Structure and reactivity studies of CoHNO+ in the gas phase

Article abstract of DOI:10.1021/jp991791y

CoHNO⁺ was generated in the gas phase by collision-induced dissociation (CID) or sustained off-resonance irradiation (SORI)-CID of CoH(CO)₃NO⁺, formed from chemical ionization of Co-(CO)₃NO with CH₅⁺. Among the three isomeric structures, HCo(NO)⁺ (1), Co(HNO)⁺ (2), and Co(NOH)⁺ (3), density functional calculations found that 2 was the global minimum on the potential energy surface with 1 and 3, 14.1 and 15.4 kcal/mole higher in energy. Co(HNO)⁺ reacted with methane by dehydrogenation to yield CoCH₃NO⁺, while CID of CoCH₃NO⁺ yielded competitive formation of CoCH₃⁺ and CoNO⁺, SORI-CID of CoCH₃NO⁺ yielded CoCH₃⁺ exclusively. These results suggested CoCH₃NO⁺ had structure CH₃CoNO⁺ rather than Co(CH₃NO)⁺. A mechanism was proposed for the reaction with methane, where a four-centered transition state is involved for C-H bond activation as opposed to direct C-H insertion.

Full text of DOI:10.1021/jp991791y

10884
J. Phys. Chem. A 1999, 103, 10884-10892
Structure and Reactivity Studies of CoHNO+ in the Gas Phase
Huiping Chen,† Denley B. Jacobson,* and Ben S. Freiser‡
H. C. Brown Laboratory of Chemistry, Purdue UniVersity, West Lafayette, Indiana 47907
ReceiVed: June 3, 1999; In Final Form: August 26, 1999
The structure and energetics of three CoHNO+ isomers [HCo(NO)+ (1), Co(HNO)+ (2), Co(NOH)+ (3)]
were probed by using density functional theory (DFT). Theory predicts that 2 is the most stable structure
with 1 and 3 14.1 and 15.4 kcal/mol less stable. We were unable to locate the transition states for 2 f 1 and
2 f 3 conversions. DFT calculations predict D°(CoH+-NO) ) 34.6 kcal/mol and D°(Co+-HNO) ) 45.5
kcal/mol. The gas-phase ion chemistry of CoHNO+ was also studied by using Fourier transform ion cyclotron
resonance (FTICR) mass spectrometry. CID and SORI-CID of CoHNO+ yield competitive formation of
Co+, CoH+, and CoNO+. These results suggest that the barrier for 2 f 1 conversion is less than 45.5 kcal/
mol. Reactions with simple molecules were dominated by NO and HNO displacements. Potential energy
surface diagrams are presented to explain these displacement reactions. Surprisingly, CoHNO+ reacts with
methane by dehydrogenation to yield CoCH3NO+. Studies suggest CH3CoNO+ rather than Co(CH3NO)+ as
the structure for this ion.
Introduction
ClRu(H)(CH2) f ClRu(CH3)
(3)
Nitric oxide (NO) is the simplest, thermally stable paramag-
netic molecule known. The binding of nitric oxide to a metal
center imparts unique chemistry to both the metal center and
the nitrosyl ligand. Despite extensive experimental and theoreti-
cal studies carried out on metal-nitrosyl complexes,12 little is
known about hydride migration to the nitrosyl ligand. Reduction
of the NO ligand on the metal center by hydride migration can
yield two products, M-nitrosyl hydride (M-HNO) and M-
hydroxyimido (M-NOH), reaction 4. In a previous study, the
coordinated HNO ligand in reaction 5 has been characterized
by X-ray methods.13 Transformation of a NO ligand to a NOH
ligand has also been observed, reaction 6.14
Intramolecular hydrogen transfer from a transition metal
center to an unsaturated ligand, and the reverse process, hydride
migration from an organic ligand back to the metal center, are
ubiquitous processes in organometallic chemistry.1 Hydride
migration to an alkene or an alkyne is a key step in homoge-
neous hydrogenation and in many alkene and alkyne alkylation
2,3
processes. Hydride migrations to simple ligands such as C2H4
and CH2O4,5 have been extensively studied, and the reverse
process (i.e., â-H elimination) is often found to be facile. In
contrast to â-H elimination/insertion reactions, R-H elimination/
insertion reactions are not so common. For example, the hydride
migration to the carbonyl ligand (i.e., CO insertion) is known
only for a few cases in transition metal complexes.6,7 One of
the known examples is the formyl formation by carbonyl
insertion into the Rh-H bond, reaction 1.8
HsMsNO f MsHNO
f MsNOH
(4)
Os(NO)Cl(CO)(PPh3)2 + HCl f Os(HNO)Cl2(CO)(PPh3)2
(5)
Rh(OPE)(H) + CO f Rh(OEP)(CHO)
(1)
[Ru3(CO)10(µ2-NO)]- + H+ f [Ru3(CO)10(µ3-NOH)] (6)
Generally the CO insertion processes have been studied by
theory because of experimental limitations. The hydride migra-
tion to a carbene type ligand and the reverse R-H elimination
have also drawn the attention of many theoreticians.9 Ziegler
and co-workers studied the CH2 insertion into a Mn-H bond,
reaction 2, and found that it is much more exothermic than the
corresponding hydride migrations to CO or CS.10 Carter and
Goddard studied reaction 3 by theory and predicted an activation
barrier of 12 kcal/mol with an exothermicity of 7 kcal/mol for
the carbene insertion into the Ru-H bond.11
Hydride migration to the NO ligand for the transition metal
complexes has not been studied in the gas phase. Investigation
of reaction 4 in the gas phase provides a unique opportunity
for studying both â-H and R-H eliminations within an individual
organometallic system without complicating effects from other
ligands. Here, we study the [Co, H, N, O]+ system by using
Fourier transform ion cyclotron resonance (FTICR) mass
spectrometry. A comprehensive theoretical investigation of
different CoHNO+ isomers [HCo(NO)+ (1), Co(HNO)+ (2), and
Co(NOH)+ (3)] was performed by using density functional
theory (DFT).
