Variability in Chemical Vapor Deposited Zinc Sulfide: Assessment of Legacy and International CVD ZnS Materials John McCloy,*a Ralph Korensteinb a Raytheon Missile Systems, 1151 E Hermans Road, Tucson, AZ, USA 85734 b Raytheon Company, 350 Lowell Street, Andover, MA, USA 01810 ABSTRACT Samples of CVD ZnS from the United States, Germany, Israel, and China were evaluated using x-ray diffraction, transmission and Raman spectroscopy, and biaxial flexure testing Visible and near-infrared scattering, μm absorption, and ultraviolet cut-on edge varied substantially in tested materials Transmission cut-on (ultraviolet edge) blue-shifts with annealing and correlates with visible color but not the μm absorption Raman scattering for CVD ZnS, presented here for the first time, was similar for all ZnS tested Crystallographic hexagonality and texture was determined and correlated qualitatively with optical scattering All CVD ZnS tested with biaxial flexure exhibit similar fracture strength values and Weibull moduli This survey suggests that despite over 30 years of production as an infrared window, the optical properties of CVD ZnS and their variability still defy easy explanation Keywords: Zinc sulfide, chemical vapor deposition, IR window INTRODUCTION Chemical vapor deposited (CVD) zinc sulfide has been an indispensable infrared window material since its introduction in the early 1970s by Raytheon.1 Development of ZnS internationally has been recently documented2 but a comparison of properties among various producers has not been previously available Samples of CVD ZnS, Hot isostatic pressed CVD ZnS (multispectral), some legacy experimental Raytheon grades (red ZnS and elemental ZnS), legacy hot-pressed ZnS, and single crystal melt-grown ZnS were obtained from various suppliers Transmission was investigated to study scattering, absorption, and color Several samples of red ZnS were annealed at various temperatures to investigate the phenomenon of coloration in ZnS X-ray diffraction (XRD) studies were used to quantitatively measure hexagonality and crystallographic texture and investigate its relation to optical transmission Finally, biaxial flexure testing was performed, generating data sets for red ZnS and elemental ZnS not previously available EXPERIMENTAL Various samples of ZnS were obtained for study as detailed in Table 1, including different suppliers of standard CVD ZnS (made from hydrogen sulfide gas and zinc vapor), legacy Raytheon elemental ZnS (made from hydrogen gas, sulfur, and zinc vapor3), legacy Raytheon red ZnS (grown ~600 °C with large excess of Zn reactant,4 and multi-spectral ZnS (CVD ZnS which has been hot-isostatic pressed in the presence of platinum metal) Single crystal Bridgeman grown crystals of ZnS and legacy hot-pressed ZnS from powder precursors were also studied Crystal structure was investigated using a Rigaku Geigerflex II x-ray diffractometer (XRD) fitted with a CuKα x-ray source, a diffracted beam monochrometer, and a sodium iodide scintillation detector Diffracted intensity of polycrystalline samples was recorded for 2θ angles from 20 to 100º in 0.05º increments Some samples were ground in an agate bowl, and powder patterns were taken of the region between 25 and 35º with 0.02º resolution to further investigate the characteristic region of ZnS polytypes * Present address: Battelle Pacific Northwest National Laboratory, john.mccloy@pnl.gov Window and Dome Technologies and Materials XI, edited by Randal W Tustison, Proc of SPIE Vol 7302, 73020M · © 2009 SPIE · CCC code: 0277-786X/09/$18 · doi: 10.1117/12.819559 Proc of SPIE Vol 7302 73020M-1 Transmission of polished samples was investigated in the ultraviolet through near-infrared (175 – 3300 nm) on a Varian Cary 5000 spectrophotometer, and data was recorded at nm wavelength increments Infrared transmission (2 – 20 μm) was measured using a Fourier transform infrared spectrometer, Bruker Equinox 55 FTIR, with a deuterated triglycine sulfate (DTGS) detector using a spectral resolution of cm-1 and signal averaging over 80 scans Raman spectroscopy was performed using a Thermo-Nicolet Almega XR single beam dispersive spectrometer with a holographic notch filter to block the 532 nm laser excitation beam Samples were placed under the 10x objective of the confocal microscope Spectral resolution was 2.2 – 2.5 cm-1 with the high resolution grating, 