cm2)

Optics & Laser Technology 50 (2013) 112–117

Contents lists available at SciVerse ScienceDirect

Optics & Laser Technology journal homepage: www.elsevier.com/locate/optlastec

Photorefractive properties of Fe, Zn co-doped near stoichiometric LiNbO3 crystals at moderate intensities (0.5–6 W/cm2) R. Bhatt a,n, S. Ganesamoorthy b, Indranil Bhaumik a, A. Sexana a, A.K. Karnal a, P.K. Gupta a, Jogy George c, K. Ranganathan c a

LMDDD, Raja Ramanna Centre for Advanced Technology, Indore 452013, India Materials Science Group, IGCAR, Kalpakkam 603102, India c SSLD, Raja Ramanna Centre for Advanced Technology, Indore 452013, India b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 2 October 2012 Received in revised form 14 January 2013 Accepted 17 February 2013 Available online 20 March 2013

Iron and zinc co-doped near stoichiometric lithium niobate (SLN:Fe,Zn) single crystals were grown by the top seeded solution growth technique from Li-rich flux. The Raman scattering analysis confirmed the near stoichiometric composition (Li/Nb  0.98) whereas the birefringence interferometry revealed the optical homogeneity to be better than 5.2  10  5/mm for the grown crystals. Two beam coupling measurements showed the writing time constant to be 26–3.6 s measured in the intensity range 0.5– 6 W/cm2. The slow and fast erasure time constants as obtained from bi-exponential fit are  22–4 and  200–24 s, respectively. Interestingly, the fast eraser time is found to be nearly the same as the writing time. The improvement in photorefractive sensitivity (  0.16 cm/J) has been observed. The light induced changes in the refractive index (Dn  5.8  10  5) are found to be in agreement with the estimated value. The intensity dependent photoconductivity estimated from the two beam coupling experiment is found to vary from 1.12  10–11 O  1 m  1 to 6.20  10  11 O  1 m  1 due to primary defects and from 1.24  10  12 O  1 m  1 to 1.04  10  11 O  1 m  1 due to secondary defects. & 2013 Elsevier Ltd. All rights reserved.

Keywords: Stoichiometric lithium niobate Photorefractive crystals Two-beam coupling

1. Introduction Iron doped lithium niobate (LiNbO3:Fe) is a promising photorefractive (PR) material because of its high diffraction efficiency, long storage time and availability in large size [1–3]. However, large response time, low sensitivity and strong light-induced scattering are limiting factors for its applications [2,4–6]. Moreover, as these crystals are usually grown from Li-deficient congruent melt (Li2O 48.6 mol%, Li/Nbo1; CLN:Fe), it contains high 5þ concentration of intrinsic defects (NbLi and V1 Li ) and disorder that limits its optical and photorefractive properties [4–12]. Defects and imperfection also result in unwanted light induced scattering, which selectively get amplified in the photorefractive process [11,13]. Suppressing these intrinsic defects is therefore essential to improve its optical properties. LiNbO3:Fe crystals grown in stoichiometric form (SLN:Fe, Li/Nb 1) show better photorefractive response time and sensitivity [5,10]. The theoretically estimated response time is of the order of few milliseconds for LiNbO3 crystals [14]. Lin et al. [15] have reported the response time to be  10 ms for heavily Fe (1 mol% Fe2O3) doped CLN crystal. But these crystals have limited practical use as it requires

n

Corresponding author. Tel.: þ91 731 2488657; fax: þ91 731 2488650. E-mail address: [email protected] (R. Bhatt).

