See discussions, stats, and author profiles for this publication at: [Link]

net/publication/258666945

Characterization and inhibition of photorefractive optical damage of swift

heavy ion irradiation waveguides in LiNbO3

Article in Journal of the Optical Society of America B · October 2012

DOI: 10.1364/JOSAB.29.003000

CITATIONS READS

3 66

5 authors, including:

Mariano Jubera Angel García-Cabañes

Centro de Láseres Pulsados Ultracortos Ultraintensos Universidad Autónoma de Madrid
14 PUBLICATIONS 283 CITATIONS 106 PUBLICATIONS 1,542 CITATIONS

SEE PROFILE SEE PROFILE

Mercedes Carrascosa
Universidad Autónoma de Madrid
197 PUBLICATIONS 2,805 CITATIONS

SEE PROFILE

All content following this page was uploaded by Mariano Jubera on 14 June 2018.

The user has requested enhancement of the downloaded file.

3000 J. Opt. Soc. Am. B / Vol. 29, No. 11 / November 2012 Jubera et al.

Characterization and inhibition of photorefractive

optical damage of swift heavy ion irradiation
waveguides in LiNbO3

Mariano Jubera,1,* Angel García-Cabañes,1 Mercedes Carrascosa,1 José Olivares,2,3 and Fabian Lüedtke4
1
Departamento de Física de Materiales, Universidad Autónoma de Madrid, Madrid 28049, Spain
2
Centro de Microanálisis de Materiales (CMAM), Universidad Autónoma de Madrid, Madrid 28049, Spain
3
Instituto de Optica, Consejo Superior de Investigaciones Científicas (CSIC), Serrano 121, Madrid 28016, Spain
4
Institute of Physics, University of Bonn, Wegelerstraße 8, Bonn D-53115, Germany
*Corresponding author: [Link]@[Link]

Received July 6, 2012; revised September 4, 2012; accepted September 5, 2012;

posted September 5, 2012 (Doc. ID 172137); published October 3, 2012
The photorefractive effect and the corresponding optical damage thresholds of novel LiNbO3 waveguides fabri-
cated by swift ion irradiation have been investigated. TE- and TM-mode operation have been characterized, and
the influence of the beam propagation length analyzed. Optical damage levels similar to those of proton-exchanged
waveguides have been found. In order to reduce optical damage, the influence of temperature has been investi-
gated. An increase of more than a factor of 100 in the optical damage threshold has been obtained by moderate
heating up to 90°C. The results are briefly discussed under the two-center model for the photorefractive effect in
undoped LiNbO3 , and compared with data from other types of LiNbO3 waveguides. © 2012 Optical Society of
America
OCIS codes: 130.3730, 190.5330, 230.4320, 230.7370.

