Kinetic and thermodynamic studies of reaction of ?Cr(CO)3C5Me5, HCr(CO)3C5Me5, and PhSCr(CO)3C5Me5 with ?NO. Reductive elimination of thermodynamically unstable molecules HNO and RSNO driven by formation of the strong CR-NO bond

Article abstract of DOI:10.1021/ic990829i

Reaction of H-Cr(CO)₃C₅Me₅ with ?NO at 1-2 atm pressure in toluene solution yields Cr(NO)(CO)₂C₅Me₅ as the sole metal-containing product in addition to N₂O and HNO₂ as the principle nitrogen-containing products. N₂O and HNO₂ are attributed to decomposition of the initial product HNO. Kinetic studies yield the rate law d[P]/dt = -k^{2nd order}[HCr(CO)₃C₅Me₅][?NO]; k^{2nd order} = 0.14 M^-1 s^-1 at 10 °C, with ΔH^≠ = 11.7 ± 1.5 kcal/mol and ΔS^≠ = -16.3 ± 3.5 cal/(mol deg). The rate of reaction is not inhibited by CO. The kinetic isotope effect for. reaction of D-Cr(CO)₃C₅Me₅ is k_H/k_D = 1.7. These observations are consistent with a first step involving direct H (D) atom transfer from the metal hydride to ?NO, forming HNO. Also supporting this mechanism is the ～150 times slower reaction of H-Mo(CO)₃C₅Me₅ and failure to observe reaction for H-W(CO)₃C₅Me₅ in keeping with metal-hydrogen bond strengths Cr < Mo < W. Reaction of PhS-Cr(CO)₃C₅Me₅ with NO at 1-2 atm pressure in toluene solution also forms Cr(NO)(CO)₂C₅Me₅ as the sole metal-containing product. The initial product is the unstable nitrosothiol PhS-NO. Kinetic studies yield the rate law d[P]/df = -k^{1st order} [PhS-Cr(CO)₃C₅Me₅]; k^{1st order} = 3.1 ± 0.3 × 10^-3 s^-1 at 10 °C, with ΔH^≠ = 21.6 ± 1.2 kcal/mol, ΔS^≠ = + 3.9 ± 1.5 cal/(mol deg). The rate of reaction is independent of both NO and CO pressure. The transition state in the first-order process is proposed to involve migration of bound thiolate to coordinated CO, forming Cr(CO)₂ (n²-C(=O)SPh)C₅Me₅. The enthalpy of reaction of ?Cr(CO)₃C₅Me₅ and NO yielding Cr(NO)(CO)₂C₅Me₅ and CO has been measured by solution calorimetry: Δ^° = -33.2 ± 1.8 kcal/mol. The Cr-NO bond strength is estimated as ～70 kcal/mol and provides the net thermodynamic driving force for the proposed elimination of the unstable molecules HNO and PhSNO.

Full text of DOI:10.1021/ic990829i

6
206
Inorg. Chem. 1999, 38, 6206-6211
Kinetic and Thermodynamic Studies of Reaction of •Cr(CO)3C5Me5, HCr(CO)3C5Me5, and
PhSCr(CO)3C5Me5 with •NO. Reductive Elimination of Thermodynamically Unstable
Molecules HNO and RSNO Driven by Formation of the Strong Cr-NO Bond
Kenneth B. Capps, Andreas Bauer, Kengkaj Sukcharoenphon, and Carl D. Hoff*
Department of Chemistry, University of Miami, Coral Gables, Florida 33124
ReceiVed July 14, 1999
Reaction of H-Cr(CO)3C5Me5 with •NO at 1-2 atm pressure in toluene solution yields Cr(NO)(CO)2C5Me5 as
the sole metal-containing product in addition to N2O and HNO2 as the principle nitrogen-containing products.
N2O and HNO2 are attributed to decomposition of the initial product HNO. Kinetic studies yield the rate law
2
nd order
2nd order
-1 -1
q
d[P]/dt ) -k
[HCr(CO)3C5Me5][•NO]; k
) 0.14 M
s
at 10 °C, with ∆H ) 11.7 ( 1.5 kcal/mol
q
and ∆S ) -16.3 ( 3.5 cal/(mol deg). The rate of reaction is not inhibited by CO. The kinetic isotope effect for
reaction of D-Cr(CO)3C5Me5 is kH/kD ) 1.7. These observations are consistent with a first step involving direct
H (D) atom transfer from the metal hydride to •NO, forming HNO. Also supporting this mechanism is the ∼150-
times slower reaction of H-Mo(CO)3C5Me5 and failure to observe reaction for H-W(CO)3C5Me5 in keeping with
metal-hydrogen bond strengths Cr < Mo < W. Reaction of PhS-Cr(CO)3C5Me5 with NO at 1-2 atm pressure
in toluene solution also forms Cr(NO)(CO)2C5Me5 as the sole metal-containing product. The initial product is the
1
st order
1st order
unstable nitrosothiol PhS-NO. Kinetic studies yield the rate law d[P]/dt ) -k
[PhS-Cr(CO)3C5Me5]; k
-
3
-1
q
q
)
3.1 ( 0.3 × 10
s
at 10 °C, with ∆H ) 21.6 ( 1.2 kcal/mol, ∆S ) + 3.9 ( 1.5 cal/(mol deg). The rate
of reaction is independent of both NO and CO pressure. The transition state in the first-order process is proposed
2
to involve migration of bound thiolate to coordinated CO, forming Cr(CO)2 (η -C(dO)SPh)C5Me5. The enthalpy
of reaction of •Cr(CO)3C5Me5 and NO yielding Cr(NO)(CO)2C5Me5 and CO has been measured by solution
calorimetry: ∆H° ) -33.2 ( 1.8 kcal/mol. The Cr-NO bond strength is estimated as ∼70 kcal/mol and provides
the net thermodynamic driving force for the proposed elimination of the unstable molecules HNO and PhSNO.
