Electronic structure of chromium oxides, CrOn- and CrOn (n=1-5) from photoelectron spectroscopy and density functional theory calculations

Article abstract of DOI:10.1063/1.1405438

A comprehensive theoretical and experimental investigation of CrO_n^- (n=1-5) and their corresponding neutral species was carried out. Photodetachment photoelectron spectra were obtained for the anions produced from a laser vaporization cluster source at various photon energies. Calculated adiabatic and vertical binding energies of CrO_n^- showed good agreement with the experimental values. It was shown that theoretical calculations are very helpful in interpretations of the photoelectron spectra of the anions, in understanding the electronic and geometrical structures of their ground and excited states, and chemical bonding for both the anions and the neutral species.

Full text of DOI:10.1063/1.1405438

Electronic structure of chromium oxides, CrO n − and CrO n (n=1–5) from
photoelectron spectroscopy and density functional theory calculations
G. L. Gutsev, P. Jena, Hua-Jin Zhai, and Lai-Sheng Wang
Citation: The Journal of Chemical Physics 115, 7935 (2001); doi: 10.1063/1.1405438
View online: http://dx.doi.org/10.1063/1.1405438
View Table of Contents: http://scitation.aip.org/content/aip/journal/jcp/115/17?ver=pdfcov
Published by the AIP Publishing
Articles you may be interested in
An all-electron density functional theory study of the structure and properties of the neutral and singly charged M
12 and M 13 clusters: M = Sc–Zn
J. Chem. Phys. 138, 164303 (2013); 10.1063/1.4799917
Photoelectron spectroscopy and density functional theory studies on the uridine homodimer radical anions
J. Chem. Phys. 137, 205101 (2012); 10.1063/1.4767053
Tungsten carbide revisited: New anion photoelectron spectrum and density functional theory calculations
J. Chem. Phys. 129, 114304 (2008); 10.1063/1.2976342
Evolution of the electronic properties of small Ni n − (n=1–100) clusters by photoelectron spectroscopy
J. Chem. Phys. 117, 9758 (2002); 10.1063/1.1519008
Electronic structure and chemical bonding of B 5 − and B 5 by photoelectron spectroscopy and ab initio
calculations
J. Chem. Phys. 117, 7917 (2002); 10.1063/1.1511184
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
129.127.200.132 On: Wed, 10 Dec 2014 12:24:19

JOURNAL OF CHEMICAL PHYSICS
VOLUME 115, NUMBER 17
1 NOVEMBER 2001
Electronic structure of chromium oxides, CrOÀn and CrOn „nÄ1–5… from
photoelectron spectroscopy and density functional theory calculations
G. L. Gutseva) and P. Jena
Physics Department, Virginia Commonwealth University, Richmond, Virginia 23284-2000
Hua-Jin Zhai and Lai-Sheng Wangb)
Department of Physics, Washington State University, Richland, Washington 99352
and W. R. Wiley Environmental Molecular Sciences Laboratory, Pacific Northwest National Laboratory,
Richland, Washington 99352
͑Received 5 July 2001; accepted 3 August 2001͒
The electronic structure of CrOϪn and CrOn (nϭ1–5) was investigated using anion photoelectron
spectroscopy and density functional theory. Photoelectron spectra of CrOϪn were obtained at several
photon energies and yielded electron affinities, vibrational and electronic structure information
about the neutral CrOn species. Density functional theory calculations were carried out for both the
neutrals and anions and were used to interpret the experimental spectra. Several low-lying electronic
states of CrO were observed and assigned from photodetachment of the CrOϪ ground state (6͚ϩ)
and an excited state (4 ), which is only 0.1 eV higher. The main spectral features of CrOϪ2 were
͟
interpreted based on a C2v CrOϪ2 (4B1). A very weak Cr͑O2)Ϫ isomer was also observed with lower
electron binding energies. Relatively simple and vibrationally resolved spectra were observed for
CrOϪ3 , which was determined to be D3h . The CrO3 neutral was calculated to be C3v with the Cr
atom slightly out of the plane of the three O atoms. The spectrum of CrOϪ4 revealed a very high
electron binding energy. Several isomers of CrOϪ4 were predicted and the ground state has a
distorted tetrahedral structure (C2) without any O–O bonding. Only one stable structure was
predicted for CrOϪ5 with a superoxo O2 bonded to a C3v CrO3. © 2001 American Institute of
Physics. ͓DOI: 10.1063/1.1405438͔
I. INTRODUCTION
thermodynamic data, but they have been difficult to obtain.
For example, the EA values of CrO3 deduced from previous
thermodynamic methods vary widely from 3.16 to 4.05 eV.6
On the other hand, PES, applied previously to many 3d TM
oxide clusters in our laboratory and Lineberger’s group,7–19
is based on direct measurements of the electron detachment
energies and allows more precise EA determinations. Since
PES of anions reflects transitions from anion states to the
corresponding neutral states, it provides information on the
electronic structure of both the anions and their neutral par-
ents. In our recent study of MnOϪn ,14 we found that theoret-
ical calculations are very helpful in the interpretation of the
photoelectron spectra, as well as in understanding the elec-
tronic and geometrical structures of both the anions and the
neutral species. A similar approach is applied here to the
CrOϪn series, especially for nϭ3–5, for which there is little
previous work.
Among the 3d TM oxide anions, theoretical studies have
been performed for ScOϪ2 ,20–22 TiOnϪ ,23 VOϪn ,24
MnOnϪ ,14,25,26 and FeOϪn ,27,28 and good agreement between
experimental data and the calculated EA’s of the correspond-
ing neutral species has generally been obtained. Detailed
studies of the electronic and geometrical structure of the neu-
tral and anionic 3d TM monoxides29 and dioxides30 have
been performed using DFT methods with different general-
ized gradient approximations for the exchange-correlation
functional. These studies showed that the BPW91 approach
͑an abbreviation for a combination of Becke’s exchange31
Transition metal ͑TM͒ oxides are technologically impor-
tant materials and have numerous applications, such as cata-
lysts in conversion processes of hydrocarbons.1,2 These ma-
terials exhibit different stoichiometries due to the existence
of various oxidation states of the TM metals. Oxides of chro-
mium, in particular, are widely used as coating materials in
magnetic recording devices.3,4 It was found that properties of
these devices depend strongly on the stoichiometry of the
chromium oxides. Investigations of the structural and elec-
tronic properties of TM oxide clusters with different metal to
oxygen ratios help to understand the physical and chemical
properties of the bulk materials and provide valuable insights
into their perspective applications. In the present article, we
report an investigation of the simplest chromium oxide spe-
cies, CrOϪn and CrOn (nϭ1–5), using anion photoelectron
spectroscopy ͑PES͒ and density functional theory ͑DFT͒ cal-
culations.