1. INTRODUCTION with heavy-mass ions with energies in the range 5–50 MeV,
Photorefractive optical damage (POD) is a main drawback for requiring much lower irradiation fluences (1–4 × 1014 cm−2 )
high power photonic devices based on LiNbO3 crystals and and, thus, much shorter irradiation times in comparison with
waveguides [1–5]. The photorefractive effect consists in a similar guides prepared by conventional ion implantation. The
light-induced refractive index change Δn arising from a cas- crystal amorphization that gives rise to the optical barrier is
cading process that combines light-induced charge transport produced by electronic excitation processes. In other words,
inside the crystal with the electro-optic effect [6]. It produces the method relies on the electronic energy deposition (stop-
beam degradation during propagation and light intensity lim- ping power) at variance with light ion implantation based on
itation effects, i.e., what is generally called POD. Thus, there elastic nuclear collisions (nuclear stopping power). Moreover,
has been much activity devoted to characterizing and redu- due to the new amorphization mechanism of electronic in-
cing it (see, for instance, [2–5,7]). Particular attention should stead of nuclear origin, the thickness of the optical barrier
be paid to waveguide configurations, because the long propa- is easily programmable and can reach much higher values
gation lengths and the high intensities reached increase opti- (up to 3–5 μm) than in the case of the usual ion implantation
cal damage effects. In fact, a large number of works are waveguides [19]. The guides support ordinarily and extraordi-
devoted to this subject, although the data from different narily polarized modes and, unlike other nonlinear LiNbO3
authors often show a considerable variability and even some waveguides, they show for both polarizations step-like,
contradictions (see [2] and references therein). Most experi- high-jump index profiles (Δne
0.1, Δno
0.2). Initially,
ments have been performed in Ti-indiffused and proton- the guides presented moderate/high propagation losses
exchanged (PE) waveguides [3,8,9] whereas the information (1–10 dB∕cm) [17,18], but very recently, using higher tem-
is scarcer for ion-implanted waveguides [3,10]. Regarding the perature annealing treatments (350°C–375°C) and thick
theoretical description of POD in undoped LiNbO3 wave- enough amorphization barriers, propagation losses were re-
guides, a full understanding of the different manifestations duced to below 0.5 dB∕cm [20]. Furthermore, good nonlinear
of this phenomenon has been lacking for a long time. Only optical (χ 33 ) and electro-optic (r 33 ) coefficients have been re-
recently has optical damage been revised [11–13] to the light ported [19,21], so that the novel waveguides have become
of a two-center band transport model previously proposed for good candidates for nonlinear devices. However, POD effects
photorefractive charge transport in LiNbO3 bulk crystals [14]. have been studied only preliminarily and using nonoptimized
This theoretical approach has been able to accurately waveguides with high propagation losses of 1–5 dB∕cm [21].
describe a variety of experimental features of POD [12,13], Moreover, those data were taken from an X-cut configuration,
allowing for a better understanding of the phenomenon. although Z-cut geometries are usually preferred for efficient
Over the past 10 years, a new method of producing non- nonlinear applications.
linear LiNbO3 optical waveguides using swift ion irradiation Therefore, the aim of this work is to address a detailed and
has been reported [15–19]. It involves substrate irradiation systematic investigation of POD of swift heavy ion irradiation

0740-3224/12/113000-06$15.00/0 © 2012 Optical Society of America

Jubera et al. Vol. 29, No. 11 / November 2012 / J. Opt. Soc. Am. B 3001

LiNbO3 planar waveguides. The POD was studied by two com-

plementary techniques: (i) interferometric measurement of
the light-induced refractive index changes responsible for
the beam damage and (ii) determination of the corresponding
optical damage thresholds, i.e., maximum light intensity
supported by the waveguide without distortion. In addition,
we have observed the distorted output beam profile/spot to
complete the information on beam degradation. TE- and
TM-mode operation have been characterized and the role
of the propagation length analyzed. The influence of increas-
ing temperature is also investigated, showing that this method
allows us to considerably reduce optical damage. The results
are discussed in the framework of the recently reported two-
center model for POD in undoped LiNbO3 waveguides [12,13]
and compared with data from other waveguides.

