Conversion of methane to methanol: Nickel, palladium, and platinum (d 9) cations as catalysts for the oxidation of methane by ozone at room temperature

Article abstract of DOI:10.1002/chem.201000627

The room-temperature chemical kinetics has been measured for the catalytic activity of Group 10 atomic cations in the oxidation of methane to methanol by ozone. Ni⁺ is observed to be the most efficient catalyst. The complete catalytic cycle with Ni⁺ is interpreted with a computed potential energy landscape and, in principle, has an infinite turnover number for the oxidation of methane, without poisoning side reactions. The somewhat lower catalytic activity of Pd⁺ is reported for the first time and also explored with DFT calculations. Pt⁺ is seen to be ineffective as a catalyst because of the observed failure of PtO⁺ to convert methane to methanol.

Full text of DOI:10.1002/chem.201000627

FULL PAPER
DOI: 10.1002/chem.201000627
Conversion of Methane to Methanol: Nickel, Palladium, and
9
Platinum (d ) Cations as Catalysts for the Oxidation of Methane by
Ozone at Room Temperature
[
a]
[a]
[a]
[b]
Andrea Bo zˇ ovi c´ , Stefan Feil, Gregory K. Koyanagi, Albert A. Viggiano,
[
c]
[c]
[c]
[a]
Xinhao Zhang, Maria Schlangen, Helmut Schwarz,* and Diethard K. Bohme*
Dedicated to Reinhart Ahlrichs on the occasion of his 70th birthday
Abstract: The room-temperature chemical kinetics has been measured for the cat-
alytic activity of Group 10 atomic cations in the oxidation of methane to methanol
+
by ozone. Ni is observed to be the most efficient catalyst. The complete catalytic
cycle with Ni is interpreted with a computed potential energy landscape and, in
Keywords: CꢀH activation · densi-
ty functional calculations · metha-
nol · oxidation · ozone · transition
metals
+
principle, has an infinite turnover number for the oxidation of methane, without
+
poisoning side reactions. The somewhat lower catalytic activity of Pd is reported
+
for the first time and also explored with DFT calculations. Pt is seen to be inef-
+
fective as a catalyst because of the observed failure of PtO to convert methane
to methanol.
ꢀ
1 [2]
Introduction
N O (OA(N )=40.0 kcalmol ) very exothermic at room
2 2
temperature, but the oxidation is impeded by activation
[3]
The economically important conversion of methane to
methanol continues to provide a chemical challenge and
energy barriers associated with CꢀH bond activation. This
has spawned a search for effective catalysts. Atomic metal
and metal cluster ions have received particular attention be-
cause they provide an extra electrostatic energy of interac-
tion that can reduce activation barriers and so lead to sub-
[1]
constitutes a “holy grail”. The CꢀH bond strength of
ꢀ
1
methane is substantial at 102.7ꢁ2.0 kcalmol , while the
thermodynamic affinity of methane for atomic oxygen, OA,
ꢀ
1 [2]
[4]
to form methanol is moderately high at 90.7 kcalmol .
This makes the direct gas-phase oxidation of methane by
stantial enhancements in reaction rates. Here we demon-
strate that atomic metal cations can catalyze the oxidation
of methane by transporting an O atom from ozone to the
methane molecule as illustrated in Scheme 1.
ꢀ
1 [2]
strong oxidants such as O (OA(O )=25.5 kcalmol ) and
3
2
[
a] Dr. A. Bo zˇ ovi c´ , Dr. S. Feil, G. K. Koyanagi, Prof. Dr. D. K. Bohme
Department of Chemistry, York University
Toronto, ON, M3J 1P3 (Canada)
Fax : (+1)416-736-5936
E-mail: dkbohme@yorku.ca
[
b] Dr. A. A. Viggiano
Air Force Research Laboratory, Space Vehicles Directorate
Scheme 1. Catalytic cycle for the homogeneous oxidation of methane
with ozone, mediated by atomic metal cations.
2
9 Randolph Rd., Hanscom Air Force Base, MA 01731-3010 (USA)
[
c] Dr. X. Zhang, Dr. M. Schlangen, Prof. Dr. H. Schwarz
Institut fꢀr Chemie, Technische Universitꢁt Berlin
Strasse des 17. Juni 135, 10623 Berlin (Germany)
Fax : (+49)30-314-21102
[5]
The thermodynamic window for catalysis by O-atom
transport from ozone to methane with atomic metal cations
E-mail: Helmut.Schwarz@mail.chem.tu-berlin.de
+
ꢀ1
requires that 25.5<OA(M )<90.7 kcalmol . Since the
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201000627.