HMn(CO)4(CH2) f Mn(CO)4CH3
(2)
H-Co+-NO Co+-HNO Co+-NOH
1
2
3
* Corresponding author. Department of Chemistry, North Dakota State
University, Fargo ND 58104.
Experimental Section
† Present address: Wyeth-Ayerst Research, Lederle Laboratories, 401
N. Middletown Road, Bldg. 222/1044, Pearl River, NY 10965.
‡ Deceased December 31, 1997.
All experiments were performed by using a Nicolet (now
Finnigan FT/MS, Madison, WI) FTMS-2000 Fourier transform
10.1021/jp991791y CCC: $18.00 © 1999 American Chemical Society
Published on Web 11/30/1999

CoHNO+ in the Gas Phase
J. Phys. Chem. A, Vol. 103, No. 50, 1999 10885
ion cyclotron resonance (FT-ICR) mass spectrometer equipped
with a dual trapping cell with a 3-T superconducting magnet.
A complete description of the instrument is given elsewhere.15
Chemicals were obtained commercially in high purity and
were used as supplied except for multiple freeze-pump-thaw
cycles to remove noncondensable gases from liquids. Methane
was introduced into the vacuum chamber via a Varian leak valve
at a static pressure of ∼1.5 × 10-7 Torr as the reagent gas for
chemical ionization. Ar was added to the cell to a total pressure
of ∼5.0 × 10-6 Torr and serves, with background methane, as
the collision gas for collision-induced dissociation (CID)16 and
sustained off-resonance irradiation (SORI)17 CID experiments.
Ar also facilitates ion thermalization for reaction studies.
The maximum translational energy acquired during CID by
the ions, Etr(max), is given in the laboratory frame and was
calculated by using the following equation applicable to a cubic
cell where E is the electric field amplitude, t is the duration of
the applied electric field, q is the ion charge, and Mion is the
mass of the irradiated ion.18,19
resulting in a (62111111|331211|3111|3) f [8s6p4d1f] contrac-
tion. The cc-pVTZ basis set24 was employed for H, N, and O
atoms. For Co+, B3LYP gives the correct ordering of atomic
states, but because of the bias toward 3dn over 4s13dn-1
5
configurations,25 the excitation energy to the F (4s13d7) state
is overestimated by 5.8 kcal/mol (calculated, 15.7; experimental,
9.9 kcal/mol). The DFT/HF hybrid method yields molecular
geometries which are qualitatively comparable with those
obtained at highly correlated levels.26 The computed thermo-
chemistry is reasonable; however, the bond dissociation energies
(BDEs) tend to be slightly overestimated. Therefore, the
accuracy of the DFT/HF approach for transition-metal com-
pounds is estimated to be within (10 kcal mol-1 for BDEs and
(5 kcal mol-1 for the relative energies of isomeric [Co, H, N,
O]+ species. All stationary points were characterized as minina
or first-order transition structures by evaluating the frequencies
and normal modes by using analytical first derivatives and the
computed force constant matrix. Corrections for zero point
energies are included, and different spin configurations have
also been considered for Co+.
E2q2t2
Etr(max) )
16Mion
Results and Discussion
Computational. Structures 1, 2, and 3 are considered for
CoHNO+ isomers. Density functional calculations were per-
formed on each structure and all were found to be true minima
on the potential energy surface. Table 1 contains the relative
energies of CoHNO+ isomers for the ground and low-lying
excited states.
CID fragment ion intensities are plotted as a fraction of the total
ion intensity at each kinetic energy. The duration of the
excitation pulse is 500 µs with the electric field amplitude varied
to control ion kinetic energy. A 50 ms delay follows ion
irradiation to allow for ion collisions and decomposition prior
to subsequent isolation or detection.
TABLE 1: Relative Energies of the Ground State and
Excited States of Structures 1, 2, and 3
In addition to conventional FTICR-CID, CID by using
sustained “off-resonance” irradiation (SORI)17 for ion activation
was also employed to determine the lowest energy fragmentation
pathways. For SORI-CID, ions are irradiated off-resonance for
500 ms. The maximum ion kinetic energy is calculated by using
the following equation:
structure
relative energy (kcal/mol)
1 (3A′)
1 (1A′)
1 (5A′′)
2 (3A′)
2 (5A′)
2 (1A′)
3 (3A′)
3 (5A′)
3 (1A′)
14.1
25.7
35.1
0
13.8
36.2
15.4
18.2
35.0
E2q2
2Mion(ω1 - ωc)2
Etr(max) )
where ω1 (rad s-1) is the excitation frequency and ωc is the
natural cyclotron frequency of the ion.
1. HCo(NO)+. Among the three reasonable isomers, we first
consider the H-Co+-NO structure, 1. The optimized geom-
etries of 1 (5A′′), 1 (3A′), and 1 (1A′) are shown in Figure 1.
Structure 1 (3A′) is the ground state, with 1 (1A′) and 1 (5A′′)
11.6 and 21.0 kcal/mol higher, respectively. The Co-N bond
lengths vary from 1.841 Å 1 (3A′), 2.174 Å in 1 (5A′′), and
1.615 Å in 1 (1A′). The extremely long Co-N distance in 1
(5A′′) suggests an ion-dipole complex. Unlike the Co-N bond,
the N-O bonds vary by no more than 0.03 Å from the N-O
bond length of 1.146 Å for free NO. In 1 (3A′), the Co-H bond
is bent toward the NO ligand with an H-Co-N bond angle of
139.0°. In 1 (1A′), the Co-H bond is bent further with an
H-Co-N bond angle of only 83.3°. Strengthening of the N-O
bond is observed as evidenced by the calculated N-O stretches
(2029 cm-1 in 1 (3A′) and 1 (5A′′) compared to 1977 cm-1
calculated for free NO). The dissociation energy to form CoH+
(4Φ) and NO (2π) from H-Co+-NO (3A′) is computed to be
34.6 kcal/mol and compares well with D°(Co+-NO) ) 40.0
kcal/mol27 calculated by Bauschlicher and co-workers.
