Tetravalent Ions Doped Lithium Niobate Crystals

1
Tetravalent Ions Doped Lithium Niobate Crystals

• Yongfa Kong,
• Shiguo Liu, Shaolin Chen, and Jingjun Xu

School of Physics and TEDA Applied Physics School
2
Outline

• 1. Introduction
• 2. Optical damage resistance
• 3. Photorefraction
• 4. Concluding remarks

3
1. Introduction

• The topic of this workshop is on Optics and New
  materials.
• Lithium niobate crystal is dull compared with the
  vast variability of todays deliberately
  engineered materials.
• Is there any news?

4
Materials Update Material of the month November
2002 Lithium niobate

• In the field of nonlinear optics there have been
  many contenders for the title of all-star
  material of the world.
• But for day-to-day applications, the most
  successful of these nonlinear materials is
  lithium niobate.

• Indeed, because of its availability, widespread
  use and versatility, it has been dubbed by many
  as the silicon of nonlinear optics.

5
Silicon of photonics

• Lithium niobate (LiNbO3), also called the
  silicon of photonics, is indispensable in
  advanced photonics and nonlinear optics.

M. Kösters1, et al., Nature Photonics 3, 510
(2009)
6
Lithium niobate (LiNbO3, LN)

• Multi-functions
• electro-optic, acousto-optic, elasto-optic,
  piezoelectric, pyroelectric, ferroelectric,
  nonlinear optic, etc.
• Multi-applications
• Waveguides, modulators, isolators, frequency
  transformers, optical parametric oscillators,
  filters, sensors, holographic storage, etc.
• Property controllability
• Good solubility to many dopants,
• Properties change with different dopants and
  doping concentrations.

7
Optical silicon

• New materials renew life for integrated optics
  Lawrence Gasman WDM Solutions, November, 2001
• Material systems based on silica on silicon,
  gallium arsenide, lithium niobate, and indium
  phosphide are contenders for the role of "optical
  silicon."

8
Workshop on Optics and New Materials II

• The topics include metamaterials, plasmonics,
  optical lattice, photonic crystals, and novel
  quantum effects of light-matter interaction.

• S. Zhu, et al., Quasi-phase-matched
  third-harmonic generation in a quasi-
• periodic optical superlattice. Science 278,
  843846 (1997).
• N. G. R. Broderick, et al., Hexagonally poled
  lithium niobate a two-
• dimensional nonlinear photonic crystal. Phys.
  Rev. Lett. 84, 43454348 (2000).
• V. Ilchenko, et al., Nonlinear optics and
  crystalline whispering gallery mode
• cavities. Phys. Rev. Lett. 92, 043903 (2004).
• C. Canalias, et al., V. Mirrorless optical
  parametric oscillator. Nature Photon.
• 1, 459462 (2007).
• A. Guarino, et al., Electro-optically tunable
  microring resonators in lithium
• niobate. Nature Photon. 1, 407410 (2007).
• R. C. J. Hsu, et al., All-dielectric
  photonic-assisted radio front-end technology.
• Nature Photon. 1, 535538 (2007).
• W. Yang, et al., Non-reciprocal ultrafast laser
  writing. Nature Photon. 2, 99
• 104 (2008).

9
What have been done on Lithium niobate crystal?

• In 1965, Ballman et al. firstly succeeded in
  growing lithium niobate single crystal
• SAW Filter 45 inch single crystals
• Electro-optic modulator 34 inch single
  crystals
• Photorefraction Fe, Cu, Mn, or Ce doped
  crystals
• Optical damage resistance Mg, Zn, In, or Sc
  doped crystals
• Property enhancement nearly stoichiometric
  crystals
• Optical waveguide H, Ti
• QPM PPLN, PPMgLN
• .