Among the small CrOn species, chromium trioxide
shows a very high electron affinity ͑EA͒, which indicates
that it should be a strong oxidizing agent.5 Species with large
EA’s have the ability to trap free electrons in excited gases
and form extremely stable negative ions. Accurate EA values
of the CrOn species can also be used for an improvement of
a͒
Current address: Mail Stop 230-3, NASA Ames Research Center, Moffett
Field, CA 94035.
b͒
Electronic mail: ls.wang@pnl.gov
0021-9606/2001/115(17)/7935/10/$18.00
7935
© 2001 American Institute of Physics
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
129.127.200.132 On: Wed, 10 Dec 2014 12:24:19

7936
J. Chem. Phys., Vol. 115, No. 17, 1 November 2001
Gutsev et al.
and Perdew–Wangs’ correlation32 functionals͒ provides simi-
lar results as the BLYP one ͑Becke’s exchange31 and Lee–
Yang–Parr correlation33͒, which are generally superior to
those obtained by a hybrid Hartree–Fock–DFT B3LYP
approach.34
The current work is aimed at understanding the elec-
tronic and geometrical structures of the monochromium ox-
ide species CrOn and CrOϪn (nϭ1–5) using a combined
experimental and theoretical study. Among these five oxide
species, CrOϪ and CrO2Ϫ have been previously studied by
PES at a lower photon energy.15,16 Several photon energies
are used in the current work because higher photon energies
are necessary for oxygen-rich species, which generally have
very high EA’s. Geometrical structures for both the neutral
and negatively charged clusters are optimized at the BPW91
level. Electron detachment energies were also calculated
from the anions and were used to help interpret the experi-
mental PES data.
the threshold value of 3ϫ10Ϫ4 au. Analytical harmonic fre-
quency calculations were performed subsequently to confirm
that the optimized geometrical configurations corresponded
to stationary states.
Density functional theory is valid for the lowest energy
states in each particular ͑spatial and spin͒ symmetry.40–43
Extensive optimizations, beginning with different spatial
configurations for each spin multiplicity for both the CrOn
and CrOϪn series, were performed to ensure that the ground-
state configurations were identified unambiguously. Com-
parison of the computed properties with experimental data
was made to further verify our theoretical results. Vertical
binding energies (Eb,i) or vertical detachment energies
͑VDE’s͒, which correspond to band maxima in photodetach-
ment spectra, are due to a sudden detachment of an electron
from an anion ͑A͒ to the corresponding neutral parent in
some particular state Ni . The quantity Eb,i can be evaluated
at the equilibrium ground-state geometry RA of the anion as:
E
b,iϭEtot N ,R ͒ϪE A,R ͒.
͑1͒
͑
͑
tot
II. EXPERIMENTAL AND COMPUTATIONAL DETAILS
A. Experimental method
i
A
A
With the present approach, only two transitions to neutral
states Nk (kϭ1 and 2͒ can be computed directly. If the
ground-state spin multiplicity of the anion is 2Sϩ1ϭM,
then we can compute the energies of electron detachment
resulting in the two lowest neutral states with spin multiplici-
ties 2Sϩ1ϭMϮ1.
In order to evaluate transition energies to other neutral
states, one could apply an approximate formula, which we
used to assign features in the PES spectra of the MnOnϪ
anions,14
Our experiments were performed with a magnetic bottle
time-of-flight photoelectron spectrometer, which has been
described in detail elsewhere.35,36 The apparatus is composed
of a laser vaporization source, a time-of-flight mass spec-
trometer and a magnetic bottle time-of-flight electron ana-
lyzer. Chromium atoms, generated from laser vaporization of
a pure Cr target, reacted with O2 seeded in a helium carrier
gas, producing a series of oxide clusters. The clusters en-
trained in the carrier gas underwent a supersonic expansion
and collimated by a skimmer. Anions from the collimated
beam were extracted perpendicularly into a time-of-flight
mass analyzer. Desired anions were mass-selected and sub-
sequently decelerated before crossing a detachment laser
beam. Both a Q-switched Nd:YAG laser ͑532 nm–2.331 eV,
355 nm–3.496 eV, and 266 nm–4.661 eV͒ and an ArF exci-
mer laser were used for photodetachment in the current ex-
periments. The energy resolution of our apparatus was better
than 30 meV for 1 eV electrons, as calibrated with CuϪ. Due
to the dependence of the resolution on electron kinetic ener-
gies, lower photon energy spectra yielded better resolved
data. But high photon energies were necessary to investigate
species with high EA’s, as well as yielding more excited
states of the neutral species.
A
A
E
b, jϭEtot Nkn ,RA͒ϪEtot A,R ͒ϩ ␧HOMO,nϪ␧ ͒, ͑2͒
͑ ͑ ͑
A
j,n
where k(ϭ1 or 2͒ denotes the neutral states accessible from
the anion ground state as in Eq. ͑1͒, jϾ2, n ͑␣ or ␤͒ denotes
the spin representation, and ␧
Kohn–Sham orbitals of the anion ground state, ␧
the eigenvalue to the highest occupied molecular orbital in
the n spin representation.
Adiabatic detachment energies (Aad), corresponding to
removal of an electron from the ground state of the anion to
states of the neutral at their respective equilibrium geom-
etries, are given by the differences in total energies of the
neutral and anion states:
A
j,n
are eigenvalues to the
A
is
HOMO,n
AadϭEtot N,R ͒ϩZ ϪE A,R ͒ϪZ ϭ⌬E ϩ⌬E ,
nuc
͑
͑
tot
N
N
A
A
el
B. Theoretical procedures
͑3͒
Our calculations were performed using an approach
where linear combinations of atomic orbitals constitute one-
electron molecular orbitals ͑MO-LCAO͒. The many-electron
potential was constructed using DFT with a generalized gra-
dient approximation for the exchange-correlation functional.
A combination of Becke’s exchange31 and Perdew–Wangs’
correlation32 functionals, referred to as BPW91, was used as
implemented in GAUSSIAN 94.37 For the atomic orbitals, we
where RN and RA denote ground-state equilibrium geom-
etries of a neutral molecule ͑N͒ and its anion ͑A͒, respec-
tively. Zero-point energies ͑Z͒ were computed within the har-
monic approximation. While values of Eb,i correspond to
band maxima in a PES spectrum, the Aad values are to be
deduced from the thresholds of PES bands or the 0–0 tran-
sitions of vibrationally resolved bands. The Aad Value ob-
tained for the ground-state transition corresponds to the EA
of the neutral molecule. Electron affinity measures the total
energy gain of a system due to attachment of an extra elec-
tron and represents an important thermodynamic quantity.