For kinetic studies, neutral reagents were introduced into the
vacuum chamber through Varian leak valves. Pressure of
neutrals was measured by using a Bayard-Alpert type ion gauge
that was calibrated to determine the pressure gradient between
the reaction cell and ion gauge. Pressure was also corrected for
ionization sensitivities. The uncertainty in the absolute pressure
of neutral reagents is less than (30%. The uncertainty in
pressure is the largest contributor to errors in reaction rate
constants. Consequently, we assign an absolute error of (30%
for reaction rate constants, while relative reaction rate constants
are more reliable.
Computational Studies
Calculations were performed with the Gaussian 94 DFT
program package20 at the Purdue University Computer Center
(PUCC). Both the ground and excited states of the [Co, H, N,
O]+ system were investigated by using density functional theory.
For the computational studies, a hybrid of DFT and Hartree-
Fock (HF) was applied in which the Becke-3-LYP (B3LYP)
functional21 was used for the exchange correlation functional.
For cobalt, the (14s9p5d) primitive set of Wachters22 supple-
mented with one diffuse p-function, one diffuse d-function, and
one diffuse f-function according to Bauschlicher23 is used,
As typical for metal-nitrosyl complexes, both bent and linear
geometries of the Co+-NO component are observed for the
ground and excited states of HCo(NO)+. Clearly, Co+ binds
NO in a bent geometry in 1 (3A′) and 1 (5A′′), and in an almost
linear fashion in the low-lying excited state, 1 (1A′). With a

10886 J. Phys. Chem. A, Vol. 103, No. 50, 1999
Chen et al.
Figure 1. Optimized geometries of the quintet, triplet, and singlet states
of HCo(NO)+ (bond lengths in Å and bond angles in degrees).
bent geometry in 1 (3A′) and 1 (5A′′), the N and O atoms are
sp2 hybridized and the coordination of NO to the metal center
involves a net donation of one electron from NO to the metal.
However, the almost linear geometry in 1 (1A′) suggests that
the coordination of NO to Co+ involves a net donation of three
electrons. In this case, strengthening of the Co+-N bond can
be seen in the shortened Co-N bond length of 1.615 Å.
2. Co(HNO)+. The optimized geometries for the ground and
two excited states of 2 are shown in Figure 2. Structure 2 (3A′)
is the global minimum on the potential energy surface of the
[Co, H, N, O]+ system. Structure 2 (3A′) has Cs symmetry with
a Co-N bond length of 1.886 Å. The N-O bond distance is
1.201 Å and the N-H bond distance is 1.044 Å. The
experimental values, are, respectively, 1.212 and 1.063 Å for
the free HNO molecule.28 The HNO unit of 2 (3A′) has a bent
geometry with the H-N-O bond angle of 112.5°, compared
to 108.7° in free HNO by calculations. The N-O bond is
probably a double bond as evidenced from the longer N-O
bond versus free NO (1.156 Å).29 The dissociation energy to
form Co+ (3F) and singlet HNO (1A′) is predicted to be 45.5
kcal/mol with zero point energy corrections. As shown in Table
1, 2 (3A′) is 14.1 kcal/mol more stable than the R-H migration
product, 1 (3A′). Although the transition state connecting 1 (3A′)
and 2 (3A′) was not located, the activation barrier for the R-H
migration is believed to be less than 40 kcal/mol with respect
to 2 (3A′), Vide infra.
Figure 2. Optimized geometries of the quintet, triplet, and singlet states
of Co(HNO)+ (bond lengths in Å and bond angles in degrees).
that in the ground state. The two excited states have a side-on
bonding with NO of the HNO molecules. This interaction can
be described by the Dewar-Chatt-Duncanson model. Donation
of π-electron density from NO along with π-back-donation from
the metal into π* orbital result in weakening of the NO bond.
This NO bond weakening is manifested in an increase in the
NO bond lengths for the two excited states (Figure 2).
3. Co(NOH)+. The triplet state (3A′) is only 2.8 kcal/mol
lower in energy than the quintet state (5A′) of Co(NOH)+. Since
they are so close together, we cannot assign which is the ground
state. Structure 3 (3A′) and 3 (5A′) both have C1 symmetry and
similar structures. The Co-N bond distance of 3 (3A′) is 1.755
Å, and the N-O distance is 1.279 Å, compared to 1.810 and
1.289 Å in 3 (5A′). The Co-N-O bond angle in 3 (3A′) is
122.9° compared to 144.2° in 3 (5A′). The singlet state, 3 (1A′),
is 19.6 kcal/mol less stable than the triplet state. Structure 3
(1A′) has a Co-N-O bond angle of 153.2° and a much shorter
Co-N bond distance of 1.561 Å. The N-O bond is also
decreased to 1.266 Å.
The ground-state HNO isomer is about 24 kcal/mol more
stable than the corresponding NOH isomer.30 On the triplet
surface, the Co(NOH)+ structure is only 1.3 kcal/mol less stable
than the H-Co+-NO structure, and is 15.4 kcal/mol less stable
than the Co(HNO)+ structure. Conversion of 3 (3A′) to 1 (3A′)
involves a â-H transfer from the hydroxyimido ligand to the
metal center and is essentially thermoneutral. Unfortunately, the
transition structure connecting 3 (3A′) and 1 (3A′) was not
located. However, the CoHNO+ ions, once generated and
thermalized in the gas phase, are most likely structure 2,
The two excited states, 2 (5A′) and 2 (1A′), are 13.8 and 36.2
kcal/mol less stable than 2 (3A′), respectively. They both have
Cs symmetry and exhibit similar structural features. A much
smaller Co-N-O bond angle (73.8° in 2 (5A′) and 79.8° in 2
(1A′)) is observed, compared to 122.9° for the ground state. The
Co+-O distance for the two excited states is much smaller than

CoHNO+ in the Gas Phase
J. Phys. Chem. A, Vol. 103, No. 50, 1999 10887
Figure 4. Energy-resolved CID breakdown curve of CoH(CO)3NO+.
