A modified Barium Gallo-Germanate glass has been developed as an exit window for high energy lasers operating in the mid-infrared wavelength region. All the physical properties, for application as a window for high energy laser systems have been measured. Absorption loss and thermo-optic coefficient were identified as key in developing the Barium Gallo-Germanate glass for high energy laser applications. A purification method was developed to reduce the absorption loss of the glass from 6×10-2 cm-1 to 2×10-3 cm-1 at 3.8 µm. Manufacturability in large size windows has been demonstrated with the fabrication of an 18″ diameter prototype window. Modified Barium Gallo-Germanate glasses have also been developed with lower thermo-optic coefficient resulting in lower optical path distortion.
©2006 Optical Society of America
The windows in High Energy Laser (HEL) systems often become the single point of failure because of absorption of the laser energy. When a high energy beam passes through a window, part of the energy is absorbed in the material and results in local heating. This heating causes thermal expansion in the material and a change in its refractive index which distorts the transmitted beam. Optical Path Distortion (OPD) is used to describe the quality of the transmitted HEL beam. The thermal effects often limit the time the HEL beam can stay ON before it is shut off for window cooling. This significantly restricts the performance of the HEL system. Sometimes these thermal effects can cause a window to fail catastrophically. It is important to understand the material properties contributing to these thermal effects so that better HEL windows can be designed. In addition to the thermal effects, the windows must also be very strong and rugged, with manufacturability in large sizes and complex shapes at low cost.
A Figure of Merit (FOM) is used by the design engineers to evaluate materials for HEL window applications. The FOM for OPD, expressed as FOM(OPD), is a combination of thermal distortion FOM(χ), Temperature Rise FOM(ΔT) and Thickness FOM(L). A low FOM is desired for better HEL performance.
The three FOM’s are expressed as:
CTE=coefficient of linear thermal expansion
dn/dT=thermo-optic coefficient (change in refractive index as a function of temperature)
q11, q12=stress optic coefficients in directions ‖ and ⊥ to the load respectively
Cp=heat capacity at constant pressure
Multi-spectral Zinc Sulfide (ZnS) is typically used as an exit window for HEL applications at a wavelength of 3.8 µm (using Deuterium Fluoride laser). Although, Multispectral ZnS is very soft and has very high dn/dT and FOM(OPD), it is still used as an HEL exit window because of lack of a better material in large sizes. Barium Gallo-Germanate (BGG) glass is 3 times harder than multi-spectral ZnS. BGG glass is isotropic, has excellent transmission in visible and mid-IR region and is also available in large sizes. This study focused on developing and evaluating BGG for HEL applications.
2. BGG Glass
The base BGG glass is made in the ternary system of BaO, Ga2O3 and GeO2. The BGG glass is made by mixing appropriate ratios of BaCO3, Ga2O3 and GeO2, melting in platinum crucible at temperatures of about 1350°C and casting in molds of the desired shape. The BaCO3, which is available in higher purity than BaO, decomposes to BaO at elevated temperatures. Figure 1 shows a broad glass forming region in the BGG system.
Physical properties of a typical BGG glass are shown in Table 1. The physical properties of BGG glass vary as a function of glass composition and can therefore be tailored [1–3]. The properties of BGG glass can also be modified by adding/substituting other components such as Al2O3, Y2O3, Gd2O3, La2O3, and In2O3 [4, 5].
3. Developing BGG glass for HEL applications
Typically BGG glasses have a strong absorption in 3–3.5 µm wavelength region because of the presence of the hydroxyl ion (OH-) as impurity. It also contributes to significantly high absorption loss at 3.8 µm. This hydroxyl absorption in BGG glass is well characterized by Jewell . Consequently, a purification method was developed to eliminate OH- impurity from the BGG glass . The concentration of hydroxyl ion in the BGG glass was reduced from a value of 30 ppm down to ~1ppm in the purified glass. The absorption coefficient of BGG glass at 3.8 µm wavelength was reduced from 6×10-2 cm-1 down to below 2×10-3 cm-1.