0030-3992/$ - see front matter & 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.optlastec.2013.02.019

a thin sample  10 mm owing to the high absorption coefficient  200 cm  1 around 500 nm. Further, response time  100 ms is reported for highly reduced near-SLN crystals with low diffraction efficiency  1% [16]. The response time 0.6 s and 1.5 s have been reported for Fe doped near SLN and CLN crystals, respectively, at power density of 10 W/cm2 [5]. Because of the near stoichiometric composition, SLN:Fe crystals contain only small 5þ concentration of NbLi defects, which can be further reduced by co-doping it with damage-resistant dopant, such as Mg2 þ , Zn2 þ , Hf2 þ etc. [4,7–11]. There are a few reports available on the photorefractive properties of near stoichiometric LiNbO3 crystals co-doped with Fe and Zn, presenting response time  10–30 s for low power densities r1 W/cm2 [9–11]. As the values are inconsistent and relatively high in comparison to 0.6 s reported for SLN:Fe in Ref. [5], there is a scope for improving the photorefractive parameters such as response time and sensitivity. In the present work, near-SLN:Fe,Zn crystals have been grown by the top seeded solution growth technique (TSSG). The grown crystals are characterized by optical absorption, Raman scattering and birefringence interferometry. The photorefractive study has been carried out by employing two beam coupling at moderate power densities in the range of 0.5–6 W/cm2. The improvements in response time and sensitivity are reported. Also, the light induced change in the refractive index and photoconductivity of the grown crystal has been evaluated.

R. Bhatt et al. / Optics & Laser Technology 50 (2013) 112–117

113

2. Experimental Double doped near SLN:Fe,Zn crystals are grown by the top seeded solution growth (TSSG) method from Li-rich melt (58 mol% Li2O and 42 mol% Nb2O5). The concentration of ZnO and Fe2O3 in the melt was 1.5 mol% and 0.045 mol% (0.05 wt%) respectively. Starting materials of Li2CO3 and Nb2O5 of 4N purity are used for the crystal growth experiments. The weighted powders are mixed thoroughly and kept for solid-state reaction at 750 1C for 10 h and subsequently at 850 1C for 15 h. Around 800 g of charge thus prepared is put in to a platinum crucible of 80 mm diameter and 80 mm height. The charge is then homogenized for 24 h before lowering the seed into the melt. Crystal growth workstation and related infrastructure for the growth of SLN crystals were established in-house. The axial temperature gradient is 10 1C/cm. The growth is initiated using a Z-oriented seed. The seeding is done as per the technique described in Ref. [17] to establish the saturation temperature quickly and to minimize the seed loss during seeding. The growth parameters are optimized to get nearly flat interface, core and inclusion free single crystals. Growth rate of 0.5–0.2 mm/h and crystal rotation rates of 20–2 rpm is used. The phase of the grown crystal is confirmed by the X-ray powder diffraction technique. Transmission spectra are recorded using a Carry 50 spectrophotometer for z-cut polished samples of 1 mm thickness. The Li content in these samples was estimated by polarized Raman spectroscopy for Z(YX)Z and Z(YY)Z scattering configurations using a micro-Raman spectrometer (Alpha300SR, Witec Instruments GmbH, Germany) with 488 nm wavelength of Ar-ion laser as excitation source. Optical homogeneity of the SLN:Fe,Zn crystals is investigated by orthoscopic birefringence interferometry, as crystal grown from off-congruent melt tends to exhibit composition variation along the length of the crystal [5]. A collimated parallel He–Ne Laser beam travels through Y-cut SLN crystal of dimension 10  10  1.2 mm3 placed between cross polarizes at 451 to the crystal axis. The delay between the ordinary and extraordinary components passing through the polarizer results in formation of interference fringes, revealing the birefringence map of the sample. The corresponding phase delay is given by f ¼2p Dnd/l, where d is the sample thickness and Dn¼ ne no is the birefringence. Y-oriented samples of dimension  10  2.5  8 mm3 are used for the photorefractive studies. The polished samples are reduced in Ar atmosphere at 900 1C for 6 h to increase the donor-toacceptor ratio (i.e., Fe2 þ /Fe3 þ ) in the crystals. The photorefractive properties of the near SLN:Fe,Zn crystals are evaluated by the two-beam coupling experiment in transmission geometry at a fixed grating spacing. The side surfaces and peripheries of the samples are short-circuited with conductive silver paste to avoid charge buildup that leads to open circuit voltage and influence the photorefractive process. A 150 mW diode-pumped frequencydoubled Nd:YVO4 laser operating at 532 nm with 2 mm beam diameter is used. The laser beam is split into two extraordinary polarized beams and made to overlap at an angle of 221 in the crystal. The interference pattern of the incident beams results in the formation of index grating ( 720 lines/mm) with period 1.39 mm. The grating vector is aligned along the c-axis to exploit the major electro-optic coefficient r33. The measurements are carried out at power densities  0.5–6 W/cm2. The evolution and decay of the grating were monitored by a Bragg angle matched weak HeNe laser ( o2 mW, 632.8 nm) beam as a probe. The intensities of diffracted and transmitted beams are measured by photodiodes and recorded in a digital storage oscilloscope.

