Probing molecular symmetry effects in the
ionization of N2 and O2 by intense laser
fields
著者
journal or
publication title
volume
number
page range
year
URL
Okunishi M., Shimada K., Pruemper
Mathur D., Ueda K.
Journal of Chemical physics
G.,
127
6
064310
2007
http://hdl.handle.net/10097/52418
doi: 10.1063/1.2764029
THE JOURNAL OF CHEMICAL PHYSICS 127, 064310 共2007兲
Probing molecular symmetry effects in the ionization of N2 and O2
by intense laser fields
M. Okunishi, K. Shimada, and G. Prümper
Institute of Multidisciplinary Research for Advanced Materials, Tohoku University, Sendai 980-8577, Japan
D. Mathura兲
Institute of Multidisciplinary Research for Advanced Materials, Tohoku University, Sendai 980-8577, Japan
and Tata Institute of Fundamental Research, 1 Homi Bhabha Road, Mumbai 400 005, India
K. Uedab兲
Institute of Multidisciplinary Research for Advanced Materials, Tohoku University, Sendai 980-8577, Japan
共Received 7 May 2007; accepted 28 June 2007; published online 13 August 2007兲
High-resolution electron spectroscopy is used to explore the role played by molecular symmetry in
determining the morphology of the energy spectra of electrons ejected when N2 and O2 are
irradiated by intense laser fields. In O2, the low-energy part of the electron spectrum is curtailed due
to the destructive interference brought about by the antibonding nature of the O2 valence orbital. The
high-energy tail of the spectrum is also suppressed by virtue of electron rescattering being of little
consequence in O2. In contrast, in N2, which has a bonding valence orbital, the electron dynamics
follow the pattern that has been established for atomic ionization in strong optical fields. © 2007
American Institute of Physics. 关DOI: 10.1063/1.2764029兴
I. INTRODUCTION
The interaction of intense laser fields with atoms and
molecules continues to be a burgeoning field of activity.
Over the last decade or so, there has been a steady stream of
discoveries of unexpected, sometimes counterintuitive, phenomena and processes that has continued to invigorate the
subject 共see, for instance, Ref. 1, and references therein兲. In
all intense field studies, the magnitude of the optical field
matches intra-atomic and intramolecular Coulombic fields,
and ionization dominates the overall dynamics. Although
electron spectroscopy is a potent weapon for studies of intense field dynamics, it remains a fact that most experimental
studies in molecules have hitherto relied on measurement of
ion yields alone; relatively little has been reported either on
electron spectroscopy2,3 or on kinematically complete measurements that encompass electron detection.4 We report here
results of high-resolution electron spectroscopy experiments
that we have conducted on diatomic nitrogen and oxygen,
with a view to explore the possible role that molecular symmetry might play in strong field dynamics.
Is molecular symmetry expected to be of consequence
when molecules undergo ionization in strong fields? A
plethora of data on strong field atomic ionization has provided evidence that the rate at which ionization occurs, and
the resulting absolute yield of ions that one obtains, depends
only on one atomic property: the first ionization energy of
the atom that is under laser irradiation. In the tunnel ionization regime, wherein the optical field distorts the atom’s radial potential function, making it possible for one or more
valence electrons to tunnel into the continuum, the resulting
a兲
Electronic mail: [email protected]
Electronic mail: [email protected]
b兲
0021-9606/2007/127共6兲/064310/4/$23.00
electron energy distribution is well accounted for by the
often-used Ammosov-Delone-Krainov 共ADK兲 theory5 in
which the only atomic parameter of concern is the ionization
energy 共IE兲 of the highest occupied atomic orbital; the symmetry of the orbitals themselves does not enter into reckoning. Convincing experimental demonstration of this comes
from measurement of nearly identical strong field ionization
rates for pairs such as Ar and N2, where the ratio
IE共Ar兲 / IE共N2兲 is almost unity 共1.01兲.6 However, in the case
of the complementary pair O2 and Xe, the ionization rate for
the molecule is an order of magnitude lower than the rate for
the atom, even though the ratio IE共Xe兲 / IE共O2兲 is even closer
to unity 共1.005兲.6,7
Rationalizations have been proffered6,7 that invoke either
the nuclear degrees of freedom in molecules or multielectron
effects. These remain to be properly vindicated. However, it
has been noted8 that calculations9 that account for vibrational
motion do not succeed in quantitatively rationalizing the observed differences between ionization rates for molecules
and of companion atoms. Similarly, incorporation of multielectron effects into computations of ionization yields10 using the ADK formula5 does not yield results that match experimentally determined values, except at laser intensities
that lie well beyond saturation.8 An entirely different insight
has emerged from intense field S-matrix calculations carried
out by Muth, Böhm et al.11 which predicted that ionization
would be suppressed in those homonuclear molecules that
possessed a valence orbital with antibonding symmetry, like
the outermost ␲g orbital in O2, but not in the case of molecules whose valence orbital had bonding symmetry, like the
␴g orbital in N2. In the case of O2, the orbital shape results in
destructive interference of the two subwaves of the fieldionized electron that originated at the two nuclei. A combined experimental and theoretical study8 has offered vindi-
127, 064310-1
© 2007 American Institute of Physics
Downloaded 01 Sep 2011 to 130.34.134.250. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/about/rights_and_permissions
064310-2
Okunishi et al.
