CopyrightJCPDS - International Centre for Diffraction Data 2003, Advances in X-ray Analysis, Volume 46.

297

APPLICATIONS OF X-RAY DIFFRACTION IN THE IMAGING INDUSTRY

Thomas N. Blanton Imaging Materials & Media, R&D, Eastman Kodak Company, Rochester, New York 14650-2106 ABSTRACT X-ray diffraction characterization of materials used in traditional silver halide and digital semiconductor photographic systems is described. Silver halide grains precipitated as multicomponent AgBr1-xIx phases were found to be comprised of a core with three shells. The composition of the core and shells was determined using lattice constant measurements. Nanocomposites for photographic antistatic applications were generated by mixing an aqueous dispersion of a 2:1 layered clay with a polyester ionomer. The extent of polymer intercalation and disorder of the clay platelets was monitored by observing any change in the clay (001) basal plane d-spacing as a function of composition. Silicon wafers used as substrates for digital photographic sensors were evaluated for the effect of depositing an epitaxial silicon layer on the wafer. A reciprocal space map constructed from high-resolution triple axis diffraction scans revealed that there was no detectable strain and the bulk silicon was comprised of a very smooth surface and few bulk-crystal defects. A gallium arsenide wafer with an epitaxial aluminum gallium arsenide film, utilized for light-emitting diodes, was also analyzed using a triple-axis diffractometer. The reciprocal space map revealed that the gallium arsenide substrate has a tilt boundary, and there is a compositional grading with a large population of dislocations. INTRODUCTION The process of photography can be divided into three primary components: capture, storage, and output. Whether the component being used is based on traditional silver halide technology, the continuously evolving digital technology, or a combination of both, materials science plays a vital role in the development of photographic systems. In traditional silver halide-based photography, over 100 individual components including photoactive materials, antistatic phases, polymers, organic dyes and couplers, cellulosics, gelatin, inorganics, developers, acids, bases, etc. are necessary to generate the prints people hold in their hands. Though digital imaging does not require as many components, the materials involved are as diverse as in silver halide systems and in some cases, are more complex to process. Continued growth in these methods of image generation requires the development of new materials and new processes. Essential to the efficient development of these materials and processes, are characterization techniques that allow for the understanding of existing and novel systems. X-ray diffraction (XRD) plays a vital role in the analysis of many of the materials used in the photographic industry. Four examples are presented that demonstrate the use of XRD for evaluation of current and future photographic materials.

CopyrightJCPDS - International Centre for Diffraction Data 2003, Advances in X-ray Analysis, Volume 46.

298

CONVENTIONAL PHOTOGRAPHIC SYSTEMS SILVER HALIDE The image capture component of traditional photographic film and paper is silver halide (AgX) dispersed in a gelatin matrix. Film usually contains silver bromide (AgBr) or silver bromoiodide (AgBr1-xIx), and paper usually contains silver chloride (AgCl) or silver chlorobromide (AgCl1xBrx). Mixed halides allow for customization of each film layer for light capture, image storage, and processing [1]. A typical photographic film can be comprised of as many as 12 layers, of which 9 or more can contain silver halide grains, coated onto a flexible or rigid support. The uniqueness of XRD is that it can distinguish between the different silver halide components due to changes in lattice constant as a function of halide composition. In the case of substitution of iodide for bromide in the AgBr lattice, the lattice constant is observed to increase due to the larger radius of iodide. A similar phenomenon is observed when bromide is substituted for chloride in the AgCl lattice. Careful studies have determined the lattice constants for silver halides and the coefficient of lattice expansion due to halide substitution [2], and are shown in Table 1. Table 1. Lattice constant calculations of mixed silver halide phases Silver halide phase Mole percent halide substitution range Lattice constant calculation() AgBr1-xIx x = 0 0.44 a = 5.77479 + 0.00368[I] AgCl1-xBrx x = 0 1.00 a = 5.55011 + 0.00225[Br] AgCl1-xIx x = 0 0.11 a = 5.55011 + 0.00635[I] In the case of AgBr1-xIx, for every mole percent iodide substituted for bromide, the mean lattice constant expands 0.00368 . In addition, note that for AgBr1-xIx, the mole percent iodide has a range of 0 to 44%. Beyond 44%, the iodide precipitates out as b or g AgI. The same is true for AgCl1-xIx when the iodide content is above 11 mol %. One might assume that X-ray fluorescence or neutron activation might be suitable to measure the halide composition in a AgX sample. However, the grains in a sample often contain multiple compositions. A core with one composition is covered with a shell with a different composition, and may be covered by a second shell of a third composition. Diffraction is unique in its ability to differentiate between these phases. Specimen preparation is critical in the analysis of silver halides by XRD. The AgX precipitation process occurs in the presence of gelatin. When this gelatin dries, it exerts compressive stress on the AgX perpendicular to the sample plane [3]. To eliminate this drying stress effect for specimens analyzed by XRD, the gelatin is washed out of the sample using warm water or an enzyme. Next, the degelled grains are mixed with NIST 640B silicon powder as an internal peak position standard. Data collection is performed using a conventional Rigaku Bragg-Brentano diffractometer, copper (Cu) radiation. The initial scan involves collecting a 5-70 2q scan to confirm what phases are present. This scan is followed by a step scan from 61-69 2q, to obtain a data set with appropriate counting statistics for analysis. In Figure 1, a diffraction pattern of a multiphase AgBr1-xIx sample mixed with Si is shown. Note that no Ka1a2 splitting is observed for these peaks (although it would be expected in this angular range). The reason for the lack of

