Optimization of Blue Photorefractive Properties and Exponential Gain of Photorefraction in Sc-Doped Ru:Fe:LiNbO₃ Crystals

Abstract

Sc:Ru:Fe:LiNbO₃ crystals were grown from congruent melt by using the Czochralski method. A series of LiNbO₃ crystals (Li/Nb = 48.6/51.4) with 0.1 wt% RuO₂, 0.06 wt% Fe₂O₃ and various concentrations of Sc₂0₃ were prepared. RF1 and RF4 refers to the samples containing 0 mol% Sc₂0₃ and 3 mol% Sc₂0₃, respectively. The photorefractive properties of RF4 were measured by ${\mathrm{Kr}}^{+}$ laser (λ = 476 nm blue light): η_s = 75.7%, τ_w = 11 s, M/# = 19.52, S = 2.85 cmJ⁻¹, Γ = 31.8 cm⁻¹ and ∆n_max = 6.66 × 10⁻⁵. The photorefractive properties of five systems (η_s, M/#, S, Γ and ∆n_max) under 476 nm wavelength from RF1 to RF4 continually increased the response time, while τ_w was continually shortened. Comparing the photorefractive properties of Sc (1 mol%):Ru (0.1 wt%):Fe (0.06 wt%): LiNbO₃ measured by ${\mathrm{Kr}}^{+}$ laser (λ = 476 nm blue light) with Sc (1 mol%):Fe (0.06 wt%):LiNbO₃ measured by He-Ne laser (633 nm red light), η_s increased by a factor of 1.9, V_w (response rate) increased by a factor of 13.9, M/# increased by a factor of 1.8 and S increased by a factor of 32. The ∆n_max improved by a factor of 1.4. A strong blue photorefraction was created by the two-center effect and the remarkable characteristic of being in phase between the two gratings recorded in shallow and deep trap centers. The photorefractive properties (η_S, τ_w, M/#, S, ∆n_max) were increased with an increase in Sc³⁺ ion concentration. Damage-resistant dopants such as Sc³⁺ ions were no longer resistant to damage, but they enhanced the photorefractive properties at the 476 nm wavelength. The experimental results clearly show that Sc-doped two-center Ru:Fe:LiNbO₃ crystal is a promising candidate blue photorefraction material for volume holographic storage. Sc-doped LiNbO₃ crystal can significantly enhance the blue photorefractive properties according to the experimental parameters. Therefore, the Sc:Ru:Fe:LiNbO₃ crystal has better photorefractive properties than the Ru:Fe:LiNbO₃ crystal.

1. Introduction

Fe-doped lithium niobate single crystals, LiNbO₃:Fe, have been grown intentionally since the 1960s as a material for the holographic recording of information [1,2,3,4,5,6]. Fe-doped LiNbO₃ crystals are one of the most widely used photorefractive materials in volume holographic storage due to their high diffraction efficiency, high data storage density and long storage lifetime. This is because the impurities of Fe can improve the photorefractive effect of LiNbO₃ crystals. However, several shortcomings, such as a long response time, low sensitivity and strong light-induced scattering, hinder the applications of Fe:LiNbO₃ crystals in optical holographic storage [2,3,4,5]. Recent studies demonstrated that a Ru ion can play a similar role [7,8]. Simultaneous doping of Fe and Ru tends to produce a better photorefraction effect of the LiNbO₃. Indeed, two kinds of such ions were incorporated into LiNbO₃ crystals to yield Ru:Fe:LiNbO₃ crystals in order to realize nonvolatile holographic storage.