3. Result and discussion Fig. 1 shows the photograph of SLN:Fe,Zn (0.05 wt% Fe2O3 and 1.5 mol% ZnO) crystal boule (grown by the top seeded solution

Fig. 1. SLN:Fe,Zn crystal boule and fabricated PR element.

growth technique using Li-rich melt) and the fabricated element. The grown crystals are of good quality, nearly flat interfaced and light brown in color with the size measuring up to 25 mm in diameter and 30 mm in length. As the crystals are grown from offcongruent melt, only less than 5% of the total charge is crystallized in order to avoid composition variations across the length of the crystal. Fig. 2 shows the Raman scattering spectra of SLN:Fe,Zn, CLN:Fe,Zn, SLN and CLN crystals for the Z(YX)Z and Z(YY)Z scattering configurations (CLN:Fe,Zn, SLN and CLN are included for comparison). The composition of CLN:Fe,Zn crystal is 0.03 wt% Fe2O3 and 4 mol% ZnO. The observed Raman peaks are due to E (TO1) (153 cm  1) and A1 (LO) (873 cm  1) phonon modes polarized along the X or Y direction. The E (TO1) phonon modes represent the vibration of Nb–O bonds and the linewidth of these modes reveals the disorder in the Nb sub-lattice, which is utilized to evaluate the stoichiometry (composition) of the crystal. The stoichiometry (Li content) of SLN:Fe,Zn crystal is evaluated from the linewidth of the 153 cm  1 Raman shift peak using the relation cLi [mol%]  53.03–0.4739 G [cm  1], where G is the linewidth of the peak [18]. The stoichiometry is also evaluated from the linewidth of 873 cm  1 peak using the relation cLi [mol%]  53.29–0.1837 G [cm  1] [18]. Referring to Fig. 2, the Raman spectra for both the peaks clearly show that the linewidth of SLN:Fe,Zn appears close to that of SLN crystal and confirms its near stoichiometric composition. The observed linewidth broadening for CLN:Fe,Zn and CLN crystals depicts the structural disorder in the lattice due to the presence of high concentration of intrinsic defects that leads to a deviation from the stoichiometry. The linewidth is estimated from the Lorentz fitting approximation and the values are 6.49, 7.92, 10.72 and 14.70 cm  1 (for the 153 cm  1 peak), and 20.90, 22.17, 29.08 and 39.07 cm  1 (for the 873 cm  1 peak) observed for SLN, SLN:Fe,Zn, CLN and CLN:Fe,Zn crystals, respectively. The estimated Li content for SLN:Fe,Zn crystal is  49.28 mol% Li2O (Li/ Nb 0.975) and  49.22 mol% Li2O (Li/Nb 0.969) as obtained from the linewidth of 153 cm  1 and 873 cm  1 peaks, respectively, which is significantly higher than the 46.12 mol% Li2O (Li/ Nb 0.86) for CLN:FeZn crystal. Fig. 3 depicts the absorption spectra of near SLN, SLN:Fe,Zn, reduced SLN:Fe,Zn and CLN:Fe,Zn crystals. It shows a shift in UV absorption edge to higher wavelength for SLN:Fe,Zn, reduced SLN:Fe,Zn and CLN:Fe,Zn in comparison to SLN crystal. The observed maximum shift for CLN:Fe,Zn crystal depicts the presence of intrinsic defects that results in a decrease in Li/Nb in this crystal. This is in agreement with the Raman analysis where maximum broadening is observed for the CLN:Fe,Zn crystal

114

R. Bhatt et al. / Optics & Laser Technology 50 (2013) 112–117

Fig. 2. Raman spectra of SLN:Fe,Zn, CLN:Fe,Zn, SLN and CLN crystals.

