until now.
Figure 2.2 Light amplification by stimulated emission of radiation in the optical resonator. (a) Spontaneous emission and the starting point of radiation. (b) Middle step of stimulated emission. (c) Amplification in resonator and laser emission.
2.3 Laser Ceramics
2.3.1 Synthesis of Garnet‐Based Materials
Representative examples and flow diagram of the production of Nd:YAG ceramics are shown in Figure 2.3. α‐Al2O3, Y2O3, and Nd2O3 (purity: >99.99%) with a particle size of 0.3, 0.05, and 0.5 μm were used as starting powder materials. About 0.5% TEOS (Tetra Ethel Ortho Silicate), an organic binder, and ethanol solvent were added to these powders and ball‐milled for 12 hours using high‐purity Al2O3 balls (purity: >99.9%) [1]. The obtained slurry was spray‐dried using a spray dryer to obtain granules of approximately 30 μm. The granules were placed in a metal mold, uniaxially pressed at 10 MPa, and further subjected to CIP (Cold Isostatic Press) machine at a pressure of 147 MPa. The molded powder compact was calcined at 800 °C for three hours and then preliminarily sintered at 1500 °C for two hours under a vacuum of 1 × 10−3 Pa. Finally, it was heat‐treated in the capsulated‐free HIP (Hot Isostatic Press, 198 MPa with Ar gas) machine at 1700 °C for two hours to obtain a transparent body.
Figure 2.3 Fabrication flow sheet of Nd:YAG ceramics.
By the way, we synthesize materials by mixing oxides of constituent elements and then reacting them by reactive sintering. Reactive sintering tends to be non‐uniform in composition and has been considered unsuitable for the development of high‐quality ceramics. Until now, high‐quality ceramics such as TYZ (Y2O3 stabilized ZrO2), Tb3Fe5O12, Yb:CaF2 have been developed using raw material powders that have been chemically homogenized using the coprecipitation method or the alkoxide method [2–4]. It is one method of material synthesis to prepare a sintered body with excellent homogeneity by using highly uniform powder obtained by the coprecipitation method. The coprecipitation method as a method of obtaining high‐quality material is widely known, and the above materials have been synthesized using this method from the 1980s. Even with this method, it is possible to produce a laser gain medium capable of laser oscillation. However, the synthesis of raw materials is complicated i.e., when synthesizing ceramic materials of various compositions, the target product cannot be synthesized unless the corresponding raw material powders are synthesized. Another problem that it is difficult to correct the composition (materials with a stoichiometric composition such as garnet cannot have a slight compositional variation because its solid solubility limit is very sharp and narrow) and hence, the production cost is very high.
On the other hand, high‐grade Nd:YAG ceramics could be prepared by reactive sintering method without using such a complicated synthesis. This fact is also contrary to the technical background in the synthesis method and the conventional development philosophy. However, the result is a fact, the powder obtained by the above‐mentioned wet synthesis is complicated and expensive, and the reactive sintering method is an extremely simple and low‐cost synthetic process obtained by mixing oxides of constituent elements. These are the advantages of the dry process. The sintered body obtained by the HIP treatment is machined to the required shape, and the input and output surfaces of the sample are optically polished into laser grade; basically, flatness λ/10 (λ = 633 nm), micro‐roughness Rms = 0.3 nm, and parallelism = 10 seconds. These samples are used for optical measurement and laser oscillation tests.
2.3.2 Laser Oscillation by Monolithic Garnet Ceramics
Figure 2.4a is a photograph of the world's first highly efficient laser oscillation using the Nd:YAG ceramic developed by the author in December 1991. On this day, because the CW (continuous oscillation) equipment was malfunctioning and also the sample was not optically polished and was not AR (Anti‐reflection) coated, an 808 nm LD light was input from the end surface of the disk‐shaped ceramic sample, and Ar gas laser excitation was applied from the side. Then, the Quasi‐CW laser was oscillated by using this setup. A pulsed oscillation signal was obtained from the oscillator, and this became the world's first invention of a high‐efficiency ceramic laser. Figure 2.4b shows the ceramic sample tested with a CW laser oscillator at a later date, and the slope efficiency was 28%. The optical loss of this ceramic was about 0.9%/cm @ 1 μm, but at that time this sample was an unprecedented ultra‐high‐quality ceramic. Since the laser gain length was not sufficiently long, the leakage of LD light was also large, and the excitation technique at that time was also insufficient, so the slope efficiency could reach only 28%. However, having the same laser oscillation characteristics comparable to the commercially available Nd:YAG single crystal which was used as a reference means that the material properties of these ceramics are “the highest level” as a laser gain medium at that time. There is no doubt about this. It is not a mistake that confirming the first laser oscillation here is 1991, because the publication in Japan was in 1994, and the publication in the US journal was in 1995. Since the late 1990s, the interest and importance of ceramics lasers have been recognized in the US, and research on ceramics lasers has been conducted worldwide. Representative results presented by the author and others are described in the following.
Figure 2.4 (a) First demonstration of Q‐CW laser oscillator using Nd:YAG ceramics in December 1991 and pulse laser oscillation spectrum from Nd:YAG ceramics. (b) Incident power vs. output of cw‐laser from Nd:YAG ceramics pumped by diode laser.
Figure 2.5a shows the oscillation characteristics of 7%Yb:YAG ceramics. Although the quantum limit efficiency of this material is 91%, the slope efficiency reaches 84% even when excited by a laser diode (LD) at 940 nm [5]. This material has a thickness of about 1 mm, and there is a leakage of pumped LD light at 940 nm; hence, if the horizontal axis is regarded as an absorbed power, the slope efficiency may reach nearly 88%. This means that the optical loss of the material is very small, and high laser conversion efficiency has been realized. Figure 2.5b shows a laser oscillation test using a thin‐disk type 10%Yb:YAG (Ф12 × t0.15 mm). The maximum output achieved is 1.8 kW, and the slope efficiency is 74%, which is very high [6].