Figure 3. Optimized geometries of the quintet, triplet, and singlet states
of Co(NOH)+ (bond lengths in Å and bond angles in degrees).
Co(HNO)+. Consequently, 3 (3A′) to 1 (3A′) conversion is
probably unimportant.
Experimental. 1. Generation and Characterization of
CoHNO+. In light of the high volatility of Co(CO)3(NO) at room
temperature, the synthesis of CoHNO+ is relatively simple.
CoHNO+ can be generated from CID or sequential SORI-CID
of protonated cobalt tricarbonyl nitrosyl [HCo(CO)3NO+],
reaction 7 or 8. HCo(CO)3NO+ was generated by chemical
+
ionization of Co(CO)3NO with CH5
.
HCo(CO)3NO+ CID8 CoHNO+ + 3CO
(7)
(8)
SORI SORI SORI
HCo(CO)3NO+
8
8
-CO8 CoHNO+
-CO -CO
Several precursor ions, CoH(CO)x(NO)+ (X ) 1-3), were
probed by CID and SORI-CID. For example, the energy-
resolved CID plot of CoH(CO)3(NO)+ is shown in Figure 4. In
the low energy range, CID results in sequential elimination of
the carbonyls. At high collision energy, formation of CoHNO+
(three CO losses) dominates over Co(CO)+ (two CO and one
HNO losses). Previous studies on some mixed carbonyl-nitrosyl
complexes have shown that the NO ligand is bonded more
strongly to the metal center than the CO ligand. UV photolysis
of Co(CO)3(NO) yields CO loss to generate the primary
photoproduct, Co(CO)2(NO), which subsequently undergoes
another CO loss to yield Co(CO)(NO).31 These results are
consistent with the observed CID fragmentation patterns. CID
of CoH(CO)2(NO)+ is qualitatively similar to that for CoH-
(CO)3(NO)+.
Figure 5. Energy-resolved CID (a) and SORI-CID (b) breakdown
curves of CoH(CO)NO+.
Competition between CO and HNO losses is observed during
CID of CoH(CO)(NO)+. The energy-resolved CID and SORI-
CID plots of CoH(CO)(NO)+ are shown in Figure 5. SORI-
CID results clearly indicate that loss of CO to yield CoHNO+
is the lowest energy decomposition channel, again, consistent
with the weaker Co+-CO bond. The complete absence of NO

10888 J. Phys. Chem. A, Vol. 103, No. 50, 1999
Chen et al.
Figure 7. Potential energy surface diagram for the decomposition of
Co(HNO)+.
TABLE 2: Percentage Abundance of Primary Products,
Rate Constants (cm3 molecule-1 s-1), and Calculated
Reaction Efficiencies for the Reactions of Co(HNO)+ and
Co(DNO)+ with Selected Neutrals
displace-
ment
efficiency
%
ion
neutral
%
kobs
kcoll
CoHNO+ C2H2
NO
NO
NO
NO
100 4.5 × 10-10 9.5 × 10-10
100 4.3 × 10-10 9.2 × 10-10
100 5.8 × 10-10 1.9 × 10-9
100 4.9 × 10-10 1.3 × 10-9
70 6.8 × 10-10 1.8 × 10-9
30
47
47
31
38
38
C2D2
H2O
C4H6
NH3
NO
HNO
HNO
HNO
NO
CO
CH3CN NO
100 4.4 × 10-10 8.8 × 10-10
100 8.2 × 10-13 7.3 × 10-10
80 1.9 × 10-9 2.7 × 10-9
20
50
0.11
70
Figure 6. Energy-resolved CID (a) and SORI-CID (b) breakdown
HNO
HNO
NO
curves of CoHNO+.
C6H6
CoDNO+ C2H2
C2D2
100 1.6 × 10-9 1.2 × 10-9
100 4.1 × 10-10 9.5 × 10-10
100 4.2 × 10-10 9.2 × 10-10
130
43
46
loss, combined with formation of Co(CO)+ via HNO loss
suggests that HNO is an intact ligand rather than separate H
and NO bound to the metal; consequently, the structure is Co-
(CO)(HNO)+. Competitive CO/HNO losses in SORI-CID
suggest that D°(Co+-HNO) is ca. 1-2 kcal/mol greater than
D°(Co+-CO) ) 41.5 ( 1.6 kcal/mol.32
NO
TABLE 3: Percentage Abundance of Primary Products,
Rate Constants (cm3 molecule-1 s-1), and Calculated
Reaction Efficiencies for the Reactions of Co(HNO)+ and
Co(DNO)+ with CH4 and CD4
ion
neutral product
%
kobs
kcoll
efficiency
The structure of CoHNO+ was further investigated by CID
and SORI-CID, with reactions 9-11 observed.
CoHNO+ CH4 CoCH3NO+ 100 7.5 × 10-12 1.0 × 10-9 0.75%
CD4 CoDNO+
82 1.5 × 10-11 1.0 × 10-9 1.5%
83 2.9 × 10-11 1.0 × 10-9 2.9%
CoCD3NO+ 18
CoDNO+ CH4 CoHNO+
CoCH3NO+ 17
CD4 CoCD3NO+ 100
CID
CoHNO+ SORI8 Co+ + HNO
(9)
-
1.0 × 10-9
-
f CoNO+ + H‚
f CoH+ + NO
(10)
(11)
2. ReactiVity of CoHNO+. The gas-phase reactivity of
CoHNO+ was explored by observing reactions of this ion with
different neutral reagents. Product distributions and kinetic data
for reaction with small molecules such as ethyne, benzene,
methane, etc. are summarized in Tables 2 and 3.