2. EXPERIMENTAL METHODS
A. Waveguide Fabrication and Basic Characterization
1. Sample Fabrication
The waveguides have been fabricated by irradiation with
fluorine ions with an energy of 30 MeV at an incidence angle Fig. 1. (a) Schematic of the interferometric setup to measure photo-
of 70° and a fluence of 3 × 1014 cm−2 , on Z-cut congruently refractive index changes, (b) single beam setup to determine light
melting lithium niobate substrates purchased from Photox IDTs. P, polarizer; M, mirror; BS, beam splitter; MO, microscope ob-
jective; D, diaphragm; L, lens.
Optical Systems (Oxford, UK). Irradiations have been carried
out in the 5 MV tandem accelerator of the CMAM at the Uni-
versity Autónoma of Madrid [22]. In order to reduce propaga-
tion losses [20] after irradiation, samples have been subjected are coupled into the waveguide through a rutile prism, excit-
to an annealing treatment in air at 350°C for 90 min. ing the fundamental mode in both cases. When the green
pump light is switched on, the effective index in the probe
2. Waveguide Characterization arm changes and a difference in the relative phase of the
The waveguides have been characterized by measuring the wavefronts is introduced. With a small diaphragm placed in
refractive index TE (ordinary refractive index no ) and TM front of the detector, it measures the intensity change due
(extraordinary refractive index ne ) profiles using the prism- to movement of the light interference pattern.
coupling m-line method with λ
632.8 nm. The fabricated A standard single-beam method using incoupling and out-
waveguides support six ordinary modes and five extraordin- coupling rutile prism couplers, as described in [9] and sche-
ary modes. Their profile is step-like, with a thickness of 2.4 μm matically shown in Fig. 1(b), was used to determine light IDTs.
and refractive index jumps of 0.1 and 0.2 for extraordinary and The IDT is defined by the incoupled intensity I in at which the
ordinary polarization, respectively. outcoupled intensity I out is no longer proportional to I in . Then,
The propagation losses were determined through the decay this magnitude determines the range of intensities at which
of the light intensity of the guided mode at λ
632.8 nm, mea- one can safely work without beam degradation along propa-
sured via the light scattered along the beam path recorded by gation. A lens is placed before the input rutile prism to facil-
a CCD camera [23]. After the annealing treatment, the wave- itate reaching high incoupled intensities. A long focal length
guides present very low propagation losses below 0.5 dB∕cm (300 mm) is chosen so that the beam is collimated along the
for both TE and TM polarizations. guide propagation length, and one single mode (the funda-
mental) is excited. While continuously illuminating the wave-
B. Optical Damage Measurements guide with light at λ
532 nm, the output power passing
Optical damage has been characterized using two comple- through a diaphragm placed 20 cm behind the waveguide is
mentary techniques: (i) interferometric measurements of monitored. The size of the diaphragm is chosen such that
the photorefractive index change and (ii) determination of at low intensities (no optical damage), about 70% of the total
the light intensity damage thresholds (IDTs) (see below). outcoupled power is transmitted. This way, the dependence
The photorefractive index change has been measured using I out versus I in loses linearity when the beam starts to distort.
a Mach–Zehnder interferometer [24] recording the time evolu- More information about the light beam quality, particularly
tion of the light-induced phase shift Δϕ between the signal above the IDT, can be obtained by a profilometer that
and reference beams until saturation. From this Δϕ the aver- captures a digital image of the outcoupled beam.
age jΔnj along the beam path l is easily obtained as The determination of the light intensity inside the wave-
jΔnj
λΔϕ∕2πl. A schematic illustration of the setup is guide is a key point in both experimental techniques. In all
shown in Fig. 1(a). Optical damage is induced by a green laser data presented in this work, we evaluate the light intensity in-
beam (λ
532 nm, a wavelength very effective in generating side the waveguide I in just at the input point, i.e., next to the
the photorefractive effect). For the probe beam, a low inten- first coupling prism, using the procedures describe in [25].
sity 633 nm He–Ne laser is chosen because of the low photo- With this method a 10%–15% error in the determination of light
refractive sensitivity at this wavelength. The two laser beams intensity is estimated.
3002 J. Opt. Soc. Am. B / Vol. 29, No. 11 / November 2012 Jubera et al.