J. Chem. Phys. 127, 064310 共2007兲
cation of these theoretical predictions: above threshold
ionization 共ATI兲 spectra were measured for the N2 and Ar
pair, and for the O2 and Xe pair, and it was demonstrated
共and theoretically rationalized兲 that, for the latter pair, lowenergy ATI peaks in O2 were suppressed 共by one order of
magnitude兲 in comparison to those in Xe, whereas the
higher-energy components in the ATI spectra were identical
in both cases. In the case of the other pair, all the ATI spectra
were essentially identical to each other.
We report here results of electron spectroscopy measurements in which we make a direct comparison of the molecular pair N2 and O2, with a view to explore the effect that
valence orbital symmetry might play on the electron dynamics. Our measurements have been conducted at higher resolution and sensitivity and over a wider range of electron kinetic energies than those hitherto reported. Distinct
differences in the morphology of the electron spectra measured using the two molecules are established. Moreover, we
have also used both linearly and circularly polarized lights to
directly establish that electron rescattering is suppressed in
O2 compared to N2.
II. EXPERIMENT
In our experiments, we detected electrons using a
264 mm long linear time-of-flight 共TOF兲 spectrometer with a
limited detection angle 关⬃0.0014共4␲兲 sr兴. The fundamental
output 共800 nm兲 from an amplified Ti:sapphire laser system
共pulse width: 100 fs; repetition rate: 1 kHz兲 was used as ionizing radiation. The 2 – 3 mm diameter laser beam was focused by an f = 60 mm lens to a field-free location between
two grounded electrodes comprising graphite-coated Al with
80% transmission Cu mesh. Linear polarization was along
the TOF axis. N2 or O2 gas was effusively introduced such
that a typical working pressure of 10−6 – 10−8 mbar 共base
pressure is less than 10−9 mbar兲 ensured that space charge
effects were of no consequence in our experiments. Only
electrons ejected in the direction of the TOF spectrometer
were detected by tandem microchannel plates. The efficacy
of our experimental method in relation to high-resolution
electron spectroscopy of strong field molecular dynamics has
been recently demonstrated in relation to a series of linear
alcohols.12
A properly defined energy scale is an important facet of
electron spectroscopy. We calibrated our spectrometer with
reference to multiphoton ionization of Xe atoms at a relatively low electron kinetic energy region 共less than 10 eV兲.
Comparing the time origin of the TOF spectrometer from this
calibration with the measured one, we estimated that the errors of the energy scale in our data is ⬍5%.
FIG. 1. Electron spectra of N2 at several different laser intensities.
We first consider the gross features of the electron spectra measured using linearly polarized light. For N2 共Fig. 1兲,
there is a sharp drop in electron yield with energy, by more
than two orders of magnitude over the energy range from
thermal to 2U p – 3U p, where U p denotes the ponderomotive
potential that is “seen” by the ionized electron. Thereafter, a
plateau region commences beyond energy values corresponding to 2U p – 3U p, with the electron yield totally disappearing as energies approach 10U p. This type of electron
energy distribution is relatively easy to rationalize on the
basis of the established wisdom for the atomic field
III. RESULTS AND DISCUSSION
Typical electron spectra measured with N2 at several different laser intensities with linear polarization are depicted in
Fig. 1. Corresponding data for O2 are shown in Fig. 2. All
these data correspond to measurements made with the
Keldysh parameter being between 0.5 and 1.4, implying the
transition region between multiphoton ionization and tunnel
ionization.