It was also demonstrated that doping other types of impurity ions can improve the resistance ability of LiNbO₃ crystals for optical damage [9,10,11,12,13,14,15]. A set of optical damage resistance dopants, including In, Mg, Zn, Sc and Hf, were doped into the crystal for such a purpose [16]. Among them, Sc³⁺ ion impurity is the most effective due to the lower threshold concentration, which was reported to be about 3.0 mol%. As a result, Sc³⁺ doping has been widely investigated to produce high-quality Sc-doped LiNbO₃ crystals with a large diameter and high optical quality. Recent studies found that the optical damage resistivity of doped ions also depends on the incident light wavelength. Qiao et al. [17] found that the UV (351 nm) photorefraction of a LiNbO₃ crystal doped with In, Zn or Na was enhanced significantly compared to the nominally pure LiNbO₃ crystal. To date, almost all enhanced photorefraction is investigated using UV light in single-center-doped LiNbO₃ crystals.

Considering the improved photorefractive ability of LiNbO₃ crystals by doping Ru and Fe, as well as the improved optical damage resistance by doping Sc, it will be of great interest to investigate the photorefractive properties of two-center-doped LiNbO₃ crystals, i.e., Sc:Ru:Fe:LiNbO₃, to reveal the role of Sc doping. Moreover, in addition to the use of UV light, using other wavelengths for photorefractive investigation will also be interesting.

Thus, based on the considerations above, two-center Sc:Ru:Fe:LiNbO₃ crystals were grown using the Czochralski method. The incident light wavelength was selected as 476 nm (blue light), which has enough energy to excite the charge carriers’ holes of the Sc:Ru:Fe:LiNbO₃ crystals [18]. It was established experimentally that the blue photorefraction of the Ru:Fe:LiNbO₃ doped with Sc was enhanced significantly compared to Sc:Ru:Fe:LiNbO₃ crystals. Sc³⁺ ions play a special role in blue photorefraction. Sc:Ru:Fe:LiNbO₃ crystals exhibit excellent properties such as a high refractive index change, high sensitivity, large exponential gain coefficient Γ and dynamic range, together with a fast response time at the 476 nm wavelength.

2. Experiments

Crystal Growth and Sample Preparation

Using the Czochralski method, a series of LiNbO₃ crystals were grown from congruent melts (Li/Nb = 48.6/51.4) in the air atmosphere along the c direction. The raw materials used for crystal growth were Li₂CO_3, Nb₂O_5, Sc₂O₃, RuO₂ and Fe₂O₃. RuO₂ and Fe₂O₃ content in the melt was 0.1 and 0.03 wt%, respectively. Sc₂O₃ content in the melt was 0, 0.5, 1.0 and 1.5 mol%. The raw materials were mixed for 12 h. Mixtures were heated at 750 °C for 2 h to remove CO₂ and then further heated up to 1150 °C for 2 h to form polycrystalline powder. The optimum growth conditions were as follows: an axial temperature gradient of 40–50 K/cm, rotating rate of 10–25 rpm and pulling rate of 0.5–2 mm/h. The grown crystals were polarized at 1200 °C with a current density of 5 mA/cm². The crystal was cut into Y-cut plates (10 × 2 × 10 mm) (X × Y × Z) with polished surfaces at a doping concentration as shown in

Table 1. The doping concentration of Ru:Fe:LiNbO₃ crystal.

Sample	RF1	RF2	RF3	RF4
[Sc 203] in melt (mol%)	0	0.5	1.0	1.5
[RuO2] in melt (wt%)	0.1	0.1	0.1	0.1
[Fe2O3] in melt (wt%)	0.03	0.03	0.03	0.03

.

3. Results and Discussion

3.1. Type of Dominant Charge Carriers

Figure 1a,b show the erasing processes in sample RF3, with both signal light (S) and reference light (R), for two different lasers with wavelengths of 476 nm and 633 nm, respectively, which were utilized to determine the dominant carrier type. The diffraction efficiency is normalized for comparison. As shown in Figure 1a, when erasing with the 476 nm light, the decay time of the R erasing is much slower than that of the S erasing, implying that the light energy transfer is unidirectional from R to S. This energy transfer direction is the same as the direction of the optical axis of the crystal, indicating holes are the dominant charge carriers. In comparison, when erasing with 633 nm light as shown in Figure 1b, the decay time of R is much faster than that of S, implying the electrons are the dominant charge carriers. The dominant charge carriers depend on the laser wavelength. At 476 nm, Ru is a shallow trap, and Fe is a deep trap center. The blue light of 476 nm has enough energy to excite holes from both the shallow and deep trap centers, so holes become the dominant charge carriers. At 633 nm, Fe is the shallow trap center, and Ru is the deep trap center. Because the red light of 633 nm can only excite electrons from the shallow trap but not from the deep trap centers, they become the dominant charge carrier [19].