Fig. 3. Absorption spectra of near SLN, SLN:Fe,Zn, reduced SLN:Fe,Zn and CLN:Fe,Zn crystals.

revealing a decrease in the Li/Nb ratio. The shift in absorption edge is attributed to inter-band transition due to the band-tailing phenomenon [19]. Further, the absorption edge is also influenced by the presence of Fe ions in the LiNbO3 lattice [1,20]. The Fe ions result in the broad absorption band at around 400 nm (3.1 eV, Cband) due to charge transfers from O2  -Fe3 þ [1,20]. As this band extends to lower wavelengths it contributes to a shift in the absorption edge. The broad absorption band centered at 480 nm (2.5 eV, D-band) in SLN:Fe,Zn and CLN:Fe,Zn is attributed to intervalence band-impurity transitions Fe2 þ -Nb5 þ [1,20–22]. As the 3d states of Fe2 þ ions are strongly coupled to the lattice it results in broad absorption in the visible and near IR regions [21]. The higher D-band absorption as observed for the CLN:Fe,Zn crystal reveals that the incorporation of Fe ions is relatively easy in CLN in comparison to SLN due to the high concentration of intrinsic defects that provides vacant Li sites and also helps in charge compensation. Further, Fe ions enter in the lattice in the form of variable valance states i.e. Fe2 þ and Fe3 þ . In a photorefractive (PR) process, Fe2 þ ion serves as a donor state (occupied trap), whereas Fe3 þ as an electron trap. The PR behavior largely depends on the donor-to-acceptor ratio (i.e., Fe2 þ /Fe3 þ ) [3,4], which can be altered by thermal annealing in reducing/oxidizing environment [3]. To increase the Fe2 þ /Fe3 þ ratio, samples are annealed in reducing (Ar) atmosphere at 900 1C for 7 h. The effect of reduction (thermal annealing in Ar atmosphere) on the absorption spectra of SLN:Fe,Zn is shown in Fig. 3. The reduced SLN:Fe,Zn show higher absorption around 480 nm absorption band which may be attributed to the increased cFe2 þ =cFe3 þ ratio. The concentration of Fe2 þ ions in SLN:Fe,Zn crystal can be determined from absorption spectra using the relation [23]: cFe2 þ  2:16  1017 cm2 a477, where, cFe2 þ denotes the Fe2 þ ions concentration and a477 (cm  1) the absorption coefficient at 477 nm. The EDX measurement reveals that the concentration of Zn2 þ ions in SLN:Fe,Zn crystal is 1.4 at%. Referring to absorption spectra (Fig. 3), a477 (cm  1) for SLN:Fe,Zn and reduced SLN:Fe,Zn crystals are 2.2 and 3.1 cm  1 and the corresponding concentration of Fe2 þ ions is  4.4  1017 cm  3 and  6.7  1017 cm  3 respectively. Moreover, the reduction process gives rise to oxygen vacancy V Odd 4þ (OO -1=2O2 þ V Odd þ 2e ) [24,25] and polarons (NbLi ) states as 4þ per the reaction OO þ2V Li’ þ NbNb -3=2O2 þNbLi þ6e [26] in LiNbO3 crystals. The reduced LiNbO3 crystals show a broad absorption around 500 nm, which is assigned to the formation of bipolar4þ 4þ 5þ ons (NbLi –NbNb ; two electrons trapped adjacent to NbLi and 5þ NbNb ) and F þ centers (one electron trapped at V Odd ) [23,26]. The optical homogeneity of SLN:Fe,Zn crystals is evaluated from optical birefringence interferometry [27]. Fig. 4 shows the recorded interference fringe pattern of near SLN:Fe,Zn crystal. The observed nearly straight-line fringe patterns correspond to contour maps of equal birefringence therefore depict that the sample is free from residual stress [28] and has nearly uniform dopant distributions (homogeneity). The slight variation in the fringe spacing and their deviation from the straightness (Fringe distortion) depicts presence of small local optical inhomogeneities due to local impurity (doping) concentration fluctuations resulting in birefringence (Dn) variation. Such inhomogeneities are quite common in case of growth from melts with added impurities (dopants) or growth from off-congruent melt as discussed earlier [5]. This occurs due to the random temperature fluctuations at the growth interface. However, the nearly straight line character of the fringes is a signature of good optical homogeneity of the grown crystals. The order of optical homogeneity can be estimated by calculating the per fringe variation in d(Dn), which is given by dðDnÞ  l=L  5:2  104 . Since the overall fringe distortion for the full aperture of the specimen (10 mm) is less than one fringe spacing, the total birefringence variation should be less than per fringe d(Dn)