CopyrightJCPDS - International Centre for Diffraction Data 2003, Advances in X-ray Analysis, Volume 46.

299

splitting is that the data were collected using a diffracted beam monochromator tuned to CuKb radiation. The use of a diffracted beam monochromator minimizes X-ray fluorescence from the sample from reaching the detector, and Kb simplifies the profile fitting of the overlapped AgBr1xIx peaks.

Figure 1. X-ray diffraction pattern for a multiphase AgBr1-xIx sample, (420) AgBr1-xIx and Si (400), (311) diffraction peaks. Collected using CuKb radiation.

The next step is to profile fit the data for determination of peak position and integrated peak intensity for each AgX phase. We have evaluated several profile-fitting routines and found that a split-Pearson VII function is best for AgX analysis. The AgX grains in the specimen used to produce the pattern in Figure 1 were designed to have a AgBr1-xIx core and two separate AgBr1xIx shells. Attempts to fit the X-ray diffraction data to this model were unsuccessful. The resulting profile fitting pattern for the AgBr1-xIx sample in Figure 1 is shown in Figure 2, revealing four individual diffraction peaks due to the core and three AgBr1-xIx shell phases.

Figure 2. Profile fitting results for a AgBr1-xIx multiphase sample using a split-Pearson VII profile. (420) diffraction peak.

CopyrightJCPDS - International Centre for Diffraction Data 2003, Advances in X-ray Analysis, Volume 46.

300

After profile fitting and use of the silicon internal standard for any sample displacement correction, the cubic lattice constant for each of the AgBr1-xIx phases can be calculated using Eq. (1). aAgBr1-xIx = (l(h2 + k2 + l2)1/2)/(2sin(2q/2)) where: aAgBr1-xIx AgBr1-xIx lattice constant () l CuKb wavelength (1.392218 ) (hkl) Miller indices 2q Diffraction angle () (1)

Given that we know the wavelength of CuKb radiation and the diffraction peak plane is (420), determination of 2q allows for calculation of the lattice constant for each phase. The simplicity of Eq. (1) is based on the mixed halide AgBr1-xIx phases being cubic in crystal structure. By rearranging the AgBr1-xIx lattice constant calculation in Table 1, the mole percent iodide in a AgBr1-xIx phase can be determined by taking the lattice constant of that phase determined in Eq. (1) and plugging into Eq. (2) %I = ((aAgBr1-xIx aAgBr)/0.00368) where: %I mole percent iodide in a AgBr1-xIx phase aAgBr1-xIx AgBr1-xIx lattice constant () aAgBr AgBr lattice constant (5.77479 ) 0.00368 coefficient of expansion () (2)

The last step of the AgBr1-xIx analysis is to take the product of the mole percent iodide in a AgX phase and the relative percent peak area of that phase, to obtain a measure of the amount of iodide present in each phase relative to the total amount of iodide present in the sample. In Table 2, the results of a total iodide evaluation for the specimen in Figure 1 are listed. Table 2. XRD results for iodide measurement in a four-phase AgBr1-xIx sample 2q Peak position () 63.495 64.452 64.914 65.121 AgBr1-xIx lattice constant () 5.9166 5.8379 5.8008 5.7843 Mole percent iodide in each phase 38.5 17.1 7.1 2.6 Relative percent (420) peak area 20 14 41 25 18.1 Relative percent of total iodide in sample 9.6 7.0 1.0 0.5

Total mole percent iodide in sample (sum of relative percent iodide in each phase)

The 38.5%I phase is the AgX core, followed by the first shell, which is the 17.1%I phase. The 2.6%I phase is the outer shell of the grain. Based on the peak profile and AgX grain architecture the 7.1%I phase is an intermediate phase that is comprised of a distribution of compositions that

CopyrightJCPDS - International Centre for Diffraction Data 2003, Advances in X-ray Analysis, Volume 46.