10
Good enough?

• Acoustic grade crystals inhomogeneous stress,
  low electricity
• Optical grade crystals graining stripes
• Photorefraction long response time, low
  sensitivity
• Optical damage resistance poor optical quality,
• 
  only in visible range
• QPMPPLN, low optical damage resistance,
• PPMgLN, hard to fabricate, poor
  thermal stability
• NS crystals very difficult to grow, very poor
  optical quality
• Defect structures
• Energy levels
• Mechanism
• .

11
What can tetravalent dopants do?

• Optical damage resistance
• Photorefraction
• Domain engineering
• Crystal growth
• Micro-mechanism of some effects
• and structural design

12
Optical damage
2. Optical damage resistance

• Light-induced optical damage,
• now also named as photorefraction,
• was discovered in LiNbO3 and LiTaO3
  crystals.

• Photorefraction (PR)
• Can be used in
• holographic storage,
• information processing,
• light control of light.
• low response speed,
• volatility.

• Optical damage
• Hinders the applications
• frequency doublers,
• optical parametric oscillators,
• Q-switches,
• optical waveguides.

A. Ashkin, et al., Appl. Phys. Lett. 9, 72(1966)
13
A solution doping

• 1980, Mg2 ions, LNMg
• Star of China

• It promotes the practical applications of LN in
  nonlinear optics at high light intensities.

• 1990, Zn2 ions, LNZn
• 1992, Sc3 ions, LNSc
• 1995, In3 ions, LNIn.

  G. Zhong et al., J. Opt. Soc. Am. 70,
631 (1980).   T. R. Volk et al., Opt. Lett.
15, 996 (1990). J. K. Yamamoto et al., Appl.
Phys. Lett. 61, 2156 (1992).    Y. Kong et al.,
Appl. Phys. Lett. 63, 280 (1995).
14
The problems of doped LN

• It is difficult to grow high optical quality
  crystals.
• Large amounts of doping concentrations
• (such as usually 5 mol Mg for CLN)
• Distribution coefficient far from 1.0
• (such as 1.2 for Mg)

• Some properties are still not satisfied
• Resistance not high enough,
• Enhanced ultraviolet photorefraction (UVPR).

15
HfO2 doped LiNbO3 (LNHf)
E.P. Kokanyan et al., J. Appl. Phys. 92 1544
(2002) Appl. Phys. Lett. 48, 1980 (2004).
16
Optical damage resistance of LNHf
(a)    2 mol Hf(b) 4 mol Hf(c) 6 mol Hf(d)
6.5 mol Mg The light intensity for (a) is 104
W/cm2 and 5105 W/cm2 for (b), (c), and (d).

• LNHf4 is able to withstand a light density of
  5105 W/cm2 without noticeable beam smearing,
• which is comparable to that observed in 6.5mol
  MgO doped LN (LNMg6.5) crystal.

S. Li et al., J. Phys. Condens. Matter. 18, 3527
(2006).
17
ZrO2 doped LiNbO3 (LNZr)

• As the doping concentration reaches 2.0 mol
  ZrO2, LNZr crystals can withstand a light
  intensity as high as 2.0?107 W/cm2.
• At the same experimental conditions, the light
  intensity that 6.5 mol Mg doped LN (LNMg6.5)
  can withstand is about 5.0?105 W/cm2.

(a)
(c)
(b)
(d)
(a), (b) and (c) LNZr1.7 (d) LN Zr2. The light
intensity for (a) is 1.3?103 W/cm2, (b) 1.3?104
W/cm2, (c) and (d) 2.0?107 W/cm2.
Y. Kong et al., Appl. Phys. Lett. 91, 081908
(2007).
18
Light-induced changes of refractive indices

• As the doping concentration of Zr above 2.0 mol,
  the refractive index changes of LNZr crystals
  are one order of magnitude smaller than that of
  LNHf and LNMg.

• Light-induced change of the refractive index in
  saturation as a function of dopants

19
The distribution coefficient of Zr

• The maximum value is 1.04 and the minimum value
  is 0.97.
• Therefore, the distribution coefficient of Zr is
  much closer to one than that of Mg.