3. RESULTS no
2
A. Photorefractive Index Changes 10 ne
The light-induced refractive index change Δn experimented

Iout (W/cm )
2
by the He–Ne probe beam versus the incoupled light intensity 1
10
I in for TM (ne ) and TE (no ) damage beams (fundamental
modes at λ
532 nm) has been measured using the Mach–
Zehnder interferometer. In each case, the reading beam has 0
10 0 1 2 3
10 10 10 10
the same polarization as the green pump beam and the pro- 2
Iin (W/cm )
pagation length is 7 mm. A logarithmic plot of jΔnj (note that
Fig. 3. (Color online) Logarithmic plot of the output intensity I out
Δn < 0 [6]) versus I in for the two polarizations is shown versus input intensity I in inside the waveguide for TE (squares)
in Fig. 2. and TM (circles) propagating fundamental modes.
In both curves one could distinguish three regions. In
the first region (I in < 40 W∕cm2 ), the values of jΔnj are
nearly independent of I in as predicted by the standard one- curves for TE [4(a)] and TM [4(b)]. A decrease of the IDT
center model of the photorefractive (PR) effect [6]. From as l increases is observed. The rate between the two extreme
I in > 40 W∕cm2 , jΔnj markedly increases (region II), and l values is about a factor of 10 for TM polarization and slightly
finally (region III) the curves tend to saturate. Comparing lower for TE modes. This dependence with the propagation
the two curves in the low intensity region, jΔno j is substan- length can be easily understood by taking into account that
tially lower than jΔne j (by roughly a factor 3). This difference for moderate and especially for high I, the beam undergoes
is consistent with the different magnitude of the electro-optic photorefractive self-defocusing due to the refractive index
effect that gives rise to jΔnj. Conversely, in the high intensity change. This is a well known phenomenon that has been
region both values are unexpectedly similar, reaching a value thoroughly studied in previous work on the subject (see,
of jΔnj ∼ 2.5 × 10−5 . These trends will be further discussed for instance, [26,27]). Therefore, the degradation of the beam
below in Section 4. is more pronounced as l increases in accordance with the
lower values of IDT.
B. Light Intensity Damage Thresholds
Light IDTs have been measured for TE and TM fundamental D. Beam Degradation above the Optical Damage
modes for nearly the same propagation length as in Fig. 2 (in Threshold
this case l
8 mm). The results are plotted in Fig. 3. The IDT In order to get more information about the degradation of the
values, for which the linearity with I in is lost, are ∼50 and beam profile above the IDT, we have monitored the output
∼150 W∕cm2 for TM (ne ) and TE (no ) beam propagation, beam profiles at a number of increasing intensities with a
respectively. Note that both IDT values are in the initial CCD camera. The obtained images for TM polarization (ne )
part of region II jΔnI in j, appearing at a similar value of and 8 mm propagation length are shown in Fig. 5. It can be
jΔnj ∼ 10−4 (see Fig. 2). The lower intensity threshold for seen that above the IDT [IDT
50 W∕cm2 from Fig. 4(b)],
TM modes is consistent with the higher photorefractive index the output beam starts to enlarge perpendicular to the z-axis
change exhibited in Fig. 2. due to a self-defocusing effect corresponding to the photore-
fractive decrease of n in the illuminated region. For higher
C. Role of the Propagation Length intensities the beam filaments show three, four, and more
As POD is a nonlinear effect, one could expect that the optical
damage threshold is affected by the propagation length. How- 3
10 (a) 5 mm
ever, this important aspect has rarely been considered in pre-
vious work. In fact, in some works the propagation distance of 8 mm
Iout (W/cm )

2
2

10 16 mm
the experiments is difficult to find or simply is not specified.
To investigate this point, optical damage thresholds have been
1
10
measured for TE and TM modes at two additional optical
lengths l
5 and l
16 mm. Figure 4 shows the obtained
0
10 0 1 2 3
10 10 10 10
2
Iin (W/cm )
no (TE)
ne (TM)
(b) 5 mm
2
10
1 8 mm
-4
|∆n| x10

Iout (W/cm )
2

16 mm

1
10

0,1 1 2 3
10 10 10 1
10 10
2
10
3
2 2
Iin(W/cm ) Iin (W/cm )

Fig. 2. (Color online) Logarithmic plot of the absolute value of the Fig. 4. (Color online) Logarithmic plot of the output intensity I out
photorefractive index change jΔne j (circles) and jΔno j (triangles) ver- versus input intensity I in inside the waveguide for three different pro-
sus the incoupled light intensity I in inside the waveguide. (The curves pagation lengths: (a) ordinary polarization (TE), (b) extraordinary
are only guides to the eye.) polarization (TM).
Jubera et al. Vol. 29, No. 11 / November 2012 / J. Opt. Soc. Am. B 3003

Fig. 5. As expected, at 90°C a totally undistorted beam is

recovered, since for this temperature the IDT appears at a
higher intensity I in ∼ 10000 W∕cm2 [see Fig. 6(a)].