R. Bhatt et al. / Optics & Laser Technology 50 (2013) 112–117

variation. Based on this, the estimated optical homogeneity of the SLN:Fe,Zn crystal is better than 5.2  10  5/mm. This is better than the birefringence variation  2.07  10  4 reported for 0.03 at% Fedoped LiNbO3 crystal grown from congruent melt by the Czochralski technique [28]. Fig. 5 shows the schematic of the two-beam coupling setup, in which, a pair of coherent laser beams intersects inside a PR medium and forms a stationary interference pattern. The periodic variations in intensity produce periodic modulation in the index of refraction. The grating thus formed is called as index grating. Fig. 6 shows the unidirectional power/energy transfer between the two interacting beams during the recording process (grating buildup). This unidirectional energy transfer depicts the non-local grating with a phase shift of (f) p/2 between the intensity patterns and refractive index grating [29]. The charge transport in this case is governed by the diffusion process with a diffusion field Ediff  1.2  103 V/cm, which is calculated by using the relation [1,29]: Ediff ¼ (kBT/e)K, where K (¼ 4p sin y/l), kB, T, and e are the grating wave vector, Boltzmann constant, the absolute temperature, and the elementary charge of electron respectively. The corresponding change in the refractive index (Dn) due to the diffusion field can be estimated from the relation: Dn1/2n3rEdiff, where n is the refractive index, and r is the electro-optic coefficient. For a field (Ediff) 1.2  103 V/cm, the change in the refractive index (Dn) due to electro-optic effect is 2.2  10  5. The grating strength is analysed by measuring the diffraction efficiency (DE, Z) which is defined as Z ¼ Id =ðId þIt Þ, where Id and It are the diffracted and the transmitted intensities, respectively for the readout beam [7]. The DE of the phase grating formed in the material is measured by blocking one of the writing beam and

115

monitoring the diffracted and transmitted intensities of the other beam. Fig. 7 shows the observed DE 65% for the unreduced SLN:Fe,Zn crystal recorded by stopping one of the writing beams. Real time gratings evolution and decay are also monitored by a weak unpolarized HeNe laser probe beam (o2 mW) at 632.8 nm incident at the Bragg matching angle to reveal the evolution and decay dynamics of the index grating. Fig. 8 shows the normalized DE for the reduced SLN:Fe,Zn recorded at different power densities in the range 0.5–6 W/cm2. It can be clearly seen that time constants i.e. grating buildup (tr) and decay (te) time are decreasing with increasing power densities. In order to evaluate response time constants diffraction efficiency curves are analysed by exponential function: ZðtÞ ¼ Zo ½ð1expt=tr Þ where Zo is the saturation DE and tr is recording time constant. It is observed that grating build up time is best approximated by a monoexponential function whereas grating decay by a bi-exponential function: ZðtÞ ¼ Zo ½ðexpt=te1 Þ þ ðexpt=te2 Þ, where Zo is the saturation DE, te1 and te2 are fast and slow time constants corresponding to two different photorefractive processes [3,30]. This is in good agreement with the finding of Xiao-Chun et al. [31], where they reported that grating erasure is best approximated by the one photo-refractive center model for the congruent crystal, while two photorefractive centers for stoichiometric crystals. The primary processes are usually governed by extrinsic defects Fe2 þ /Fe3 þ that result in electron grating and dominate in the photorefraction, whereas the secondary process may be attributed to intrinsic secondary defects such as bipolarons

Intensity (a.u.)

