Charge transfer of Ti 4+ with Ar and N 2 at electron-volt energy

Article abstract of DOI:10.1016/s0009-2614(98)00433-3

The charge-transfer reactions of ground state Ti⁴⁺ (3p⁶, ¹S) with Ar and N₂ were studied in a quadrupole r.f. ion trap at the mean collision energy of 4.55 eV. The rate coefficients were measured to be 1.06(0.14)×10^-9 cm³/s for Ti⁴⁺ with Ar at an equilibrium temperature of 1.6×10⁴ K and 7.45(0.38)×10^-10 cm³/s for Ti⁴⁺ with N₂ at an equilibrium temperature of 1.3×10⁴ K. Both results are comparable with the Langevin rate coefficients.

Full text of DOI:10.1016/s0009-2614(98)00433-3

12 June 1998
Ž
.
Chemical Physics Letters 289 1998 298–302
1
Charge transfer of Ti4q with Ar and N2 at electron-volt energy
2
Min Yan , Mang Feng, Kelin Gao, Xiwen Zhu
Laboratory of Magnetic Resonance and Atomic and Molecular Physics, Wuhan Institute of Physics and Mathematics, The Chinese Academy
of Sciences, P.O. Box 71010, Wuhan 430071, PR China
Received 10 December 1997; in final form 27 February 1998
Abstract
1
q
6
The charge-transfer reactions of ground state Ti4 3p , S with Ar and N2 were studied in a quadrupole r.f. ion trap at
Ž
.
the mean collision energy of 4.55 eV. The rate coefficients were measured to be 1.06 0.14 =10 cm3rs for Ti4q with Ar
y
9
Ž
.
at an equilibrium temperature of 1.6=104 K and 7.45 0.38 =10
cm3rs for Ti4q with N2 at an equilibrium
y
10
Ž
.
temperature of 1.3=104 K. Both results are comparable with the Langevin rate coefficients. q 1998 Elsevier Science B.V.
All rights reserved.
1. Introduction
collision energies, only a few have been made in the
electronvolt energy range.
Charge transfer processes between multiply
charged ions and neutrals are important atomic pro-
cesses in many plasmas present in astrophysical ob-
To gain some insight into the energy balance,
charge- and excitation-state distribution and radiation
power loss mechanisms of fusion plasmas due to the
strong influence of the electron-capture processes,
one has to study the charge transfer processes involv-
ing some specific multi-charged ions. The multi-
charged or even highly stripped complex ions of
tungsten, molybdenum, titanium and iron can be
originated and released from materials of the wall of
Ž
.
jects planetary nebulae, interstellar medium, etc.
and created in the laboratories for various purposes
w x
1 . The experimental measurements of the charge
transfer rate coefficients will help to understand
plasma equilibrium which was strongly influenced
by these processes and will also help to verify the
basic theoretical treatment of these complicated
quantum systems. Several experimental techniques
have been developed for these studies, such as
Ž
some fusion plasma apparatus the limiter and reac-
.
tor, in the case of tokamaks and be present as the
major impurities of the plasma. The charge-transfer
processes of these ions with neutral atoms and
molecules may bring the most important practical
w x
w
x
merged-beams 2 , ion storage 3–6 , and time-of-
w
x
flight mass spectrometers 7,8 , but most of these
experimental works have been carried out at keV
w x
implications 9 . Few experimental results have been
reported for charge-transfer reactions of multi-
charged ions with electron-volt energy from these
refractory elements which are present at the edge of
the fusion plasma, only W2q and Mo6q have been
1 Project 19674066 supported by National Natural Science
Foundation of China.
Ž
.
2 Email: m.yan@nmr.whcnc.ac.cn.
0009-2614r98r$19.00 q 1998 Elsevier Science B.V. All rights reserved.
Ž
.
PII: S0009-2614 98 00433-3