Nitric Oxide and Carbon Monoxide. Nitric oxide (NO)
reacts efficiently (50% of the collision rate) with CoHNO+ by
HNO displacement. Isotopically labeled NO (15NO) reacts with
CoHNO+ in a similar fashion to yield Co15NO+ exclusively,
reaction 12. No simple nitric oxide displacement (i.e., exchange)
is observed, reaction 13. These results strongly support structure
2 or 3 for CoHNO+, with 2, Co(HNO)+, most likely due to
energetics. If CoHNO+ consists of structure 1, then we would
expect to observe significant NO displacement, reaction 13,
which is not observed.
CID of CoHNO+ (Figure 6) yields predominant Co+ (HNO loss)
over the entire kinetic energy range studied (0-35 eV).
Formation of CoH+ and CoNO+ are competitive fragmentation
channels. In contrast to CID, SORI-CID (Figure 6) yields CoH+
(NO loss) in significant amount suggesting that NO loss is the
lowest energy-demanding fragmentation channel. Competitive
H and HNO losses are also observed at low energy.
A potential energy surface diagram for the decomposition of
CoHNO+ is shown in Figure 7. DFT calculations predict that
Co(HNO)+ (2) is the most stable species on the triplet surface,
with HCo(NO)+ (1) 14 kcal/mol higher. Upon collisional
activation, 2 could rearrange to 1 where competitive ligand
losses yield CoH+ and CoNO+, respectively. Alternatively, 2
could decompose to yield Co+ and HNO, which requires about
45 kcal/mol. Under SORI-CID conditions, competitive forma-
tion of Co+, CoH+, and CoNO+ suggests an activation barrier
of about 40 kcal/mol for 2 f 1 conversion smaller than or very
close to the 45 kcal/mol energy asymptote for HNO loss. As a
result, rearrangement of 2 could be kinetically constrained under
CID conditions and CID of 2 would then give predominant Co+.
CoHNO+ + 15NO f Co15NO+ + HNO
N CoH15NO+ + NO
(12)
(13)
The potential energy surface diagram for reactions 12 and
13 is shown in Figure 8. While D°(CoHNO+-NO) is not

CoHNO+ in the Gas Phase
J. Phys. Chem. A, Vol. 103, No. 50, 1999 10889
SCHEME 1
SCHEME 2
SCHEME 3
Figure 8. Potential energy surface diagram for the reaction of Co-
(HNO)+ with 15NO.
Figure 9. Potential energy surface diagram for the reaction of Co-
(HNO)+ with C2H2.
known, D°(Co+-NO) ) 40.0 kcal/mol27 provides a reasonable
estimation. 15NO coordination results in formation of 4 with
roughly 40 kcal/mol excess energy. HNO elimination from 4
yields Co(15NO)+, reaction 12. Reaction 13 involves R-hydride
migration in structure 2 to Co+ to yield 5 with ∼26 kcal/mol
excess energy. The reverse process involving hydride migration
to 15NO leads to the formation of 6 at the same energy level as
4. NO elimination from 6 would yield Co(H15NO)+. Since
reaction 13 is thermoneutral, then there must be a prohibitive
barrier (>40 kcal/mol) for R-hydride migration to convert 4 to
5.
Carbon monoxide reacts slowly (efficiency ) 0.11%) with
Co(HNO)+ to yield HNO displacement (Table 2). This slow,
direct HNO displacement suggests that it may be slightly
endothermic. This result is consistent with SORI-CID of Co-
(CO)HNO+ where both HNO and CO losses were observed,
with CO loss dominating, Figure 5. Similar to the reaction with
nitric oxide, there may be a substantial barrier for hydride
migration from HNO to the metal center and from the metal
center to CO.
Figure 10. Potential energy surface diagram for the reaction of Co-
(HNO)+ with H2O.
Co(CHCH2)+ is ∼17 kcal/mol more stable than HCo(CHCH)+.
The activation barrier for HCo(CHCH)+ f Co(CHCH2)+
conversion, although not obtained, is estimated to be less than
20 kcal/mol (ca. 16 kcal/mol).34 Hence, 9 has 40 kcal/mol excess
internal energy and can eliminate NO to yield Co(CHCH2)+.
The energy for NO elimination from 9 must be less than the
barrier for hydrogen migrations (7 f 8 f 9) since no H/D
exchange was observed. This requires that D°(CoC2H3+-NO)
be <30 kcal/mol.
The process analogous to 8 to 9 conversion (Scheme 1) could
also occur for 1,3-butadiene (Scheme 2) and benzene (Scheme
3). The potential energy surface diagram for Co(HNO)+ with
1,3-butadiene should be qualitatively similar to that for Co-
(HNO)+ with ethyne where only NO displacement is observed.
In the case of benzene, HNO and NO displacements are both
exothermic because D°(Co+-C6H6) ) 61.1 ( 2.5 kcal/mol.35
That benzene yields only HNO displacement indicates that HNO
elimination is kinetically favored even though intramolecular
hydride migration to eliminate NO is energetically feasible.
Hence, there must be a prohibitive barrier to hydride migration
for the benzene collision complex.
Ethyne, 1,3-Butadiene, and Benzene. Reactions with ethyne,
1,3-butadiene, and benzene are discussed together even though
there is no fixed reaction pattern among them. Ethyne and 1,3-
butadiene yield NO displacement exclusively, while benzene
yields HNO displacement exclusively, Table 2.