*
used the standard 6-311ϩG basis (Cr: 15s11p4d1f/
͓
10s7p4d1f and O: 13s6p1d/5s4p1d ) due to Wachters
͔
͓
͔
and Hay.38,39 Geometry optimizations were carried out by the
steepest descent method until the gradient forces fell below
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
129.127.200.132 On: Wed, 10 Dec 2014 12:24:19

J. Chem. Phys., Vol. 115, No. 17, 1 November 2001
Electronic structure of chromium oxides
7937
given. Generally, the current data agree well with the previ-
ous high resolution spectrum, with minor variations in rela-
tive intensities, due to the different experimental techniques
employed ͑the current experiment is angle-integrated and
measures electrons from all 4␲ solid angles͒. In particular,
the 2.13 eV feature (5⌬) is much stronger than the previous
spectrum, whereas the 1.82 eV vibrational progression
ϩ
(3͚ ) appears to be weaker but with stronger hot band
*
transitions ͑HB͒. Two more additional and relatively weak
features could also be identified from the current spectra at
355 and 266 nm: the feature at 2.64 eV (3
͟
) whose inten-
*
sity was enhanced in the 266 nm spectrum and a vibrational
3
*
progression starting at 3.03 eV ( ⌽ ). More transitions were
observed in the 266 nm ͓Fig. 1͑c͔͒, but could not be defi-
nitely identified due to limited spectral resolution and con-
gestion. The assignments were made on the basis of our the-
oretical calculations, as discussed later. The observed states
are labeled in Fig. 1 ͑ indicates transitions from an excited
*
anion͒ and compared with the theoretical calculations in
Table I. The observed detachment transitions and obtained
spectroscopic constants are given in Fig. 2.
The spectra of CrOϪ2 , shown in Fig. 3, were obtained at
three detachment energies as well. The 355 nm spectrum
͓Fig. 3͑a͔͒ was vibrationally resolved with the 0–0 transition
at 2.43 eV. There appears to be two vibrational progressions.
The main progression has a vibrational spacing of about 900
cmϪ1, and the other progression has a spacing of about 220
cmϪ1. Very weak signals (X ) were observed at ϳ1.5 eV.
Ј
These signals completely disappeared when the carrier gas
was seeded with N2O, instead of O2, as was the case for the
193 nm spectrum ͓Fig. 3͑c͔͒. The weak features were likely
due to a Cr͑O2)Ϫ isomer, as discussed later. The 266 nm
spectrum ͓Fig. 3͑b͔͒ revealed two more intense features ͑A
and B͒, compared to the 355 nm spectrum. The broadness of
these features indicates that there must be significant geom-
etry changes between the anions and the corresponding neu-
tral states or there might be overlapping and unresolved elec-
tronic transitions. At 193 nm, there seemed to be unresolved
features throughout the higher energy side beyond feature B.
The relative intensities of the first three detachment bands in
the 193 nm spectrum were changed compared to the 266 nm
spectrum.
FIG. 1. Photoelectron spectra of CrOϪ at ͑a͒ 532 nm ͑2.331 eV͒, ͑b͒ 355 nm
*
͑3.496 eV͒, and ͑c͒ 266 nm ͑4.661 eV͒. HB stands for hot bands. indicates
transitions from an excited state of CrOϪ.
In order to estimate the thermodynamic stability of the
chromium oxides, we also computed fragmentation energies
as the differences in total energies of fragments Fi formed in
a particular decay channel and the total energy of an initial
compound M:
The weak low-binding energy feature around 1.5 eV
completely disappeared in the 193 nm spectrum because an
N2O-seeded He carrier gas was used in this experiment.
Similar behavior was observed in the 266 and 355 nm spec-
tra taken under the same conditions. This feature is likely
due to a peroxo species Cr͑O2)Ϫ, which is consistent with
the results of our previous DFT calculations performed for
oxo, peroxo, and superoxo isomers of chromium dioxide and
its anion.30 The low EA of this isomer is also in agreement
with a general anticipation that a 3d-metal oxygen peroxo
complex M͑O2) has a lower EA than the corresponding 3d-
metal oxo complex OMO.7 A higher resolution spectrum of
CrOϪ2 obtained at 351 nm was previously reported by
Lineberger et al.16 However, the 351 nm photon energy al-
lowed only the first detachment band to be observed. The
current spectrum at 355 nm is consistent with the previous
data, although the previous spectrum showed a much broader
De M͒ϭ
E
F ͒ϪE M͒,
͑4͒
͑
͑
tot
͑
͚
i
tot
i
where the zero-point energy corrections are neglected be-
cause they are relatively small.
III. EXPERIMENTAL AND THEORETICAL RESULTS
A. Experimental results
Figure 1 shows the spectra of CrOϪ measured at three
photon energies. Well-resolved bands with vibrational struc-
tures were obtained. Many excited-state features were ob-
served in the 266 nm spectrum ͓Fig. 1͑c͔͒, but they could not
be clearly resolved and assigned beyond ϳ3.5 eV owing to
limited resolution and spectral congestion. A high resolution
spectrum of CrOϪ obtained at 351 nm was reported previ-
ously by Lineberger et al.,15 but no detailed assignment was
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
129.127.200.132 On: Wed, 10 Dec 2014 12:24:19

7938
J. Chem. Phys., Vol. 115, No. 17, 1 November 2001
Gutsev et al.
TABLE I. Experimental ͑Expt.͒ and calculated vertical electron detachment energies ͑eV͒ from the two lowest-
lying states of CrOϪ.
Electron
configuration
Final
state
BPW91
6-311ϩG
BPW91
BLYP
6-311ϩG
BLYP
Augmen.a
Augmen.a
Expt.
*
*
6
ϩ
2
CrOϪ͑ ͚ ͒9␴11␦24␲
→9␴11␦24␲
͟
⌬
ϩ
1.22
2.17
2.47
3.84
4.54
4.49
4.89
ϩ0.31b
0.87
1.97
2.65
3.22
1.24
2.22
2.41
3.89
4.56
4.52
4.91
ϩ0.34b
0.86
1.97
2.62
3.21
1.06
1.93
2.23
3.69
4.24
4.34
4.62
Ϫ0.05b
1.06
1.81
2.38
2.93
1.11
2.02
2.20
3.79
4.32
1.22Ϯ0.01
2.13Ϯ0.02
2.28Ϯ0.03
1
5
→9␴11␦14␲
2
5
→9␴01␦24␲
͚
2
5
→3␲39␴11␦24␲
͟
͟
ϩ
2
7
5
2
7
→8␴19␴11␦24␲
͚
͚
4.41
4.70
5
ϩ
CrOϪ(4
͟
)9␴21␦24␲
Ϫ0.003b
1.08
ϩ0.096b
1
1
5
→9␴11␦24␲
→9␴21␦24␲
→9␴11␦24␲
→9␴21␦14␲
͟
ϩ
1.12Ϯ0.01
1.82Ϯ0.02
2.64Ϯ0.03
3.03Ϯ0.03
0
1
1
3
͚
1.83
2.37
2.94
3
͟
3
⌽
3
or
͟
1
1
5
→8␴19␴21␦24␲
͟
Ϫ
4.16
4.83
4.21
4.87
4.22
4.87
4.26
4.90
5
→3␲39␴21␦24␲
͚
aBasis Cr 19s13p11d6f is obtained by adding exponents intermediate between those of the standard
͓
͔
*
6-311ϩG basis and five diffuse f-functions. For O, the standard 6-311ϩG(3df) basis is used.
bEnergy in eV relative to the
͚
state.
6
ϩ
Franck–Condon envelope for the X band, apparently due to
intense hot band transitions.
Both CrOϪ4 and CrOϪ5 were found to have very high
electron detachment energies and their spectra could only be
observed at 193 nm, as shown in Fig. 5. Two well-resolved
and intense features were observed for CrOϪ4 ͓Fig. 5͑a͔͒.