+
upper boundary condition for OA(M ) favors late over
Chem. Eur. J. 2010, 16, 11605 – 11610
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
11605

early transition metal cations, we have chosen to investigate
9
+
the catalytic activity of d (Group 10) cations Ni (OA =
ꢀ
1
[6]
+
ꢀ1 [7]
6
3.2ꢁ1.2 kcalmol ),
Pd
(OA =33.6ꢁ2.5 kcalmol ),
+
ꢀ1 [8]
and Pt (OA =75.1ꢁ1.6 kcalmol ).
Previous measurements have provided some insight into
the kinetics of the second leg of the cycle shown in
+
+
Scheme 1 with NiO and PtO (but apparently not with
PdO ). NiO has been observed to produce Ni
CH OH exclusively with an efficiency of 20%, whereas
Pt + CH OH production was reported to proceed with
PtO on 25% of the reactive collisions along with a major
75%) production of PtCH2 + H O with a total reaction
rate coefficient of 1.05ꢃ10 cm molecule s . We have
investigated these two methane reactions, as well as that
with PdO , in a flow tube experiment and also report here
+
+
+
+
[9]
3
+
3
+
+
(
2
ꢀ
9
3
ꢀ1 ꢀ1 [10]
+
the first measurements of the kinetics of the first leg of the
catalytic cycles. Furthermore, we have computed the poten-
+
tial energy surfaces (PES) for the M /O /CH couples (M=
3
4
Ni, Pd) using DFT to assess the experimental results.
Results and Discussion
The reactions of the bare metal cations with ozone all were
ꢀ
10
3
ꢀ1 ꢀ1
observed to be fast, kꢂ4.0ꢃ10 cm molecule s , and
clean”, they produced only the metal oxide cation (see
“
[2]
Table 1). Since IE(O ) is quite high (12.53ꢁ0.08 eV), elec-
3
9
Table 1. Rate coefficients and efficiencies for reactions of d atomic-
+
+
metal cations M with O
with CH at room temperature using ESI/qQ/SIFT/QqQ mass spectrome-
try. Also included are the overall efficiencies Fcycle for the catalytic reduc-
3
and of their metal monoxide cations MO
4
tion of O
3
4
by CH .
+
[a]
[b]
+
[a]
[b]
[b]
[c]
M
k
(
/F
)
MO
k
/F
Products
F
cycle
[
b]
F
ox
(Fred)
+
+
+
Ni
4.0/0.41
(0.41)
Pd 4.8/
.55
(0.55)
NiO
1.7/0.16
ACHTUNGTRNE(NUNG 0.16)
Ni
0.066
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
+
+
+
PdO 0.083/
0.0083
Pd (78%)
0.0036
+
0
PdOCH
2
(15%)
+
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
A
H
U
G
R
N
U
G
PdO
7%)
+
PtCH
A
H
U
G
R
N
U
G
4
)
(
+
+
Pt
7.8/0.96
(0.96)
PtO
2
(98%)
(2%)
–
+
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
PtH
2
[
a] k is the total reaction rate coefficient measured in units of
ꢀ
10
3
ꢀ1 ꢀ1
1
0
cm molecule
s
with an uncertainty estimated to be ꢁ30%.
, where k is the collision rate coef-
ficient, calculated by using the algorithm of the modified variational tran-
[
b] The efficiency, F, is defined as k/k
c
c
[
13]
sition state/classical trajectory theory developed by Su and Chesnavich.
Subscripts “ox” and “red” indicate the oxidation of the bare metal ion
and the reduction of the metal oxide ion, respectively, by O-atom trans-
[
5]
fer. [c] Fcycle is defined as Fox ꢃ Fred
.
+
+
Figure 1. Reaction profiles measured for the reactions of NiO , PdO ,
+
and PtO with CH
4
in helium buffer gas at 295ꢁ3 K and 0.35ꢁ
tron transfer to these metal cations is endothermic by at
least 3.57 eV. The reaction kinetics for the three metal oxide
cations with methane were found to be much more diverse.
Ion profiles recorded with methane addition are shown in
Figure 1. Up to five different channels, those given in reac-
tion (1), were observed in the proportions indicated.
0
.01 Torr.
MO þ CH4 ! Mþ þ CH OH
þ
3
ð1aÞM ¼ Nið100%Þ; Pdð78%Þ
11606
www.chemeurj.org
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 11605 – 11610

Atomic Cation Catalysts for the Oxidation of Methane by Ozone
FULL PAPER
þ
!
