Multiplexing of 100 holograms

4.3 Experimental setup

4.4.3 Multiplexing of 100 holograms

In order to verify that the M/# is real and achievable for practical applications, 100 holograms have been multiplexed in sample S1 (extraordinary light polarization, light wavelength 458 nm). Figure 4-4 shows the comb function of the 100 multiplexed holo-

Fig. 4-4. Comb function of multiplexing 100 holograms in sample S1: LiNbO₃ doped with 0.2 atomic% Mn. The M/# calculated from this comb function is 5.

grams. We used a pre-calculated exposure schedule to equalize the diffraction efficiency.

The selectivity of holograms is about 0.185^o and we chose the angle between two neigh- boring holograms to be 0.4 degrees. The M/# we got from this multiplexing, which was cal- culated by

, (4-4)

is 5.0/mm, where η_n is the diffraction efficiency of the nth hologram. The loss of some M/#

is due to the non-ideal exposure schedule.[4-12] The larger diffraction efficiencies around the center of the comb function are due to some back reflection. The M/#’s obtained from both -- single-hologram recording and erasure, and multiplexing -- agree very well.

4.4.4 M/# and sensitivity vs. oxidation state

The oxidation state of LiNbO₃ crystals can be changed by annealing at elevated temperature in appropriate atmosphere, typically oxygen for oxidation and argon for reduc- tion. It is well known that M/# and sensitivity in LiNbO₃:Fe crystals are strong functions of oxidation state. Typically, the more the crystal is reduced, the larger the sensitivity and the smaller the M/#, and vice versa. We also measured M/# and sensitivity in sample S3, a Table 4-1. Summary of thermal treatments of sample S3.

Oxidation state Thermal treatment

State 1 Highly-oxidized, starting with state 6, in oxygen at 930°C for 24 hours.

State 2 Starting with state 1, in argon at 780°C for 1 hour.

State 3 Starting with state 2, in argon at 780°C for 3 hours.

State 4 Starting with state 3, in argon at 780°C for 4 hours.

State 5 Starting with state 4, in argon at 780°C for 11 hours.

State 6 Highly-reduced, in vacuum at 1000°C for 14 hours, then in oxygen at 925°C for 4 hours.

M #⁄ η_n

n

∑

=

LiNbO₃ crystal doped with 0.2 atomic% Mn, with different oxidation states. Sample S3 was cut from the same boule as S1 and was also proton-reduced and with the same size as sample S1. Table 1 contains information about the thermal treatment for each oxidation state of S3. The method of quantitatively calibrating the oxidation/reduction state of LiNbO₃:Mn is still missing, therefore the ratio C_Mn²⁺/C_Mn³⁺ could not be determined for each oxidation state. The measured M/#’s and sensitivities for different oxidation states are shown in Figure 4. Surprisingly, the sensitivity in sample S3 is almost the same, 0.5 cm/J,

Fig. 4-5. Measured sensitivity and M/# vs. oxidation state in sample S3: LiNbO₃ doped with 0.2 atomic% Mn.

and it is almost independent of the oxidation states, while the M/# drops by a factor of 15 from the highly oxidized to the highly reduced state. The highest M/# was obtained for the highly-oxidized state. This independence of sensitivity on the oxidation state in LiNbO₃:Mn is in strong contrast to the trade-off between M/# and sensitivity in LiNbO₃:Fe and is good for holographic applications.

One possible way to account for the lack of a trade-off between M/# and sensitivity in the LiNbO₃:Mn crystal is by assuming that a large majority of the Mn traps are occupied by electrons even in highly oxidized crystals. Therefore the sensitivity, which is propor- tional to the filled trap density, does not change much when the LiNbO₃:Mn crystal is oxi- dized and a small percentage change in the Mn²⁺ concentration takes place. On the other hand, the percentage change of Mn³⁺ is large, which leads to the increase in M/# as the crystal is oxidized. Since we assume most of the Mn traps are Mn²⁺, considering the doping level (0.2 atomic% Mn), the sensitivity could be high. Nevertheless, this assumption and the one-center model are not self-consistent. In Chapter 5 an alternative charge transport model will be proposed to explain all the photorefractive phenomenon of LiNbO₃:Mn crys- tals.

4.5 Holographic recording in LiNbO ₃ : 0.5 wt% MnCO ₃

We have shown that the dark decay in manganese-doped lithium niobate crystals with doping level as high as 0.2 atomic% Mn is still dominated by proton compensation.

The measured M/# and sensitivity in LiNbO₃: 0.2 atomic% Mn are pretty large compared to those we got from LiNbO₃:Fe. Now we would like to know what the limit is, i.e., what is the highest practical doping level in LiNbO₃:Mn for the application of holographic stor-

age, and what kind of M/# and sensitivity we can obtain from these crystals. Fortunately, we have several manganese-doped lithium niobate crystals with doping level of 0.5 wt%

MnCO₃ available in our labs. It turned out that the highest practical doping level in man- ganese-doped lithium niobate for holographic storage is about 0.5 wt% MnCO₃. The larg- est M/# and sensitivity measured in these crystals are 90/cm and 1.4cm/J with the wavelength of 458nm and extraordinary polarization and the optimal oxidation state for these crystals is highly oxidized. Moreover, some other advantages of manganese-doped lithium niobate crystals for holographic storage, such as excellent recording stability and repeatability, no holographic scattering, have been observed.