4

3

2

1

0

Fig. 4. Birefringence fringe patterns of SLN:Fe,Zn crystal.

0

2

4

6 8 Time (s)

10

12

14

Fig. 6. Power exchange (coupling) between the two interacting beams due to the formation of index grating in the SLN:Fe,Zn crystal.

Fig. 5. Schematic of the two beam coupling experimental setup.

116

R. Bhatt et al. / Optics & Laser Technology 50 (2013) 112–117

Table 1 Grating time constants of SLN:Fe,Zn crystal. Power density (W/cm2)

Buildup time (tr) (s)

Decay time (te1) (s)

Decay time (te2) (s)

0.5 1 2 4 5 6

26 21 12 7.2 4.2 3.5

22 18 8 10 6 4

200 115 75 61 57 24

Fig. 7. Diffraction efficiency of unreduced SLN:Fe,Zn crystal measured during the grating readout process by stopping one of the writing beams.

Fig. 9. Variation of grating buildup and decay time constants with power density.

Table 2 The photorefractive properties of the near SLN:Fe,Zn crystals. Sample

Fig. 8. Normalized square root of diffraction efficiency of SLN:Fe,Zn crystal probed with a weak HeNe laser beam measured at different power densities.

4þ

4þ

4þ

(NbLi –NbNb ) and small-polarons NbLi [31], hole gratings, etc., which give rise to different time constants. The estimated time constants values are listed in Table 1 and the variation of grating buildup and decay time constants with power density is shown in Fig. 9. It is interesting to see that decay time constant te1 is nearly close to the grating build-up tr time constant. The observed response time constant (grating build-up time) values are shorter than the value presented in Ref. [10], wherein writing time of 22 s has been reported for near SLN:Fe,Zn (with Zn 2 mol% and Fe 0.02 wt%) measured using 488 nm wavelength at 1.28 W/cm2 (Table 2). The photorefractive sensitivity is the other important parameter to characterize a photorefractive material. Sensitivity describes the recording speed of the medium. It is determined by the absorption coefficient of the recording light and the transport length of the excited carriers [2]. Sensitivity (S) of the medium is determined by [7] pffiffiffi S ¼ ð1=IdÞð@ Z=@tÞ9t ¼ 0 , where Z is the DE, I is the sum of the intensity of incident beams, L is the crystal thickness, and t is recording time. Referring to the diffraction efficiency curve in Fig. 8,

Fe (wt%) Zn (mol%)

SLN:Fe,Zn 0.05 (melt) SLN:Fe,Zn 0.03 SLN:Fe,Zn 0.02 CLN:Fe:Zn 0.03

1.5 2 2 6

tr

te (s)

(s) 3.5 te1  4 te2  24 10.2 567 22 – 30 95

S Dn  10  5 (cm/J) 0.16

3.7

– 0.065 0.11

– – 0.12

Present work Ref. [8] Ref. [9] Ref. [32]

it clearly shows that sensitivity is increasing (as slope of DE increasing) with increasing power density. In the present study, the photorefractive sensitivity for near SLN:Fe,Zn crystal estimated from DE data is  0.12–0.16 cm/J measured at 2–6 W/cm2. These values are better than the reported values of 0.065 cm/J measured at 1.28 W/ cm2 for near SLN:Fe,Zn crystal [10]. The light induced change in the refractive index (Dn) is evaluated from the measured diffraction efficiency data using the relation: Z ¼ sin2 ðpdDn=l cos yÞ, where Z is the diffraction efficiency, l is the wavelength of recording light, d is the crystal thickness and y is the angle of incident light in the crystal [4,10]. Fig. 10 shows the light induced change in the refractive index of near SLN:Fe,Zn crystal with exposure time. The maximum refractive index change (Dn) obtained from these measurement is 5.8  10  5. This is in good agreement to the refractive index value (Dn) 2.2  10  5 estimated theoretically using the value of diffusion field as discussed earlier. The photoconductivity (sph) of near SLN:Fe,Zn crystal is evaluated from holographic measurement (i.e. by the two beam coupling technique at 532 nm wavelength with different intensities) using