!
!
!
MCH2 þ H O
ð1bÞM ¼ Ptð98%Þ
ð1cÞM ¼ Pdð15%Þ
ð1dÞM ¼ Pdð7%Þ
ð1eÞM ¼ Ptð2%Þ
+
2
the Berlin group that reported PtCH as a major product
2
+
[10]
ion, 75%, and bare Pt as a minor product ion, 25%.
þ
MOCH2 þ H2
However, much better agreement is found with the more
+
þ
recent results of new experiments in which PtO was pro-
MO ðCH Þ
4
1
95
+
duced by pulsing N O into mass-selected Pt ions that
2
þ
MH2 þ CH O
were generated by laser desorption/ionization from the solid
metal. After additional mass selection and subsequent ther-
2
+
Channel (1a), the second leg of the catalytic cycle under
investigation, is exothermic by ꢀ27.5, ꢀ57.1, and ꢀ15.6 kcal
mol for the reactions of NiO , PdO , and PtO , respec-
malization with pulsed Ar buffer gas, reactions of PtO with
ꢀ
9
leaked-in CH were studied at a pressure of 8ꢃ10 mbar
4
ꢀ
1
+
+
+
and short reaction times of 0.5 to 2 s. The branching ratios
[2,6–8]
+
+
+
+
tively.
Only NiO exhibited pure formation of the
for the product ions PtCH2 , Pt , and PtH were deter-
2
atomic cation along with methanol, as indicated previous-
mined to be 94, 3, and 3%, respectively.
[
9]
ly; we are able to report that it reacts fast with a 41% effi-
The gas-phase kinetics measured with our selected-ion
flow tube tandem mass spectrometer at room temperature
for the catalytic cycles leading to methanol formation from
ozone and methane are summarized in Table 1. The experi-
mental results indicate an interesting unambiguous trend in
+
ciency (see Table 1). PdO also produced mainly the atomic
cation (78%) with minor channels of H elimination (15%)
2
and methane addition (7%). The platinum oxide cation was
observed to react with methane at the collision limit, pre-
dominantly to dehydrate the intermediate (98%) and not to
9
catalytic efficiency going down the periodic table with the d
+
+
+
generate bare Pt . Formation of PtH and CH O was also
transition-metal cations. Ni appears to be best suited as a
2
2
+
observed but this channel is almost negligible, at 2%. The
catalyst with no ozone side reactions, Pt is not suitable as
+
PtCH2 reacts further with CH to eliminate hydrogen and
it is not regenerated in the reaction of its oxide cation with
4
[11]
+
+
to add methane.
0.1%) initially present in the spectrum does not increase
with the addition of CH into the reaction region (see
The very small signal at m/z of Pt
methane, and Pd is intermediate in its performance as a
(
catalyst.
A computed potential energy landscape for the catalysis
4
[12]
+
Figure 1).
mediated by M (M=Ni and Pd) is given in Figure 2. The
+
Our result for the products of the PtO reaction with
methane differs from the previous FT-ICR observations by
profile for the uncatalyzed oxidation of methane, overall
very exothermic (ꢀ70.8 and ꢀ65.2 kcalmol according to
ꢀ1
+
+
4 3
Figure 2. Potential-energy surfaces computed for the oxidation of CH with O in the absence (black lines) and presence of Ni (red line) or Pd (blue
line) calculated at B3LYP/def2-TZVP. The dashed lines represent the pathways for the species in their higher electronic states. For clarity, only the elec-
tronic states in the lowest energy are shown for metal-mediated reactions.
Chem. Eur. J. 2010, 16, 11605 – 11610
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
11607

