute energies
for complexes are not accurate, differences between
energies in pairs of species structurally related like
those here are considered to be qualitatively meaning-
ful.24 Pertinent data appear in Table 6. Two entries
appear for the Ph2P(o-anisyl) nitrate 9 to show the effect
of chelation by the o-OMe group. The values in general
are quite similar, but two points stand out: (i) the
considerably higher value for the P(OPh)3 system (the
least reactive nitrate) and (ii) the significantly lower
value for the chelated Ph2P(o-anisyl) system.
To address the considerably greater reactivity of the
Ph2P(o-anisyl) nitrate 9 in more detail, we computed a
series of structures. (η5-Cp)Ru(CO)(η1-L)ONO2 with
unconstrained Ru-ONO2 distance afforded an un-
expected species in which the nitrate anion is appar-
ently dissociated without significant bonding of the
o-OMe group to Ru: Ru-ONO2 ) 4.73 Å and Ru-OMe
) 4.149 Å. Constraining Ru-ONO2 to 3.50 Å, a distance
observed without constraint for the PPh3 analogue,
raised the energy by 10 kcal/mol for the geometry-
optimized structure but reduced Ru-OMe only to 4.13
Å. Removing the nitrate anion completely without
mechanically forming a Ru-OMe bond afforded a co-
ordinatively unsaturated cation with Ru-OMe distance
of 3.233 Å and an energy of -42.93 kcal/mol. Mechani-
cally forming a Ru-OMe bond yielded an η2 coordina-
F igu r e 5. Thermal ellipsoid plot of complex 2 with 50%
probablility enclosures.
5, and pertinent structural data are presented in Tables
3 and 4. Complete crystallographic data are available
in the Supporting Information.
Sem iem p ir ica l P M3(tm ) Com p u ta tion a l Stu d ies.
A comparison of X-ray and PM3(tm) semiempirical
structural data (Table 5) for the complexes available (1,
2, and 6) showed reasonable agreement between the
two. The computed structures display Ru-CO slightly
short, CtO somewhat too long, Ru-Cl moderately too
short, and bond angles quite acceptable. For 6 it was
necessary to constrain the Ru-O bond to keep the
nitrate group from dissociating to noncovalent distances
(ca. 4.6 Å). The Ru-ONO2 distance was set to 2.113 Å,
the value observed in its crystal structure. With the
same minimization routine in place, the other nitrato
species did not dissociate the nitrate group but placed
it considerably farther away than seen in the AsPh3
complex crystal structure (from 2.33 Å for 8 to 2.18 Å
for 10). To get a better prediction of structures, we also
constrained the Ru-O bond distances in 7-10 to
essentially that in the crystal structure of 6.
Within the set of computed structures, very small
differences in Ru-C and Ru-Cl bond distances were
observed for chlorides 1-4, and Ru-P distances were
very similar for 2-4. The computed P(OPh)3 chloride
5, however, displayed Ru-C 0.02 Å longer and Ru-P
0.03 Å shorter than did its phosphine analogues. The
set of computed nitrates showed modest differences in
two bond distances relative to the computed chlorides:
Ru-C ca. 0.01 Å longer and Ru-E slightly longer
(0.005-0.01 Å longer). Computed bond angles around
Ru were similar for the two sets. Pertinent data are
summarized in the Supporting Information.
(24) (a) Adam, K. R.; Atkinson, I. M.; Lindoy, L. F. J . Mol. Struct.
1996, 384, 183. (b) Børve, K. J .; J ensen, V. R.; Karlsen, T.; Støvneng,
J . A.; Swang, O. J . Mol. Model. 1997, 3, 193. (c) Decker, S. A.; Donini,
O.; Klobukowski, M. J . Phys. Chem. A 1997, 101, 8734. (d) Cundari,
T. R.; Saunders:, L. C.; Sisterhen, L. L. J . Phys. Chem. 1998, 102, 997.
(e) Donovan-Merkert, B. T.; Clontz, C. R.; Rhinehart, L. M.; Tijong,
H. I.; Carlin, C. M.; Cundari, T. R.; Rheingold, A. L.; Guzei, I.
Organometallics 1998, 17, 1716. (f) Decker, S. A.; Klobukowski, M.
Can. J . Chem. 1999, 77, 65. (g) Cundari, T. R.; Deng, J . J . Chem. Info.
Comput. Sci. 1999, 39, 376. (h) Cundari, T. R.; Deng, J . Intern. J .
Quantum Chem. 2000, 77, 421.
If one assumes that the ion-pairing electrostatic
interactions between (η5-Cp)Ru(CO)(ER3)+ and nitrate
ion are very similar in strength for all ER3, relative ∆H‡

2278 Organometallics, Vol. 20, No. 11, 2001
Ta ble 7. Com p u ted Effects for η1 to η2 Con ver sion of (η5-Cp )Ru (CO)(P h 2P (o-XC6H4))+
η1 w η2 changes
∆(Ru-X) (Å) ∆(Ar-X) (Å) ∆(∆Hf) (kcal/mol)
Cao et al.