6 W/cm2 74 W/cm2 148 W/cm2 526 W/cm2 1200 W/cm2 4. DISCUSSION

Fig. 5. (Color online) Output beam spot images at increasing inten- The wide set of experimental data presented in this work pro-
sities for the fundamental TM mode and a propagation length vides an in depth characterization of POD in SHI LiNbO3
l
8 mm. The vertical direction is parallel to the guide plane and per-
pendicular to the propagation direction. waveguides, including photorefractive index changes, optical
damage thresholds, and beam degradation profiles for inten-
sities above the IDT. As high intensities are reached in the ex-
spots while progressively degrading. This surprising behavior periments, a first point to clarify is the possible contribution of
has been already observed in Z-cut α-phase PE waveguides thermo-optic effects to the refractive index change. However,
[28] and successfully explained by the two-center model [13]. since the thermo-optic coefficient of undoped LiNbO3 is po-
sitive [32], thermal refractive index changes should be also
E. Influence of Waveguide Temperature: Optical positive, leading to self-focusing instead of the observed
Damage Inhibition self-defocusing behaviors. Therefore, in the range of intensi-
Previous results show moderate POD levels in comparison ties of our experiments, relevant thermo-optic contributions
with other types of LiNbO3 waveguides, but they are still can be discarded.
too high for efficient high power operation. Heating the sam- The obtained data (both jΔnj and IDT) for the extraordin-
ple has been reported as a method of reducing the POD in ary polarization are roughly similar to those reported for
LiNbO3 crystals [29] and PE waveguides [30,31]. Furthermore, α-phase PE waveguides [9,13,31] that only support this polar-
in the two-center model this reduction is well explained ization. Note that this implies that such SHI waveguides as
through the effect of the enhanced thermal excitation of elec- α-phase PE guides [8,33] are better than Ti-indiffused and bulk
trons from NbLi centers [12]. Thus, we have investigated crystals regarding POD. For TE modes (no ) the photorefrac-
whether this method also applies to swift heavy ion (SHI) wa- tive index change (see Fig. 3) is smaller at lower and moderate
veguides. The dependence of the IDT with temperatures in the intensities, although it saturates at higher intensities at
range 25°C–90°C has been determined in an SHI LiNbO3 wa- roughly the same Δn as TM modes (ne ). To explain this non-
veguide [see Fig. 6(a)] for extraordinary polarization. We can trivial behavior, one has to consider that the electro-optic
observe an increase of the IDT of more than a factor of 100. In effect writes
fact, for moderate temperatures of 90°C, POD appears at re-
markably higher intensities of about 10 KW∕cm2 . −1 3
Δn
n rE; (1)
To further illustrate the role of temperature, we show in 2
Fig. 6(b) how a highly distorted output beam of I in

1200 W∕cm2 at RT (corresponding to that of the last image E being the photorefractive space charge field, r the electro-
in Fig. 5) evolves as the temperature of the waveguide in- optic coefficient, and n the average refractive index for each
creases without changing intensity. One can clearly appreci- polarization. In principle, one could expect a similar photore-
ate that beam degradation reduces following a sequence fractive field E [6] for both polarizations, because the photo-
roughly the opposite of that appearing on increasing I in in voltaic effect, the main charge transport mechanism giving
rise to E in LiNbO3 [6], is nearly the same for both polariza-
tions [34]. Then, using the values of the electro-optic coeffi-
(a) 90ºC cients and refractive indexes of LiNbO3 , one obtains
n3e r 33 ∕n3o r 13
2.8, in good accordance with the factor of ∼3
3
75ºC
10 observed between the two curves for low intensity (see Fig. 2).
Iout (W/cm2)

50ºC
On the other hand, the saturating region at high I values ap-
2 25ºC
pears when the negative Δn profile is high enough to induce
10
on the beam an important self-defocusing, and thus a decay of
the light intensity along propagation. In other words, satura-
1
10
1 2 3 4
tion is produced by the POD. As can be seen in Fig. 2, this
10 10 10
2
Iin (W/cm )
10
effect occurs for a value of jΔnj ∼ 2.5–3 × 10−4 , and, as ex-
(b) pected, it does not depend on the beam polarization. Ob-
viously, from Eq. (1) the space charge field E responsible
for this jΔnj should be higher for ordinary polarization. A si-
milar saturating behavior of jΔnj has been reported for PE
waveguides, but in holographic experiments [35].
The shape of the obtained dependence of jΔnj on I (see
25ºC 50ºC 75ºC 90ºC Fig. 2), already observed in α-phase PE waveguides [24], is
Fig. 6. (Color online) (a) Logarithmic plot of the output intensity I out well explained with the two-center model for the photorefrac-
versus input intensity I in inside the waveguide for TM polarization and tive effect in LiNbO3 waveguides recently reported [13].
for four different temperatures in the range 25°C–90°C (l
8 mm),
(b) corresponding output beam spot images at increasing tempera-
According to this model, in the low intensity region, only
tures. The vertical direction is parallel to the guide plane and perpen- the Fe impurity plays a role (one-center approach) [6], but
dicular to the propagation direction. as the intensity grows, a secondary center (niobium in the
3004 J. Opt. Soc. Am. B / Vol. 29, No. 11 / November 2012 Jubera et al.