D. K. Bohme, H. Schwarz et al.
[2]
+
+
the calculations and to published enthalpies of formation,
respectively), indicates a two-step reaction. Similar to si-
global minimum M ACHTUNGERTNNGNU( CH OH) (5), regenerating bare M
3
and thus closing the catalytic cycle. One of the challenges of
selective methane/methanol conversion is to prevent a fur-
ther oxidation of methanol into formaldehyde. As shown
[
14]
lanes and germanes, methane only forms a van der Waals
s
[21]
complex with O without stabilization. The structure n-TS1
3
corresponds to the concerted insertion of O into a CꢀH
in Figure 2, the methanol complex 5 may undergo oxidative
3
ꢀ
1
bond of methane with a barrier of 34.3 kcalmol . The hy-
addition of the OꢀH bond to the metal center (5 ! 6) fol-
s
+
drotrioxide intermediate CH OOOH decomposes into
lowed by a dehydrogenation to yield (H )M
A
H
U
G
R
N
U
G
3
2
2
s
methanol and singlet O , via a 1,3 H-shift transition state n-
TS2. Given the fact that singlet O is generated by the ozo-
nation of triethylsilane via triethylsilyl hydrotrioxide, the
spin inversion is not believed to occur along the reaction
path, even though the high-spin H-abstraction transition
is then easily evaporated from 7, giving the thermodynami-
+
cally more favorable product M ACHNUTGERTNNUNG( OCH ) . The barrier from
2
2
1
2
[15]
+
5 to M
ACHTUNGTNERNGNU( OCH ) is determined by TS6–7, which lies closely
2
+
below the entrance channel of MO /CH , amounts to
5.5 kcalmol for Pd and 1.1 kcalmol for Ni. Thus oxida-
4
ꢀ
1
ꢀ1
t
3
+
state n-TS1 and the triplet product ( O + CH OH) are
tion of 5 to generate Pd ACHTUNGTRENNUG( OCH ) takes place slowly for
2
+
2
3
+
lower in energy. So the uncatalyzed oxidation of methane by
ozone to form methanol is kinetically forbidden due to the
PdO /CH , whereas it hardly occurs for NiO /CH , in line
4 4
+
+
with the experimental observation that NiO /CH gives Ni
CH OH exclusively, whereas PdO /CH affords
Pd ACHTUNGERTNNUN(G OCH ) as side product.
2
4
ꢀ
1
+
overall barrier of 34.3 kcalmol on the singlet state surface
+
3
4
[
16]
+
and the expected rather inefficient spin-orbit coupling.
+
Since methane activation by PtO has been extensively
investigated both experimentally and computationally,
Our computations are in good qualitative agreement with
[10,17]
+
previous studies for the NiO /CH toward methanol by
Shiota and Yoshizawa.
4
+
[18a]
we focus here on Ni and Pd. The relevant computed M -
catalyzed pathways are presented in Figure 2. In contrast to
the uncatalyzed reaction, spin inversion can take place due
to the efficient spin-orbit coupling of the metal-mediated re-
Additional reaction pathways
were extensively investigated and these were compared with
+
the Pd /O /CH system. which has not been studied previ-
3
4
[
22]
ously. Our results explain the selective oxidation of meth-
[18]
+
action. For the sake of clarity, the ionic pathways (in blue
and red) are described as continuous surfaces consisting of
two spin states and only the electronic states in the lowest
energy are shown (see the Supporting Information for more
details of other reaction pathways). The activation barrier
for the O-atom insertion into methane is drastically lowered
ane to methanol by NiO ; formation of the side products
+
+
NiOH + CCH or Ni
A
H
U
G
R
N
U
G
3
2
2
+
ly unfavorable, and the side product Ni ACHTUNGNRTEUNG( OCH ) + H is
2 2
not kinetically accessible. In comparison, slight difference
+
on the PES gives different reaction pattern for Pd /O /CH .
3
4
While there exist some parallels for the “Group 10” metal
+
+
by the addition of a singly charged bare Ni or Pd cation.
The first step involves an O-atom transfer from ozone to
hydride chemistry in their reactions with CH , there are
4
subtle as well as fundamental differences, and as stated in a
+
+
[23]
M
on a doublet surface with the formation of MO in
different context, “the same and not the same” holds true
their lowest (quartet) states and O (triplet) with relative en-
once more.
2
ꢀ
1
+
+
+
ꢀ1
ergies of ꢀ38.3 and ꢀ33.8 kcalmol for Ni and Pd , re-
Our computed OA
the value of 33.6ꢁ2.5 kcalmol determined experimentally
from the kinetic energy onset of PdO by reaction of Pd
with O and in agreement with the, albeit highly uncertain,
ACHTUNGTRNENUNG( Pd ) of 47.8 kcalmol is higher than
ꢀ
1
spectively. The transfers proceed smoothly since the transi-
ꢀ
1
+
+
tion states TS1–2 are only a few kcalmol higher in energy
than the energies of the initial adducts 1.
2
ꢀ
1
The next elementary reaction starts with an entrance
value of 48ꢁ23 kcalmol determined from available ther-
+
4
[7]
channel of MO ( S) + CH . The similarity in the shapes of
mochemical data. The experimental value derived from
4
+
+
+
the PES of NiO and PdO also has been reported for
methane activation by neutral NiO and PdO. The chan-
nels toward M(OH) + CCH3 or M
the reaction of Pd with O might be underestimated due to
the neglect of a possible impulsive pairwise mechanism.