Introduction
The chemistry of metal-nitrosyl complexes has been inten-
in eq 2. The first step in reaction 2 was proposed to involve
•
Cr + PhS-SPh h Cr-SPh + •SPh + •Cr f Cr-SPh (2)
1,2
sively investigated in recent years. Despite a wealth of
synthetic and structural observations, kinetic and thermodynamic
studies of this simple radical analogue of CO are rare. The
authors can find no reliable thermodynamic data regarding the
M-NO bond strength for organometallic complexes in solution.
Nitric oxide is a “weak” radical in the sense that its bond
reversible generation of the •SPh radical in a thermodynamically
uphill reaction. The activation energy for the first step, ∼10
kcal/mol, was found to be only ∼2 kcal/mol above the calculated
thermodynamic estimate on the basis of measured bond dis-
4
sociation energies. In addition, the •SPh radicals generated in
3
4,5
strength to hydrogen (∼50 kcal/mol ) is low enough to make it
reaction 2 were shown to rapidly abstract an H atom (eq 3).
thermodynamically unstable with respect to eq 1. We have
•
SPh + H-Cr f H-SPh + •Cr
(3)
2
(H-NO) f H + 2•NO
(1)
6
2
The bond strengths to hydrogen for Cr-H (62 kcal/mol) and
7
H-SPh (79 kcal/mol) can be used as a scale for the “strength”
recently investigated reactions of the stable chromium-centered
radical •Cr(CO)3C5Me5 () •Cr) with phenyl disulfide as shown
of the corresponding radicals •Cr and •SPh. The “weak” natures
of the •NO, •Cr, and •SPh radicals are also reflected in the low
4
8
9
10
enthalpies of dimerization [•NO e 4, •Cr ) 15, •SPh ) 43
(
(
(
1) Richter-Addo, G. B.; Legzdins, P. Metal Nitrosyls; Oxford University
kcal/mol] as shown in eq 4. Interest in radical reactions involving
Press: New York, 1992.
2) Lancaster, J. Nitric Oxide, Principles and Actions; Academic Press:
Limited, London, 1996.
•
A + •A f A-A
(4)
3) The H-NO bond strength of 50 kcal/mol is calculated on the basis of
the best existing experimental data (ref 3a). Recent high level
theoretical calculations suggest a value of it ∼48 kcal/mol (ref 3b).
In the authors’ view, a solution phase estimate of 50 ( 4 kcal/mol
seems reasonable. Discussion of decomposition pathways of HNO can
be found in ref 3c. (a) Chase, M. W.; Davies, C. A.; Downey, J. R.;
Frurip, D. J.; McDonald, R. A.; Syverud, A. N.; J. Chem. Phys. Ref.
Data 1985, 14, Supplement 1. (b) Lee, T. J.; Dateo, C. E. J. Chem.
Phys. 1995, 103, 9110. (c) Bunte, S. W.; Rice, B. M.; Chabalowski,
C. F. J. Phys. Chem. A 1997, 101, 9430. (d) Nakamoto, K. Infrared
and Raman Spectra of Inorganic and Coordination Compounds; Wiley
and Sons: New York, 1997.
such weak radicals11 and bound substrates prompted us to
(4) Ju, T. D.; Capps, K. B.; Lang, R. F.; Hoff, C. D. Inorg. Chem. 1997,
36, 614.
(5) The rate of reaction of phenyl thiyl radicals and H-Mo(CO)3C5H5 is
one of the fastest H atom transfers to this radical that have been
measured: Franz, J. Private communication.
(6) Kiss, G.; Zhang, K.; Mukerjee, S. L.; Hoff, C. D.; Roper, G. C. J.
Am. Chem. Soc. 1990, 112, 5657.
(7) Bordwell, F. G.; Zhang, X. M.; Satish, A. V.; Cheng, J. P. J. Am.
Chem. Soc. 1994, 116, 6605.
1
0.1021/ic990829i CCC: $18.00 © 1999 American Chemical Society
Published on Web 12/09/1999

Reductive Elimination Reactions with •NO
Inorganic Chemistry, Vol. 38, No. 26, 1999 6207
Rate of Reaction of •NO and H-Cr(CO) Me . Nitric oxide was
investigate reactions of nitric oxide. The concomitant formation
of strong metal-nitrosyl bonds is the driving force in its
reactivity. This paper reports the first in a series of kinetic and
thermodynamic study of reactions of •NO and group VI metal
complexes.12
3
C
5
5
passed through coiled copper tubing held at -78 °C in a dry ice/acetone
trap to remove •NO /N and used to fill a 1 L stainless gas cylinder
2
2 4
O
fitted with a calibrated pressure transducer obtained from Omega
Scientific. The previously evacuated container was filled to a total
absolute pressure of ∼3 atm. A thermostated glass reactor at 10.0 °C
in the FT-IR microscope/reactor system11 was evacuated and filled with
nitric oxide three times and then loaded with 25 mL of freshly distilled
toluene. Reaction was initiated by syringe addition of 2.0 mL of a
Experimental Section
All manipulations were carried out using standard Schlenk tube/
inert atmosphere techniques. Toluene was purified by distillation from
sodium/benzophenone ketyl under argon. Methylene chloride was
purified by distillation from phosphorus pentoxide under argon.
