Ionization suppression of Cl
2
molecules in intense laser fields
E. P. Benis,
*
J. F. Xia, X. M. Tong, M. Faheem, M. Zamkov, B. Shan, P. Richard, and Z. Chang
J. R. Macdonald Laboratory, Department of Physics, Kansas State University, Manhattan, Kansas 66506-2604, USA
(Received 30 October 2003; published 26 August 2004)
The strong field ionization of Cl
2
molecules is investigated by using an ultrashort pulse Ti:sapphire laser. A
spatial imaging technique is used in such measurements to reduce the effect of spatial integration. Cl
2
shows
strong ionization suppression as do other diatomic molecules having valence orbitals with antibonding sym-
metry O
2
,S
2
when compared with the field ionization of atoms with nearly identical ionization potential. A
more general molecular tunneling ionization model is proposed, and the calculations are in reasonable agree-
ment with the measurements. Our results support that antibonding leads to ionization suppression, a trend that
only F
2
goes against and that needs to be further investigated.
DOI: 10.1103/PhysRevA.70.025401 PACS number(s): 33.80.Rv, 32.80.Rm, 34.50.Gb, 42.50.Hz
Perhaps the most fundamental process in the interaction
of strong laser fields with molecules is single electron ion-
ization. Even though atomic ionization has been extensively
investigated both theoretically and experimentally, the study
of molecular ionization is still far from complete. Early ex-
perimental data [1,2] indicated that the ionization rates of
diatomic molecules were similar to atoms with nearly iden-
tical ionization potentials (IPs). A simple tunneling ioniza-
tion model suggested by Perelomov et al. [3], and simplified
by Ammosov et al., [4](referred to as the ADK model) were
used to explain the atomic tunneling ionization. However,
further investigations on diatomic molecules showed that the
ionization is strongly suppressed for the cases of
D
2
IP of 15.467 eV and O
2
IP of 12.070 eV in compari-
son to their companion atoms Ar IP of 15.764 eV and
Xe IP of 12.13 eV, respectively, while the ionization rate
for N
2
IP of 15.581 eV and F
2
IP of 15.697 eV is compa-
rable to their companion Ar atom [5–9].
Early theoretical interpretations proposed the mechanisms
of molecular alignment [5] and dissociative recombination
[6] to explain the suppression in the cases of D
2
and O
2
,
respectively. However, they proved to be inadequate by later
theoretical [10] and experimental studies [7,11]. A possible
effect on ionization suppression from the vibrational motion
of the molecules was shown to be too small by Saenz [10].
Guo proposed a larger “effective” ionization potential im-
posed by a larger effective charge experienced by the open-
shell valence electron [12]. However, the proposed model
cannot be obtained theoretically or empirically from other
experiments and does not account for the behaviors of F
2
and
D
2
, as discussed by Wells et al. [9]. Muth-Bohm et al. [13]
explained the ionization suppression in O
2
as the result of
destructive interference in the electron emission probability
from the two centers. The model fails to explain the suppres-
sion of D
2
. Moreover, it predicts suppression for F
2
, a result
which contradicts experiment [8,9]. Tong et al. [14] devel-
oped a molecular ADK (MO-ADK) which is based on the
assumptions of the ADK model for tunneling ionization of
atoms, but suitably modified to account for the difference in
the electronic wave functions in atoms and molecules. Re-
sults from their MO-ADK model are in good agreement with
measured ratios of ionization signals for pairs with suppres-
sion D
2
:Ar,O
2
:Xe and pairs without suppression
N
2
:Ar,CO:Kr. Results are also in agreement with the
measured ionization signals of NO, S
2
, and SO. However,
the model predicts suppression for F
2
: Ar, which is in dis-
agreement with experiment. Is F
2
an exception to the MO-
ADK model or is the MO-ADK model not valid for the
full-filled antibond orbital?
To answer the question, in this Brief Report, we present a
measurement of the field ionization of Cl
2
IP of 11.48 eV,
which has a full-filled antibond orbital, in comparison to its
closest companion atom, Xe. A spatial imaging technique
was used to measure the ionization of mixed Cl
2
and Xe gas
target. The basic advantage of the technique is that the ion-
ization signal can be obtained at one laser pulse energy. A
more general molecular tunneling ionization model is pro-
posed and calculations are compared to the measurements.
So far, only O
2
and S
2
showed strong suppression predicted
by the MO-ADK model, leaving F
2
as a puzzling exception.
By including Cl
2
in the single ionization suppression studies
we expect to shed more light on the issue.
The experiments were performed at the J. R. Macdonald
Laboratory at Kansas State University using the Kansas
Light Source (KLS), i.e., a Ti:sapphire laser delivering
790 nm, 25 fs pulses with an energy of 4 mJ at a repetition
rate of 1 kHz. A beam splitter for ultrafast optics has been
used to pick up 20% of the the total output to this experi-
ment. The on-target laser power was tuned by a variable
neutral density filter. The linearly polarized pulses were fo-
cused by a 25 cm focal length lens into the UHV chamber.
The UHV chamber maintained a base pressure of 2
10
-9
Torr. A mixture of equal pressures of the two gases
(Cl
2
and Xe) was prepared in a small volume bottle and then
introduced into the chamber through the needle valve, reach-
ing pressures up to 1 10
-5
Torr. Special care was taken in
order to minimize any side effects caused by the increased
reactivity of Cl
2
with H
2
O and other contaminants in the
vacuum chamber. A residual gas analyzer was utilized to
*Present address: Institute of Electronic Structure and Laser, P.O.
Box 1527, 71110, Heraklion, Crete, Greece; electronic address:
benis@iesl.forth.gr
PHYSICAL REVIEW A 70, 025401 (2004)
1050-2947/2004/70(2)/025401(4)/$22.50 ©2004 The American Physical Society 70 025401-1