20
SnO2 doped LiNbO3 (LNSn)
Distortion of transmitted argon laser beam spots
after 5 min of irradiation. (a)-(d) for Sn1LN,
Sn2LN, Sn2.5LN, and Sn5LN, respectively. The
light intensities are (a) 2.5102 W/cm2, (b)
4.7103 W/cm2, (c) 4.8105 W/cm2, and (d) 5.4105
W/cm2.
L. Wang et al., Opt. Lett. 35, 883 (2010).
21
(No Transcript)
22
The distribution coefficient of LNSn
Dependence of the distribution coefficient of
Sn4 ions in SnLN crystals on the doping levels
of SnO2.
23
Ultraviolet photorefraction (UVPR)

• Enhancement of UVPR in LNMg

J. Xu, et al., Opt. Lett. 25, 129(2000)
24
Pulsed UV image amplification for programmable
laser marking
A laser at 355 nm, with 5 mJ, 10 ns pulse
duration, a repetition rate of 20 Hz.
25
The UVPR of LNZn and LNIn
H. Qiao, et al., Phys. Rev. B 70, 094101(2004).
26
The resistance of LNZr to UVPR
Fig.2 Beam distortion of the transmitted UV light
passing through LiNbO3 crystals (wavelength 351
nm, intensity 1.6105 W/cm2). (a) PLN (b)
LNZr1 (c) LNZr2 (d) LNZr5.

• Fig.1 The dependence of UV photorefractive
  diffraction efficiency and saturated refractive
  index change of LNZr on the doping concentration
  of Zr.
• The open symbols show the data for LNMg5.

F. Liu, et al., Opt. Lett. 35, 10 (2010)
27
(No Transcript)
28
(No Transcript)
29
The UVPR of LNHf
Fig.1 Distortion of transmitted UV beam spots
after irradiation of 5 min (wavelength 351nm,
intensity 18.5 kW/cm2) ae correspond to LN
doped with 2, 2.5, 3, 4, and 6 mol. Hf.
W. Yan, et al., Opt. Lett. 35, 601 (2010)
30
Comparison of LNMg, LNHf, LNZr and LNSn
Crystals Properties LNMg LNHf LNZr LNSn
Optical damage resistance (W/cm2, 514.5nm) 5?105 5?105 gt2?107 4.8?105
Saturation refractive index change (514.5nm) 7.8 ?10-6 8.4 ?10-6 7.1 ?10-7 7.65 ?10-6
Doping threshold (mol in melt) 4.6 2.5 2.0 2.5
Distribution coefficient 1.2 0.93 0.97 0.98
UV Photorefraction (351nm) 2.1 ?10-5 ____ 1.1 ?10-6 ______

• 6.5 mol MgO 5 mol MgO in melt.

31
Fe2O3 doped LiNbO3 (LNFe)
3. Photorefraction

• By now, Fe2O3 doped LiNbO3 (LNFe) is one of the
  most excellent candidate materials for optical
  data storage due to its
• high diffraction efficiency,
• high data storage density,
• long storage lifetime.

• The problems
• low response speed,
• strong light-induced scattering,
• volatility.

32
A solution to increase the response speed

• Co-doping with damage-resistant elements such as
  Mg, Zn, In and Sc, has been found to be a useful
  way to increase the response speed and resistance
  to scattering.
• When the doping concentrations are above the
  threshold, Fe3 ions and part of Fe2 ions on Li
  sites will be repelled to Nb sites,
• improves the response speed.
• apparently decreases the diffraction efficiency.