There were also weak features tailing the lower energy side
Figure 4 shows the spectra of CrOϪ3 taken at two photon
energies. The 266 nm spectrum ͓Fig. 4͑a͔͒ exhibited a well-
resolved vibrational progression with an average spacing of
890 cmϪ1. Two more well-separated and well-resolved bands
͑A and B͒ were observed in the 193 nm spectrum ͓Fig. 4͑b͔͒.
The weak features might be due to vibrational structures or
possible overlapping transitions as revealed from our theo-
retical results. The weak feature a to the left of the A feature
was due to a discrete detachment transition. As discussed
later, several transitions might have contributed to the B fea-
ture.
FIG. 3. Photoelectron spectra of CrOϪ2 at ͑a͒ 355 nm, ͑b͒ 266 nm, and ͑c͒
193 nm ͑6.424 eV͒. The vertical lines in ͑a͒ indicate resolved vibrational
progressions.
FIG. 2. The observed detachment transitions from the two anionic states of
CrOϪ and the spectroscopic constants of CrO.
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
129.127.200.132 On: Wed, 10 Dec 2014 12:24:19

J. Chem. Phys., Vol. 115, No. 17, 1 November 2001
Electronic structure of chromium oxides
7939
TABLE II. Observed vertical electron detachment energies ͑VDE͒ and spec-
troscopic constants for CrOn and CrOϪn (nϭ2–5) compared with theoreti-
cal results ͑Theor.͒ computed according to Eqs. ͑1͒ and ͑2͒.
a
VDE ͑eV͒
Vib. freq.a
State
Expt.
Theor.b
͑cmϪ1
͒
X5B2
Cr͑O2͒
CrO2
1.50͑6͒ 1.37
2.43͑2͒ 2.35
3.41͑5͒ 2.73
4.25͑5͒ 4.26
d
900͑80͒c; 220͑40͒
X3B1
A3A1 , 3B1
B5A2 , 3A2
5A1 , 3B1 , 5B1
X1A1Ј
Ͼ4.5
4.66
3.66͑2͒ 3.61
4.61͑2͒ 4.42
4.70͑2͒ 4.61
5.43͑2͒ 5.57, 5.75
5.07͑4͒ 4.95
6.04͑3͒ 6.13
e
CrO3
890͑60͒
a1A2Ј
A3A1Ј
1A2Љ , 3A2Љ
X3A2
CrO4
CrO5
A1A
X3A
4.7͑1͒
4.69
Љ
Ј
A1A
5.33͑4͒ 5.17
aThe numbers in the parentheses represent uncertainties in the last digits.
bTheoretical data for CrO2 are from Ref. 30.
cThe symmetric stretching mode (␯1).
dThe bending mode (␯2).
eFinal states are labeled according to the D3h point group of the CrOϪ3 anion.
fact due to direct detachment transitions. The spectrum of
CrOϪ5 revealed a very broad band ͑X͒ at lower binding ener-
gies and two features ͑A and B͒ at higher binding energies.
Our experimental results for CrOϪn (nϭ2–5) are summa-
rized in Table II and compared to our theoretical results, as
discussed below.
FIG. 4. Photoelectron spectra of CrOϪ3 at ͑a͒ 266 nm and ͑b͒ 193 nm. The
vertical lines in ͑a͒ indicate the resolved vibrational progression.
of the X band and between the X and A bands. The low-
energy tail could be due to minor isomers presented in the
anion beam. As shown later from our theoretical calcula-
tions, the weak features between the X and A bands were in
B. Theoretical results
Optimized ground-state geometries of CrO, CrO2, CrO3,
and CrO5 and their anions are presented in Fig. 6. Many
low-lying isomers and excited states of CrO4 and CrOϪ4 are
found and are shown in Fig. 7. Theoretical investigations of
CrOϪ2 with several isomers were reported in more detail
previously,30 and Fig. 6 only displays the oxo ground state
for CrO2 (3B1) and CrO2Ϫ (4B1). The structures of low-lying
FIG. 6. Ground-state geometrical configurations of CrO, CrO2, CrO3, and
CrO5 along with their anions.
FIG. 5. Photoelectron spectra of ͑a͒ CrOϪ4 and ͑b͒ CrO5Ϫ at 193 nm.
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
129.127.200.132 On: Wed, 10 Dec 2014 12:24:19

7940
J. Chem. Phys., Vol. 115, No. 17, 1 November 2001
Gutsev et al.
C2 symmetry with two opposite pairs of oxygen atoms
slightly rotated around the C2 axis ͑Fig. 7͒. The upper
ЄOCrO bond angle is 67°, which is intermediate between
typical values of peroxo ͑40°–45°͒ and oxo ͑100°–120°͒
bond angles. A peroxo isomer is higher in energy by only
0.12 eV. The CrOϪ4 anion has at least three states, which are
stable thermodynamically relative to detachment of the extra
electron. The ground state of CrOϪ4 is B1 and has an oxo
2
form of C2v symmetry. This geometrical configuration is
similar to the isoelectronic MnO4.25 A peroxo doublet (2A1)
and a half-peroxo quartet (4B1) isomer of CrO4Ϫ were found
to be above the ground state by 1.04 and 2.27 eV, respec-
tively.
An extensive search for a thermodynamically stable state
of CrO resulted in only one configuration (3A ) of the su-
Љ
5
peroxo type, as shown in Fig. 6. The ground state (2A ) of
Љ
the CrOϪ5 anion has the same superoxo shape and is below
the ground state of the neutral parent by 4.41 eV. We found
another isomer for CrOϪ5 , which also has a doublet spin mul-
tiplicity with a double peroxo geometrical shape. This isomer
is above the ground state anion by 1.81 eV.
IV. DISCUSSION AND COMPARISONS OF
EXPERIMENTAL AND THEORETICAL RESULTS
A. CrO and CrOÀ
FIG. 7. Geometric configurations of the ground and lowest excited states of
CrO4 and CrOϪ4 .
5
The ground state of CrO is known to be
͟ with an
electron configuration, 1␦24␲19␴1.46,47 Upon formation of
the anion, the extra electron can enter either the 9␴ MO to
4
6
ϩ
give a
͟
state or the 4␲ MO to give a
͚
state. Our
peroxo and superoxo isomers of CrO2 and CrOϪ2 are not
shown. In the current theoretical work, we focus on the mon-
oxide and the higher oxides starting with CrO3.
theoretical results, as given in Table I, suggest that the two
states are nearly degenerate in total energy. Direct computa-
tions of VDE’s from the two anions to the neutral CrO
ground state were performed according to Eq. ͑1͒ and VDE’s
to excited states of CrO were computed according to Eq. ͑2͒.