(
)
M. Yan et al.rChemical Physics Letters 289 1998 298–302
299
w
x
studied by Kwong et al. 5,6 in their cylindrical r.f.
trap. Here we report a measurement of charge trans-
ions were created by laser ablation using the focused
Ž
light two lens with focus length of 45 cm were
fer coefficients of Ti4q average energy of about
4.55 eV with Ar and N2. No theoretical and other
used from the output of the second harmonic 532
Ž
.
Ž
.
.
nm of a pulsed Nd:YAG laser. With the laser
experimental values are available for these reactions
at electron-volt energy. Our experiment has been
performed in a quadrupole r.f. ion trap, the low-en-
ergy Ti4q has been created and then confined in the
trap by laser ablation combined with the technique of
crossed-beam cooling for ions introduced by Kwong
energy per pulse of approximately 5 mJ correspond-
ing to the pulse power density above 1=109 Wrcm2
with laser pulse duration of about 15 ns, other
2q
charge-state ions such as Ti3q mrqs16 , Ti
Ž
.
q
q
Ž
.
Ž
.
Ž
.
mrqs24 , Ti
mrqs48 , O
mrqs16 and
4q
O2q mrqs8 were produced with Ti
ions and
Ž
.
w x
et al. 4 . Due to the flexibility of these techniques
may be confined in the trap. So operating the trap
this work can be extended to varieties of charge-
transfer reactions of multi-charged ions with neutral
atoms and molecules.
near the cross point of bz s1 and br s0 of the first
Ž
stability region i.e. the operating point of az s0.15,
.
qz s0.78 was chosen for selective-storage of the
specific ions. In our measurement for Ti4q ions with
the r.f. frequency of 1.1MHz the mass-selective-con-
finement operating voltage are about 115 V for r.f.
voltage and 11 V for DC voltage. We can maintain
an appropriate density of Ti4q ions for measurement
and eliminate all other charge state ions by keeping
the trap working near this operation point for about
several milliseconds. The ions were extracted for
analysis from the trap axially due to the axial desta-
bility by scanning the r.f. voltage applied to the ring
electrode, and the ions were ejected through a hole in
the center of an endcap electrode to the surface of a
channel electron-multiplier quite near the endcap
electrode. With an appropriate r.f. voltage scanning
2. Experiments and results
The apparatus of our experiments consists of a
quadrupole r.f. ion trap with a minimum radius of
the ring electrode r0 s0.99 cm, and the minimum
separation between two endcap electrodes 2 z0 s1.4
Ž
.
cm. Two solid TiO2 99.9% purity targets are held
in position in the plane of the ring electrode and at
the same fixed distance from its center. The arrange-
ment of the apparatus and the method of laser abla-
tion and crossed-beam cooling for creating low-en-
ergy multi-charged ions were similar to that demon-
w x
Ž
strated in Ref. 4 . Here we briefly describe the
rate corresponding to a mass scanning rate about
.
different approach of mass-selective confinement and
rejection detection of the ions in our measurements.
The timing diagram for production, trapping and
detection of the ions is shown in Fig. 1. The Ti4q
2000 aurs we can give the best mass resolution of
Mrd MG6 at 12 au.
The experimental procedure for measuring the
charge transfer rate coefficients is summarized as
y7
Ž
.
follows. A small amount about 10 torr of reac-
tant gas Ar or N2 was introduced to the vacuum
chamber through a variable leak valve. The Ti4q
ions and other ion species were created by laser
ablation and cooled due to crossed-beam collision.
The Ti4q ions were selectively stored in the trap by
the method mentioned above, and then extracted
from the trap by scanning the r.f. voltage at a delay
Ž
time t0 qd t t0 refers to the initial delay time of
.
about 50 ms in our measurement relative to the
selective-confinement, and collected by the electron
multiplier due to the influence of the negative poten-
Fig. 1. Timing diagram for ion trap and laser ablation. The
selective voltage of VAC and VDC are set for mass-selective
storage of the Ti4q ions. The delay time from the end of the
selective-storage operation and the beginning of the analytical
scan is changeable.
Ž
.
tial about y2500 V in our experiments which is
applied to the surface of the multiplier. The mass
spectrum from the multiplier was amplified by a