FIG. 2. Electron spectra of O2 at several different laser intensities.
Downloaded 01 Sep 2011 to 130.34.134.250. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/about/rights_and_permissions
064310-3
Ionization of N2 and O2 by laser fields
J. Chem. Phys. 127, 064310 共2007兲
FIG. 3. Electron spectra of 共1兲 N2 and
共2兲 O2 for linearly and circularly polarized lights of intensities such that the
optical field remains essentially the
same.
ionization.13 Single electron promotion to the continuum is
independent of the phase of the optical field, and its subsequent motion is governed by the quiver motion that is generated by the oscillating field and the drift velocity. A free
electron that is “born” at the peak of the applied field does
not gain drift energy while birth at the null of the field will
allow it to have 2U p worth of kinetic energy. In the tunneling
picture, the ionization rate maximizes when the laser intensity is at its peak value and, hence, the electron energy distribution for single ionization has a peak at zero energy14 and
a subsequent drop towards 2U p. This is commensurate with
what is also observed in the case of O2 ionization by linearly
polarized light 共Fig. 2兲.
Electrons that are born after the peak of the laser intensity is reached return to the ionic core where they undergo
elastic backscattering. This gives rise to the long tail extending to 10U p that we observe in Fig. 1. Backward scattering is
efficient in producing the high-energy tail because the laser
field and the electron velocity vector have opposite signs,
leading to additional acceleration being imparted to the backward scattered electron. Electron spectra of O2 共Fig. 2兲 also
show some similar structures. Careful inspections, however,
reveal two significant differences between N2 and O2 spectra.
First, there is another plateau structure in the ATI spectra that
appears in the low-energy region below U p, in vindication of
the expectations of the strong field S-matrix theory:11 destructive interference effects related to the antibonding ␲g
orbital of O2 suppress the low-energy region in the electron
spectrum. Such suppression is not observed in the case of the
bonding ␴g orbital in N2. The sharp contrast observed at low
kinetic energies becomes more pronounced at higher laser
intensities. The second significant difference between N2 and
O2 spectra is that the high-energy plateau region that arises
from electron rescattering is very strongly suppressed in the
case of O2.
In Fig. 3 we compare N2 and O2 spectra that are measured with linear and circular polarizations at laser intensity
values corresponding to the Keldysh parameter being near
unity for both molecules. The laser intensity values were
appropriately adjusted to ensure that the magnitude of the
optical field experienced by the molecules was more or less
the same for both polarization states. The electron energy
distribution measured with the circular polarization shows no
evidence for a plateau beyond 2U p, an observation that is
commensurate with the fact that no rescattering is now possible. In the case of N2, the difference between the linear
polarization and circular polarization spectra in the energy
region beyond 3U p is quite stark. The corresponding starkness in difference is not there in the case of O2, in harmony
with the above observation that the high-energy plateau due
to the rescattering is suppressed in the case of O2.
The overall morphology of the spectra is also polarization dependent. There is a broad peak that is formed at about
1U p in the electron spectra measured with circular polarization. We have the following observations to make. As has
been discussed for atoms,15 the discovery of ATI was made
on the basis of electron spectroscopy of rare gases using
linearly polarized intense laser light.16 Subsequent measurements, made with circular polarization,17 revealed that the
lowest-energy part of the electron spectrum was suppressed
at sufficiently high laser intensities. The envelope of ATI
peaks, in fact, assumed a bell-shaped spectrum, with lowerorder peaks becoming suppressed compared with the situation pertaining to the linear polarization case. Moreover,
higher-order ATI peaks were observed to become somewhat
more prominent with circular polarization. Bell-shaped energy distributions have also been computed for circular polarization by Delone and Krainov14 within the framework of
strong field atomic ionization theory and have been rationalized by Corkum et al. on the basis of classical strong field
electron dynamics.18 Reiss15 has theoretically shown that
higher-order ATI becomes more possible with increase of
intensity as more channels open up because of the extra photons that come into play in the ionization dynamics. Moreover, with circularly polarized light, angular momentum
needs to be conserved in the final electron states and, conse-
Downloaded 01 Sep 2011 to 130.34.134.250. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/about/rights_and_permissions
064310-4
J. Chem. Phys. 127, 064310 共2007兲
Okunishi et al.
quently, low-energy electron states are suppressed. This provides a rationalization for the bell-shaped component of the
electron energy spectrum that is observed in the case of molecular ionization in our experiments 共Fig. 3兲.