Furthermore, a higher value also is expected for the O-atom
affinity of Pd from a comparison with the experimental
2
[19]
[7]
+
[20]
+
[10]
A
H
U
G
E
N
N
(OCH ) + H2 are
2
+
found to be thermodynamically less favorable than the ob-
+
ꢀ1 [24]
served products (see the Supporting Information). The CH /
values for the S-atom affinities of Ni (56.7 kcalmol )
4
+
+
ꢀ1 [25]
CH OH conversion by MO proceeds through two steps,
and Pd (54.4 kcalmol ).
3
+
+
MO insertion into a CꢀH bond and reductive elimination
The experiments show that the reaction of PtO with
+
from HOMCH3 (4) to generate 5. The former step is rate-
CH prefers another pathway which leads to the formation
4
+
determining for the overall pathway. The small energy dif-
of Pt
A
H
U
G
R
N
U
G
2
2
[18]
ferences between TS3–4 and the entrance channel, that is,
been reported by Schwarz and co-workers. We have in-
cluded a comparison of the PES reported for this reaction
ꢀ
1
+
ꢀ1
+
4
.6 kcalmol for NiO and 3.4 kcalmol for PdO , indi-
+
cate a significant competition between the formation of 4
and the dissociation back to the reactants. So dissociation
back to reactants is predicted to be more pronounced for
with those computed here for the reactions of NiO and
+
PdO in the Supporting Information (Figure S2). In com-
paring the rate-determining steps, namely the first CꢀH
+
+
the reaction of PdO and this is qualitatively consistent
bond activation to give M(OH)
A
H
U
G
R
N
U
G
3
with the experiments which recorded an efficiency for the
PdO reaction lower than that for NiO (see Tables 1 and
S1). The desired product, methanol, dissociates from the
ative energies of the rate-determining transition states are
+
+
ꢀ1
+
+
+
ꢀ5.9, ꢀ4.6, and ꢀ3.4 kcalmol for PtO , NiO , and PdO ,
respectively. This is in line with the experimentally observed
11608
www.chemeurj.org
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 11605 – 11610

Atomic Cation Catalysts for the Oxidation of Methane by Ozone
FULL PAPER
+
+
order in overall reaction efficiency: PtO (1.0)> NiO
ion signals were monitored as a function of the flow of the reagent gas.
Primary rate coefficients were determined from the observed semi-loga-
rithmic decay of the primary reactant ion intensity using pseudo first
order kinetics an accuracy estimated to be ꢁ30%.
+
(
0.16)> PdO (0.0083).
The new Berlin measurements were performed by means of a Spectro-
Conclusion
[30]
spin CMS 47X FT-ICR mass spectrometer equipped with a Smalley-
[
31]
type cluster-ion source developed by Bondybey, Niedner-Schatteburg
[
32,33]
and co-workers.
In brief, the fundamental of a pulsed Nd:YAG laser
The room-temperature chemical kinetics has been measured
for the catalytic activity of d atomic cations in the oxidation
of methane to methanol by ozone. Ni is observed to be the
most efficient catalyst and exhibits, in principle, an infinite
turnover number for the oxidation of methane, without poi-
soning side reactions. The complete catalytic cycle with Ni
is interpreted with a computed potential energy landscape.
A somewhat lower catalytic activity is observed for Pd and
9
(l=1064 nm, Spectron Systems) is focused onto a rotating platinum
target to generate a metal plasma from which ion formation occurs by
synchronization of a He pulse and subsequent supersonic expansion.
After passing a skimmer, the ionic components of a molecular beam are
transferred into the analyzer cell where they are trapped in the field of a
+
+
+
7.05 T superconducting magnet. PtO was produced by pulsing-in N O
2
1
95
+
into Pt ions that were mass selected by means of the FERETS ion-
[34]
ejection technique.
+
reported for the first time. The DFT calculations account
for this reduced efficiency in terms of a higher-energy rate-
+
determining transition state. Pt is seen to be ineffective as
Computational Section
+
a catalyst because of the observed failure of PtO to con-
[
35]
The calculations were performed using the Gaussian03 package. Geo-
vert methane to methanol.
Both our experimental results and the interpretive poten-
tial energy landscapes in Figure 2 clearly point toward the
efficient catalytic action of gaseous Ni and Pd ions in the
conversion of methane to methanol in the presence of
ozone in the gas phase. The challenge now becomes the gen-
eration of such gaseous (or surface?) ions in devices that
provide practical applications.