toluene solution of HCr(CO)
3 5 5
C Me , yielding a starting concentration
of HCr(CO) Me of 0.0058 M. Pressure and temperature were
3
C
5
5
recorded continuously throughout the reaction. The reaction was
conveniently monitored by following the decay of peaks at 1996 and
Research grade H
2
, CO, and •NO were obtained from Matheson gas;
[99% isotopic purity] were obtained from Isotec. The
complexes RS-Cr(CO) Me were prepared as reported in the
-
1
13
1913 cm due to HCr(CO)
3 5 5 5 5 2
C Me or production of C Me Cr(CO) -
CO and D
2
-
1
(
NO) with peaks at 2000, 1930, and 1684 cm . Production of N O
was monitored by its band at 2219 cm , and HNO
its band at 3443 cm . Samples of N
decomposition of NH NO
2
3
C
5
5
-
1
4
,26
2
was monitored by
prepared by thermal
and acidification of NaNO with anhydrous
literature.
-1
2 2
O and HNO
4
3
2
(
8) The value for the enthalpy of dimerization of nitric oxide is taken
from ref 18. It is of interest to estimate enthalpies of bond formation
for mixed species assuming a low metathesis enthalpy A-A + B-B
f 2 A-B. Using these data produces the following yields estimates
for the radical pairs: C5Me5(CO)3Cr-SPh ) 29, PhS-NO ) 23.5, and
C5Me5(CO)3Cr-NO ) 9.5 kcal/mol. The value for Cr-SPh is lower
than the measured value of 35 kcal/mol, and values for the other two
estimated bond strengths are not known.
HCl were observed to match our band assignments in toluene solution
in our reactor system. Study of the rate of reaction as a function of
added CO pressure [up to 2 atm] showed no influence on rate of
reaction. Plots of ln[HCr(CO) C Me ] versus time were linear for 2-4
3 5 5
half-lives. Rates of reaction shown in Table 1 in the Supporting
Information typically had experimental errors of ∼10%. The rate of
3 5 5
reaction of DCr(CO) C Me was shown to be 1.7 ( 0.2 times slower
(9) Watkins, W. C.; Jaeger, T.; Kidd, C. E.; Fortier, S.; Baird, M. C.;
Kiss, G.; Roper, G. C.; Hoff, C. D. J. Am. Chem. Soc. 1992, 114,
than that measured for HCr(CO)
CO) Me and H-W(CO) Me
Rates of Reactions of H-Mo(CO)
with •NO. The rates of reaction of H-Mo(CO)
Me were studied under identical conditions and in the same reactor
system described above for H-Cr(CO) Me . The rate of reaction of
H-Mo(CO) Me was initially studied at 10 °C, where it underwent
no apparent reaction. Study of the rate of reaction at 30 °C led to
observed production of Mo(NO)(CO) Me by a first-order process
3
C
5
5
Me
with •NO.
Me
5
.Rates of Reactions of H-Mo-
(
3
C
5
5
3
C
5
907.
(
10) This value for the sulfur-sulfur bond strength is derived on the basis
of literature values for the enthalpy of hydrogenation of phenyl
disulfide as discussed in detail in ref 4.
3
C
5
5
and H-W(CO)
3 5 5
C Me
3 5
C
Me and H-W(CO) C -
5
3 5
5
(11) Ju, T. D.; Capps, K. B.; Roper, G. C.; Hoff, C. D. Inorg. Chim. Acta
3
C
5
5
1
998, 270, 488.
12) (a) Ju, T. D.; Lang, R. F.; Hoff, C. D. J. Am. Chem. Soc. 1996, 118,
328. (b) Lang, R. F.; Ju, T. D.; Kiss, G.; Hoff, C. D.; Bryan, J. C.;
3
C
5
5
(
5
Kubas, G. J. J. Am. Chem. Soc. 1994, 116, 7917. (c) Capps, K. B.;
Bauer, A.; Kiss, G.; Hoff, C. D. J. Organomet. Chem. 1999, in press.
2
C
5
5
with a rate of reaction ∼150 times slower than that observed for reaction
of the chromium hydride under these conditions. Nitrous oxide was
produced as the reaction proceeded, as monitored by its characteristic
(
d) Capps, K. B.; Bauer, A.; Sukcharoenphon, K.; Hoff, C. D.
Unpublished results.
(13) Solubility Data Series Vol 8 Oxides of Nitrogen; Young, C. L., Ed.;
Pergamon Press: Oxford, 1981.
-
1
3 5 5
band at 2219 cm . Reaction of H-W(CO) C Me with •NO was also
studied at 30 °C and a nitric oxide pressure of 2.8 atm over a 30 min
(
14) The only literature report the authors could find for reaction of nitric
oxide and the group VI metal hydrides is a brief statement in the
Experimental Section of ref 14a. Several routes to preparation of Cr-
period. No reaction was observed to take place under these conditions.
Rate of Reaction of •NO and PhS-Cr(CO)
reaction were obtained in a manner strictly analogous to that described
above for H-Cr(CO) Me . Reaction rates measured based on decay
of PhS-Cr(CO) Me at 2008, 1954, 1920 (sh) cm or growth of
peaks due to C Me Cr(CO) (NO) were identical. First-order plots were
3 5 5
C Me . Data for this
(
1
1
NO)(CO)2C5R5, as well as its crystal structure, are reported in ref
4b. (a) Piper, T. S.; Wilkinson, G. J. Inorg. Nucl. Chem. 1956, 3,
11. (b) Atwood, J. L.; Shakir, R.; Malito, J. T.; Herberhold, M.;
3
C
5
5
5
-
1
3
C
5
Kremnitz, W.; Bernhagen, W. P. E.; Alt, H. G. J. Organomet. Chem.