IV. SUMMARY
We have used electron spectroscopy to probe the role
that molecular symmetry effects play in the ionization dynamics that ensue when the molecular pair N2 and O2 is
irradiated by intense laser fields. In the case of O2, which has
a nonbonding outermost molecular orbital, the low-energy
part of the spectrum of the electrons is severely suppressed
due to destructive interference effects. Suppression of the
low-energy component also serves to suppress electron rescattering in O2 and, consequently, the high-energy tail of the
electron spectrum is also significantly curtailed. In contrast,
in N2, which has a bonding valence orbital, the electron dynamics follow the pattern that has been well established for
atomic ionization in strong optical fields.
ACKNOWLEDGMENTS
We acknowledge useful comments from Andreas Becker.
The work was supported in part by CREST.
1
Progress in Ultrafast Intense Laser Science, edited by K. Yamanouchi, S.
L. Chin, P. Agostini, and G. Ferrante 共Springer, Berlin, 2006兲.
2
M. J. DeWitt and R. J. Levis, Phys. Rev. Lett. 81, 5101 共1998兲.
3
E. E. B. Campbell, K. Hansen, K. Hoffmann, G. Korn, M. Tchaplyguine,
M. Wittmann, and I. V. Hertel, Phys. Rev. Lett. 84, 2128 共2000兲.
4
H. Rottke, C. Trump, M. Wittmann et al., Phys. Rev. Lett. 89, 013001
共2002兲.
5
M. V. Ammosov, N. B. Delone, and V. P. Kranov, Zh. Eksp. Teor. Fiz.
91, 2008 共1986兲 关Sov. Phys. JETP 64, 1191 共1986兲兴.
6
C. Guo, M. Li, J. P. Nibarger, and G. N. Gibson, Phys. Rev. A 58, R4271
共1998兲.
7
A. Talebpour, C.-Y. Chien, and S. L. Chin, J. Phys. B 29, L677 共1996兲.
8
F. Grasbon, G. G. Paulus, S. L. Chin, H. Walther, J. Muth-Böhm, A.
Becker, and F. H. M. Faisal, Phys. Rev. A 63, 041402 共2001兲.
9
A. Saenz, J. Phys. B 33, 4365 共2000兲.
10
C. Guo, Phys. Rev. Lett. 85, 2276 共2000兲.
11
J. Muth-Böhm, A. Becker, and F. H. M. Faisal, Phys. Rev. Lett. 85, 2280
共2000兲.
12
T. Hatamoto, M. Okunishi, T. Lischke, G. Prümper, K. Shimada, D.
Mathur, and K. Ueda, Chem. Phys. Lett. 439, 296 共2007兲.
13
G. G. Paulus, W. Nicklich, H. Xu, P. Lambropoulos, and H. Walther,
Phys. Rev. Lett. 72, 2851 共1994兲; Europhys. Lett. 27, 267 共1994兲.
14
N. B. Delone and V. P. Krainov, Phys. Usp. 41, 469 共1998兲 关Usp. Fiz.
Nauk 168, 531 共1998兲兴.
15
H. R. Reiss, J. Phys. B 20, L79 共1987兲.
16
P. Agostini, F. Fabre, G. Mainfray, G. Petit, and N. K. Rahman, Phys.
Rev. Lett. 42, 1127 共1979兲.
17
F. Yergeau, G. Petite, and P. Agostini, J. Phys. B 19, L663 共1986兲; P. H.
Buksbaum, M. Bashkansky, R. R. Freeman, T. J. McIlrath, and L. F.
DiMauro, Phys. Rev. Lett. 56, 2590 共1986兲.
18
P. B. Corkum, N. H. Burnett, and F. Brunel, Phys. Rev. Lett. 62, 1259
共1989兲.
Downloaded 01 Sep 2011 to 130.34.134.250. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/about/rights_and_permissions