3.2. Measurement of Exponential Gain Coefficient in Sc:Ru:Fe:LiNbO₃ Crystal

The exponential gain coefficient of a photorefractive crystal reflects the ability of the energy to transition from the pump light to the signal light during the information storage process. The larger the coefficient is, the greater the multiplicity of the signal light that is amplified. Thus, the signal light becomes clearer during its reappearance, and greater diffraction efficiency can be obtained. Therefore, the exponential gain coefficient is one of the important guidelines for evaluating the photorefractive effect.

3.3. Calculation Formulas of the Exponential Gain Coefficient

The intensity of the signal beam and pump beam with coupling can be achieved by solving the coupling-wave equation. The result is expressed below:

$$I_1(\delta )=\frac{I_1(0)+I_2(0)}{I_1(0)+I_2(0)\exp (\mathsf{\Gamma }\delta )}\exp (\mathsf{\Gamma }-a)\delta$$

(1)

$$I_2(\delta )=\frac{I_1(0)+I_2(0)}{I_1(0)+I_2(0)\exp (\mathsf{\Gamma }\delta )}\exp (-a\delta )$$

(2)

where Γ is the exponential gain coefficient for the signal beam in the coupling process, a is the exponential attenuation coefficient for the beam in the medium, δ is the thickness of the medium, I₁(0) and I₂(0) are the intensity of the incident signal beam and pump beam at the front surface of the medium, and I₁(δ) and I₂(δ) are the intensity of the two emergent beams at the back surface of the medium. It is worth noting that only when Γ > α can the signal beam have a positive gain. The formulas above represent the energy exchange that results from the two-light coupling.

Disregarding the Fresnel reflection losses and pump depletion, the pump beam intensities with and without coupling are approximate (I₂(δ) ≈ I′₂(δ)), and the exponential gain coefficient Γ can be expressed as Formula (3) [20,21]:

$$\mathsf{\Gamma }=\frac{1}{\delta }ln\frac{I_1\left(\delta \right)I_2^{\prime }\left(\delta \right)}{ I_1^{\prime }\left(\delta \right)I_2\left(\delta \right)} \approx \frac{1}{\delta }ln\frac{I_1\left(\delta \right)}{ I_1^{\prime }\left(\delta \right)}$$

(3)

where I′₁(δ) and I′₂(δ) are the transmitted intensity of the signal and pump beam without coupling.

Using the ${\mathrm{Kr}}^{+}$ laser as a light source, with λ = 476 nm, the polarization direction exists in the plane of incidence (e). The crossing angle between the incidence signal beam $I_1\left(0\right)$ and pump beam$I_2\left(0\right)$ is 2θ, the crystal medium thickness is δ = 1.0 mm and the diameters of the signal light and pump light are D = 0.8 mm and D = 3 mm, respectively. The light passes through the γ face of the LiNbO₃ crystal.$I_2\left(0\right)$ = 1.52 w/cm² and $I_2\left(0\right)$/$I_1\left(0\right)$ = 1240. The experimental setup of two-wave coupling is shown in Figure 2.

The measurement results of the exponential gain coefficient Γ and response time τ_W of the RF1, RF2, RF3 and RF4 samples are listed in

Table 2. Experimental results of two-wave coupling of exponential gain coefficient of Sc:Ru:Fe:LiNbO₃ crystal.