979, 165, 65.
15) Reaction of HCl with OsCl(CO)(NO)(PPh3)2 yields a stability complex
of HNO, OsCl2(CO)(HNO)(PPh3)2 , the crystal structure of which has
been determined: Wilson, R. D.; Ibers, J. A. Inorg. Chem. 1979, 18,
1
5
5
2
-
1
(
found to be linear for 2-4 half-lives. A broad band at 1565 cm in
the FT-IR spectrum was observed to rise during reaction and undergo
6 6
slow decay and is assigned to PhSNO. A reaction done in C D [also
3
36.
distilled from Na/benzophenone] allowed better observation of this peak.
Due to the fact that this peak rose to a maximum value and began to
decay prior to the completion of the experiment, no kinetic analysis
was performed regarding generation of PhSNO. Reactions done using
a gas mixture of 1 atm nitric oxide and 2 atm carbon monoxide occurred
at the same rate as reactions using a gas mixture of 1 atm nitric oxide
and 2 atm argon pressure.
(
16) Kristjansdottir, S. S.; Norton, J. R. In Transition Metal Hydrides;
Dedieu, A., Ed.; VCH Publishers: New York, 1992; p 309.
(
17) Such radical combinations are typically very fast. For example, the
second-order rate constant for 2•Cr(CO)3C5Me5 f C5Me5(CO)3Cr-
7
-1 -1
Cr(CO)3C5Me5 has been determined to be 1.7 × 10 M
s
at 228
K in CH2Cl2: T. C.; Geiger, W. E.; Baird, M. C. Organometallics
1
994, 13, 4494.
(
18) Cotton, F. A.; Wilkinson, G.; Murillo, C. A.; Bochmann, M.; AdVanced
Reaction of •NO and HS-Cr(CO)
3 5 5
C Me . The rate of reaction of
HS-Cr(CO) Me with •NO was studied at 10 °C using techniques
Inorganic Chemistry; Wiley: New York, 1999.
3
C
5
5
(
(
19) Franz, K. J.; Lippard, S. J. J. Am. Chem. Soc. 1998, 120, 9034.
20) It should be noted that the activation parameters are calculated using
rate constants with all species in toluene solution. The enthalpy and
entropy of a solution of nitric oxide in toluene are -0.5 kcal/mol and
(23) The value for νRS-NO for a number of nitrosothiols are in this region:
(a) Oae, S.; Kim, Y. H.; Fukushima, D.; Shinhama, K. J. Chem. Soc.,
Perkin Trans. 1 1978, 913. (b) Oae, S.; Shinhama, K.; Fujimori, K.;
Kim, Y. H. Bull. Chem. Soc. Jpn. 1980, 53, 775.
1
3
-
15.2 cal/(mol deg), respectively. The entropy of activation based
on reaction of nitric oxide gas would be substantially more negative
[
ca. -31.5 cal/(mole deg)], while the enthalpy of activation is little
(24) Barnett, K. W.; Slocum, D. W. J. Organomet. Chem. 1972, 44, 1.
(25) (a) There is little evidence in related chemistry of this system that
changed.
3
(
21) This system would be expected to exhibit an inverse equilibrium
isotope effect because νM-H < νH-NO. However, if the reaction
transition state depends on breaking the M-H versus M-D bond in
an early transition state prior to establishment of the H-NO bond, a
normal kinetic isotope effect is expected. For an excellent discussion
of isotope effects in related transfers see: Bullock, R. M. In Transition
Metal Hydrides; Dedieu, A., Ed.; VCH Publishers: New York, 1992;
p 263.
ring slippage to form coordinatively unsaturated (η Cp*)Cr(CO)3X
is a low energy pathway, but this cannot be entirely discounted. Work
aimed at testing this and other mechanistic possibilities is planned.
(b) A referee has pointed out that the proposed intermediates in these
reactions might be trapped by other species in addition to nitric oxide.
Experiments to test this using suitable trapping agents are also planned.
The proposed reaction pathways have reasonable support but are
tentative, as in all mechanistic work.
(22) Barnett, D. J.; McAninly, J.; Williams, D. L. H. J. Chem. Soc., Perkin
Trans. 2 1994, 1131.
(26) Bauer, A.; Capps, K. B.; Wixmerten, B.; Abboud, K. A.; Hoff, C. D.;
Inorg. Chem. 1999, 38, 2136.

6208 Inorganic Chemistry, Vol. 38, No. 26, 1999
Capps et al.
identical to those described above. The rate of reaction was found to
be first order in HS-Cr(CO) Me . The observed first-order rate
constant at two pressures of nitric oxide were measured: k ) 5.2 ×
at 2.35 atm nitric oxide and k ) 5.3 × 10 at 1.26 atm
nitric oxide, indicating zero-order dependence on nitric oxide. FT-IR
spectral changes confirmed that Cr(NO)(CO) Me was the sole metal-
3
C
5
5
-
4
-1
-4 -1
1
0
s
s
C
2 5
5
containing product in the reaction. No sign of production of N
seen on the time scale of the experiment.