Our results obtained at the BPW91 and BLYP levels of
The ground state of CrO3 (1A1) is closed shell with C3v
symmetry and an out-of-plane angle of 15.6° ͑Fig. 6͒. Previ-
ous calculations using a local-spin-density-approximation
͑LSDA͒ approach and the LSDA with nonlocal corrections
*
theory with the use of two basis sets ͑the 6-311ϩG basis
and an augmented 6-311ϩG one͒, are given in Table I. The
44
͑NLSDA͒ yielded similar geometric structures with the
*
Cr–O bond length being 1.59 and 1.62 Å at the LSDA and
NLSDA levels, respectively. Our computed harmonic fre-
quency for the symmetric stretching mode (a1) of 959.6
cmϪ1 is in good agreement with the value of 890 cmϪ1 mea-
sured from the PES spectrum of CrOϪ3 ͑Table II͒. A value of
968.4 cmϪ1 was obtained for this mode by Andrews et al.45
from infrared spectra of CrO3 trapped in inert matrices. The
lowest-energy isomers of CrO3 were found to be well-
separated in total energy from the ground state: a triplet oxo
state (3A2) by 1.22 eV and a singlet peroxo state (1A1) by
3.15 eV. The ground state of CrOϪ3 has D3h symmetry with a
2A1Ј electronic state. This state is quite floppy and its out-of-
plane vibrational mode (a2Љ) is only 43 cmϪ1, indicating a
rather flat potential-energy curve for the out-of-plane motion.
We did not find any stable isomer for CrOϪ3 below the de-
tachment continuum, consistent with the experiment ͑Fig. 4͒,
where, unlike the other oxides, no trace of isomers was ob-
served.
variations at the different levels of theory are rather small
and the ordering of the energy levels obtained is consistent.
The strong peak at 1.22 eV is assigned to be the transi-
tion from the ground state of the anion to the ground state of
the neutral. This assignment is consistent with that of
Lineberger et al.15 The obtained EA is also in excellent
agreement with the previous value of 1.221 eV. The peaks at
lower binding energies appeared as hot band transitions.
However, as was pointed out previously by Lineberger
et al.,15 the peak at 1.12 eV is too intense to be due to a hot
band transition. It was assigned to be from an excited state of
the anion. Comparing the current spectra to that of
Lineberger et al., we note that the excited anion state was
even more populated in the current experiment. The obser-
vation of excited anions was consistent with our theoretical
results, which indicated that there are two nearly degenerate
anion states depending on which orbitals the extra electron
enters. Experimentally, the two anionic states differ by 0.096
eV, according to the above assignment. Such a small energy
difference could not be definitively distinguished at the cur-
rent levels of theory, as one can see from Table I. Lineberger
For both CrO4 and CrOϪ4 , we found several stable iso-
mers closely spaced in energy ͑Fig. 7͒. The ground state of
neutral CrO4 has a rather unusual geometrical shape: it has
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
129.127.200.132 On: Wed, 10 Dec 2014 12:24:19

J. Chem. Phys., Vol. 115, No. 17, 1 November 2001
Electronic structure of chromium oxides
7941
et al. assumed that the ground state of CrOϪ was 4
not provide any spectral assignments.
͟
, but did
lent MOs as given in Table II. It was found that the VDE’s
for detachment from the 4b1(3A1) and 10a1(3B1) orbitals
are very close ͑ϳ2.73 eV͒ and they were assigned to corre-
spond to the broad feature A observed around 3.4 eV. The
193 nm spectrum showed almost continuous signals above
feature B, consistent with the presence of many closely
spaced transitions due to detachments of electrons from the
inner valent orbitals.
The two strongest excited-state transitions correspond to
the feature at 2.13 eV and the vibrational progression starting
at 2.28 eV. The binding energies of these two states are in
good agreement with the computed vertical binding energies
5
5
ϩ
6
ϩ
for the ⌬ and ͚ states from the ͚ anion, respectively,
as shown in Table I. The features observed at 1.82, 2.64, and
3.03 eV are in good agreement with the three low-lying
4
states derived from the
͟ state of the anion ͑Table I͒. The
C. CrO3 and CrOÀ3
weak intensities of these transitions suggested that they are
more likely to be from the excited anion. Thus, the combined
experimental and theoretical results allowed us to conclude
that the ͚ state should be the ground state of the anion.
Figure 2 shows the spectroscopic transitions and the obtained
spectroscopic constants. The more accurate vibrational fre-
quencies for the corresponding transitions observed by
Lineberger et al. are also given in Fig. 2.
The ground state of neutral CrO3 has C3v symmetry
͑Fig. 6͒ with
a closed-shell electronic configuration,
6
ϩ
9a127e41a22(10a1)0 corresponding to a 1A1 state, whereas
the CrOϪ3 anion has D3h symmetry and an electronic con-
figuration (1e )4(3aЉ)2(6e )4(1aЈ)2(7aЈ)1 which corre-
Љ
Ј
2
2
1
2
sponds to a A1Ј state. Direct computations of detachment
energies according to Eq. ͑1͒ to the singlet and triplet states
provide values of 3.61 and 4.61 eV, respectively, in excellent
agreement with the first two PES features at 3.66 eV (X) and
4.70 eV (A) ͑Fig. 4͒. The triplet neutral state ͑at the equilib-
rium anion geometry͒ has the electronic configuration
. . . (1a2Ј)1(7aЈ1)1, which could be identified as correspond-
ing to a 3A2 state of the neutral after the geometry relaxation
takes place. The latter state is the lowest in total energy
among the optimized triplet states of neutral CrO3 and is
above the ground 1A1 state by 1.22 eV. The broad vibrational
progression observed for the X band indicates that there is a
large geometry change between the ground states of the an-
ion and the neutral. The 890 cmϪ1 vibrational spacing is in
reasonable agreement with the calculated symmetric stretch-
ing frequency of 959.6 cmϪ1. The bond length change be-
tween the anion and the neutral ground state ͑Fig. 6͒ is also
consistent with the observed vibrational progression. The
change from D3h CrO3Ϫ to C3v CrO3 suggests that the out-
of-plane bending mode should also be active in the X band.
However, our calculations gave a very low bending fre-
quency ͑43 cmϪ1͒, which is too small to be resolved under
our current experimental conditions.
The vibrational frequency (␻e) of CrO computed29 at
various levels of theory is 911 cmϪ1 ͑BPW91͒, 903 cmϪ1
͑BLYP͒, and 870 cmϪ1 ͑B3LYP͒, which are all in good
agreement with the average vibrational spacing of 920 cmϪ1
obtained from the present spectra, as well as with the previ-
ous more accurate gas-phase experimental value46 of 898
cmϪ1. Theoretical frequencies29 were also computed for the
two anionic states: 827 cmϪ1 ͑BPW91͒, 810 cmϪ1 ͑BLYP͒,
and 815 cmϪ1 ͑B3LYP͒ for
876 cmϪ1 ͑BLYP͒, and 861 cmϪ1 ͑B3LYP͒ for 4
are also in reasonable agreement with the experimental value
of 780Ϯ80 cmϪ1
͚
and 879 cmϪ1 ͑BPW91͒,
6
ϩ
͟
. The latter
.