Samples	τw (s)	λ (nm)	δ (mm)	2θ (deg)	Γ (cm−1)
RF1	104	476	1.0	24.4	22.5
RF2	73	476	1.0	25.6	26.8
RF3	42	476	1.0	28.8	29.6
RF4	16	476	1.0	31.8	34.8

and

Table 3. Experimental results of two-wave coupling of exponential gain coefficient of Sc:Ru:Fe:LiNbO₃ crystal.

Samples	τw (s)	λ (nm)	δ (mm)	2θ (deg)	Γ (cm−1)
RF1	125	633	1.0	22.3	20.5
RF2	94	633	1.0	18.2	18.2
RF3	72	633	1.0	16.1	14.3
RF4	36	633	1.0	14.3	12.4

.

From Table 2 and Table 3, we can conclude that using the ${\mathrm{Kr}}^{+}$ laser (476 nm blue light), the exponential gain coefficient Γ in Sc:Ru:Fe:LiNbO₃ with Sc³⁺ ion concentration increased from 24.4 cm⁻¹ for RF1 to 31.8 cm⁻¹ for RF4. Using the He-Ne laser (633 nm red light), the exponential gain coefficient Γ in Sc:Ru:Fe:LiNbO₃ with Sc³⁺ ion concentration decreased from 22.3 cm⁻¹ for RF1 to 14.3 cm⁻¹ for RF4.

In Figure 3, comparing the Γ at 476 nm of RF1 with the Γ at 633 nm of RF1, it increased by a factor of 1.09; the Γ at 476 nm of RF2 compared with the Γ at 633 nm of RF2 showed an increase by a factor of 1.40; the Γ at 476 nm of RF3 compared with the Γ at 633 nm of RF3 showed an increase by a factor of 1.78; and the Γ at 476 nm of RF4 compared with the Γ at 633 nm of RF4 showed an increase by a factor of 2.2.

4. Blue Photorefractive Properties of the Sc:Ru:Fe:LiNbO₃ Crystal

The photorefractive properties of the Sc:Ru:Fe:LiNbO₃ crystal using the ${\mathrm{Kr}}^{+}$ laser (λ = 476 nm blue light) as a light source were investigated. The intensity of each beam of the ${\mathrm{Kr}}^{+}$ laser was about 140 mw/cm². The incidence angle was θ = 14°. The photorefractive properties were investigated using the two-beam coupling setup (see Figure 2). During recording, one of the writing beams was blocked intermittently, while the other writing beam served as a readout beam to measure the diffraction efficiency of the written grating. The diffraction efficiency η_s of the grating is defined as $\raisebox{1ex}{$I_S^{\prime }$}\!\left/ \!\raisebox{-1ex}{$I_S$}\right.\times 100\%$, where $I_S^{\prime }$ and I_S are the diffracted and transmitted light intensity, respectively. The temporal behavior of η_s during recording and erasing could be well described by the following functions [22]:

$$\sqrt{\eta (t)}=\sqrt{{\eta }_{sat}}\left[1-\exp (-1/\tau w)\right]$$

(4)

$$\eta (t)=\sqrt{{\eta }_{sat}}\exp (-1/\tau w)$$

(5)

where η_sat is the saturation diffraction efficiency during recording, and τ_w and τ_e are the recording (response time) and erasing time constants, respectively. Holographic data storage systems have long held promise for their high storage capacity, short access times and high data transfer rates. For holographic storage systems, two of the most important system parameters are dynamic range, M/#, and sensitivity, S. The larger the M/#, the higher the storage density and the better the signal-to-noise ratio. Sensitivity determines the recording speed. The higher the sensitivity, the shorter the recording time required. M/# and S can be measured by single-hologram recording and erasure experiments. From the single hologram recording and erasure curve, we can calculate the M/# and S values using the following formulas:

$$M/\#=\frac{d\sqrt{\eta }}{dt}|{}_{t=o}{\tau }_e\approx \frac{{\tau }_e\sqrt{{\eta }_s}}{{\tau }_w}$$