2
O was
Enthalpy of Reaction of •NO and •Cr(CO)
of reaction in toluene solution of •Cr(CO) Me
3
C
5
Me
5
. The enthalpy
3
C
5
5
was measured using
a high-pressure Calvet cell at a pressure of 2.5 atm nitric oxide (10
measurements) and in a Guild solution calorimeter at 1.0 atm (5
measurements). The average value of -33.0 ( 1.8 kcal/mol is the
average of both types of measurements and corresponds to the enthalpy
of reaction with all species in solution. Because of the highly exothermic
nature of reactions of trace oxygen with both the metal radical and
nitric oxide, these measurements required additional purification of nitric
oxide by exposing it to a solution of the metal radical to ensure nitrogen
dioxide was not present during the calorimetric runs. Spectroscopic
analysis by FT-IR of the reactions done in toluene and NMR analysis
of two experiments done in deuteriobenzene confirmed the clean
conversion of •Cr(CO)
3
C
5
Me
5
to Cr(NO)(CO)
3
C
5
Me
. The complex Cr-
was prepared by stirring a solution of 0.3 g of •Cr-
5
.
1
2
13
Reaction of CO with Cr(NO)( CO)
2 5 5
C Me
1
3
(
NO)( CO) Me
2
C
5
5
13
(
CO) Me in 60 mL of toluene under at a pressure of 2 atm CO
3
C
5
5
for 30 min. Injection of 50 mL of nitric oxide gas using a high-pressure
stainless steel syringe obtained from Harvard Scientific to this solution
and continued stirring resulted in quantitative conversion of the labeled
1
3
radical complex to Cr(NO)( CO)
spectroscopy: the bands of C Me
1
2
C
5
Me
5
as determined by FT-IR
5
5
Cr(CO)
2
(NO) at 2000, 1930, and
3 5
Figure 1. Infrared spectroscopic data for reaction of H-Cr(CO) C -
-
1
-1
684 cm are shifted to 1958, 1887, and 1682 cm for the labeled
-1
Me
5
(peaks decreasing at 1996 and 1912 cm ) and •NO in toluene
complex. This solution was transferred under argon to a high-pressure
sample cylinder directly attached to a high-pressure FT-IR cell obtained
from Harrick Scientific. The cell was filled with 14 atm of 12CO and
heated over a period of about 2 h to 60 °C. No detectable change or
solution at 20 °C at 2.17 atm, with •NO pressure producing Cr(NO)-
(CO) C Me (peaks increasing at 2001, 1929, and 1682 cm ). Inset
peak is band at 2019 cm due to increase in N O.
2
-
1
2
5
5
-1
1
2
broad band at 3442 cm-1 in toluene. As described in the
incorporation of CO occurred during this time. The cell was allowed
to cool and left overnight. A spectrum run the next day also showed
Experimental Section, prepared samples of N O and HNO2
2
1
2
no apparent incorporation of CO.
matched the experimental peaks in the kinetic runs. As shown
in Figure 1 in the Supporting Information, under pseudo-first-
order conditions in [•NO], first-order plots of decay of [HCr-
(CO)3C5Me5], buildup of [Cr(NO)(CO)2C5Me5], and buildup of
[N2O] are nearly parallel. The similar rates of decay of the metal
hydride and buildup of nitrous oxide and metal nitrosyl complex
do not prove that the reaction is occurring through intermediate
HNO formation but do support that generation of N2O is linked
Results and Discussion
There is little literature data available to indicate the reaction
products of metal hydrides and nitric oxide beyond the fact that
metal nitrosyls can be formed.14,15 This work was begun to see
if •NO was capable of direct attack on the metal-hydrogen
bond of the complexes H-M(CO)3C5Me5 (M ) Cr, Mo, W).
Kinetic Study of Reaction of •NO and H-Cr(CO)3C5Me5.
In toluene solution under pseudo-first-order conditions of 1.18-
-
1
to reaction of the hydride. The broad band at 3442 cm
-1
assigned to HNO2 was shifted to 2550 cm when D-Cr(CO)3C5-
Me5 was studied, indicating that DNO2 was formed by interac-
tion with the hydride and not solvent or a contaminant such as
trace water.
2
.81 atm of NO, reaction of the chromium hydride occurs
cleanly as monitored by FT-IR spectroscopy (eq 5). The fate
H-Cr(CO) C Me + 2•NO f
As shown in Table 1 in the Supporting Information, the rate
of reaction was found to be first order in •NO pressure over
the 1.18-2.81 atm pressure range studied. The values of the
second-order rate constants derived from these data are 7.7 ×
3
5
5
Cr(NO)(CO) C Me + CO + “H-NO” (5)
2
5
5
of the chromium complex is unequivocally followed by the
characteristic νCO and νNO bands as shown in Figure 1.
Production of nitrosyl hydride (“HNO”) is based on detection
-
4
-3
-3
1
0
(0 °C), 1.69 × 10 (10 °C), and 3.50 × 10 (20 °C)
-1 -1
3
atm
s
. The rates of reaction with all species in toluene
solution were calculated using published data on the solubility
of its decomposition products N2O and HNO2 by FT-IR
spectroscopy. Information concerning the decomposition of
HNO is available in the gas phase and under various conditions
1
3
of nitric oxide in toluene, yielding derived second-order rate
constants based on reaction of nitric oxide in solution: 6.3 ×
-
2
-1
-1
3
10 (0 °C), 1.37 × 10 (10 °C), and 2.85 × 10 (20 °C)
in aqueous solution. The authors were unable to find literature
-1 -1
M
s
. Activation parameters derived from these data (for
data on its decomposition in toluene solution under a pressure
of •NO. A reasonable expectation based upon observations in
the literature is shown in eq 6. The presence of N2O is readily
2
0
q
all species in solution) are ∆H ) 11.7 ( 1.5 kcal/mol and
q
∆
S ) -16.3 ( 3.5 cal/(mol deg).
