Reduction of laser-induced dark traces in LiNbO
3
:Mg and LiNbO
3
:Zn
by heat treatment
Jiachun Deng,
a)
Yongfa Kong,
b)
Jiang Li,
c)
Jinke Wen, and Bing Li
Department of Physics, Nankai University, Tianjin 300071, People’s Republic of China
͑Received 8 December 1995; accepted for publication 6 March 1996͒
The dark traces induced by high-power laser radiation in LiNbO
3
:Mg and LiNbO
3
:Zn are found to
be reduced by annealing the crystals in an O
2
atmosphere. Most of the decrease of the
second-harmonic generation efficiency
for 1064–532 nm conversion due to dark traces can be
recovered in heat-treated crystals. The transmission loss spectra ⌬T() of dark traces show a
complex absorption band ranging from 365 nm to near infrared, which can be decomposed into
three main peaks at 3.1, 2.5, and 1.6 eV. Maximum transmission loss due to dark traces after 1200
shots with an input power density of 18 MW/cm
2
is 17% and 25% for as-grown LiNbO
3
:Zn and
LiNbO
3
:Mg crystals respectively, and it is reduced to 4.6% and 3.8% respectively for heat-treated
crystals. © 1996 American Institute of Physics. ͓S0021-8979͑96͒01912-3͔
I. INTRODUCTION
Lithium niobate ͑LiNbO
3
) crystal is an excellent mate-
rial for linear and nonlinear optical devices, but its utility in
some applications such as a second-harmonic generator is
limited by the photorefractive effect. Nominally congruent
LiNbO
3
doped with Mg higher than the concentration thresh-
old ( Ͼ 4.6mol% in the melt͒ was found to exhibit a re-
marked reduced photorefractive response compared with un-
doped LiNbO
3
.
1–3
Furthermore, LiNbO
3
:Zn ( Ͼ 6.2mol% in
the melt͒ has also been found to be able to withstand much
higher laser intensity than pure LiNbO
3
.
4,5
However, there is
another type of optical damage, i.e. dark trace ͑also named
bulk darkening trace or grey track͒, which is induced by a
high-power laser beam in LiNbO
3
:Mg.
4,6
For LiNbO
3
:Zn,
Volk et al.
4,5
claimed that no dark trace occurs under an
incident intensity up to 120 MW/cm
2
.
In a previous paper,
7
we demonstrated that both
LiNbO
3
:Zn and LiNbO
3
:Mg crystals suffer from pulsed-
laser-induced dark traces, which reduce the second-harmonic
generation ͑SHG͒ efficiency, though the effect is slightly
weaker in the former than in the latter. The dark trace is
induced by 532 nm radiation in the 1064-532 nm SHG, and
can be recovered after the laser beam is shut off. The trans-
mission loss spectra of dark traces in LiNbO
3
:Mg and
LiNbO
3
:Zn show a broad absorption band ranging from the
UV absorption edge to near infrared. Li et al.
8
recently stud-
ied laser-induced dark traces in doped LiNbO
3
crystals more
carefully, with the 1064 nm fundamental radiation and the
532nm second-harmonic radiation of the Nd:YAG laser op-
erating in a pulse mode. Their results show that the dark
traces are induced only by the 532 nm radiation, and are not
due to two-photon absorption. The mechanism responsible
for dark traces in LiNbO
3
crystals is not understood yet. The
aim of this work is to find a way to reduce the dark traces in
LiNbO
3
:Mg and LiNbO
3
:Zn occurring under intensive laser
radiation. Following the results of Deng et al.
7
and Li et al.,
8
one can easily see that the transmission loss spectra of dark
traces in doped LiNbO
3
crystals are very similar to those of
codoped congruent materials by means of reduction treat-
ment in an argon atmosphere at a temperature near 1100 °C.
In this article, we therefore intend to investigate the influence
of heat treatment in an oxygen atmosphere on the induced
dark traces in our doped LiNbO
3
samples. The transmission
loss spectra has also been discussed.
II. EXPERIMENT
The LiNbO
3
:Mg ͑6.5 mol %͒ and LiNbO
3
:Zn
͑7.5 mol %͒ samples used in the experiments are the same as
used in Ref. 7. The crystals measured 5 mmϫ5mmϫ15 mm
with the y axis parallel to the longest dimension. Angle
phase matching for 1064–532 nm SHG can be achieved at
room temperature, and was realized with the propagation di-
rection of the input beam at
m
ϭ78.6° for LiNbO
3
:Zn and
82.3° for LiNbO
3
:Mg. Here the samples were first placed in
an open alumina tube and annealed in 1 atm pressure of pure
oxygen at 1050 °C for 24 h, and then cooled slowly to room
temperature. After this process, the samples were polished to
optical quality again.
The SHG efficiency
was measured using a Q-switched
Nd:YAG laser operating at 1 Hz with nominal 20-ns pulse
width and the crystals ͑15 mm long͒ as a SHG component.
The dimension of the fundamental input beam was 4 mm.
The crystals were examined for SHG efficiency
for the
first shot and after 3600 shots. The optical transmission spec-
tra were measured after 1200 shots by Shimadzu Spectro-
photometer ͑UV-365͒.
III. RESULTS
The variation of SHG efficiency
with the input power
density for as-grown and heat-treated samples is listed in
Table I. The
of the first shot is about the same for as-
grown and heat-treated samples. No visible dark traces are
observed at this moment, and the maximum value of
is
about 48% for LiNbO
3
:Zn and 45% for LiNbO
3
:Mg at an
input power density of 27.5 MW/cm
2
. After irradiating 3600
a͒
Present address: Department of Physics, Tianjin Normal University,
300074, Tianjin, P.R. China.
b͒
Present address: Department of Materials Science, Tianjin University,
300070, Tianjin, P.R. China.
c͒
Corresponding author; Electronic mail: Fengcb@bepc2.ihep.ac.cn
9334 J. Appl. Phys. 79 (12), 15 June 1996 0021-8979/96/79(12)/9334/4/$10.00 © 1996 American Institute of Physics