Figure 2 shows the absorption loss in BGG glass before and after purification. The absorption loss, on 0.5″ thick samples, was measured using an infrared spectrometer. As is evident from the absorption loss plot of the sample before purification, the loss at 3.8 µm is a strong function of OH- impurity and CO2 impurity. The plot for the current sample exhibits significantly reduced OH- impurity. The loss curve at about 3.5 µm goes to the base line, representing lower loss, before rising again at about 3.7 µm due to the CO2 impurity. This is because of the tail of the CO2 absorption peak contributes to the loss at 3.8 µm. We concluded that further loss reduction at 3.8 µm wavelength could be achieved by glass purification to remove CO2 impurity. If the CO2 impurity is eliminated, the absorption loss at 3.8 µm can be reduced to approach the theoretical limit of 5×10-5 cm-1. The theoretical minimum loss of BGG glass was obtained as the intersection between the extrapolated multi-phonon edge and the calculated Rayleigh scattering loss on a log loss vs. 1/wavelength plot. Vis-IR transmission through a 0.5″ thick purified BGG glass is shown in Fig. 3. The transmission trace includes about 13.6% reflection loss from the two surfaces which can be minimized by the application of anti-reflective (AR) coatings.
4. Characterization of BGG glass for HEL applications
In order to evaluate BGG glass for HEL applications, all the material properties, identified in Eqs. (2)–(4), were measured. All the measurements were performed at University of Dayton Research Institute (UDRI) as per the ASTM standards. The results are summarized in Table 1 where the values are compared with multi-spectral ZnS. The values for ZnS were obtained from Klocek . As can be seen from Table 1, BGG glass is about 3 times harder and 2 times stronger then multispectral ZnS. BGG glass also has about 5 times lower dn/dT compared to multispectral ZnS.
Using the physical properties from Table 1, the thermal distortion FOM(χ), Temperature Rise FOM(ΔT) and Thickness FOM(L) and overall FOM(OPD) were calculated. These FOM’s are plotted in Fig. 4. The FOM(χ), FOM(L) and FOM(OPD) for BGG are lower than those for multispectral ZnS. The FOM(ΔT) for BGG is higher compared to multispectral ZnS because of its higher absorption loss. Although the absorption loss in BGG glass has been reduced from 6×10-2 cm-1 to 2×10-3 cm-1, it is still high compared to an absorption loss of 6×10-4 cm-1 for multispectral ZnS. As discussed earlier, further reduction of absorption loss in BGG glass is possible with the removal of CO2 impurity and can be significantly lower than ZnS. The overall FOM(OPD) of BGG is about 40% lower than that for multispectral ZnS. If the absorption loss in BGG can be reduced to 6×10-4 cm-1 (comparable to multispectral ZnS) the FOM for overall OPD would be 5 times (80%) lower.
BGG glass can also be fabricated in large sizes and complex shapes, including complex optics. Figure 5 shows an 18″ diameter BGG glass window and a 3.5″ dome. Because of the ease in manufacturability and processing, BGG glass windows are estimated to be significantly cheaper than ZnS windows.
5. ATHERMAL BGG Glass
Additional improvements can be made to further lower the overall OPD of the BGG glass. These improvements include reducing the dn/dT of BGG glass to lower FOM(χ), reducing the absorption loss to lower FOM(ΔT) and increasing the strength of BGG glass to reduce FOM(L). It is also possible to make the dn/dT term negative (-dn/dT) such that the dn/dT term in Eq. (2) balances the sum of the first term [(n-1).(1+v).CTE] and the last term [n3.Y.CTE.(q11+q12)/4]. In that case the FOM(χ) would be zero making the window athermal (FOM(OPD)=0).
The dn/dT of BGG glass can be reduced by changing the glass composition. Negative dn/dT in glasses can be accomplished by tailoring the composition. The dn/dT is expressed in terms of more fundamental properties:
β=volume expansion coefficient=3×CTE
The refractive index increases with temperature as the electronic polarizability (ϕ) increases and it decreases as the volume expansion (β) increases. The increase in volume expansion effectively reduces the charge per unit volume. In order to make a negative dn/dT glass, the composition must be modified such that the β is greater than ϕ. Even in the event where negative dn/dT is not achieved, simply reducing the refractive index of the glass can lower the dn/dT effectively reducing the OPD.