lithium site) contributes increasingly to jΔnj [11–13]. Thus, 4. S. M. Kostritskii, “Photorefractive effect in LiNbO3 -based
these data can be considered a further support for the validity integrated-optical circuits at wavelengths of third telecom
window,” Appl. Phys. B 95, 421–428, (2009).
of the two-center photorefractive model to describe POD in 5. M. Kösters, B. Sturman, P. Werheit, D. Haertle, and K. Buse,
undoped LiNbO3 waveguides. “Optical cleaning of congruent lithium niobate crystals,” Nat.
The obtained results for the dependence of IDT on the Photonics 3, 510–513 (2009).
propagation length l prove the key role of this geometrical 6. F. Agulló-López, G. F. Calvo, and M. Carrascosa, “Fundamentals
of photorefractive phenomena,” in Photorefractive Materials
parameter that should be specified in any experiment of op- and Their Applications I, P. Günter and J. P. Huignard, eds.
tical damage of waveguides. This result should be general and, (Springer, 2006), pp. 43–77.
thus, applicable to all types of LiNbO3 guides. In fact, the large 7. J. R. Schwesyg, M. Falk, C. R. Phillips, D. H. Jundt, K. Buse, and
dispersion of optical damage data between some papers M. M. Fejer, “Pyroelectrically induced photorefractive damage
[2,8,9] might be explained, at least partially, as a consequence in magnesium-doped lithium niobate crystals,” J. Opt. Soc.
Am. B 28, 1973–1987 (2011).
of the different propagation length of each experimental con- 8. A. Yamada, H. Tamada, and M. Saltoh, “Photorefractive damage
figuration. in LiNbO3 thin-film optical waveguides grown by liquid phase
Finally, the 100 enhancement factor of IDT from RT to 90°C epitaxy,” J. Appl. Phys. 76, 1776–1783 (1994).
is an important result in order to avoid optical damage for 9. O. Caballero-Calero, A. Alcázar, A. García-Cabañes, J. M.
Cabrera, and M. Carrascosa, “Optical damage in X-cut proton
practical applications. This effect has been previously inves-
exchanged LiNbO3 planar waveguides,” J. Appl. Phys. 100,
tigated for bulk crystals [29] and waveguides [36], but IDT 093103 (2006).
changes seem to be smaller (the enhancement factor is 10. F. Cheng, “Photonic guiding structures in LiNbO3 crystals
∼10) for similar or even larger temperature changes. Other produced by energetic ion beam,” J. Appl. Phys. 106, 081101
inhibition methods, such as substrate doping with damage re- (2009).
11. J. Carnicero, O. Caballero, M. Carrascosa, and J. M. Cabrera,
sistant impurities (Mg, Zn, etc), could also be applied to SHI “Superlinear photovoltaic currents in LiNbO3 : analyses under
waveguides, although they have not been used so far. Further the two-center model,” Appl. Phys. B 79, 351–358 (2004).
work in this direction should also be relevant and may further 12. M. Carrascosa, J. Villarroel, J. Carnicero, A. García-Cabañes,
reduce optical damage effects. and J. M. Cabrera, “Understanding light intensity thresholds
for catastrophic optical damage in LiNbO3 ,” Opt. Express 16,
115–120 (2008).
5. SUMMARY AND CONCLUSIONS 13. J. Villarroel, J. Carnicero, F. Ludtke, M. Carrascosa, A.
García-Cabañes, J. M. Cabrera, A. Alcazar, and B. Ramiro,
In summary, the photorefractive effect of SHI LiNbO3 wave- “Analysis of photorefractive optical damage in lithium niobate:
guides has been systematically characterized, finding an inten- aplication to planar waveguides,” Opt. Express 18, 20852–20861
sity response in accordance with the recently reported (2010).
14. E. Jermann and J. Otten, “Light-induced charge transport in
two-center charge transport model [12,13]. In addition, the
LiNbO3 :Fe at high light intensities,” J. Opt. Soc. Am. B 10,