Xa
r(Ru-X) (Å) r(Ar-X) (Å) ∆Hf (kcal/mol)
Ru-X-Ar (deg)
X
η1-Br
η1-Cl
η1-F
3.169
4.463
4.179
5.385
2.654
2.304
3.588
3.552
1.894
1.698
1.347
1.376
1.908
1.723
1.347
1.406
3.245
-6.298
-43.415
-42.929
-2.579
-29.693
-43.275
-47.627
NA
NA
NA
NA
94.3
108.4
94.0
97.5
Br
Cl
F
OMe
-0.515
-2.519
-0.591
-1.833
0.014
0.025
0.000
0.027
-5.824
-23.395
0.140
η1-OMe
η2-Br
η2-Cl
η2-F
-4.698
η2-OMe
a
Calculated metrics and heats of formation determined using the PM3(tm) semiempirical Hamiltonian.
reported in crystallographic data for η2-bound haloarene
ligands15) than for X ) OMe. Pertinent parameters are
presented in Table 7.
A path by which complex 9 might dissociate the
nitrate anion was suggested from computed structures
for the two isomers (Figure 7) of coordinatively satu-
rated (η3-Cp)Ru(CO)(η2-Ph2P(o-MeOC6H4))ONO2, which
could be generated from 9 by Cp ring slippage.8b [Both
isomers displayed no imaginary frequencies when
vibrations were calculated, a fact showing that neither
corresponds to a transition state.] The P-equatorial
form, 19 kcal/mol more stable than its P-axial isomer,
had elongated Ru-ONO2 (3.50 Å) and Ru-OMe (3.40
Å). Return of the Cp ring from η3- to η5-form could expel
nitrate from the coordination sphere to generate the
putative ion-paired intermediate (η5-Cp)Ru(CO)(η2-Ph2P-
(o-MeOC6H4))+.
F igu r e 6. Computed η2 coordinatively saturated cation
with a very weak Ru-OMe bond.
tively saturated cation (Figure 6) with a very weak Ru-
OMe bond (3.55 Å) about 5 kcal/mol lower in energy
than the former. The η2-isomer is essentially isoelec-
tronic to the halo-substituted aryl species (η5-Cp)Ru(CO)-
(η2-Ph2P(o-XC6H4))+ characterized by Kulawiec et al.15
Computations of structures for the η1 and η2 isomers
for X ) Br, Cl, and F showed no stabilization upon η1
to η2 conversion for X ) F (matching their experimental
observations), considerable stabilization for X ) Cl, and
stabilization for X ) Br comparable to that for X ) OMe.
Considerably shorter Ru-X bonds were computed for
X ) Cl and Br (reasonably close to values such as those
Con clu sion
The kinetics studies on the chloride-for-nitrate me-
tathesis carried out for this set of five d6 (η5-Cp)Ru(CO)-
(ER3)ONO2 showed that the stereoelectronic nature of
the ancillary pnictogen ligand has a relatively minor
effect upon reactivity. Four factors argue strongly for a
mechanism in which nitrate ion dissociates from the
ruthenium coordination sphere to generate a coordina-
F igu r e 7. Computed isomers of coordinatively saturated (η3-Cp)Ru(CO)(η2-Ph2P(o-MeOC6H4))ONO2 generated via Cp
ring slippage.

Kinetic Studies of (η5-Cp)Ru(CO)(ER3)ONO2 in CH2Cl2
Organometallics, Vol. 20, No. 11, 2001 2279
tively unsaturated highly solvated contact ion-pair: (a)
the form of rate law observed (first-order in [CpRu-
ONO2], zero-order in chloride and bromide), (b) the rate
enhancement without change in form of rate law caused
by traces of water, (c) the moderately large negative
entropy of activation, (d) and the very large rate
enhancement stemming from the presence of the o-
methoxy group in the Ph2P(o-anisyl) ligand. Semi-
empirical PM3(tm) calculations modeled the reactants
and products reasonably well on the basis of compari-
sons of geometry-optimized structures with X-ray crys-
tal structures, and computed putative intermediates
were logical in terms of energetics and structural
parameters.
and thanks Dr. Alan Salter for helpful discussions.
T.R.C. wishes to acknowledge the National Science
Foundation for their support of equipment (CHE-
9785017) and facilities (STI-9602656) improvements for
the CROMIUM research laboratories. E.A.V. thanks the
Office of Naval Research for instrumentation support.
Su p p or tin g In for m a tion Ava ila ble: Complete structure
reports including atomic coordinates, thermal parameters, and
bond lengths and angles for the crystal structures of 1, 2, and
6 and the Eyring plot for conversion of nitrate 6 into chloride
1. This material is available free of charge via the Internet at
http://pubs.acs.org. Complete atomic coordinates, thermal
parameters, and bond lengths have been deposited at the
Cambridge Crystallographic Data Centre. CCDC refs: 157617,
157697, and 157618. This information is also available free of
charge via the Internet at http://pubs.acs.org.
Ack n ow led gm en t. N.W.H. acknowledges support
from Research Corporation and the Petroleum Research
Fund, administered by the American Chemical Society,
OM000670D