[
36]
metries were optimized at the unrestricted UB3LYP level of theory
[
37]
with the def2-TZVP basis set. Frequency calculations were carried out
for all optimized structures with the same method to verify the nature of
the stationary points on the potential-energy surfaces (PESs) and to
obtain the zero-point energy corrections. The connections between transi-
tion states and corresponding minima were verified using the intrinsic re-
+
+
[
38]
action coordinate technique (IRC). Finally, all relative energies (cor-
ꢀ
1
rected for ZPE contributions) are reported in kcalmol
.
Acknowledgements
Experimental Section
D.K.B. acknowledges continued financial support from the National Re-
search Council (Canada), the Natural Science and Engineering Research
Council (Canada) and MDS SCIEX. As holder of a Canada Research
Chair in Physical Chemistry, D.K.B. also thanks the contributions of the
Canada Research Chair Program to this research. The research in Berlin
was supported by the Fonds der Chemischen Industrie and the Deutsche
Forschungsgemeinschaft “Cluster of Excellence: Unifying Conception in
Catalysis”. X.Z. acknowledges financial support from the Alexander von
Humboldt-Stiftung.
Experiments at York University were performed with a multi-source,
multi-sector selected-ion flow tube tandem mass spectrometer, symbol-
+
[
26]
+
+
ized as ESI ACHTUNGTRENNUNG( ICP)/qQ/SIFT/QqQ. The atomic cations Ni , Pd , and Pt
were generated from solutions of their salts by using an electrospray ioni-
zation (ESI) source with a microspray emitter needle. Nickel nitrate (Al-
drich, p. a. ꢂ 99.999%) and platinum chloride (Aldrich, p. a. ꢂ 99.99%)
were dissolved in a mix of HPLC grade methanol (Aldrich) and Milli-
pore (18.2 mW) water (20:80 v/v). Acetonitrile (Aldrich, p. a. ꢂ 99.8%)
was used as a pure solvent in the ESI experiments of palladium acetate.
The salt solutions were introduced directly into the electrospray source
ꢀ
1
at a flow rate of 7 mL min . A declustering potential of 200 V was ap-
plied to obtain sufficient amounts of bare metal ions. Metal oxide cations
were formed in the orifice-skimmer region of the atmosphere-vacuum in-
terface prior to the mass selection quadrupole by doping the interface
[
1] See, for example: a) W. Buijs, Top. Catal. 2003, 24, 73; b) G. A.
Olah, A. Goeppert, G. K. S. Prakash, Beyond Oil and Gas: The
Methanol Economy, Wiley-VCH, Weinheim, 2009; c) R. Palkovits,
C. v. Malotki, M. Baumgarten, K. Mꢀllen, C. Baltes, M. Antonietti,
P. Kuhn, J. Weber, A. Thomas, F. Schꢀth, ChemSusChem 2010, 3,
ꢀ
1
gas, normally pure nitrogen, with 50 mLmin of 4% ozone in O
2
. Ozone
was generated in a O3V–0, OREC Inc. model laboratory ozone genera-
2
77.
[
27]
tor. Oxygen from Liquid Carbonic was used directly.
[
2] NIST-JANAF Themochemical Tables, NIST Chemistry Web Book;
NIST Standard Reference Database, http://webbook.nist.gov/
chemistry/.
The bare atomic or atomic oxide cations were mass selected and injected
through a Venturi type aspirator into the flow tube that is flushed with
helium at 0.35ꢁ0.01 Torr. Before reaching the reaction region, the ions
[3] See, for example: F. J. Dillemuth, D. R. Skidmore, C. C. Schubert, J.
Phys. Chem. 1960, 64, 1496.
5
undergo multiple collisions with helium (ca. 4ꢃ10 ) to ensure thermaliza-
tion. The large number of collisions with the helium buffer gas atoms
should be sufficient to ensure that the atomic ions reach a translational
temperature equal to the tube temperature of 292ꢁ2 K prior to entering
the reaction region. The bare atomic cations were allowed to react with
ozone (4% ozone in oxygen), in a manner described in detail previous-
[4] a) D. K. Bçhme, H. Schwarz, Angew. Chem. 2005, 117, 2388; Angew.
Chem. Int. Ed. 2005, 44, 2336; b) R. A. J. OꢄHair, G. N. Khairallah,
J. Cluster Sci. 2004, 15, 331; c) G. E. Johnson, R. Mitric, V. Bonacic-
Koutecky, A. W. Castleman, Jr., Chem. Phys. Lett. 2009, 475, 1;
d) D. Schrçder, H. Schwarz, Proc. Natl. Acad. Sci. USA 2008, 105,
18114; e) G. E. Johnson, E. C. Tyo, A. W. Castleman, Jr., Proc. Natl.