301

have a mean lattice spacing of 5.8088 . The presence of this intermediate phase indicates that the precipitation process needed to be adjusted to create a clean interface between the first and final shell. Neutron activation analysis found the total iodide composition in this sample to be 18.4 mol %, which is in good agreement with the XRD result of 18.1 mol % iodide in Table 2. This comparison serves as a check to ensure that the results of the XRD analysis are reasonable. CLAY-POLYMER NANOCOMPOSITE Clay-polymer nanocomposite materials have received significant interest in the industrial community for use in a number of novel applications [4]. Dispersion of clay in a polymeric matrix can result in three general types of morphology: (1) macroscopic clay aggregates in the polymer, (2) intercalated clay with one or more molecular layers of the polymer inserted into the clay gallery, and (3) exfoliated clay with singular clay platelets randomly distributed in the polymer [5]. It is the latter two arrangements of the clay in polymer that generate a variety of interesting properties in the resultant clay-polymer nanocomposites. Using X-ray diffraction (XRD), the extent of polymer intercalation and disorder of the clay platelets in these coatings can be monitored. When using hydrophobic web materials of low surface conductivity, typically used as support for photographic film and paper, these supports can readily become electrostatically charged [6]. This static build up occurs because of friction between dielectric materials and triboelectrically chargeable transport means, such as rollers, during high-speed conveyance of the web. An electrically charged web can result in static discharge through the generation of sparks that may pose fire hazards in the presence of flammable solvents at a coating site. For a web containing unprocessed photographic emulsion, sparking can cause additional problems, such as, irregular fog pattern or static marks and degradation of image quality. Electrostatic charge can be dissipated effectively by incorporating an electrically conductive antistatic layer into the support. Because of its ionic conductivity and nanoparticulate characteristic, synthetic clay is a suitable candidate for antistatic coating applications for photographic supports. The clay material used in this work is a synthetic smectite, which closely resembles the natural clay mineral hectorite (montmorillonite), in both structure and composition. The particular commercial clay used in this work is known as Laponite, manufactured by Southern Clay Products, a division of Laporte Industries Ltd. Laponite is a 2:1 layered hydrous magnesium lithium silicate with the following chemical formula: Na0.7 [(Si8 Mg5.5 Li0.3) O20 (OH)4] Samples were prepared by first dispersing Laponite RDS clay powder in deionized water using a high shear mixer to form a clay sol. An aqueous solution or dispersion of a polyester ionomer AQ55 (Eastman Chemicals) was added to this clay sol in various weight ratios to obtain the clay/polymer mixture in its final aqueous form. The solid content was maintained at 4% by weight. Specimens for XRD analysis were prepared by placing ~10 drops of the clay/polymer aqueous mixture onto microscope slides and dried in ambient air to form a thin film of the sample. All XRD data were collected using a conventional Rigaku Bragg-Brentano diffractometer, copper radiation, diffracted beam graphite monochromator tuned to CuKa radiation, and a scintillation detector.

CopyrightJCPDS - International Centre for Diffraction Data 2003, Advances in X-ray Analysis, Volume 46.

302

Laponite RDS:polymer samples with weight ratios of 100:0 to 10:90 were formulated for XRD measurement. The diffraction patterns from some of these samples are plotted in Figure 3, showing the clay (001) basal plane diffraction peak shift to higher d-spacing as a result of increasing concentration of polymer. The (001) basal plane spacing corresponding to different clay:polymer ratios, as measured at ~ 47% RH, is listed in Table 3 for this system.

Figure 3. X-ray diffraction patterns obtained from Laponite RDS:polyester ionomer films with clay:polymer weight ratios of (a) 30:70, (b) 60:40, (c) 70:30, (d) 80:20, (e) 90:10, (f) 100:0. Clay (001) basal plane diffraction peak.