(
)
300
M. Yan et al.rChemical Physics Letters 289 1998 298–302
narrow-banded DC amplifier and then transferred by
a fast ADrDA converter, and finally recorded in the
computer for later analysis. The storage time was
increased from the initial delay time t0 with a fixed
increment d t until the ion signal intensity had
dropped to one-tenth of that at the initial delay time.
More than ten such time delay points were used to
obtain the data. For each delay time more than five
data points were acquired. Then we repeated such
measurements by varying the storage time in the
opposite direction. The average of the data in each
time delay gave the mean relative ion density that
remained in the trap and could minimize the mea-
surement error due to the random change of the laser
4q
Ž .
Fig. 3. a The decay of relative ion intensity of Ti
vs. storage
time in different N2 pressures. The solid lines are the least-square
Ž .
fits to a single exponential function.
decay rate vs. N2 pressure. The slope of the straight line fit gives
the charge-transfer rate coefficient.
b Variations of the Ti4q
power and the effects of changing conditions on the
target surface. The approach was performed for five
different Ar or N2 pressures.
Data points with different shapes in Fig. 2a and
Fig. 3a represents the results obtained under different
gas pressures. Each set of data points denotes varia-
tions of the relative ion intensities versus storage
time t. The solid line is a least-squares fit to the data
by a single exponential decay function in natural
logarithm scale. Its slope gives the charge transfer
rate at a given Ar or N2 pressure.
4q
Ž .
Fig. 2. a The decay of relative ion intensity of Ti
vs. storage
time in different Ar pressures. The solid lines are the least-square
Ž .
Fig. 2b plots the charge transfer rate between
Ti4q ions and Ar as a function of Ar pressure while
Fig. 3b plots that between Ti4q ions and N2. The
fits to a single exponential function.
b
Variations of the Ti4q
decay rate vs. Ar pressure. The slope of the straight line fit gives
the charge-transfer rate coefficient.

(
)
M. Yan et al.rChemical Physics Letters 289 1998 298–302
301
depth for Ti4q was about 45.5 eV. We estimate that
the mean energy of the stored Ti4q ions is 4.55 eV
which is approximately a tenth of the potential well
solid line in each figure is the least-squares fit to a
linear function of the experimental data and its slope
gives the charge transfer rate coefficient. For Ti4q
Ž
.
w
x
ions and Ar the value is deduced to be 1.06 0.14 =
10 . This mean energy corresponds to a nearly
10y9 cm3rs, and for Ti4q ions and N2 , 7.45 0.38
Ž
.
thermal distribution of ion temperature of about 3.52
=104 K. A equilibrium temperature of Ti4q ions
with corresponding room temperature reaction gas
could be obtained. For Ti4q ions and Ar this temper-
ature is about 1.6=104 K and that of Ti4q with N2
is about 1.3=104 K. From these data we could
derive the mean relative velocity of the Ti4q and the
=10y10 cm3rs. The uncertainty presented here is
due to the uncertainty of the gas pressure measure-
ments and the uncertainty of the ion signal intensity
results from the fluctuation of laser power and other
nonlinear effects of the detection circuit. We ne-
glected the ions escape loss and its uncertainty in our
measurements.
1
3
reactant gas Ar or N2 according to 2 mÕ2 s 2 kBTeq
Ž
kB is the Boltzman constant, Teq is the equilibrium
.
temperature for each collision reaction , the corre-
sponding values are 9.1=103 mrs and 8.2=103
mrs, respectively.
Since there is no available theoretical value at this
energy range for charge transfer of Ti4q with neutral
atoms or molecules, we may compare the measured
results with the Langevin cross sections 11 . The
static average electric dipole polarizability for Ar
and N2 is 1.64=10y24 cm3 and 1.74=10y24 cm3,
respectively 12 . The corresponding Langevin
charge-transfer rate coefficients for Ti4qqAr can be
estimated to be 5.43=10y9 cm3rs and that for
Ti4qqN2 to be 5.59=10y9 cm3rs. These values
are about five and seven times larger than our mea-
surements, respectively, and can be considered as to
be comparable with the measured results. Since the
Langevin model is based on the classical kinetic
theory, there always exists a discrepancy between the
charge-transfer rate coefficients and corresponding
3. Discussion
The argon-like Ti4q ion has no low-lying
metastable state. If the ions produced by laser abla-
tion were in a variety of excited electronic states,
they will rapidly cascade to their ground state through
allowed transitions and collisional deexcitations by
plasma electrons. After the selective-confinement and
a given initial delay time of about 50 ms, the stored
w
x
w
x
Ž1
.
Ti4q ion should be in its 3p6 S ground state. With
91 eV above its neutral ground state the Ti4q in
ground state can undergo charge transfer reaction in
following possible exothermic channels with Ar or
N2 ,
Ti3qqArqqD E
Ti2qqAr2qqD E
ŽD EF27.49eV.
ŽD EF27.34eV.
Ti4qqAr™
1
Ž .
½
Ti3qqNq2 qD E
ŽD EF27.66eV.
ŽD EF19.05eV.
ŽD EF 4.52eV.
ŽD EF31.99eV.
°
Ti3qqNqNqqD E
w
x
experimental values 6,13 . More precise theoretical
calculations need complex quantal models which are
still under development for the treatment of these
charge-transfer collisions at electron-volt energies
Ti4qqN2 ™
~
Ti3qq2NqqeqD E
¢
Ti2qq2NqqD E
2
Ž .
w
x
14–16 .
For Ti4qq Ar , two channels are possible with
similar exothermicity. The first one is a single elec-
tron transfer process but the second is a double
electrons transfer reaction. For Ti4qq N2 , four
channels stated above are possible with different
exothermicity. Since we did not measure the product
ion signal for different channels in the experiment,
the values we gave were the total charge-transfer rate
coefficients of all these reaction channels.
References
w x
Ž
.
1
w x
2
S.E. Butler, T.G. Heil, A. Dalgarno, Ap. J. 241 1980 442.
C.C. Havener, M.S. Huq, H.F. Krause, P.A. Schulz, R.A.
Phaneuf, Phys. Rev. A39 1989 1725.
Ž
.
w x
3
Ž
.
D.A. Church, H.M. Holzsheiter, Chem. Phys. Lett. 76 1980
109.
w x
Ž
.
4
w x
5
D.A. Church, H.M. Holzsheiter, Phys. Rev. A40 1989 54.
V.H.S. Kwong, T.T. Gibbons, Z. Fang, J. Jiang, H. Knocke,
Y. Jiang, B. Ruger, S. Huang, E. Braganza, W. Clark, Rev.
With the AC voltage of 114 V and no DC bias
after the selective-storage the axial potential well
Ž
.
Sci. Instrum. 61 1990 1931.

