NONLINEAR CRYSTALS FOR GENERATING UV LIGHT
INTRODUCTION
Generating coherent light in the ultraviolet (UV) is important for a broad range of applications including materials processing, medical surgery, diagnostics, scientific experiments, and many others. A practical way to generate UV light is by converting more easily obtained infrared or visible laser beams into the UV light using nonlinear optics--harmonic generation processes in crystals. This works well for producing light in nearly all of the UV A range (315 nm – 400 nm), the longest of the UV wavelengths. There are several crystals that can do this job. Finding suitable crystals for generating shorter wavelengths, including crystals for the deep UV, becomes more difficult.
WHAT A NONLINEAR CRYSTAL MUST BE ABLE TO DO
A crystal that can efficiently generate harmonics needs to be meet many criteria [See references 1, 2]. To be efficient it must provide a strong nonlinear interaction. Also, the nonlinear interaction needs to be cumulative over the length of the crystal, a condition called phase matching, which depends on the wavelengths and polarizations of the light involved as well as the crystal’s properties [See reference 3]. To allow phase matching, the crystal’s refractive indices must be unequal for light polarized along different crystal axes--this is called birefringence. This is to balance the change in refractive index due to chromatic dispersion—shorter wavelengths see larger refractive indices. The birefringence must be greater for generating shorter wavelength UV light because dispersion is greater at shorter wavelengths, so nonlinear crystals for generating UV light need especially large birefringence. This further limits the number of available crystals that can generate shorter wavelength UV light.
Crystals for generating UV light also need other characteristics: They must be transparent in the desired range of the UV light. They need to be resistant to laser-induced optical damage at all wavelengths in use, especially in the UV where damage is more likely. The crystals need to be able to be grown to sizes that are sufficient for their applications, with good optical homogeneity, and with chemical and mechanical stability. They must support precision optical fabrication, and it is highly desirable that the crystals are not hygroscopic. In addition, it is problematic if there are safety concerns regarding the growth, fabrication, or handling of the crystals.
CRYSTALS FOR GENERATING UV
Currently, there are many research efforts underway to create new nonlinear crystals, particularly those capable of generating short wavelength UV light. However, at this time there are only a very limited number of practical UV nonlinear crystals. It can be illustrative to discuss nonlinear materials based on what harmonics they can produce of the well-known infrared Nd:YAG laser line at 1064 nm. This 1064 nm line is itself important, and is representative of many other important solid-state laser lines with similar wavelengths. For example, Yb-doped fiber lasers typically operate between about 1030 nm and 1090 nm.
Table 1. A brief compilation of some crystals that can be used to generate the harmonics of the Nd:YAG laser line at 1064 nm.
It is harder to find materials to generate the shorter UV wavelengths. Some of the more constraining challenges at shorter wavelengths are that at these wavelengths fewer crystals transmit well, crystals suffer optical damage more easily, and there are more limitations on phase matching, as discussed earlier. Shorter wavelengths are more difficult. [See our Blog on Wavefront Distortion]
SOME SPECIFIC WAVELENGTHS AND CRYSTALS FOR GENERATING THEM
For second harmonic generation starting with 1064 nm infrared light and generating 532 nm green light, there are multiple good options. Some work better with relatively low power beams (Mg:LiNbO3 or KTP); others work well with high power beams. Lithium Triborate crystals (LBO crystals) are damage resistant, and LBO is great material for high average power applications. KDP is included in this list since it is a standard by which other materials are evaluated because it has been highly useful in low repetition rate, high energy scientific applications where the fact that KDP can be grown in large sizes is enabling. However, KDP does not excel in terms of conversion efficiency, thermal conductivity, or handling of high average powers.
For generating 355 nm UV light using sum frequency generation, there are fewer crystals that work. LBO crystals give excellent results when used with high peak power beams, but do not work efficiently in low peak power beams. There are other options in Table 1, which like LBO, have superior conversion efficiency (higher nonlinearity) compared to KDP.