Acad. Sci. USA 2008, 105, 18108.
[5] V. Blagojevic, G. Orlova, D. K. Bohme, J. Am. Chem. Soc. 2005,
127, 3545.
[6] Organometallic Ion Chemistry (Ed.: B. S. Freiser), Kluwer, Dor-
drecht, 1996.
[
28]
ly, whereas the atomic oxide cations were allowed to react with meth-
ane (Matheson, C.P. Grade (99.97%)) added into the reaction region and
then sampled along with product ions and analyzed in a triple quadru-
pole mass spectrometer. The contribution of reactions of the atomic ions
with oxygen to those with ozone was negligible (<1%); we have previ-
[
29]
ously studied the oxygen reactions separately.
Reactant and product
Chem. Eur. J. 2010, 16, 11605 – 11610
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
11609

D. K. Bohme, H. Schwarz et al.
[
[
[
7] Y.-M. Chen, P. B. Armentrout, J. Chem. Phys. 1995, 103, 618.
8] X-G. Zhang, P. B. Armentrout, J. Phys. Chem. A 2003, 107, 8904.
9] M. F. Ryan, H. Schwarz, unpublished results, cited by D. Schrçder,
H. Schwarz, Angew. Chem. 1995, 107, 2126; Angew. Chem. Int. Ed.
Engl. 1995, 34, 1973.
[28] S. Feil, G. K. Koyanagi, A. A. Viggiano, D. K. Bohme, J. Phys.
Chem. A 2007, 111, 13397.
[29] G. K. Koyanagi, D. Caraiman, V. Blagojevic, J. Phys. Chem. A 2002,
106, 4581.
[30] K. Eller, W. Zummack, H. Schwarz, J. Am. Chem. Soc. 1990, 112,
621.
[31] S. Maruyama, L. R. Anderson, R. E. Smalley, Rev. Sci. Instrum.
1990, 61, 3686.
[32] C. Berg, T. Schindler, M. Kantlehner, G. Niedner-Schatteburg, V. E.
Bondybey, Chem. Phys. 2000, 262, 143.
[33] M. Engeser, T. Weiske, D. Schrçder, H. Schwarz, J. Phys. Chem. A
2003, 107, 2855.
[
10] R. Wesendrup, D. Schrçder, H. Schwarz, Angew. Chem. 1994, 106,
232; Angew. Chem. Int. Ed. Engl. 1994, 33, 1174.
11] K. K. Irikura, J. L. Beauchamp, J. Phys. Chem. 1991, 95, 8344.
1
[
[
+
12] We speculate that the very small initial signal at m/z 196 (Pt ) may
+
+
arise from incomplete separation of Pt and PtO in the mass selec-
+
tion prior to the flow tube or come from the reaction of PtO with
background impurities (e.g., CO).
[
[
13] T. Su, W. J. Chesnavich, J. Chem. Phys. 1982, 76, 5183.
14] J. Cerkovnik, T. Tuttle, E. Kraka, N. Lendero, B. Plesni cˇ ar, D.
Cremer, J. Am. Chem. Soc. 2006, 128, 4090.
[34] R. A. Forbes, H. F. Laukien, J. Wronka, Int. J. Mass Spectrom. Ion
Processes 1988, 83, 23.
[35] Gaussian 03, Revision E.01, M. J. Frisch, G. W. Trucks, H. B. Schle-
gel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, J. A. Montgom-
ery, Jr., T. Vreven, K. N. Kudin, J. C. Burant, J. M. Millam, S. S.
Iyengar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani,
N. Rega, G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K.
Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, C. Adamo,
J. Maramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J.
Austin, R. Cammi, C. Pomelli, J. W. Ochterski, P. Y. Ayala, K. Moro-
kuma, G. A. Voth, P. Salvador, J. J. Dannenberg, V. G. Zakrewski, S.
Dapprich, A. D. Daniels, M. C. Strain, O. Farkas, D. K. Malick,
A. D. Rabuck, K. Raghavachari, J. B. Foresman, J. V. Ortiz, Q. Cui,
A. G. Baboul, S. Clifford, J. Cioslowski, B. B. Stefanov, G. Liu, A.
Liashenko, P. Piskorz, I. Komaromi, R. L. Martin, D. J. Fox, T.
Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara, M. Challa-
combe, P. M. W. Gill, B. Johnson, W. Chen, M. W. Wong, C. Gonza-
lez, J. A. Pole, Gaussian Inc., Wallingford CT, 2004.