G. Zhang, Proc. SPIE 2529, 14 (1995).
33
HfO2 and Fe2O3 co-doped LiNbO3(LNFe,Hf)
Samples Doping concentrations Doping concentrations Doping concentrations Photorefractive properties Photorefractive properties Photorefractive properties
Samples Fe (wt.) Mg (mol) Hf (mol) ?sat () tr (s) S (cm/J)
LNFe 0.01 70 160
LNFeMg2 0.01 2 70 60
LNFeMg6 0.01 6 15 15
LNFeHf2 0.03 2 68.0 17.2 3.99
LNFeHf4 0.03 4 47.6 12.6 4.36
LNFeHf5 0.03 5 55.4 10.7 5.23
S. Li, et al., Appl. Phys. Lett. 89, 101126 (2006)
34
ZrO2 and Fe2O3 co-doped LiNbO3(LNFe,Zr)
Samples Doping concentrations Doping concentrations Doping concentrations Photorefractive properties Photorefractive properties Photorefractive properties
Samples Fe (wt.) Mg (mol) Zr (mol) ?sat () tr (s) S (cm/J)
LNFe 0.01 70 160
LNFe,Zr1 0.03 1 25.5 2.2 13.46
LNFe,Zr2 0.03 2 32.0 1.8 12.87
LNFe,Zr3 0.03 3 32.7 1.8 13.48
LNFe,Zr4 0.03 4 32.5 1.8 13.40
LNFe,Zr5 0.03 5 42.2 2.2 12.61
Y. Kong et al., Appl. Phys. Lett. 92, 251107
(2008)
35
The OH- absorption spectra of LNFe,Zr
3507 cm-1 Fe3 in Nb-site
LNFe,Zr from top to bottom are for 1, 2, 3, 4,
and 5 mol Zr, respectively 0.03 wt Fe
LNFeMg
36
The UV-Visible spectra of LNFe,Zr and LNFe,Hf
400700 nm Fe2?Nb5 intervalence transfer

• Fe2/3 ions still remain at Li sites when the
  doping concentration of ZrO2 or HfO2 goes above
  its threshold value!

LNFe, Zr A, B, C, D, and E are for 1, 2, 3, 4,
and 5 mol Zr, and X and Y are for 2 and 5 mol
Hf, respectively 0.03 Fe.
37
Comparison of LNFe, LNFe,Mg, LNFe,Hf and
LNFe,Zr
Samples Doping concentrations Doping concentrations Doping concentrations Doping concentrations Photorefractive properties Photorefractive properties Photorefractive properties
Samples Fe (wt.) Mg (mol) Hf (mol) Zr (mol) ?sat () tr (s) S (cm/J)
LNFe 0.01 70 160
LNFeMg6 0.01 6 15 15
LNFeHf5 0.03 5 55.4 10.7 5.23
LNFeZr2 0.03 2 32.0 1.8 12.87
S. Li, et al., Appl. Phys. Lett. 89, 101126
(2006) Y. Kong et al., Appl. Phys. Lett. 92,
251107 (2008)
38
Nonvolatile holographic storage

• LiNbO3Fe,Mn

one-center two-center
K. Buse, et al., Nature 393, 665 (1998)
39
Energy level diagram of LiNbO3
Conduction band
Conduction band
1.6 eV
NbLi4/5
2.5 eV
2.6 eV
2.6 eV
2.8 eV
2.8 eV
NbLi4/5 NbNb4/5
EFermi
EFermi
Fe2/3
Fe2/3
Mn2/3
Mn2/3
CLNMn,Fe
LNZr,Fe,Mn

• The co-doping of Zr eliminates undesired
  intrinsic electron traps, which greatly enhances
  the charge transition speed for nonvolatile
  holographic storage

