A set of twelve specially doped lithium niobate crystals were grown to test the effect of the dopant on holographic recording in the crystals via the photorefractive effect. The crystals were doped with Ce, Co, Cr, Cu, Fe, Mn, Ni, Rh, Tb, Fe:Ce, Fe:Cr, and Fe:Mn. The transmission spectra was measured for each crystal and holograms have been written in each of the crystals with wavelengths from 457 nm to 671 nm. The wavelength sensitivity, scattering, and stability of the holograms varied substantially among the crystals. A qualitative description of the hologram’s properties and a comparison of sensitivities between the crystals will be presented.
© 1998 Optical Society of America
Lithium niobate has been investigated for use as a volume holographic data storage material for nearly two decades, beginning soon after Ashkin et al.  reported on optical damage in lithium niobate crystals. Lithium niobate’s advantages are the large crystal sizes available which allow for increased capacity and high angular selectivity, long storage lifetime among photorefractive crystals, and high diffraction efficiencies. It is a good candidate for holographic data storage in systems like optical correlators where large numbers of analog images must be accessed at high speed. Angle multiplexing holograms in lithium niobate allows parallel access to up to a thousand images which can be correlated against an input scene in a fraction of a second. However, traditional iron doped lithium niobate has a peak sensitivity for writing holograms in the blue-green portion of the visible light spectrum. A sensitivity to red and infrared wavelengths would be desirable so small inexpensive diode lasers could be used for writing and reading the holograms.
The optical damage first reported in lithium niobate was determined to be caused by iron impurities in the crystals. Iron has been the primary dopant used to enhance the photorefractive effect in lithium niobate but other dopants and crystal oxidation/reduction have been shown to alter the photorefractive properties. In the last few years research has been conducted on the effects of transition metals other than iron as dopants with the possibility that the wavelength sensitivity and other holographic properties of lithium niobate may be changed., The experiments described below were done with twelve lithium niobate crystals, each with a different single or double dopant combination. The dopants were chosen among the transition metals for the ions’ similar size to iron, and having two neighboring common oxidation states. The investigation of these crystals focused on how the dopant effected the crystals’ ability to record and store holograms.
2. Crystal composition
The composition of each of the twelve doped lithium niobate crystals is unique. The crystals were grown at Nankai University in Tianjin, China. The dopant and form of the dopant added to the starting melt of each crystal is listed in Table 1 along with the three letter name given to each crystal. The names were used to easily identify data files during experiments. EPR analysis of the crystals identified impurities which likely affected the test results. The primary impurity in all the crystals was manganese.
3. Transmission spectra
A Shimadzu spectrophotometer was used to measure each crystal’s transmission spectrum from 350–900 nm. The spectrophotometer uses two beams of light, one passes through the sample while a reference beam passes through air. The beam intensities are then compared to determine the percentage of transmission. A mask was used with each crystal to ensure no stray light went around the crystal. An identical mask was used in the reference leg of the spectrophotometer to ensure accuracy.
The crystals have a wide range of colors as can be seen in Figure 1. The variety in color shows up in the transmission spectra shown in Figures 2–13. The dopants strongly affect the spectra. The variation in transmission indicates that the wavelength sensitivity of the crystals should also vary.. A fifty percent absorption is considered good for photorefractive sensitivity because it absorbs enough energy to record a hologram, but still passes sufficient light that a signal beam can pass through and good diffraction efficiency can be achieved. The wavelengths where the transmission spectra pass through the 50% mark vary from under 400 nm to 900 nm.
Figure 9 shows that copper doping increases absorption around 900 nm suggesting that Ned may have photorefractive sensitivity in the near infrared. Buse et al. has written holograms in copper doped lithium niobate using 1064 nm using a two photon process. They used a 532 nm beam to excite the electrons to a higher level. The use of two wavelengths also allowed for nondestructive readout. Moe, the nickel doped crystal, also shows good absorption around 730 nm indicating potential red and near infrared sensitivity.
4. Holographic recording
Attempts were made to record holograms in each of the crystals at the wavelengths of 457 nm and 514 nm from an argon ion laser, 647 nm from a krypton laser, and 671 nm from a diode laser. An identical experimental arrangement was used for each crystal with a given wavelength. However, the setup, including angle between beams, beam ratios, total power, and exposure time, varied from wavelength to wavelength. While holograms were successfully written in most crystals, some crystals would not record a hologram with the wavelengths and powers used during these experiments. If a hologram was not recorded with the standard exposures used, additional attempts were made with longer exposures and more power, if available, before determining no hologram could be recorded. The holographic sensitivity was measured as the relative strength of gratings formed when each crystal was given an identical exposure to two plane waves. No attempt was made during these experiments to determine the maximum diffraction efficiency of any of the crystals; only relative sensitivity.
To record a hologram, each crystal was placed in a setup similar to Figure 14. A hologram was recorded in each crystal by exposing it to beams A and B for the specified time. After the exposure, beam B was blocked. The crystal was illuminated with beam A and the intensity of the diffracted beam was immediately read at the detector using a Newport model 835 power meter. The standard exposure times and powers used for each wavelength are listed in Table 2.