[
15] E. J. Corey, M. M. Mehrotra, A. U. Khan, J. Am. Chem. Soc. 1986,
1
08, 2472.
[
16] a) J. N. Harvey, M. Aschi, Phys. Chem. Chem. Phys. 1999, 1, 5555;
b) J. N. Harvey, S. Grimme, M. Woeller, S. D. Peyerimhoff, D. Dano-
vich, S. Shaik, Chem. Phys. Lett. 2000, 322, 358; c) M. Tashiro, R.
Schinke, J. Chem. Phys. 2003, 119, 10186; d) A. M. Mebel, M. Haya-
shi, V. V. Kislov, S. H. Lin, J. Phys. Chem. A 2004, 108, 7983; e) Y.
Dede, X. Zhang, M. Schlangen, H. Schwarz, M.-H. Baik, J. Am.
Chem. Soc. 2009, 131, 12634.
[
[
17] M. Pavlov, M. R. A. Blomberg, P. E. M. Siegbahn, R. Wesendrup, C.
Heinemann, H. Schwarz, J. Phys. Chem. 1997, 101, 1567.
18] a) Y. Shiota, K. Yoshizawa, J. Am. Chem. Soc. 2000, 122, 12317;
b) J. N. Harvey, Phys. Chem. Chem. Phys. 2007, 9, 331.
[
[
19] D.-Y. Hwang, A. M. Mebel, J. Phys. Chem. A 2002, 106, 12072.
20] a) D. Schrçder, H. Schwarz, Angew. Chem. 1990, 102, 1468; Angew.
Chem. Int. Ed. Engl. 1990, 29, 1433; b) A. Bo zˇ ovi c´ , D. K. Bohme,
Phys. Chem. Chem. Phys. 2009, 11, 5940.
[36] a) A. D. Becke, J. Chem. Phys. 1993, 98, 5648; b) C. Lee, W. Yang,
R. G. Parr, Phys. Rev. B 1988, 37, 785.
[
21] For recent experimental and computational details on the selective
[37] a) F. Weigend, R. Ahlrichs, Phys. Chem. Chem. Phys. 2005, 7, 3297;
b) Effective core potential (ECP) for Pd: B. Metz, H. Stoll, M.
Dolg, J. Chem. Phys. 2000, 113, 2563; c) the basis sets were obtained
from the Gaussian Basis Set Library EMSL at https://bse.pnl.gov/
bse/portal; d) D. Feller, J. Comput. Chem. 1996, 17, 1571; e) K. L.
Schuchardt, B. T. Didier, T. Elsethagen, L. Sun, V. Gurumoorthi, J.
Chase, J. Li, T. L. Windus, J. Chem. Inf. Model 2007, 47, 1045.
[38] a) K. Fukui, J. Phys. Chem. 1976, 80, 4161; b) K. Fukui, Acc. Chem.
Res. 1981, 14, 363; c) C. Gonzalez, H. B. Schlegel, J. Chem. Phys.
1989, 90, 2154; d) C. Gonzalez, H. B. Schlegel, J. Phys. Chem. 1990,
94, 5523.
CꢀH or OꢀH bond activation of CH
3
OH, see: a) M. Schlangen, H.
Schwarz, Chem. Commum. 2010, 46, 1878; b) M. Schlangen, H.
Schwarz, ChemCatChem. 2010, 2, 799, and references therein.
22] J. Roithovꢅ, D. Schrçder, Chem. Rev. 2010, 110, 1170.
23] M. Schlangen, H. Schwarz, Angew. Chem. 2007, 119, 5711; Angew.
Chem. Int. Ed. 2007, 46, 5614.
[
[
[
24] C. Rue, P. B. Armentrout, I. Kretzschmar, D. Schrçder, H. Schwarz,
J. Phys. Chem. A 2002, 106, 9788.
[
[
25] P. B. Armentrout, I. Kretzschmar, Inorg. Chem. 2009, 48, 10371.
26] G. K. Koyanagi, V. Baranov, S. Tanner, J. Anichina, M. J. Y. Jarvis, S.
Feil, D. K. Bohme, Int. J. Mass Spectrom. 2007, 265, 295.
27] See, for example: A. I. Fernandez, A. J. Midey, T. M. Miller, A. A.
Viggiano, J. Phys. Chem. A 2004, 108, 9120.
[
Received: March 10, 2010
Published online: September 8, 2010
11610
www.chemeurj.org
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 11605 – 11610