There are fewer options yet for generating 266 nm UV light, which is the second harmonic of 532, which is itself the second harmonic of 1064. LBO does not have enough birefringence to phase match this 532nm + 532nm process. LBO will phase match 355 + 1064 → 266nm, but that approach has other drawbacks. It requires an additional nonlinear step, and each step typically reduces efficiency and boosts costs. BBO crystals are the common solution for generating 266 nm. BBO has good nonlinearity, but it has a lot of walkoff, which may reduce both the conversion efficiency to the UV and the UV beam quality. BBO is often the best choice for generating 266 nm light.
As noted above, 213 nm can be generated in BBO (266nm+1064nm → 213nm), but we are not aware of many applications using 213 nm light, and this process requires three nonlinear steps, one more than is needed for generating 266 nm, and the same as would be required to generate 177 nm.
People are very interested in making deeper UV light; the 177 nm 6th harmonic of Nd:YAG is desired. Currently, KBe2BO3F2 (KBBF) and RbBe2BO3F2 are the only reported materials to successfully generate 177 nm, with most of the results being from KBBF. This material has promise, but is strictly experimental—it has not been grown in dimensions of more than about 5 mm. With a nonlinearity similar to KDP (which is not exceptional), it would be desirable to have much longer KBBF parts. Also, BeO is used in the growth of these crystals, and BeO is toxic.
FUTURE POSSIBILITIES
There are other approaches for research on crystals that can generate deep UV through nonlinear conversion. Periodically poled materials might work [See reference 4]. These are equivalently called quasi-phase-matched materials. We are unaware of such crystals for use in the UV, and making periodically poled materials for the UV will be a challenge, particularly making high damage materials or large aperture materials. Organic crystals may offer potential, and are being investigated [See reference 5]. They are often limited by a narrow transparency window, challenges in growing the materials, and issues with mechanical instability.
CONCLUSIONS
There is an active research community working on nonlinear crystals for generating shorter wavelength UV light. Currently, LBO is an excellent choice for generating wavelengths near 355 nm, and BBO is a preferred choice for generating wavelengths near 266 nm. The options for producing shorter wavelength UV light are very limited and are the subject of ongoing investigations.
DR. WILLIAM GROSSMAN, AUTHOR
Will Grossman is a consultant retained by GAMDAN, and his role is to help our customers be more successful with nonlinear optics. His technical expertise includes laser design, nonlinear optics, and laser reliability. Dr. Grossman’s laser designs are used around the world in commercial products. More on the author can be found here.
If you enjoyed this article, consider subscribing to our NONLINEAR CRYSTAL BLOG by signing up below.
REFERENCES
1. Halasyamani, P. Shiv, and Weiguo Zhang. "inorganic materials for UV and deep-UV nonlinear-optical applications." Inorganic Chemistry 56, no. 20 (2017): 12077-12085.
2. Kang, Lei, and Zheshuai Lin. "Deep-ultraviolet nonlinear optical crystals: concept development and materials discovery." Light: Science & Applications 11, no. 1 (2022): 201.
3. For a good reference on nonlinear optics, including historical citations, see: Boyd, R. W. "Nonlinear Optics–3rd Edition, wyd." AP, New York, USA (2007). Academic Press, Burlington, Mass, USA (2008).
4. The following is a paper on the status of using quasi phase-matching (periodic poling) for UV nonlinear optics: Pai Shan, Zujian Wang, Rongbing Su, Chao He, Xiaoming Yang, and Xifa Long. "Research progress of quasi-phase matching deep-ultraviolet nonlinear optical crystals." Chinese Journal of Quantum Electronics 38, no. 2 (2021): p. 180.
5. An example of a recent article on creating organic crystals for UV nonlinear optics is: Barhum, Hani, Cormac McDonnell, Tmiron Alon, Raheel Hammad, Mohammed Attrash, Tal Ellenbogen, and Pavel Ginzburg. "Organic Kainate Single Crystals for Second-Harmonic and Broadband THz Generation." ACS Applied Materials & Interfaces 15, no. 6 (2023): 8590-8600.