R. Bhatt et al. / Optics & Laser Technology 50 (2013) 112–117

117

and two-beam coupling experiments. The concentration of Fe2 þ ions in the grown crystal is 4.4  1017 cm  3 as determined from optical absorption spectra whereas the concentration of Zn 1.4 at% as evaluated by EDX. The two beam coupling experiment revealed diffraction efficiency of 70%. The intensity dependent recording and erasure time constants are evaluated for the photorefractive process. An improvement in the photo-refractive properties of SLN:Fe,Zn crystal is observed in comparison to the previously reported values. The light induced change in refractive index and the photoconductivity of SLN:Fe,Zn crystal are also evaluated by the holographic technique.

Acknowledgment The authors would like to thank M.K. Swami and H.S. Patel for the Raman measurements and Mrs. Pragya Tiwari for EDX measurements. Fig. 10. Light induced change in the refractive index of the SLN:Fe,Zn crystal.

the relation: sph ¼ er eo =te (er  28) with the assumption that the dark conductivity sd 5 sph. The estimated photoconductivity due to primary defects is found to vary from 1.12  10  11 O  1 m  1 to 6.20  10  11 O  1 m  1 depending on light intensities from 0.5 to 6 W/cm2. While due to secondary defects it varies from 1.24  10  12 O  1 m  1 to 1.04  10  11 O  1 m  1 with light intensities from 0.5 to 6 W/cm2. These values are in reasonably agreement with the reported photoconductivity of undoped SLN crystal  2.7  10  12 O  1 m  1, measured by applying a dc field along the z-direction in presence of an extraordinary polarized laser beam at 532 nm with 10 W/cm2 light intensity [4]. The higher sph as observed for near SLN:Fe,Zn crystal is attributed to Zn doping which is supposed to increase the photoconductivity [7], and to the Fe2þ ions (donor), which results in direct photo-excited charge transfer to conduction 5þ band (d-orbitals of Nb) ions and to the intrinsic NbLi defects [3]. At low or normal light intensities (Io1 W/cm2) the charge transport is mainly governed by the primary Fe2þ /3 þ (extrinsic) defects as per the one center charge transport model [3]. However, at moderate to high light intensities (1–10 W/cm2) role of secondary 4þ 4þ defects (chemical reduction induced secondary defects) NbLi –NbNb [3,24] and oxygen defects V Odd related hole traps [24] are expected to 4þ 4þ contribute. As NbLi –NbNb and F þ -centers are characterized by the broad absorption band centered at 500 nm, these defects are also expected to contribute to the photorefractive processes and particularly they will be more effective in the moderate to high intensity. For example, the bipolarons get dissociated into small polarons and charge carrier (e  ) when illuminated with visible (532 nm) radiation, and recombine in dark, therefore contributes to photorefraction processes [26,31]. Moreover, the photoelectrons can be directly 4þ 4þ excited from Fe2þ or NbLi –NbNb to conduction band (CB) or to 5þ intrinsic traps NbLi . Electrons can also be excited from the valence band into Fe3 þ that lead to holes in the valence band. These holes migrate in the valance band and then get retrapped at the nearest extended states (hole traps) or hole polarons (V Li2O ) in valance forming hole grating. The observed higher decay time constant (te2) may be attributed to these hole grating (secondary processes). All these defects therefore contribute to the photorefraction process and leads to the formation of electron and hole gratings.