(6)

$$S=\frac{d\sqrt{\eta /dt|{}_{t=o}}}{il}\approx \frac{\sqrt{{\eta }_{\mathrm{s}}}}{{\tau }_{w}IL}$$

(7)

where τ_e is the erasure time constant and I and L are the total recording intensity and the thickness of the crystal, respectively. η_s is the maximal diffraction efficiency that is defined as the ratio of the diffraction light intensity to the transmission signal light intensity without a grating. The maximum refractive index change, ∆n_max, can be calculated using the following equation:

$$\mathsf{∆}{n}_{\max }=\frac{\arcsin ,\sqrt{{\eta }_s}\lambda \cos \theta }{180\times d}$$

(8)

where d is the thickness of the crystal, θ is the incident angle inside the crystal and λ is the wavelength outside the crystal. The maximum refractive index change ∆n_max can be calculated from the measured maximum diffraction efficiency.

From the results in

Table 4. The blue (λ = 476 nm) photorefractive properties of Sc:Ru:Fe:LiNbO₃ crystal.

Samples	${\mathit{\eta }}_{\mathit{S}}(\%)$	${\mathit{\tau }}_{\mathbf{w}}(\mathit{S})$	${\mathit{\tau }}_{\mathbf{e}}(\mathit{S})$	$\mathit{M}/\#$	S (cmJ−1)	∆nmax (10−5)
RF1	62.8	242	870	2.85	0.12	5.94
RF2	66.7	104	610	4.79	0.28	6.02
RF3	70.4	46	430	7.84	0.65	6.16
RF4	75.7	11	244	19.52	2.85	6.66

and

Table 5. The red (633 nm) photorefractive properties of Sc:Fe:LiNbO₃ crystal.

Samples	${\mathit{\eta }}_{\mathit{S}}(\%)$	${\mathit{\tau }}_{\mathbf{w}}(\mathit{S})$	$\mathit{M}/\#$	S (cmJ−1)	∆nmax × 10−5
Sc(1 mol%):Fe(0.06 wt%):LiNbO3	34.7	1450	2.66	0.0087	4.28

, we can see that the Sc:Ru:Fe:LiNbO₃ crystals and Ru:Fe:LiNbO₃ crystals both represent strong blue photorefraction. Moreover, increasing the Sc³⁺ concentration could enhance the blue photorefraction significantly. For the photorefractive properties of RF4 in the Sc:Ru:Fe:LiNbO₃ crystal, the diffraction efficiency η_s reaches 75.7%, the response time τ_w shortens to 11 (s), the dynamic range M/# reaches 19.52, the photorefractive sensitivity reaches 2.85 (cm J⁻¹) and the maximum refractive index change ∆n_max reaches 6.66 × 10⁻⁵. With an increase in Sc³⁺ ion concentration, the diffraction efficiency (η_s), response time (τ_w), dynamic range (M/#), sensitivity (s) and maximum refractive index change (∆n_max), from RF1 to RF4 under blue light irradiation, continually increase. Comparing Sc (3.0 mol%):Ru (0.1 wt%):Fe (0.06 wt%):LiNbO₃ (RF4) blue light with Ru: (0.1 wt%):Fe (0.06 wt%):LiNbO₃ (RF1), η_s is increased by a factor of 1.2, V_w (response rate) has a large increase by a factor of 22, M/# increases significantly by a factor of 6.8, S has a sharp increase by a factor of 23 and ∆n_max increases by a factor of 1.1. Comparing the ${\mathrm{Kr}}^{+}$ laser (476 nm blue light) as a light source to record the holographic storage parameter of Sc (1 mol%):Ru (0.1 wt%):Fe (0.06 wt%):LiNbO₃ with a He-Ne laser (633 nm red light) as a source to record holographic storage parameters of Sc (1 mol%):Fe (0.06 wt%):LiNbO₃, η_s is increased by a factor of 1.9, V_w (response rate) is increased by a factor of 13.9, M/# increased significantly by a factor of 1.8, S sharply increased by a factor of 32 [23] and ∆n_max improved by a factor of 1.4.