light IDTs for TE and TM polarizations have been determined, 2085–2092 (1993).
obtaining values that decrease with the propagation length 15. H. Hu, F. Lu, F. Chen, B. Shi, K. Wang, and D. Shen, “Monomode
and that are roughly similar to those of PE waveguides for TM optical waveguide in lithium niobate formed by MeV Si ion
(extraordinary polarization). Finally, moderate heating (90°C) implantation,” J. Appl. Phys. 89, 5224–5226 (2001).
16. G. G. Bentini, M. Bianconi, M. Chiarini, L. Correa, C. Sada, P.
allows a remarkable reduction in photorefractive damage, in- Mazzoldi, N. Argiolas, M. Bazzan, and R. Guzzi, “Effect of low
creasing the IDT by a factor 100, i.e., reaching values compar- dose high energy O3 implantation on refractive index and
able to or even better than those presented by other LiNbO3 linear electro-optic properties in x-cut LiNbO3 : planar optical
waveguides. Therefore, the set of reported data indicate that waveguides formation and characterization,” J. Appl. Phys.
regarding POD, SHI LiNbO3 waveguides are really competitive 92, 6477–6483 (2002).
17. J. Olivares, G. García, A. García-Navarro, F. Agulló-López,
with other types of LiNbO3 guides. This property, together O. Caballero, and A. García-Cabañes, “Generation of high-
with other advantages, such as high refractive index profile confinement step-like optical waveguides in LiNbO3 by swift
jumps for both polarizations, opens the door for their success- heavy ion-beam irradiation,” Appl. Phys. Lett. 86, 183501 (2005).
ful application for a variety of photonic devices. 18. J. Olivares, A. Garcia-Navarro, A. Méndez, F. Agulló-López, G.
García, A. García-Cabañes, and M. Carrascosa, “Novel optical
waveguides by in-depth controlled electronic damage with swift
ACKNOWLEDGMENTS ions,” Nucl. Instrum. Methods B 257, 765–770 (2007).
19. J. Olivares, A. García-Navarro, G. García, F. Agulló-López, F.
This work was supported by the Ministerio de Economia y Agulló-Rueda, A. García-Cabañes, and M. Carrascosa, “Buried
Competitividad (MINECO) under grants MAT2008-06794- amorphous layers by electronic excitation in ion-beam irra-
C03-01 and MAT2011-28379-C03-01. An Formación de Per- diated luthium niobate: structure and kinetics,” J. Appl. Phys.
101, 033512 (2007).
sonal Investigador (FPI) fellowship is acknowledged by 20. M. Jubera, J. Villarroel, A. García-Cabañes, M. Carrascosa, J.
M. Jubera. Olivares, F. Agullo-López, A. Méndez, and J. B. Ramiro, “Analy-
sis and optimization of propagation losses in LiNbO3 optical
waveguides produced by swift heavy-ion irradiation,” Appl.
REFERENCES Phys. B 107, 157–162 (2012).
1. L. Arizmendi, “Photonic applications of lithium niobate 21. J. Villarroel, M. Carrascosa, A. García-Cabañes, O.
crystals,” Phys. Stat. Sol. A 201, 253–283 (2004). Caballero-Calero, M. Crespillo, and J. Olivares, “Photorefractive
2. D. Kip and M. Wesner, “Photrefractive waveguides,” in Photore- response and optical damage of lithium niobate optical wave-
fractive Materials and Their Applications I, P. Gunter and J. P. guides produced by swift-heavy ion irradiation,” Appl. Phys.
Huignard, eds. (Springer, 2006), pp. 289–316. B 95, 429–433 (2009).
3. T. Volk, M. Wolecke, and N. Rubinina, “Optical damage resis- 22. [Link]
tance in lithium niobate,” in Photorefractive Materials and 23. Y. Okamura, S. Yoshinaka, and S. Yamamoto, “Measuring mode
Their Applications II, P. Günter and J. P. Huignard, eds. propagation losses of integrated optical waveguides. a simple
(Springer, 2007), pp. 165–203. method,” Appl. Opt. 22, 3892–3894 (1983).
Jubera et al. Vol. 29, No. 11 / November 2012 / J. Opt. Soc. Am. B 3005