B. CrO2 and CrOÀ2
The ground state of CrOϪ2 has a C2v structure and
a quartet electronic state (4B1) with an electronic config-
uration, 9a213b126b2210a114b1111a11. Detachment of the elec-
tron from the 11a1 HOMO leads to neutral CrO2 in its
ground state (3B1) with an electronic configuration,
9a213b126b2210a114b11. The VDE corresponding to this process
computed according to Eq. ͑1͒ is 2.35 eV, which is in good
agreement with the first PES feature (X) of CrOϪ2 at 2.43 eV
͑Fig. 3, Table II͒. Although both the ground states of CrO2Ϫ
and CrO2 have C2v symmetry, the neutral has a shorter Cr–O
bond length and smaller ЄO–Cr–O angle, as shown in Fig.
6. These geometry changes between the anion and the neutral
are consistent with the PES spectrum ͓Fig. 3͑a͔͒, which re-
vealed vibrational excitations in two modes. The 900 cmϪ1
mode should be due to the Cr–O symmetric stretching and
the 220 cmϪ1 mode the bending mode. These vibrational
frequencies and assignments are consistent with those by
Lineberger et al.16
Using Eq. ͑2͒ with two reference energies E (A,1AЈ)
͓
tot
1
ϪE (A,2AЈ) ϭ3.61 eV and E (A,3AЈ)ϪE (A,2AЈ)
͔
͓
͔
1
tot
tot
tot
1
2
ϭ4.61 eV, we obtained a set of energies corresponding to
detachment of an electron with subsequent formation of sin-
glet and triplet states: 4.42 eV (1AЈ), 5.18 eV (1E ), 5.57
Ј
2
1
eV ( A2
Љ), 5.33 eV (3E ), and 5.75 eV (3AЉ). Such closely
Ј
2
spaced energies are related to the nearly same orbital ener-
gies in both majority and minority spin representations. The
4.42 eV transition (1A2Ј) may correspond to the feature la-
beled a at 4.61 eV ͓Fig. 4͑b͔͒. The third main detachment
feature labeled B may then be due to detachment from the
6e and 3a2Љ orbitals, corresponding to the last four calcu-
Ј
The second direct computation of the detachment energy
corresponds to removal of an electron from the 6b␤2 MO ͓the
topmost MO in the ␤ ͑minority͒ spin representation͔ and
yielded a VDE of 4.26 eV, which corresponds to the vertical
lated detachment transitions. The large intensity of feature B
is consistent with the overlapping nature of the assigned tran-
sitions. The detachment from the 1e MO was computed to
Љ
occur at 6.68 eV, which is beyond the 193 nm photon energy,
consistent with our PES spectra. However, all the latter as-
signments can only be viewed as tentative. With our current
one-electron approach, it is difficult to reach a quantitative
assignment. Multireference theoretical models would be re-
5
detachment to a A2 final neutral state. This computed VDE
value is in excellent agreement with the VDE of the B band
͑4.25 eV͒ in the PES spectra ͑Fig. 3͒. Applying Eq. ͑2͒, we
estimated approximate detachment energies from other va-
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
129.127.200.132 On: Wed, 10 Dec 2014 12:24:19

7942
J. Chem. Phys., Vol. 115, No. 17, 1 November 2001
Gutsev et al.
TABLE III. Experimental and theoretical adiabatic electron affinities ͑eV͒ for CrOn (nϭ1–5).
CrO
CrO2
Cr͑O2)
CrO3
CrO4
CrO5
Expt.
1.22Ϯ0.01a
2.43Ϯ0.02b
1.50Ϯ0.06
3.66Ϯ0.02
4.98Ϯ0.09
4.4Ϯ0.1
Theor.
1.17
2.22
1.37
3.38
4.61
4.41
a1.221Ϯ0.006 eV from Ref. 15.
b2.413Ϯ0.008 eV from Ref. 16.
agreement with the broad band X, which should be due to
two overlapping features. The broad nature of this band is
quired to pin down the excited states of the CrO3 neutral
species. The seeming simplicity of the CrOϪ3 PES spectra
may in fact contain complicated and overlapping detachment
transitions.
consistent with the geometry changes between the anion and
1
neutral ground state and the dissociative nature of the A
Ј
state.
Using Eq. ͑2͒ with the two reference energies
D. CrO4 and CrOÀ4
E (A,3A )ϪE (A,2A ) ϭ4.69 eV and E (A,1A )
͓
͔
͓
tot
Љ
Љ
Ј
tot
tot
1
ϪE (A,2A ) ϭ5.17 eV, we estimated VDE’s from other
The ground state of CrO4 is A (C2 symmetry͒ and its
͔
Љ
tot
electronic configuration is 15a213b2. But there are two low-
lying triplet states, as shown in Fig. 7. The triplet peroxo
state is only 0.12 eV above the ground state CrO4. The anion
ground state has an electronic configuration of
valence MO’s: 23a (1A )–5.60 eV, 9a (3A )–5.45 eV,
Ј
Ј
Љ
Ј
9a (1A )–5.66 eV, 22a (3A )–6.21 eV, 22a (1A )–6.37
Љ
Ј
Ј
Љ
Ј
Љ
eV, 8a (3A )–6.21 eV, 8a (1A )–6.45 eV. The 5.45 eV
Љ
Ј
Љ
Ј
detachment energy from the 9a orbital is in excellent agree-
Љ
2
2
2
1
2
13a12a27b27b1, corresponding to a B1 (C2v) state, which
ment with the A band at 5.33 eV. The 22a and 8a orbitals
Ј
Љ
2
is degenerate in total energy with a B2 state. Detachment
are degenerate and the 6.21 eV detachment energies from
these two MO’s are in good agreement with the B band at
6.08 eV. The large number of transitions closely spaced in
energy is consistent with the complicated PES spectrum ob-
served for CrOϪ5 .
from the 7b1 MO results in the neutral 1A1 ground state with
E (A,1A )ϪE (A,2B ϭ6.13 eV, whereas detachment
͓
͔
1
tot
1
tot
from 7b2 leads to
ϪE (A,2B ) ϭ4.95 eV, which is slightly above the peroxo
a
3A2 state with E (A,3A )
͓
tot 2
͔
1
tot
3A2 state, the lowest triplet state of the neutral ͑Fig. 7͒. These
two detachment channels are in excellent agreement with the
two main PES features at 5.07 eV (X) and 6.04 eV (A).
Here, we met the case where the first vertical detachment of
an extra electron from the singly occupied MO results in
formation of a neutral at C2v symmetry, which would relax
to C2 symmetry. This is a really unusual situation: this first
F. Adiabatic electron affinities
The experimental and theoretical adiabatic electron af-
finities of the chromium oxides are summarized in Table III.
The theoretical EA’s computed according to Eq. ͑3͒ are: 1.17
eV ͑CrO͒,29 2.22 eV (CrO2),30 3.38 eV (CrO3), 4.61 eV
(CrO4), and 4.41 eV (CrO5). These computational data are
in good agreement with the experimental values. The EA’s
increase with the number of O atoms and saturate at CrO4.