To explore the possibility of a negative dn/dT glass, a series of glass composition changes were made with the goal to reduce the refractive index and increase the CTE. These compositional changes include addition or partial substitution of sodium oxide, potassium oxide, rubidium oxide and cesium oxide for barium oxide. These substitutions decreased the refractive index of the BGG glass and increased the CTE. Some of the modified BGG glasses are shown in Table 2. The refractive index of the glasses was measured using ellipsometry at 1.06 µm wavelength. The CTE of the samples was measured using a dilatometer. Poisson’s ratio and the stress optic coefficients (q11, q12) are not expected to change significantly for different compositions. Although the Young’s modulus will change from composition to composition, the change is expected to be small. For the calculations purpose it is assumed that the Poisson’s ratio, Young’s modulus, and stress-optic coefficient do not change from the parent BGG glass. Polarizability is calculated for various glasses based on the types of ions present in each glass and their electronic polarizabilities. The dn/dT terms reported in Table 2 are calculated using Eq. (5).
The dn/dT for BGG glass was measured to be 9.0×10-6/K at 1.06 µm. This dn/dT has been lowered by compositional modification (labeled as MBGG) as shown in Table 2. The MBGG-5 glass is estimated to have a dn/dT value of -2.4×10-6/K. Even though the dn/dT of MBGG-5 glass is expected to be negative the FOM(χ) for this glass is still expected to be only half of the parent BGG glass. This is because the CTE term and the stress optic terms in Eq. 2 are still greater than the dn/dT term. Using all the values of MBGG-5 glass as shown in Table 2 and changing its dn/dT value it was determined that if the glass had a dn/dT value of -10.25×10-6/K the FOM(χ) would have been zero making the glass athermal for HEL applications. Additional work will progress in pursuit of an athermal BGG glass with a dn/dT value of -10.25×10-6/K.
In conclusion, we present BGG glass as a new window material for HEL window applications with better mechanical, thermal and optical properties than the currently used multispectral ZnS. BGG glass is easily scalable to large sizes and complex shapes. Large conformal windows can be made at low cost. It was demonstrated by glass compositional modifications that negative dn/dT in BGG glass is possible in order to further lower its OPD. There is also a possibility of making athermal BGG windows for HEL applications.
References and links
1. P.L. Higby and I.D. Aggarwal, “Properties of Barium Gallium Germanate Glasses,” J. Non. Cryst. Solids 163303–308 (1993). [CrossRef]
2. J.M. Jewell, “Alkaline Earth Gallogermanate Glasses, in Rare Elements in Glasses,” edited by J.E. Shelby, Trans Tech Publications, 317–343 (1994).
3. J.M. Jewell, L.E. Busse, K. Crahan, B.B. Harbison, and I.D. Aggarwal, “Optical Properties of BaO-Ga2O3-GeO2 Glasses for Fiber and Bulk Optical Properties,” SPIE Proc. 2287, 154–163 (1995). [CrossRef]
4. J.M. Jewell, P.L. Higby, and I.D. Aggarwal, “Properties of BaO-R2O3- Ga2O3-GeO2 (R=Y, Al, La, and Gd) Glasses,” J. Am. Ceram. Soc. 77  697–700 (1994). [CrossRef]
5. S.S. Bayya, B.B. Harbison, J.S. Sanghera, and I.D. Aggarwal, “BaO-Ga2O3-GeO2 Glasses with Enhanced Properties,” J. Non. Cryst. Solids 212198–207 (1997). [CrossRef]
6. J. M. Jewell and Aggarwal I D. Structural influences on the hydroxyl spectra of barium gallogermanate glasses[J].J Non-Cryst Solids 181, 189–199 (1995) [CrossRef]
7. I. D. Aggarwal, S. S. Bayya, G. D. Chin, and J. S. Sanghera, “Vis-IR Transmitting BGG Glass Windows,” Proceedings of DoD Electromagnetic Symposium 2004.
8. J.S. Browder, S.S. Ballard, and P. Klocek, “Physical Properties of Crystalline Infrared Materials” in “Handbook of Infrared Materials,” edited byP. Klocek, Mercel Dekker Inc. Publications, 193–425, (1991).