40
LiNbO3Zr,Fe,Mn
Oxidation time Irec/Isen (mW/cm2) ?s () ?f () S (cm/J) ?r (s)
24h 800/40 54.3 14.9 0.65 2.4
24h 600/40 52.1 14.5 0.88 2.2
24h 400/40 62.0 13.6 1.78 1.2
48h 400/40 57.0 7.8 1.13 2.0
20h 400/40 62.5 14.0 2.10 0.88
Y. Kong et al., Opt. Lett 34, 3896 (2009)
41
Comparison of LNZr,Fe,Mn, LNMg,Fe,Mn, and
LNIn,Fe,Mn
42
LiNbO3Zr,Cu,Ce
Oxidation Time Isen/Irec (mW/cm2) ?sat () ?non () S (cm/J) S (cm/J)
13h 40/400 62.4 6.3 0.312 0.099
24h 40/400 72.6 6.6 0.079 0.024
24h 40/600 74.2 4.3 0.033 0.008
24h 40/800 72.7 3.0 0.025 0.005
The light intensity dependence of the measured
light-induced scattering in the samples of triply
doped LiNbO3 crystals. The lines are guides to
the eyes.
F. Liu et al., Opt. Express 18, 6333 (2010)
43
The sensitivity of LiNbO3 co-doped with different
ions for nonvolatile holographic storage
Crystal component S(cm/J) S(cm/J) Reference
LiNbO3Fe,Mn - 0.07 K. Buse, et al., Nature 393, 665 (1998)
sLN(Li/Nb49.65/50.35) 0.03 L. Hesselink, et al., Science 282, 1089 (1998)
LiNbO3Cu,Ce 0.022 - Y. Liu, et al., Opt. Lett. 25, 908 (2000).
LiNbO3Fe,Cu 0.035 - D. Liu, et al., Appl. Opt. 41, 6809 (2002).
LiNbO3Ce,Mn 0.0025 - Q. Dong, et al., Appl. Opt. 43, 5016 (2004).
sLiNbO3Cu,Ce (Li/Nb49.57/50.43) - 0.107 X. Li, et al, Appl. Opt. 46, 7620 (2007).
LiNbO3Mg,Fe,Mn 0.047 - W. Zheng, et al., Cryst. Res. Tech. 43, 526 (2008).
LiNbO3Zr,Fe,Mn 3.47 1.31 Y. Kong et al., Opt. Lett 34, 3896 (2009)
LiNbO3Zr,Cu,Ce 0.312 0.099 F. Liu et al., Opt. Express 18, 6333 (2010)
44
4. Concluding remarks

• The above results indicate that tetravalent ions
  are excellent choice for the control of optical
  damage or photorefraction of LN.
• These results also open a door for us to
  understand the micro-mechanism of optical damage
  resistance.
• These results give us suitable choices for
  crystal design.

The question remains Why LNZr has such
outstanding properties as compared with LNHf,
LNSn, and LNMg?
45
Silicon single crystal
Fig. 2. Cross-sectional view of the defect-free,
near-surface region of a silicon wafer. The lower
portion of the figure shows silicon dioxide
precipitates used for impurity gettering.
Fig. 1. Range of electrical resistivities of pure
and donor-doped silicon single crystals shown in
comparison with metals and insulators.
H. Queisser, et al., Science 281, 945 (1998)
46
Optical fiber

• In 1966, Prof. Kao and Hockham proposed that when
  the loss of glass fiber was less than 20 dB/km it
  could be used as a conductor for optic
  communication, however at that time the loss of
  the best optical glass in the world was as large
  as 1000 dB/km.

• In 1970,Corning Incorporated made optical fibers
  with loss of 20dB/km.
• In 1974, the loss of optical fiber reduced to 2
  dB/km as the purity of raw materials increased to
  8N.
• In 1976, the loss of optical fiber reduced to 0.5
  dB/km as the concentration of OH in raw materials
  controlled in the order of ppm.
• In 1980, the transport loss of optical fiber
  dropped to only 0.2 dB/km, which is closed to the
  theoretical value of 0.15dB/km.

47
How about lithium niobate crystals?

• Though lithium niobate has been dubbed as
  optical silicon or photonic silicon, compared
  with silicon single crystal and optical fiber,
  its research is rather preliminary.
• We do not exactly know
• the defect structures, even the intrinsic
  defects,
• the function of every dopant,
• the relationship between defects and optical or
  photonic properties.
• We are far from what we expect
• The control of defects
• The growth of high quality single crystals.

• Our dream!

48
Thank you for your attention!