25 μm slit, and 79 to 1290 cm-1 spectral range Laser power was 13% and spectra were averaged over 32 scans A few ZnS samples were also surveyed with the low resolution grating out to 4248 cm-1, with spectral resolution 6.4 – 10.5 cm-1 Biaxial flexure strength testing was performed in accordance with ASTM C1499.5 The characteristic diameters were as follows: 10.67 mm (load ring), 20.32 mm (support ring), and 25.4 mm (sample) Sample thickness was 1.9 to 2.0 mm All samples had the CVD growth axis perpendicular to the plane of the large area of the sample which was placed under stress Various samples were tested, with some datasets having more data points than others due to sample availability All sample surfaces were polished using a controlled grind procedure to minimize the effects of the surface flaw population differences, using successive steps of grinding and polishing with smaller grits Table 1: ZnS samples investigated; all polycrystalline unless otherwise noted Nomenclature Material type Supplier Bridgeman IRTRAN RH II-VI Raytran PS Melt grown crystal ZnS Hot pressed powder ZnS Standard CVD Standard CVD Standard CVD Standard CVD Vitron Rafael CVD RedZnS Standard CVD Standard CVD CVD MTI Kodak Rohm & Haas II-VI Infrared Raytheon Princeton Scientific (US) importer Vitron Rafael Raytheon eZnS PS msZnS CVD (H2 + S + Zn) CVD (H2 + S + Zn) CVD + (Pt?) HIP PS HH CVD + (Pt?) HIP Vitron msZnS Vitron HH Rafael msZnS Ray msZnS, msZnS CVD + Pt HIP CVD + HIP (no Pt) CVD + HIP (no Pt) CVD + Pt HIP Chinese Country of Origin USA USA USA USA USA Germany Germany Israel USA Notes (110) plane various material lots Raytheon USA “FLIR grade” “FLIR grade” ~600 °C deposition, very Zn rich reactant; “elemental ZnS” Research institute synthetic crystals Princeton Scientific (US) importer Princeton Scientific (US) importer Vitron Vitron Rafael Raytheon China See Reference Germany “multispectral ZnS” Germany special short HIP (unknown details) “clear grade” “special clear grade” “multispectral ZnS” “multispectral ZnS” Germany Germany Israel USA RESULTS 3.1 Atomic structure Diffraction patterns of polycrystalline samples were taken from 2θ angles 20 – 100° Some of these are shown normalized to their maximum peak in Figure 1a Generally, the texture of as-deposited CVD samples including eZnS is Proc of SPIE Vol 7302 73020M-2 predominantly {100} while that of HIPped samples is predominantly {111} , presumably due to plastic deformation and recrystallization Nucleation and growth rates are highest on “atomically rough” planes which have higher densities of kinks and steps, like {100} planes, and would be expected to dominate as-deposited materials,7 though it should be noted that RedZnS had a significantly higher fraction of {111} A quantitative analysis of the texture was performed using the scattered intensities for textured planes (hkl) in the polycrystalline sample (txt superscript), calculated as8 ( )( txt txt iso iso f hkl = I hkl / ∑ I hkl / I hkl / ∑ I hkl ) (1) In an isotropic polycrystalline sample fhkl=1 for all (hkl) planes The larger the value, the larger the volume fraction of {hkl} planes orientated in the Bragg diffraction position The powder pattern for cubic ZnS #05-0566 was used as the “isotropic” sample (iso superscript) Furthermore, the % {111} and % {100} was calculated as follows (Figure 1b): f + f 222,msZnS 1.60 + 1.46 (2) = = 0.5 = 50% %{111}msZnS = 111,msZnS 6.18 ∑ f hkl ,msZnS a) b) 066) :6 H ((C) ((9) 666) (006) (000) 660) 06t6) - - 5O msZnS 4O RafmsZnS Red 30 S S x (0) / 9-i 'isotropic' (randomly oriented) S Sd 20 IRTRAN Various CVD ZnS 10% H eZnS 00 06 06 09 09 06 OL 06 o OOI 10% 0% 20 40% 30 5O 6O (100) texture (°) c) 0.1 d) Specific sample Rayth sZnS Rafae rzrs Oht (%) 3% % 13106.4 cm1) a A) 1104 5.411 : 41 Vtrn o Pt HI 13% 1126 RLI CH nee 3% 6% - 54111 22 414 54:4 EIemnt 32% : 5411 'IIVI Ef3% 202 5.410 Rafae CVD E 5% 251 112 IRTRArJ2 26% Brdçjen-in S 7% : ':' 5411 0.08 - 0.06- 0.04 E z 0.02 - 26 27 28 29 30 31 20 Figure X-ray diffraction data for ZnS samples; (a, UL) normalized polycrystalline diffraction data; (b, UR) quantification of textures, with various CVD ZnS samples in dotted box; (c, LL) hexagonality, calculated extinction, and lattice parameter; (d, LR) XRD powder pattern from which hexagonality data was calculated Lattice parameter measurements were made, using a NIST traceable