This trend is consistent with our previous observations of
other oxide species and indicates a sequential oxidation of
the Cr atom. Both CrO4 and CrO5 possess rather large EA’s
exceeding that of atomic Cl ͑3.62 eV͒48 and formally belong
to the class of superhalogen compounds.49 The reason of
such high EA’s for the 3d TM oxides was first proposed on
the basis of calculations by an X␣-method.50 Our recent
calculations26,27 of FeO4 and FeOϪ4 by the BPW91 method
confirmed that the reason is in the structure of the topmost
MO, which accepts the extra electron. This MO is almost
entirely composed of oxygen AO’s and is bonding with re-
spect to oxygens. Note that the classical superhalogen MnO4
1
transition to the singlet A state has an energy higher by 1.2
eV than the transition to the triplet 3A2 state, which is higher
in total energy but has a similar geometry as the anion. This
suggests that the A band corresponding to removal of the 7b1
electron should have a long low-energy tail extending to the
left of the X band, not inconsistent with our PES spectrum
͓Fig. 5͑a͔͒. Using Eq. ͑2͒ with two reference energies, we
estimated energies of several other vertical transitions. The
estimated singlet states are: 5.29 eV (1A2), 5.60 eV (1B2),
6.53 eV (1A2), and 6.71 eV (1B1); and the triplet states are:
5.03 eV (3B2), 6.06 eV (3B1), and 6.09 eV (3A2). All these
transitions except the 6.53 and 6.71 eV singlet states may be
contained in the PES spectrum of CrOϪ4 ͓Fig. 5͑a͔͒.
E. CrO5 and CrOÀ5
*
has the Aad of 4.96 eV computed at the BPW91/6-311ϩG
We found CrO5 to possess only one thermodynamically
level of theory, in good agreement with the experimental
value of 4.8 eV obtained from the PES spectra of MnOϪ4 .25
The topmost MO of MnO4 is a pure oxygen one as is re-
quired by the superhalogen theory.49
3
stable state of a superoxo type (C ) with a triplet state A
Љ
s
(8a )2(22a )2(9a )2(23a )1(10a )1 . The ground state of
͓
͔
Љ
Љ
Ј
Љ
Ј
the CrOϪ5 anion is 2A . . . (23a )2(10a )1 and has a simi-
͓
͔
Љ
Љ
Ј
lar Cs geometrical shape as its neutral parent, as shown in
Fig. 6. According to Eq. ͑1͒, detachment of an electron from
the 23a orbital results in the 3A neutral ground state with a
G. Fragmentation patterns
Ј
Љ
VDE of 4.69 eV, whereas detachment from the 10a orbital
Table IV reports the fragmentation energies of CrOϪn and
CrOn computed according to Eq. ͑4͒. Experimental data are
available for CrO and CrO2,51,52 and they are in fair agree-
Љ
1
reaches a dissociative A neutral state with a VDE of 5.17
Ј
eV. These two computed detachment channels are in good
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
129.127.200.132 On: Wed, 10 Dec 2014 12:24:19

J. Chem. Phys., Vol. 115, No. 17, 1 November 2001
Electronic structure of chromium oxides
7943
TABLE IV. Fragmentation energies (De in eV͒ of CrOn and CrOϪn , com-
puted according to Eq. ͑4͒.
sored by Department of Energy’s Office of Biological and
Environmental Research and located at Pacific Northwest
National Laboratory, which is operated for the U.S. Depart-
ment of Energy by Battelle.
CrOn
CrOϪn
Channel
Channel
De
De
CrO→CrϩO
4.94a
CrOϪ→CrOϩe
→CrϩOϪ1
1.17
4.78
2.22
6.11
6.22
3.35
6.44
7.66
4.56
4.15
4.76
4.41
1.55
3.22
1 H. H. Kung, Transition Metal Oxides: Surface Chemistry and Catalysis
͑Elsevier, New York, 1989͒.
CrOϪ2 →CrO2ϩe
→CrOϪϩO
2 P. C. Thune, R. Linke, W. J. H. van Gennip, A. M. de Jong, and J. W.
Niemantsverdriet, J. Phys. Chem. B 105, 3073 ͑2001͒.
3 H. Brandle, D. Weller, S. S. Parkin et al., Phys. Rev. B 46, 13889 ͑1992͒.
4 M. A. Korotin, V. I. Anisimov, D. I. Khomskii, and G. A. Sawatzky, Phys.
Rev. Lett. 80, 4305 ͑1998͒.
CrO2→CrϩO2
→CrOϩO
5.11
5.98b
→CrϩOϪ2
CrOϪ3 →CrO3ϩe
→CrOϪ2 ϩO
CrO3→CrOϩO2
→CrO2ϩO
5.48
5.31
5 C. W. Walter, C. F. Hertzeler, P. Devynck, G. P. Smith, and J. R. Peterson,
J. Chem. Phys. 95, 824 ͑1991͒.
→CrOϪϩO2
CrOϪ4 →CrO4ϩe
→CrOϪ3 ϩO
CrO4→CrO2ϩO2
→CrO3ϩO
2.42
2.94
6 E. B. Rudnyi, O. M. Vovk, and E. A. Kaibicheva, J. Chem. Thermodyn.
21, 247 ͑1989͒.
→CrOϪ2 ϩO2
CrOϪ5 →CrO5ϩe
→CrOϪ3 ϩO2
→CrOϪ4 ϩO
7 H. Wu, S. R. Desai, and L. S. Wang, J. Chem. Phys. 103, 4363 ͑1995͒.
8 H. Wu, S. R. Desai, and L. S. Wang, J. Am. Chem. Soc. 118, 5296 ͑1996͒.
9 H. Wu, S. R. Desai, and L. S. Wang, J. Phys. Chem. 101, 2103 ͑1997͒.
10 H. Wu and L. S. Wang, J. Chem. Phys. 107, 16 ͑1997͒.
11 L. S. Wang, H. Wu, and S. R. Desai, Phys. Rev. Lett. 76, 4853 ͑1996͒.
12 L. S. Wang, H. Wu, S. R. Desai, J. Fan, and S. D. Colson, J. Phys. Chem.
100, 8697 ͑1996͒.
CrO5→CrO3ϩO2
→CrO4ϩO
0.49
3.37
aExperimental D0 value is 4.78Ϯ0.09, see Ref. 51.
bExperimental D0 value is 5.47ϩϪ00..635 , see Ref. 52.
13 L. S. Wang, Photodetachment Photoelectron Spectroscopy of Transition
Metal Oxide Species, in Advanced Series in Physical Chemistry, Photo-
ionization and Photodetachment, Vol. 10, edited by C. Y. Ng ͑World Sci-
entific, Singapore, 2000͒, p. 854.
ment with our computational results. Since Cr has a 4s13d5
electronic configuration, it has a formal valence of six. How-
ever, the Cr–O bond is stronger in CrO2 than that in CrO3 in
contrast to the anticipation based on the notion of maximum
saturation of formal vacancies. CrO4 is stable by about 55
kcal/mol while CrO3 is stable by only 10 kcal/mol toward
evolution of molecular oxygen.
14 G. L. Gutsev, B. K. Rao, P. Jena, X. Li, and L. S. Wang, J. Chem. Phys.
113, 1473 ͑2000͒.