A potential energy surface diagram for (CoHNO)+ with
ethyne is presented in Figure 9. Coordination of ethyne to Co-
(HNO)+ forms the collision complex, 7, which has roughly 37
kcal/mol excess internal energy using D°(Co+-C2H2) ) 37 kcal/
mol.33 Direct cleavage of the Co+-HNO bond requires ∼45
kcal/mol; consequently, it is roughly 8 kcal/mol endothermic.
However, 7 can undergo hydride migration to form 8, a nitrosyl
complex ∼14 kcal/mol higher in energy than 7. Direct NO
elimination from 8 to yield HCo(CHCH)+ is unlikely since it
should be ∼7 kcal/mol endothermic. However, 8 can rearrange
to 9, a vinyl-nitrosyl complex. DFT calculations predict that
Water, Ammonia, and Acetonitrile. Reactions with am-
monia and acetonitrile yield both HNO and NO displacements
with NO displacement dominating (Table 2). Reaction with
water, however, yields NO displacement exclusively.
A potential energy surface diagram for the reaction of Co-
(HNO)+ with H2O is presented in Figure 10. Coordination of
water to Co(HNO)+ forms the collision complex, 10, with 40

10890 J. Phys. Chem. A, Vol. 103, No. 50, 1999
Chen et al.
Figure 11. Potential energy surface diagram for the reaction of Co-
(HNO)+ with NH3.
kcal/mol excess internal energy using D°(Co+-H2O) ) 40.1
kcal/mol.36 HNO displacement is simply endothermic by ∼5
kcal/mol. However, 10 to 11 conversion involves R-hydride
migration from the HNO ligand to the metal center with a low
activation barrier. NO elimination from 11 yields HCo(H2O)+
and is estimated to be exothermic by ∼5 kcal/mol.
Figure 12. Pseudo-first-order plot of the reaction of Co(HNO)+ with
methane at 2.0 × 10-7 Torr.
A potential energy surface diagram for the reaction of Co-
(HNO)+ with ammonia is presented in Figure 11. The collision
complex, 12, is formed with 59 kcal/mol excess internal energy
using D°(Co+-NH3) ) 58.8 kcal/mol.36 Direct HNO elimina-
tion (30%) from 12 is exothermic by ∼14 kcal/mol. Structure
12 could rearrange to 13 with 45 kcal/mol excess internal
energy. NO elimination (70%) from 13 to yield HCo(NH3)+ is
estimated to be exothermic by 20 kcal/mol, 6 kcal/mol more
exothermic than the HNO displacement channel.
Reaction with acetonitrile yields 80% NO displacement and
20% HNO displacement. The potential energy surface for this
reaction should be qualitatively similar to that for ammonia.
Methane. Co(HNO)+ reacts with methane to form CoCH3-
NO+ (H2 loss), reaction 14.
CoHNO+ + CH4 f CoCH3NO+ + H2
(14)
Methane activation is rare for the first-row transition metal
complexes. Some second-row and third-row transition-metal ions
or ion complexes, such as Ta+ and RhCH2+, react with methane
by dehydrogenation.37 Many cobalt ion complexes, for example,
+
CoCH2+, CoCH3+, CoCF3 etc., are inert toward methane.38
Therefore, activation of methane by Co(HNO)+ is surprising.
Pseudo-first-order kinetics is observed for reaction 14, and a
kinetic plot for the reaction of Co(HNO)+ with CH4 is shown
in Figure 12. The primary product distributions and kinetic
information for the reactions of Co(HNO)+/Co(DNO)+ with
CH4/CD4 are summarized in Table 3. In addition to dehydro-
genation, H/D exchange was also observed. For example,
CoHNO+ reacts with CD4 to yield CoDNO+ predominantly
along with some CoCD3NO+, reactions 15 and 16.
Figure 13. Energy-resolved CID (a) and SORI-CID (b) breakdown
curves of CH3CoNO+.
CoHNO+ + CD4 f CoDNO+ + CD3H
f CoCD3NO+ + HD
(15)
53.3 ( 2 kcal/mol39 > D°(Co+-NO) ) 40.0 kcal/mol.27 These
results suggest that CoCH3NO+ has the structure CH3CoNO+,
containing two intact ligands, CH3 and NO. Although Co-
(HNO)+, 2, is predicted to be 14 kcal/mol more stable than HCo-
(NO)+, 1, CH3CoNO+ apparently is thermodynamically more
stable than the corresponding Co(CH3NO)+ structure. A com-
parison of relevant fundamental bond dissociation energies
supports this idea.40 In any case, the ion formed in reaction 14
has the structure, CH3CoNO+.
(16)
CoCH3NO+, generated in reaction 14, was subjected to CID
and SORI-CID for structural investigation. CID of CoCH3-
NO+ (Figure 13) yields CoCH3+ (NO loss), CoNO+ (CH3 loss),
and Co+ (CH3NO loss). SORI-CID of CoCH3NO+ (Figure 13)
+
yields CoCH3 exclusively, consistent with D°(Co+-CH3) )

CoHNO+ in the Gas Phase
J. Phys. Chem. A, Vol. 103, No. 50, 1999 10891
SCHEME 4
Summary
CoHNO+ is generated in the gas phase by CID or SORI-
CID of CoH(CO)3NO+, formed from chemical ionization of Co-
(CO)3NO with CH5+. Among the three isomeric structures,
HCo(NO)+ (1), Co(HNO)+ (2), and Co(NOH)+ (3), density
functional calculations found that 2 is the global minimum on
the potential energy surface with 1 and 3 14.1 and 15.4 kcal/
mol higher in energy. Reaction of CoHNO+ with isotopically
labeled NO (15NO) yields HNO displacement exclusively, which
strongly supports a structure with an intact HNO ligand, either
2 or 3. Owing to energetics, the structure is most likely 2. CID
of Co(HNO)+ yields Co+ predominantly along with CoNO+
and CoH+. In contrast, SORI-CID of Co(HNO)+ yields
significant CoH+. Although the transition state for the 2 f 1
conversion is not located, CID and SORI-CID of CoHNO+
suggest that the activation barrier for 2 f 1 conversion is less
than 45.5 kcal/mol.