The light transmitted by beam B through the crystal was measured just as the exposure began. Using the incident and transmitted power of beam B, a comparison can be made against the transmission spectra for each wavelength. The amount of transmitted light varied among the crystals by approximately an order of magnitude as shown in Tables 3 through 6 and expected from the transmission spectra. The third column of the tables show the power in the diffracted beam, the reconstruction of beam B. Table 4, for the holograms written with 514 nm, also shows the trend of the diffracted power as the hologram was continuously read out with beam A. The holograms appeared to decay rapidly in some crystals, most strongly Moe, while increasing slightly in others. Since there is no direct comparison between the results for the different wavelengths due to unavoidable variations in the setups available, Table 7 shows the average diffracted power for each crystal normalized to the diffracted power of Pat, the iron single-doped crystal.
For the 457 nm wavelength, the iron single-doped crystal had the strongest response. The Fe:Ce and Fe:Mn double-doped crystals’ appear to depress the crystals ability to record a hologram since they had only half as much diffracted power as the single Fe doped crystal. However, the Ce and Mn single-doped crystals had the strongest response among the single-doped crystals followed by Cu. The Tb doped crystal’s diffracted power appears higher than is justified because it has such a high percentage transmission. Its hologram was very weak compared to the others and the reading was partly due to scattered light. The Cr double-doped crystal was more than an order of magnitude more absorbing than the other crystals and so the diffracted beam was weak even though there was a strong hologram recorded.
The Fe:Mn doped crystal is the most sensitive to the 514 nm wavelength. The Ce and Fe:Ce doped crystals also showed very strong holograms. Eve’s apparent strong response is partly due to its low absorption. The Fe doped crystal is less sensitive than several of the other single-doped crystals. Again the Fe:Cr doped crystal recorded a strong hologram, but had a weak diffracted beam due to high absorption.
For the 647 nm wavelength, manganese seemed to suppress hologram formation. The Mn single-doped crystal had no holograms while the Fe:Mn doped crystal had good holograms, but they were only 1/5 the strength of the Fe single-doped crystal. The Ni, Ce, and Cu crystals were the best of the single dopant crystals after Fe.
The Fe:Ce double-doped crystal had the strongest holograms at the 671 nm wavelength. The Ni crystal had the strongest holograms among the non-Fe doped crystals. It had five times stronger holograms than the Ce single-doped crystal. This result suggests that a Fe:Ni doped crystal may be much more sensitive than even the Fe:Ce crystal. Again, Mn seemed to suppress hologram formation. The Mn single doped crystal had no holograms and the holograms in the Fe:Mn crystal were an order of magnitude weaker than with the single doped Fe crystal.
The set of twelve crystals exhibited a wide range of responses within the portions of the visible spectrum investigated. The iron single and double-doped crystals as a group had the strongest response. Adding other dopants in conjunction with iron can reduce or enhance the ability to write holograms. For recording at 514 nm, the addition of manganese seem advantageous. The single nickel doped crystal had a strong response to 671 nm and deserves further investigation for applications that need the use of low powered diode lasers. While the holograms were strong in the nickel doped crystal, they were also unstable and degraded extremely quickly which is diadvantageous for long term data storage butmay be suitable for dynamic holography. The Fe:Cr double-doped crystal recorded strong holograms with every wavelength, but had very high absorption and scattering problems. The dopant combination may work better with half the amount of Cr added to the crystal and probably less Fe too. Several crystals did not record good or any holograms. The Tb and Rh doped crystals had weak responses at best.
The unintended impurities also influenced the results and must not be ignored. It must also be kept in mind that the holograms can be compared for a given wavelength, but no quantitative comparisons can be made between the wavelengths because of the variations in the experimental arrangements for each wavelength.
References and links
1. A. Ashkin, G.D. Boyd, J.M. Dziedzic, R.G. Smith, A.A. Ballman, J.J. Levinstein, and K. Nassau, “Optically-induced refractive index inhomogeneities in LiNbO3 and LiTaO3,” Appl. Phys. Lett. 9 (1), 72–74 (1966). [CrossRef]
3. S. Yin and F.T.S. Yu, “Specially doped LiNbO3 crystal holography using a visible-light low-power laser diode,” IEEE Phot. Technol. Lett. 5 (5), 581–582 (1993). [CrossRef]
4. K. Buse, F. Jermann, and E. Kratzig, “Infrared holographic recording in LiNbO3:Fe and LiNbO3:Cu,” Opt. Mater. 4, 237–240 (1995). [CrossRef]
5. A. M. Darwish, M. D. Aggarwal, J. C. Wang, R. Copeland, P. Venkateswarlu, P. P. Banerjee, D. K. McMillen, and T. D. Hudson, “Investigation of the charge transfer and the photosensitivity in single- and double-doped LiNbO3 single crystals: an optical-electron paramagnetic resonance study,” Proc. SPIE 3137, 63–74 (1997). [CrossRef]
6. K. Buse, F. Jermann, and E. Kratzig, “Infrared holographic recording in LiNbO3:Fe and LiNbO3:Cu,” Opt. Mater. 4, 237–240 (1995). [CrossRef]