alumina standard internal to the instrument, using the alumina (0.2.10) plane reflection (expected at 2θ = 88.995° and measured at 89.02°) to perform the angular correction Diffracted intensity was measured from 87.7 to 89.3° for ZnS at 0.02° intervals with second integration time The sphalerite (422) reflection, measured at 2θ = 88.4 to 88.6°, was corrected by the standard, then used to calculate the interplanar spacing (dhkl) and the cubic lattice parameter.9 The estimated accuracy of lattice constant measurements is better than ±0.001 Å Measured lattice parameters are shown in Figure 1c Proc of SPIE Vol 7302 73020M-3 A simple measure of the hexagonality in polycrystalline ZnS powders and hot-pressed compacts has been proposed using the ratios of characteristic peaks in the wurtzite and sphalerite XRD spectra.10-12 The atomic fraction of hexagonal stacking (hexagonality) can be calculated as {w (0002) + s (111)} I 28.53 ⎛ α hex = 1.84 ⎜⎜ − ⎝ {w (0002) + s (111)} I 28.53 +I w (10 10) 26.91 ⎞ ⎟⎟ ⎠ (3) where w is the 2H hexagonal phase (wurtzite), s is the 3C cubic phase (sphalerite), the subscripts are 2θ angles, and the superscripts are the characteristic planes, since the 2H(0002) and the 3C(111) overlap at 2θ=28.53° Intensity values were obtained using the peak height of the higher resolution, smaller angular range data Representative data is shown in Figure 1c and 1d for HIPped ZnS (msZnS, similar spectra for all HIPped samples in this study), eZnS (similar spectra for Chinese and red), CVD ZnS (Raytran, similar for all standard CVD samples in this study), and Bridgeman meltgrown crystal Note the extra peaks (~29.6˚ and 30.1˚) for the Bridgeman crystal, which can be most closely indexed by assuming a 72R polytype structure (even though the crystals were sold as 3C pure cubic crystals) 3.2 Optical properties Composite spectra of FTIR and UV-VIS transmission measurements for various samples are shown in Figure Sample thicknesses, UV edge (defined as the last spectrometer data point on the short wavelength edge of the transparency window before a negative transmission value was recorded), and transmission and calculated extinction for various wavelengths and samples are shown in Table Extinction is calculated from measured transmission as: (4) β ext = (1 / L ) * ln ⎡ (1 − R ) / 2T + R + (1 − R ) / 4T ⎤ ⎣ ⎦ -1 where β is the extinction coefficient (in cm ), L is the sample thickness, T is the measured transmittance, and R is the single surface reflectivity in air calculated from the index of refraction (from the Sellmeier equation for CVD ZnS3) Table 2: Measured transmittance, UV edge, and calculated extinction (cm-1) at key wavelengths; 1.064, 3.39, and 10.6 μm are laser wavelengths and μm gives an indication of the zinc hydride absorption Errors on the calculated extinction are larger for thinner samples Theoretical (no scatter, phonon abs) Raytheon Elemental Raytheon Red Chinese Vitron FLIR Rohm & Haas Rafael CVD II-VI Infrared Princeton Scientific FLIR Kodak IRTRAN2 Raytheon Multispectral Bridgeman single crystal T1.064/ α1.064 T3.39/ α3.39 T6.0/ α6.0 T10.6/ α10.6 340 n = 2.288 73.5% n = 2.255 74.1% n = 2.240 74.5% n = 2.192 75.5% 4.64 3.18 4.62 2.00 4.64 2.49 4.89 5.13 386 400 386 380 382 361 383 390 55.3% / 0.59 67.0% / 0.27 62.1% / 0.35 28.7% / 4.59 16.9% / 3.11 50.9% / 1.42 26.8% / 2.02 8.4% / 4.18 72.9% / 0.03 72.2% / 0.08 71.8% / 0.07 68.5% / 0.38 56.9% / 0.55 65.9% / 0.45 61.0% / 0.38 54.3% / 0.59 59.6% / 0.46 48.1% / 1.33 73.1% / 0.04 62.8% / 0.82 60.8% / 0.42 70.4% / 0.22 60.9% / 0.40 50.9% / 0.72 68.7% / 0.14 67.9% / 0.32 68.4% / 0.21 72.5% / 0.20 68.4% / 0.21 70.8% / 0.25 64.8% / 0.30 64.7% / 0.29 6.39 4.64 445 344 1.1% / 6.54 72.0% / 0.04 63.3% / 0.24 73.6% / 0.01 71.3% / 0.06 74.4% / ... 13 % 11 26 RLI CH nee 3% 6% - 5 411 1 22 414 54:4 EIemnt 32% : 5 411 'IIVI Ef3% 202 5. 410 Rafae CVD E 5% 2 51 112 IRTRArJ2 26% Brdçjen-in S 7% : ':' 5 411 0.08 - 0.06- 0.04 E z 0.02 - 26 27 28 29 30 31. .. superscript) Furthermore, the % {11 1} and % {10 0} was calculated as follows (Figure 1b): f + f 222,msZnS 1. 60 + 1. 46 (2) = = 0.5 = 50% % {11 1}msZnS = 11 1,msZnS 6 .18 ∑ f hkl ,msZnS a) b) 066) :6... rrutspotri 5h rr1ct3r1Sc HfHed stressed area iii m2 L - 2 Sample Description -2 31 :j4 11 5 41 21 2398 511 240 20.4 Weibull rIodulus (unbiased Weibull Modulus biased) 2 21 6 21 91 1 01 655 scale factor