4. Conclusions Near stoichiometric SLN:Fe,Zn crystal is grown and characterized by optical absorption, Raman, birefringence interferometry

References [1] Gunter P, Huignard J-P, editors, Photorefractive materials and their applications II: Materials, Springer series in optical sciences. 2007; 134. [2] Hesselink L, Orlov SS, Liu A, Akella A, Lande D, Neurgaonkar RR. Science 1998;282:1089. [3] Adibi A, Buse K, Psaltis D. Physical Review A 2001;63:023813. [4] Jermann F, Simon M, Kratzig E. Journal of the Optical Society America B 1995;12:2066. [5] Kitamura K, Furukawa Y, Ji Y, Zgonik M, Medrano C, Montemezzani G, et al. Journal of Applied Physics 1997;82:1006. [6] Sun X, Luo S, Wang J, Jiang Y, Shi H. Journal of Physics D: Applied Physics 2009;42:115413. [7] Kong Y, Wu S, Liu S, Chen S, Xu J. Applied Physics Letters 2008;92:251107. [8] Ling FR, He ZH, Yao JQ, Wang B, Jiang S. Optics & Laser Technology 2008;40: 941. [9] Li H, Sun Z, Ye S, Cai W, Zhao L. Journal of Physics and Chemistry of Solids 2005;66:990. [10] Zheng W, Zhangb N, Zhaoa L, Xua Y. Materials Chemistry and Physics 2004;84:7. [11] Li S, Liu S, Kong Y, Xu J, Zhang G. Applied Physics Letters 2006;89:101126. [12] Kityk IV, Makowska-Janusik M, Fontana MD, Aillerie M, Fahmi A. Journal of Applied Physics 2001;90:5542. [13] Falk M, Woike T, Buse K. Applied Physics Letters 2007;90:251912. [14] Yeh P. Applied Optics 1987;26:602. [15] Lin SH, Hsieh ML, Hsu KY, Hsieh TC, Lin S, Yeh T, et al. Journal of the Optical Society America B 1999;16:1112. [16] Chen XJ, Zhu DS, Li B, Ling T, Wu ZK. Optics Letters 2001;26:998. [17] Bhaumik I, Ganesamoorthy S, Bhatt R, Sundar R, Karnal AK, Wadhawan VK. Journal of Crystal Growth 2002;243:522. [18] Wohlecke M, Corradi G, Betzler K. Applied Physics B 1996;63:323. [19] Bhatt R, Bhaumik Indranil, Ganesamoorthy S, Karnal AK, Swami MK, Patel HS, et al. Physica Status Solidi (A) 2012;209:176. [20] Dischler B, Herrington JR, Rauber A. Solid State Communications 1974;14: 1233. [21] Kurz H, Kratzig E, Keune W, Engelmann H, Gonser U, Dischler B, et al. Applied Physics A: Materials Science & Processing 1977;12:355. [22] Basun SA, Evans DR, Bunning TJ, Guha S, Barnes JO, Cook G, et al. Journal of Applied Physics 2002;92:7051. [23] Schirmer OF, Imlau M, Merschjann C, Schoke B. Journal of Physics: Condensed Matter 2009;21:123201. [24] Qiao H, Tomita Y. Optics Express 2006;14:5773. [25] Deleo GG, Dobson JL, Masters MF, Bonjack LH. Physical Review B 1988;37: 8394. [26] Donnerberg H, Tomlinson SM, Catlow CRA, Schirmer OF. Physical Review B 1989;40:11909. [27] Henningsen T, Singh NB. Journal of Crystal Growth 1989;96:114. [28] Pracka I, Bajor AL, Kaczmarek SM, Wirkowicz M, Kaczmarek B, Kisielewski J, et al. Crystal Research and Technology 1999;34:627. [29] Xin F, Zhang G, Bo F, Sun H, Kong Y, Xu J, et al. Journal of Applied Physics 2010;107:33113. [30] Adibi A, Buse K, Psaltis D. Optics Letters 1999;24:652. [31] Xiao-Chun Li, Yong-Fa Kong, Li-Zhong Wang, Hong-De Liu. Chinese Physics B 2008;17:1014. [32] Zheng W, Zhang N, Zhao L, Xu Y. Optics Communications 2003;227:259.