Blue Photorefractive Mechanisms in the Sc:Ru:Fe:LiNbO₃ Crystals

In the Sc:Ru:Fe:LiNbO₃ crystal, Ru and Fe for photorefraction play the dominant role. Sc³⁺ contributes to the photoconductivity, σ_ph. It is known that writing time is inversely proportional to the photoconductivity σ_ph according to the following equation [20]:

$${\tau }_w\approx \frac{\epsilon {\epsilon }_0}{4\pi {\sigma }_{ph}}$$

(9)

Doping Sc³⁺ ions into the LiNbO₃ crystal induces an increase in photoconductivity σ_ph at the 476 nm wavelength, leading to a fast photorefractive response. It is worth emphasizing that blue light is short enough to excite charge carriers in both shallow and deep centers; thus, a grating can be directly recorded in the two centers with the same phase. Ascribed to the two centers’ effect and in-phase merit at the 476 nm wavelength, the total intensity of the recorded grating is the sum of the two gratings recorded in the deep and shallow trap centers. Therefore, the photorefractive characteristics are improved dramatically compared with those in the single-center-doped LiNbO₃ crystals at the 476 nm wavelength. The mechanism of the blue photorefractive enhancement can be described as follows: doping Sc³⁺ ions into the Ru:Fe:LiNbO₃ crystal changes the occupied sites of the ions in the crystals. As a result, the concentrations of ${\mathrm{Ru}}_{\mathrm{Li}}^{2+}$, ${\mathrm{Fe}}_{\mathrm{Li}}^{+/2+}$ and ${\mathrm{Sc}}_{\mathrm{Li}}^{2+}$ are changed. Moreover, a large coupling gain coefficient is observed, and the energy transferring direction is always in the +c-axis direction in blue photorefraction, indicating diffusion is the dominant charge transport mechanism and the light-excited holes are the dominant charge carriers. Such phenomena are different from those in red and green photorefraction. The factors mentioned above contribute to the significant enhancement of blue photorefractive properties when comparing the Ru:Fe:LiNbO₃ with the Sc:Ru:Fe:LiNbO₃ crystals.

5. Conclusions

Doped with 0.1 wt% RuO₂, 0.03 wt% Fe₂O₃ and 0, 0.5, 1.0 and 1.5 mol% Sc₂O₃, Sc:Ru:Fe:LiNbO₃ crystals were grown by using the Czochralski method. The types of dominant charge carriers were determined, and the holes were the dominant charge carriers under blue light irradiation, which is short enough to excite the charge carriers’ holes in both shallow and deep trap centers. Therefore, the grating can be directly recorded in the two centers with the same phase. The maximum exponential gain coefficient Γ of RF4 under 476 nm irradiation reached 31.8 cm⁻¹. The Γ at 476 nm (blue light) of RF4 compared with the Γ at 633 nm (red light) of RF4 increased by a factor of 2.8. The exponential gain coefficient Γ at 476 nm (RF1, RF2, RF3 and RF4) continually increased. The photorefractive properties of η_S, ${\tau }_w$, M/#, S and ∆n_max reached 75.7%, 11 s, 19.52, 2.85 cmJ⁻¹ and 6.66 × 10⁻⁵. Comparing the ${\mathrm{Kr}}^{+}$ laser as a light source to record the holographic storage parameters of Sc(1 mol%):Ru(0.1 wt%):Fe(0.06 wt%):LiNbO₃ with the He-Ne laser as a source to record the holographic storage parameters of Sc (1 mol%):Fe (0.06 wt%):LiNbO₃, η_S increased by a factor of 1.9, V_w (response rate) increased by a factor of 13.9, M/# increased by a factor of 1.8, S increased by a factor of 32 and ∆n_max improved by a factor of 1.4. This short wavelength recording in the two-center LiNbO₃ also offers a possible improvement for high-density volume holographic storage.