Table 3. Laponite RDS basal plane spacing for different Laponite RDS:polyester ionomer ratio Laponite RDS:polyester ionomer weight (001) basal plane spacing () ratio 100:0 13.4 90:10 14.2 80:20 15.5 70:30 16.9 60:40 18.5 30:70 not observed 10:90 not observed As the polymer concentration increased, an increase in the basal plane spacing up to 40% of polymer was observed. This increase in spacing is nearly linear with the polymer concentration. At much higher levels of the polymer (70 %) the (001) diffraction peak was not observed, indicating that the clay had become highly disordered. This state is believed to be caused by a loss of plane-to-plane ordering in the clay. Although the insertion of the polymer in between basal planes causes lattice expansion, there is a limit to an ordered expansion. Once this limit is reached the lattice order is lost and the basal plane spacing is no longer observed. In the case of polyester ionomer, this apparent threshold for Laponite RDS clay disorder is around 70 weight percent of polymer content. Above 70% polymer, the clay is observed to be exfoliated. This

CopyrightJCPDS - International Centre for Diffraction Data 2003, Advances in X-ray Analysis, Volume 46.

303

physical state of the clay is desirable as it provides the greatest amount of clay surface area per amount of clay in the nanocomposite. DIGITAL PHOTOGRAPHIC SYSTEMS SILICON Digital photography uses an image sensor composed of silicon (Si) covered with a network of electrodes. There are two types of silicon imagers, CCD (charge-coupled device) and CMOS (complementary metal oxide semiconductor) [7]. Regardless of which imager is used, both are manufactured using similar equipment but different manufacturing processes and device architectures that make the imagers quite different in both capability and performance. Imagers based on CCD technology provide superior image quality, whereas CMOS imagers are more easily manufactured and require less power consumption. For image sensor fabrication, it is standard practice for Si wafers to be received with one epitaxial Si layer deposited onto the Si single-crystal substrate. At issue is whether these wafers are acceptable for manufacturing of sensors. XRD topography methods are well suited to assess the quality of the epitaxial layer and wafer in regards to lattice strain, surface smoothness, and bulk-crystal defects. Incoming wafers are analyzed as received, with no additional specimen preparation. Specimen handling is important as induced stress and/or trace impurities can be introduced that will render the wafer unusable. Data collection was performed using a Bede D3 high-resolution diffractometer in triple-axis mode, CuKa1 radiation. Figure 3 shows a triple axis q/2q scan of a P+/P (boron doped) epitaxial Si/(100) Si wafer specimen. No peak splitting between the substrate and epilayer was observed giving a preliminary indication that there is no observable lattice strain.

Figure 3. Triple-axis q/2q scan of boron-doped epitaxial silicon film deposited on a boron-doped (100) Si singlecrystal wafer. Data collected using the (004) lattice plane.

A more complete picture of the scattering from a specimen can be obtained by collecting a reciprocal space map [8]. A series of q/2q (analyzer) scans are collected with plus and minus w (sample tilt) offsets. Any trace of strain resulting from the doped epitaxial layer would show up

CopyrightJCPDS - International Centre for Diffraction Data 2003, Advances in X-ray Analysis, Volume 46.

304

as a peak or shoulder in the q/2q direction. A reciprocal space map of the P+/P (boron doped) epitaxial Si/(100) Si wafer specimen is shown in Figure 4.

Figure 4. Reciprocal space map of boron-doped epitaxial silicon film deposited on a boron-doped (100) Si singlecrystal wafer. The x-axis is q/2q (arc sec) and the y-axis is w (arc sec). Data collected using the (004) reflection.

Note the long streak in the q/2q direction. This streak represents a truncation rod, caused by the finite distance that the X-rays travel in the Si for the (004) reflection. The actual depth of penetration is not large enough for complete destructive interference of the diffracted X-rays around the peak. The presence of the streak is an indication that the surface is very smooth and is also an indication of a high-quality (few defects) bulk single-crystal wafer. Diffuse scatter from bulk defects and surface roughness, if present, will greatly reduce the surface streak. The quality of the crystal is further defined by the presence of an intense analyzer streak (rotated ~45 clockwise) and a weak beam conditioner streak (rotated ~45 counter clockwise) [9]. This wafer was found to be acceptable for silicon image sensor fabrication. GALLIUM ARSENIDE In some hybrid photographic systems, a combination of digital and traditional photographic materials are used to optimize the output of an image. One such concept uses a printer equipped with light-emitting diodes (LEDs) to expose AgX photographic paper. The LEDs are composed of gallium arsenide (GaAs) substrates with III-V compound(s) epitaxial layers. As with Si wafers, it is also common practice to purchase the GaAs with the epilayer(s) already deposited, finish the processing in the fabrication line, dice and package for the printer applications. Again, XRD topographic methods are essential to assess the quality of these as-received wafers. Figure 5 shows a triple axis w scan of the (004) GaAs lattice plane from an AlGaAs epitaxial film grown on a (100) GaAs wafer. Whereas a q/2q triple axis scan measures strain, the w scan measures crystalline tilts. The scan in Figure 5 clearly shows two substrate peaks indicating that the X-ray beam has crossed over a tilt boundary in the substrate. The reciprocal space map for this specimen is shown in Figure 6.