15 P. G. Wenthold, R. F. Gunion, and W. C. Lineberger, Chem. Phys. Lett.
258, 101 ͑1996͒.
16 P. G. Wenthold, K.-L. Jonas, and W. C. Lineberger, J. Chem. Phys. 106,
9961 ͑1997͒.
17 P. C. Engelking and W. C. Lineberger, J. Chem. Phys. 66, 5054 ͑1977͒.
18 T. Andersen, K. R. Lykke, D. M. Neumark, and W. C. Lineberger, J.
Chem. Phys. 86, 1858 ͑1987͒.
The low-energy decay channels of CrOϪn anion corre-
spond to detachment of an extra electron for nϭ1–3. CrO4Ϫ
is almost equally stable toward detachment of an extra elec-
tron or dissociation of atomic and molecular oxygen. While
CrOϪ3 is rather stable towards detachment of an extra elec-
tron and dissociation of an atomic oxygen, it is much less
stable toward evolution of molecular oxygen, as would be
expected due to the superoxo O2 bonding in CrOϪ5 .
19 R. F. Gunion, S. J. Dixon-Warren, W. C. Lineberger, and M. D. Morse, J.
Chem. Phys. 104, 1765 ͑1996͒.
20 C. W. Bauschlicher, Jr., M. Zhou, L. Andrews, J. R. T. Johnson, I. Panas,
A. Snis, and B. O. Roos, J. Phys. Chem. 103, 5463 ͑1999͒.
21 M. Rosi, C. W. Bauschlicher, Jr., G. V. Chertihin, and L. Andrews, Theor.
Chem. Acc. 99, 106 ͑1998͒.
22 G. V. Chertihin, L. Andrews, M. Rosi, and C. W. Bauschlicher, Jr., J. Phys.
Chem. 101, 9085 ͑1997͒.
23 M. B. Walsch, R. A. King, and H. F. Schaefer III, J. Chem. Phys. 110,
5224 ͑1999͒.
V. SUMMARY
24 S. F. Vyboishchikov and J. Sauer, J. Phys. Chem. A 104, 10913 ͑2000͒.
25 G. L. Gutsev, B. K. Rao, P. Jena, X. B. Wang, and L. S. Wang, Chem.
Phys. Lett. 312, 598 ͑1999͒.
We report a comprehensive theoretical and experimental
investigation of CrOϪn (nϭ1–5) and their corresponding
neutral species. Photodetachment photoelectron spectra were
obtained for the anions produced from a laser vaporization
cluster source at various photon energies. Calculated adia-
batic and vertical binding energies of CrOϪn are in good
agreement with the experimental values. We show that theo-
retical calculations are very helpful in interpretations of the
photoelectron spectra of the anions, in understanding the
electronic and geometrical structures of their ground and ex-
cited states, and chemical bonding for both the anions and
the neutral species.
26 G. L. Gutsev, B. K. Rao, and P. Jena, J. Phys. Chem. A 103, 10819 ͑1999͒.
27 G. L. Gutsev, S. N. Khanna, B. K. Rao, and P. Jena, Phys. Rev. A 59, 3681
͑1999͒.
28 G. L. Gutsev, S. N. Khanna, B. K. Rao, and P. Jena, J. Phys. Chem. 103,
5812 ͑1999͒.
29 G. L. Gutsev, B. K. Rao, and P. Jena, J. Phys. Chem. A 104, 5374 ͑2000͒.
30 G. L. Gutsev, B. K. Rao, and P. Jena, J. Phys. Chem. A 104, 11961 ͑2000͒.
31 A. D. Becke, Phys. Rev. A 38, 3098 ͑1988͒.
32 J. P. Perdew and Y. Wang, Phys. Rev. B 45, 13244 ͑1991͒.
33 C. Lee, W. Yang, and R. G. Parr, Phys. Rev. B 37, 785 ͑1988͒.
34 A. D. Becke, J. Chem. Phys. 98, 5648 ͑1993͒.
35 L. S. Wang, H. S. Cheng, and J. Fan, J. Chem. Phys. 102, 9480 ͑1995͒.
36 L. S. Wang and H. Wu, in Advances in Metal and Semiconductor Clusters.
IV. Cluster Materials, edited by M. A. Duncan ͑JAI, Greenwich, CT,
1998͒, p. 299.
37 M. J. Frisch, G. W. Trucks, H. P. Schlegel et al., GAUSSIAN 94, Revision
B.1, Pittsburgh, PA.
ACKNOWLEDGMENTS
38 A. J. H. Wachters, J. Chem. Phys. 52, 1033 ͑1970͒.
39 P. J. Hay, J. Chem. Phys. 66, 4377 ͑1977͒.
This work was supported in part by a grant to Virginia
Commonwealth University by the Department of Energy
͑Grant No. DE-FG02-96ER45579͒. The experimental work
was supported by NSF ͑L.S.W.͒ under grant CHE-9817811
and performed at the W. R. Wiley Environmental Molecular
Sciences Laboratory, a national scientific user facility spon-
40 W. Kohn and L. J. Sham, Phys. Rev. A 140, 1133 ͑1965͒.
41 O. Gunnarsson and B. I. Lundqvist, J. Chem. Phys. 13, 4274 ͑1976͒.
42 E. K. U. Gross, L. N. Oliveira, and W. Kohn, Phys. Rev. A 37, 2809
͑1988͒.
43 A. Gorling, S. B. Trickey, P. Gisdakis, and N. Rosh, Topics in Organome-
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
129.127.200.132 On: Wed, 10 Dec 2014 12:24:19

7944
J. Chem. Phys., Vol. 115, No. 17, 1 November 2001
Gutsev et al.
tallic Chemistry, edited by P. Hoffman and J. M. Brown ͑Springer, New
York, 1999͒, Vol. 4, pp. 109–165.
47 K. P. Huber and G. Herzberg, Constants of Diatomic Molecules ͑Van
Nostrand Reinhold, New York, 1979͒.
48 H. Hotop and W. C. Lineberger, J. Phys. Chem. Ref. Data 14, 731 ͑1985͒.
49 G. L. Gutsev and A. I. Boldyrev, Adv. Chem. Phys. 61, 169 ͑1985͒.
50 G. L. Gutsev and A. I. Boldyrev, Chem. Phys. Lett. 108, 255 ͑1984͒.
51 H. Kang and J. L. Beauchamp, J. Am. Chem. Soc. 108, 5663 ͑1986͒.
52 R. T. Grimley, R. P. Burns, and M. Inghram, J. Chem. Phys. 34, 664
͑1961͒.
44 S. Veliah, K.-H. Xiang, R. Pandey, J. M. Recio, and J. M. Newsam, J.
Phys. Chem. 102, 1126 ͑1998͒.
45 G. V. Chertihin, W. D. Bare, and L. Andrews, J. Chem. Phys. 107, 2798
͑1997͒.
46 A. J. Merer, Annu. Rev. Phys. Chem. 40, 407 ͑1989͒.
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
129.127.200.132 On: Wed, 10 Dec 2014 12:24:19