SCHEME 5
The reactivity of Co(HNO)+ was explored by observing its
reactions with simple molecules. These reactions are dominated
by NO and HNO displacements. For example, Co(HNO)+ reacts
with ethyne, water, and 1,3-butadiene to yield NO displacement
exclusively. Reaction with nitric oxide, carbon monoxide, and
benzene yields HNO displacement exclusively. Reaction with
ammonia and acetonitrile yields both HNO and NO displace-
ments with NO displacement dominating. Potential energy
surface diagrams for these NO and HNO displacement reactions
are presented which consistently involve R-H migration from
the HNO ligand to the metal center. Co(HNO)+ reacts with
methane by dehydrogenation to yield CoCH3NO+. While CID
We now consider the mechanism for reaction 14, as well as
that for reactions 15 and 16. A proposed mechanism for reaction
14 is shown in Scheme 4. Several routes can lead to the
formation of CH3CoNO+. In route A, the collision complex,
14, may undergo dehydrogenation through the four-centered
transition state, 15, to yield CH3CoNO+. Routes B and C share
the same intermediate, 16, formed by an initial R-hydride
migration from the HNO ligand to the metal center. In route B,
dehydrogenation proceeds through the four-centered transition
state, 17, to yield CH3CoNO+. Route C, however, involves direct
C-H insertion yielding 18 followed by dehydrogenation to yield
CH3CoNO+.
+
of CoCH3NO+ yields competitive formation of CoCH3 and
CoNO+, SORI-CID of CoCH3NO+ yields CoCH3+ exclusively.
These results suggest CoCH3NO+ has structure CH3CoNO+
rather than Co(CH3NO)+. A mechanism is proposed for the
reaction with methane where a four-centered transition state is
involved for C-H bond activation as opposed to direct C-H
insertion.
Acknowledgment. This paper is dedicated to the memory
of Dr. Ben S. Freiser. Acknowledgment is made to the Division
of Chemical Sciences in the Office of Basic Energy Sciences
in the United States Department of Energy (Grant No. DE-
FG02-87ER13766) for supporting this research.
Using D°(Co+-CH4) ) 22.9 ( 0.7 kcal/mol,41 the collision
complex, 14, is formed with roughly 20 kcal/mol excess internal
energy. Both Co+ and CoH+ are inert with methane even though
reaction 17 is exothermic by ∼12 ( 4 kcal/mol.40
References and Notes
CoH+ + CH4 f CoCH3+ + H2
(17)
(1) (a) Collman, J. P.; Hegedus, L. S.; Norton, J. R.; Finke, R. G.
Principles and Applications of Organotransition Metal Chemistry; Univer-
sity Science Book: Mill Valley, CA, 1987. (b) Cotton, F. A.; Wilkinson,
G. In AdVanced Inorganic Chemistry; Wiley: New York, 1988.
(2) Roe, D. C. J. Am. Chem. Soc. 1983, 105, 7770.
(3) Sakaki, S.; Kato, H.; Kanai, H.; Tarama, K. Bull. Chem. Soc. Jpn.
1975, 48, 813.
(4) Dombek, B. D. J. Am. Chem. Soc. 1980, 102, 6855.
(5) Nakamura, S.; Morokuma, K. Abstracts, 33rd Symposium on
Organometallic Chemistry, Japan: Tokyo, Japan, October 1986; paper A109.
(6) (a) Herrmann, W. A. Angew. Chem., Int. Ed. Engl. 1982, 21, 117.
(b) Blackborow, J. R.; Daroda, R. J.; Wilkinson, G. Coord. Chem. ReV.
1982, 43, 17.
(7) (a) Grimmett, D. L.; Labinger, J. A.; Bonfiglio, J. N.; Masuo, S.
T.; Shearin, E.; Miller, J. S. Organometallics 1983, 2, 1325. (b) Masters,
C. AdV. Organomet. Chem. 1979, 17, 61. (c) Fahey, D. R. J. Am. Chem.
Soc. 1981, 103, 136.
(8) (a) Wayland, B. B.; Woods, B. A. J. Chem. Soc., Chem. Commun.
1981, 700. (b) Farnos, M. D.; Woods, B. A.; Wayland, B. B. J. Am. Chem.
Soc. 1986, 108, 3659.
These results suggest a prohibitive barrier associated with direct
C-H insertion for 14 f 15 or 16 f 18 conversion. However,
activation of methane through the four-centered transition state,
17, precludes direct C-H bond insertion. This type of four-
centered transition state has been proposed in previous studies.
Watson observed an interesting methane exchange reaction
proceeding through a four-centered transition state.42 In addition,
+
Jacobson has studied the dehydrogenation reaction of RhCH2
with methane proceeding through a four-centered transition
state.37b Therefore, we think the mechanism associated with
route B is more likely.
A proposed mechanism for reactions 15 and 16 is presented
in Scheme 5 and is based on pathway B in Scheme 4. The
collision complex 19 first converts to 20 by R-hydride migration.
HD loss may proceed by the four-centered transition state, 21.
H/D exchange may involve the four-centered transition state,
22, which yields 23 followed by methane reductive elimination.
H/D exchange through 22 dominates over HD loss through 21.
(9) Canestrari, M.; Green, M. L. H. J. Chem. Soc., Dalton Trans. 1982,
1789.
(10) Ziegler, T.; Versluis, L.; Tschinke, V. J. Am. Chem. Soc. 1986,
108, 612.
(11) Carter, E. A.; Goddard, W. A. J. Am. Chem. Soc. 1987, 109, 579.

10892 J. Phys. Chem. A, Vol. 103, No. 50, 1999
Chen et al.