24. F. Luedtke, J. Villarroel, A. García-Cabañes, K. Buse, and M. exchanged waveguides formed on MgO-doped lithium niobate
Carrascosa, “Correlation between photorefractive index crystals,” Jpn. J. Appl. Phys. 44, 1407–1409 (2005).
changes and optical damage thresholds in z-cut proton- 31. B. Chen, J. Fonseca-Campos, W. Liang, Y. Wang, and C. Q. Xu,
exchanged-LiNbO3 waveguides,” Opt. Express 17, 658–665 “Wavelength and temperature dependence of photorefractive ef-
(2009). fect in quasi-phase-matched LiNbO3 waveguides,” Appl. Phys.
25. O. Caballero, J. Carnicero, A. Alcazar, G. de la Paliza, A. Lett. 89, 043510 (2006).
García-Cabañes, M. Carrascosa, and J. M. Cabrera, “Light inten- 32. L. Moretti, M. Iodice, F. G. Della Corte, and I. Rendira, “Tem-
sity measurements in optical waveguides using prism couplers,” perature dependence of the thermo-optic coefficient of LiNbO3,
J. Appl. Phys. 102, 074509 (2007). from 300 to 515 K in the visible and infrared regions,” J. Appl.
26. A. Ashkin, G. D. Boyd, J. M. Dziedzic, R. G. Smith, A. A. Ballman, Phys. 98, 036101 (2005).
J. J. Levinstein, and K. Nassau, “Optically induced refractive in- 33. O. Caballero, A. Alcazar, J. Herrero, J. Carnicero, C. Ong, M.
dex inhomogeneities in LiNbO3 and LiTaO3 ,” Appl. Phys. Lett. 9, Domenech, G. Lifante, A. García-Cabañes, J. M. Cabrera, and
72–74 (1966). M. Carrascosa, “Comparative study of optical damage and
27. Y. Kong, J. Wen, and H. Wang, “New doped lithium niobate photovoltaic currents in planar LiNbO3 waveguides,” Proc. SPIE
crystal with high resistance to photorrefraction,” Appl. Phys. 5840, 695–702 (2005).
Lett. 66, 280–282 (1995). 34. R. S. Weis and T. K. Gaylord, “Lithium niobate: summary of
28. J. Villarroel, O. Caballero-Calero, B. Ramiro, A. Alcázar, A. physical properties and crystal structure,” Appl. Phys. A 37,
García-Cabañes, and M. Carrascosa, “Photorefractive non-linear 191–283 (1985).
beam propagation in lithium niobate waveguides above the 35. J. Villarroel, M. Carrascosa, A. García-Cabañes, and J. M.
optical damage threshold,” Opt. Mater. 33, 103–106 (2010). Cabrera, “Light intensity dependence of the photorefractive ho-
29. J. Rams, A. Alcazar-de-Velasco, M. Carrascosa, J. M. Cabrera, lographic response and dark decay of α-phase PE waveguides,”
and F. Agulló-López, “Optical damage inhibition and threshold- J. Opt. A 10, 104008 (2008).
ing effects in lithium niobate above room temperature,” Opt. 36. A. Ikeda, T. Oi, K. Nakayama, Y. Otsuka, and Y. Fujii, “Tempera-
Commun. 178, 211–216 (2000). ture and electric field dependences of optical damage in proton-
30. A. Ikeda, T. Oi, K. Nakayama, Y. Otsuka, and Y. Fujii, “Tempera- exchanged waveguides formed on MgO-doped lithium niobate
ture and electric field dependences of optical damage in proton- crystals,” Jpn. J. Appl. Phys. 44, 1407–1409 (2005).

View publication stats