CopyrightJCPDS - International Centre for Diffraction Data 2003, Advances in X-ray Analysis, Volume 46.

305

Figure 5. Triple-axis w scan of the (004) GaAs lattice plane. Specimen is an AlGaAs epitaxial film grown on (100) GaAs.

Figure 6. Reciprocal space map of an AlGaAs epitaxial film grown on (100) GaAs. The x-axis is q/2q (arc sec) and the y-axis is w (arc sec). Data collected using the (004) lattice plane.

The two red nodes at a q/2q value of 0 arc sec are attributable to tilt boundaries in the GaAs substrate. The intensity distribution to the left of the substrate shows that there is a compositional grading (d-space changes). The width of this grading layer in the w direction, over 100 arc sec full width at half maximum (FWHM), indicates that the layer is full of dislocations. At the end of the compositional grading at 1500 arc sec is an AlGaAs epilayer with a tilt boundary. The data in Figure 6 demonstrate the usefulness of the reciprocal space map. Further processing of this lot of AlGaAs/GaAs wafers would be pointless because the GaAs substrates are defective. The presence of the tilt boundaries will inhibit high quality epilayer growth, regardless of the deposition process.

CopyrightJCPDS - International Centre for Diffraction Data 2003, Advances in X-ray Analysis, Volume 46.

306

SUMMARY X-ray diffraction as a tool for imaging materials characterization has been demonstrated. Conventional techniques using Bragg-Brentano instrumentation and specialized techniques requiring systems with high-resolution optics will continue to play a role in the development of new imaging chemical, sensors, and hybrid elements. ACKNOWLEDGMENTS The author would like to acknowledge Craig Barnes of Eastman Kodak Company for assistance in collection of diffraction data, and Dr. Kevin Matney of Bede Scientific Incorporated for collection of triple-axis diffraction data. REFERENCES [1] Pouradier, J.; Pailliotet, A.; Berry C. R., Properties of Silver Halides, Chapter 1, The Theory of the Photographic Process, Fourth edition, T. H. James editor, Macmillan Publishing Co., Inc.: New York, 1977, 1-12. [2] Blanton, T., Photography: Image Capture and Image Storage Materials, Chapter 25, Industrial Applications of X-ray Diffraction, F. H. Chung and D. K. Smith editors, Marcel Dekker, Inc: New York, 2000, 589-608. [3] Blanton, T. N.; Watkins, T. R.; Howey, M. A., The Effect of Gelatin on Silver Halide Strain, Adv. X-ray Anal., Vol. 40, 1998, 556-566. [4] Giannelis, E., A new strategy for synthesizing polymer-ceramic nanocomposites, J. Minerals Metals Mater. Soc., 1992, 44(3), 28-30. [5] Lan, T.; Kaviratna, P. D.; Pinnavia, T. J., Mechanism of clay tactoid exfoliation in epoxyclay nanocomposites, Chem. Mater., 1995, 7(11), 2144-2150. [6] Majumdar, D.; Blanton, T. N.; Melpolder, S., Coatings of Clay-Polymer Nanocomposite Materials, Chapter 22, Specialty Polymer Additives Principles and Applications, S. AlMalaika, A.; Golovoy, C. A., Wilkie editors, Blackwell Science Ltd.: Oxford, 2001, 415-428. [7] http://www.kodak.com/US/en/corp/researchDevelopment/technologyFeatures/cmos.shtml [8] Bowen, D. K.; Tanner, B. K., High Resolution X-ray Diffractometry and Topography, Taylor & Francis: United Kingdom, 1998, 149-171. [9] Iida, A.; Kohra, K., Separate measurements of dynamic and kinematic x-ray diffractions from silicon crystals with a triple axis diffractometer, Phys. Status Solidi A, 1979, 51(2), 533-542.