(12) Richter-Addo, G. B.; Legzdins, P. In Metal Nitrosyls; Oxford
University Press: New York, 1992.
(13) Wilson, R. D.; Ibers, J. A. Inorg. Chem. 1979, 18, 336.
(14) Stevens, R. E.; Guettler, R. D.; Gladfelter, W. L. Inorg. Chem.
1990, 29, 451.
(15) Gord, J. R.; Freiser, B. S. Anal. Chim. Acta 1989, 225, 11.
(16) Freiser, B. S. Talanta 1985, 32, 697.
(17) Gauthier, J. W.; Trautman, T. R.; Jacobson, D. B. Anal. Chim. Acta
1991, 246, 211.
(18) Grosshans, P. B.; Marshall, A. G. Anal. Chem. 1991, 63, 2057.
(19) Freiser, B. S. Techniques for the Study of Ion Molecule Reactions;
Farrar, J. M., Saunders, W. H., Jr., Eds.; Willey: New York, 1988; p 61.
(20) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Gill, P. M. W.;
Johnson, B. G.; Robb, M. A.; Cheeseman, J. R.; Keith, T.; Petersson, G.
A.; Montgomery, J. A.; Raghavachari, K.; Al-Laham, M. A.; Zakrzewski,
V. G.; Ortiz, J. V.; Foresman, J. B.; Peng, C. Y.; Ayala, P. Y.; Chen, W.;
Wong, M. W.; Andres, J. L.; Replogle, E. S.; Gomperts, R.; Martin, R. L.;
Fox, D. J.; Binkley, J. S.; Defrees, D. J.; Baker, J.; Stewart, J. P.; Head-
Gordon, M.; Gonzalez, C.; Pople, J. A. Gaussian 94, Revision D.1 Gaussian,
Inc.: Pittsburgh, PA, 1995.
(27) Thomas, J. L. C.; Bauschlicher, C. W.; Hall, M. B. J. Phys. Chem.
A 1997, 101, 8530.
(28) Clement, M. J. Y.; Ramsay, D. A. Can. J. Phys. 1961, 39, 205.
(29) Huber, K. P.; Herzberg, G. Molecular Spectra and Molecular
Structure Constants of Diatomic Molecules; Van Nostrand: New York,
1979.
(30) Smolyar, A. E.; Zaretskii, N. P.; Charkin, O. P. Zh. Neorg. Khim.
1979, 24, 3160.
(31) Gadd, G. E.; Poliakoff, M.; Turner, J. J. Organometallics 1987, 6,
391.
(32) Goebel, S.; Haynes, C. L.; Khan, F. A.; Armentrout, P. B. J. Am.
Chem. Soc. 1995, 117, 6994.
(33) Sodupe, M.; Bauschlicher, C. W. J. Chem. Phys. 1991, 95, 8640.
(34) Chen, H.; Jacobson, D. B.; Freiser, B. S., paper in preparation.
(35) Meyer, F.; Khan, F. A.; Armentrout, P. B. J. Am. Chem. Soc. 1995,
117, 9740.
(36) Squires, R. R. J. Am. Chem. Soc. 1989, 111, 4101.
(37) (a) Buckner, S. W.; MacMahon, T. J.; Byrd, G. D.; Freiser, B. S.
Inorg. Chem. 1989, 28, 3511. (b) Jacobson, D. B.; Freiser, B. S. J. Am.
Chem. Soc. 1985, 107, 5870.
(21) (a) Becke, A. D. Phys. ReV. 1988, A38, 3098. (b) Becke, A. D. J.
Chem. Phys. 1993, 98, 1372. (c) Becke, A. D. J. Chem. Phys. 1993, 98,
5648.
(22) Wachters, A. J. H. J. Chem. Phys. 1970, 52, 1033.
(23) Bauschlicher, C. W.; Langhoff, S. R., Jr.; Barnes, L. A. J. Chem.
Phys. 1989, 91, 2399.
(38) (a) Jacobson, D. B.; Freiser, B. S. J. Am. Chem. Soc. 1985, 107,
4373. (b) Jacobson, D. B.; Freiser, B. S. J. Am. Chem. Soc. 1984, 106,
3891. (c) Chen, Q.; Lin, C.-Y.; Chen, H.; Freiser, B. S. Organometallics
1997, 16, 4020.
(39) van Koppen, P. A. M.; Kemper, P. R.; Bowers, M. T. J. Am. Chem.
Soc. 1993, 115, 5616.
(24) Dunning, T. H., Jr. J. Chem. Phys. 1989, 90, 1007.
(25) Ziegler, T.; Li, J. Can J. Chem. 1994, 72, 783.
(26) (a) Bauschlicher, C. W., Jr.; Patridge, H. Chem. Phys. Lett. 1995,
240, 533. (b) Ricca, A.; Bauschlicher, C. W., Jr. Chem. Phys. Lett. 1995,
245, 150. (c) Heinemann, C.; Hertwig, R. H.; Wesendrup, R.; Koch, W.;
Schwarz, H. J. Am. Chem. Soc. 1995, 117, 495. (d) Holthausen, M. C.;
Fiedler, A.; Schwarz, H.; Koch, W. J. Phys. Chem. 1996, 100, 6236.
(40) Calculations are based on thermochemical data taken from the
reference: Lias, S. G.; Bartmess, J. E.; Liebman, J. F.; Holmes, J. L.; Levin,
R. D.; Mallard, W. G. Gas-Phase Ion and Neutral Thermochemistry J. Phys.
Chem. Ref. Data. 1988, 17, Suppl. No. 1.
(41) Kemper, P. R.; Bushnell, J. E.; van Koppen, P. A. M.; Bowers, M.
T. J. Phys. Chem. 1993, 97, 1810.
(42) Watson, P. L. J. Am. Chem. Soc. 1983, 105, 6491.