The rate of reaction was found to be unchanged when 2-3
HNO + 2•NO f N O + HNO2
(6)
atm of CO pressure were added, discounting reaction mecha-
nisms based on reversible loss of CO. The rate of reaction of
D-Cr(CO)3C5Me5 was found to be 1.7 times slower than that
of H-Cr(CO)3C5Me5, giving further evidence that the transition
2
established by its characteristic IR absorption3d at 2219 cm-1
in toluene solution, and HNO2 was identified on the basis of a

Reductive Elimination Reactions with •NO
Inorganic Chemistry, Vol. 38, No. 26, 1999 6209
state involves breaking the metal-hydrogen bond. This normal
kinetic isotope effect for a reaction expected to have an inverse
equilibrium isotope effect implies an early transition state for
the H atom transfer reaction.21
in toluene solution also forms Cr(NO)(CO)2C5Me5 as the sole
metal-containing product, and the unstable nitrosothiol PhS-
NO is also proposed to be formed as an initial product (eq 11).
The activation parameters are in reasonable agreement with
a mechanism involving rate-limiting H atom transfer from
chromium to nitric oxide as shown in eq 7, followed by rapid
trapping of the chromium radical as shown in eq 8.
PhS-Cr(CO) C Me + 2•NO f
3
5
5
Cr(NO)(CO) C Me + PhS-NO (11)
2
5
5
Nitrosothiols22 are known to decompose according to eq 12.
RS-NO f RS-SR + 2•NO (12)
Characterization of PhS-NO is based on the observed FT-IR
H-M(CO) C Me + •NO f •M(CO) C Me + H-NO (7)
3
5
5
3
5
5
•
M(CO) C Me + •NO f M(NO)(CO) C Me + CO (8)
3 5 5 2 5 5
2
Available data on M-H bond strengths for these complexes
6,16
(
Cr ) 62, Mo ) 65, W ) 73 kcal/mol) indicate that H atom
-
1
23
transfer shown in eq 7 would be uphill by 12, 15, and 23 kcal/
mol for Cr, Mo, and W, respectively. Any metal radical
generated in eq 7 would be expected to combine rapidly with
band at 1565 cm , assigned for νPhS-NO, which grew in and
then decayed during the course of reaction.
The rate of reaction 11 was studied under pseudo-first-order
conditions of constant pressure of •NO. Spectroscopic data
showed clean and quantitative conversion of PhSCr(CO)3C5-
Me5 to Cr(NO)(CO)2C5Me5 . Plots of ln[PhSCr(CO)3C5Me5]
versus time were linear through at least three half-lives. As
shown in Table 2 in the Supporting Information, the rate of
reaction was found to be independent of the pressure of •NO
when studied between 1.0 and 2.84 atm. Values for the first-
1
7
nitric oxide (eq 8).
Reactions involving •NO always present the possibility that
NO2 may be involved instead. This more powerful radical,
capable even of removing a hydrogen atom from R-H, is
readily formed when •NO is exposed to O2 or from the well-
known disproportionation reaction18 (eq 9). Whereas reaction
•
3
•NO h N O + •NO2
(9)
2
-4
-3
order rate constant of 8.0 × 10 (10 °C), 3.1 × 10 (20 °C),
-
2
-1
and 1.1 × 10
(30 °C) s
yield calculated activation
7
is endothermic, reaction 10 is readily calculated to be
exothermic for all of the metal hydrides (eq 10).
q
q
parameters of ∆H ) 21.6 ( 1.2 kcal/mol and ∆S ) +3.9 (
.5 cal/(mol deg). The rate of reaction was shown to be
1
independent of added CO pressure; reactions performed at a
total pressure of 3 atm (1 atm •NO and 2 atm CO) yielded the
same first-order rate constant as those done under 3 atm (1 atm
H-M(CO) C Me + •NO f •M(CO) C Me + H-NO
3
5
5
2
3
5
5
2
(10)
•
NO and 2 atm Ar).
The sequence of reactions 9 and 10 followed by 8 presents
an alternative pathway for which generation of •NO2 in eq 9
would be rate-limiting. Reactions in which •NO2 may build up
either from slow infusion of oxygen or metal-catalyzed dispro-
portionation may lead to false conclusions regarding apparent
reactivity of metal hydrides if care is not taken to exclude this
possibility. The measured value for the enthalpy of activation,
The fact that the rate of reaction was independent of both
•NO and CO suggested two possible mechanisms. The first is
rate-determining fragmentation of the starting materials to
radicals by complete cleavage of the Cr-SPh bond (eq 13).
19
4
PhS-Cr(CO) C Me f PhS• + •Cr(CO) C Me (13)
∼
12 kcal/mol, is identical to estimates made earlier on the basis
3
5
5
3
5
5
∆
H ) + 35 kcal/mol
of bond strength data. A small overbarrier of 1-4 kcal/mol is
expected for radical reactions of this type. The close agreement
between the calculated energy difference based on bond strength
data and the observed enthalpy of activation is likely attributable
to experimental errors on the order of 2-3 kcal/mol in each of
these bond strength estimates.3
Rapid trapping of the two radicals (eq 13) by •NO would yield
products. However, the activation parameters listed above do
not support this mechanism. The reported value for the Cr-
,6
4
SPh bond of 35 kcal/mol is too high to support full dissociation
Rate-determining H atom transfer from metal to nitric oxide
is also supported by investigation of reactions of the analogous
molybdenum and tungsten hydrides. As discussed earlier,
molybdenum and tungsten hydride complexes are predicted to
have activation energies (based on the metal-hydrogen bond
dissociation energies) 3 and 11 kcal/mol, respectively, higher
than that for chromium if the mechanism involves breaking the
M-H bond as a rate-determining step. Qualitative studies
showed that under comparable conditions the molybdenum
hydride reacted ∼150 times slower than the chromium hydride
and the tungsten hydride did not react to any extent. The fact
that the tungsten hydride did not react to any detectable amount
shows that under the conditions utilized reactions based on •NO2
are probably not responsible. Although no mechanism can be
completely proven, eqs 7 and 8 are the simplest reaction scheme
to radicals. In addition, the entropy of activation would be
expected to be more positive for a dissociative mechanism.