(
)
302
w x
M. Yan et al.rChemical Physics Letters 289 1998 298–302
w
w
w
x
6
V.H.S. Kwong, Z. Fang, Y. Jiang, T.T. Gibbons, Phys. Rev.
13 J. Wang, Ph.D. Thesis, University of Nevada, Las Vegas,
Ž
.
Ž .
NV 1997 .
A46 1992 201.
w x
7
x
R.A. Phaneuf, I. Alvarez, F.W. Meyer, D.H. Crandall, Phys.
14 M. Gargaud, Bacchus-Montabonel, R. McCarroll, J. Chem.
Ž
.
Ž .
Phys. 99 1993 4495.
Rev. A26 1982 1892.
w x
Ž
.
x
8
J. Wang, V.H.S. Kwong, Rev. Sci. Instrum. 68 1997 3712.
15 P. Honvault, M. Gargaud, Bacchus-Montabonel, R. McCar-
roll, R. 1995, in Proceedings of Nineteenth Conference on
the Physics of Electronic and Atomic Collision, eds., J.B.A.
Mitchell, J.W. McConkey, C.E. Brion, 790.
w x
Ž .
9
R.V. Jensen, D.E. Post, Nucl. Fusion 17 1977 1187.
w
w
w
x
Ž
.
10 D.A. Church, H.G. Dehmelt, J. Appl. Phys. 40 1969 3421.
x
Ž
.
11 L.H. Thomas, Proc. Roy. Soc. A114 1927 561.
x
w
x
12 R.L. David, CRC Handbook of Chemistry and Physics. 78th
16 Y. Sun, H.R. Sadeghpour, K. Kirby, A. Dalgarno, G.P.
Ž
.
Ž .
Lafyatis, Int. Rev. Phys. Chem. 15 1996 53.
edn. CRC Press, Boca Raton, FL, 1997–1998 .