Any other mechanism independent of [•NO] and not involving
either dissociation of CO (ruled out because the rate of reaction
is independent of added CO) or •SPh (ruled out because the
enthalpy of activation is too low) must be a rate-limiting
rearrangement of the metal complex. Rate-determining migration
of alkyls to coordinated carbon monoxide have been shown to
yield rate laws for reaction 14 which are independent of [L]:24
rds
fast
8
2
R-Mo(CO)
Cp
9
8 Mo(CO)
(η -C(dO)R)Cp + L
3
2
R-C(O)Mo(CO) (L)Cp (14)
2
2
5
accounting for all of the observations described above.
Kinetic Study of Reaction of •NO and PhS-Cr(CO)3C5Me5.
Reaction of PhS-Cr(CO)3C5Me5 with •NO at 1-2 atm pressure
Analogous reactivity with phenyl thiyl migration and •NO as
the incoming ligand is shown in eq 15. Thermochemical data

6210 Inorganic Chemistry, Vol. 38, No. 26, 1999
Capps et al.
Me5, the total Cr-NO bond strength is estimated to be 70 kcal/
mol for loss of •NO as a net three-electron donor (eq 19). This
Cr(NO)(CO) C Me f •NO + •Cr(CO) C Me (19)
2
5
5
2
5
5
∆H ) + 70 kcal/mol
value provides a “benchmark” for how strong thermodynami-
cally disfavored loss of •NO is when bound to a low valent
saturated complex. As discussed above, the exothermic nature
of reaction 18 provides the driving force for the calculated
thermoneutral production of phenyl thiyl radicals according to
reaction 16. Additional work is in progress to further define
(
∆
subtraction of eq 18 reported below from eq 13) indicate that
H° ≈ 2 kcal/mol for reaction 16. Because it can be estimated
24
29
the thermodynamic activating power of •NO.
1
2
13
Rate of Reaction of CO with Cr(NO)( CO)2C5Me5.
These studies were begun to try to obtain some estimate the
enthalpy of binding of •NO as a one-electron versus three-
electron donor. The measured net enthalpy of reaction 18 (-33
kcal/mol) can be considered the sum of the two steps shown in
eq 20. The first step shown in eq 20 is formation of a radical
PhS-Cr(CO) C Me + •NO f
3
5
5
PhS• + CO + Cr(NO)(CO) C Me (16)
2
5
5
that ∆S° ≈ 30 cal/(mol deg) for reaction 16, any net process
leading to it should have ∆G° e 0. Other possibilities with
literature precedent, namely, involvement of the Cp* ligand
either in migration of the •SPh radical to the ring or in η to η
“
(i)
(ii)
5
3
ON• + •Cr(CO) C Me
98 ON•‚‚‚•Cr(CO) C Me 98
3
5
5
3 5 5
2
5
ring slippage”, are viewed by the authors as less likely.
Cr(NO)(CO) C Me + CO (20)
2
5
5
Kinetic Study of Reaction of •NO and HS-Cr(CO)3C5Me5.
Recently the sulfhydryl complexes HS-M(CO)3C5Me5 have been
prepared in our laboratory.26 Reaction of •NO with H-S-Cr-
pair or bent nitrosyl complex in which •NO is effectively a one-
electron donor. The second step is loss of a CO ligand and
conversion to a bent nitrosyl complex that is effectively a three
electron donor. Kinetic data for either the forward or reverse
steps in reaction 20 could provide information about how the
(
CO)3C5Me5 was investigated to see if reaction might occur by
attack at the sulfur-hydrogen bond (eq 17). It was thought that
?
•
NO + H-S-Cr(CO) C Me
9
8 H-NO + •S-Cr(CO) C Me
3
5
5
3
5
5
-
33 kcal/mol is partitioned between steps i and ii.
There is considerable evidence, particularly for phosphine
(17)
reaction 17 might be feasible if the H-S bond strength in the
substitution, that conversion of linear NO complexes to bent
facilitates ligand substituion. The reverse of step ii for eq 20
1
sulfhydryl complex was reduced as a result of stability of the
•
S-Cr(CO)3C5Me5 radical.26 However, no evidence for reaction
would provide a mechanism for exchange of isotopically labeled
1
3
1
7 was found: (i) the rate of reaction was zero-order in •NO
CO. The labeled complex Cr(NO)( CO)2C5Me5 was prepared,
and the rate of reaction 21 investigated. Heating the reactants
pressure and hence resembles reaction of PhS-Cr(CO)3C5Me5
instead of H-Cr(CO)3C5Me5, and (ii) decomposition products
expected from reaction 17 [N2O and HNO2 from HNO or
chromium sulfido products from •S-Cr(CO)3C5Me5] were not
detected. Reaction proceeded cleanly to give Cr(NO)(CO)2C5-
Me5 as the only metal-containing product. Reaction of HS-Cr-
1
3
12
Cr(NO)( CO) C Me + CO f
2
5
12
13
Cr(NO)( CO) C Me + CO (21)
2
5
5
from room temperature to 60 °C under a pressure of 15 atm
(CO)3C5Me5 was found to be slower than that of PhS-
1
2
CO resulted in no detectable substitution over a period of 2
Cr(CO)3C5Me5, in keeping with the fact that the Cr-SPh bond
is weaker than the Cr-SH4,26 bond, which would presumably
h. These data imply a substantial barrier to the reverse of the
second step in eq 20. All attempts on our part to detect the
proposed intermediate ON•‚‚‚•Cr(CO)3C5Me5 have been unsuc-
cessful to date and indicate vary low barriers to reaction 20 in
present a higher barrier to migration.
Enthalpy of Reaction of •NO and •Cr(CO)3C5Me5. The
enthalpy of reaction 18 was measured in toluene solution at 30
2
9
the forward direction. These two observations imply that the
majority of the -33 kcal/mol is “spent” on the second reaction
and that the enthalpy of forming the initial adduct is small.
Additional work in progress is aimed at better understanding
of the thermochemistry of •NO coordination.
°
C:
•
NO + •Cr(CO) C Me f Cr(NO)(CO) C Me + CO
3
5
5
2
5
5
∆
H° ) - 33.0 ( 1.8 kcal/mol
18)
(
Reaction 18 includes loss of CO as well as coordination of •NO.
The Cr-CO bond strength in Cr(CO)6 is 36.8 kcal/mol.
Assuming this bond strength can be transferred28 to •Cr(CO)3C5-
Conclusion
2
7
Reaction of •NO and H-Cr(CO)3C5Me5, RS-Cr(CO)3C5Me5,
and •Cr(CO)3C5Me5 have been studied. Combination of the two
radicals leads rapidly to Cr(NO)(CO)2C5Me5 and is exothermic
by 33 ( 1.8 kcal/mol. The Cr-NO bond strength of ∼70 kcal/
mol is one of the highest ligand-to-metal bonds measured to
(
27) Smith, G. P. Polyhedron 1988, 7, 1605.
(
28) There is always risk involved in assuming that bonds strengths in one
complex can be transferred to another. The formal oxidation state of
Cr in •Cr(CO)3C5Me5 is +1 compared to Cr(CO)6 where the formal
oxidation state is 0. Higher oxidation states normally result in weaker
M-CO bonds, but the strong donating power of the Cp* ligand offsets
that such that the average CO stretching frequencies of the A and E
3
0
date for a first row metal.
(29) At temperatures as low as - 30 °C, even under CO atmosphere, no
intermediate such as Cr(NO)(CO)3C5Me5 could be detected. The
sampling time in these experiments was 10 s. Work in progress is
aimed at repeating this measurement on the millisecond time scale.
(30) Hoff, C. D. Prog. Inorg. Chem. 1992, 40, 503.
-
1
bands of •Cr(CO)3C5Me5 (1994 and 1886 cm ) are lower than the
-
1
observed A band of Cr(CO)6 (1980 cm ). In short, the authors feel
that the bond strengths are comparable for the two low valent
complexes discussed here.

Reductive Elimination Reactions with •NO
Inorganic Chemistry, Vol. 38, No. 26, 1999 6211
Ultimately it is this driving force that is responsible for
generation of the unstable molecules HNO and RSNO when
nitric oxide reacts with the chromium hydride and thiolate
complexes. The mechanisms of these two apparently similar
reductive eliminations are in fact quite different. Reaction of
H-Cr(CO)3C5Me5 is proposed to proceed by rate-limiting H atom
transfer from chromium to nitric oxide, whereas reaction of RS-
Cr(CO)3C5Me5 was shown to proceed by a rate-limiting
intramolecular rearrangement of the chromium thiolate, possibly
ligand is displayed. For the saturated complexes, H-Cr(CO)3C5-
Me5 and RS-Cr(CO)3C5Me5, the observed reactions are rela-
tively slow, especially when compared to the rate of reaction
of •Cr(CO)3C5Me5 . Additional work is in progress to further
explore thermodynamic and kinetic aspects of how •NO interacts
with complexes of the group VI metals.
Acknowledgment. Support of this work by the Petroleum
Research Fund administered by the American Chemical Society
and the National Science Foundation is gratefully acknowledged.
A. B. thanks the Deutsche Forschungsgemeinschaft (DFG) for
the granted postdoctoral fellowship.
2
through a complex such as Cr(CO)2 (η -C(dO)SPh)C5Me5 . In
neither of these reactions does interaction of nitric oxide with
the metal atom play an initial decisive role. Reaction of the
hydride is viewed as abstraction of an H atom in which •NO
reacts simply as a weak radical in the first step. In reaction of
RS-Cr(CO)3C5Me5, the apparent role of •NO is to trap a reactive
intermediate formed in the first step of the reaction, which is
viewed as an intramolecular rearrangement. It is in the final
steps of these reactions that the unique character of •NO as a
Supporting Information Available: Experimental data and cal-
culated rate constants for reactions of PhS-Cr(CO)
CO) Me with •NO and first-order plots of decay of [HCr(CO)
Me ], buildup of [Cr(NO)(CO) Me ], and buildup of [N O]. This
3 5 5
C Me and H-Cr-
(
C
3 5
5
3 5
C -
5
2
C
5
5
2
material is available free of charge via the Internet at http://pubs.acs.org.
IC990829I