Residual strains in amorphous silicon films measured by x-ray double crystal topography

135

The coefficient of thermal expansion ͑CTE͒, biaxial modulus, and stress of some amorphous semiconductors ͑a-Si:H, a-C:H, a-Ge:H, and a-GeC x :H͒ and metallic ͑Ag and Al͒ thin films were studied. The thermal expansion and the biaxial modulus were measured by the thermally induced bending technique. The stress of the metallic films, deposited by thermal evaporation ͑Ag and Al͒, is tensile, while that of the amorphous films deposited by sputtering ͑a-Si:H, a-Ge:H, and a-GeC x :H͒ and by glow discharge (a-C:H) is compressive. We observed that the coefficient of thermal expansion of the tetrahedral amorphous thin films prepared in this work, as well as that of the films reported in literature, depend on the network strain. The CTE of tensile films is smaller than that of their corresponding crystalline semiconductors, but it is higher for compressive films. On the other hand, we found out that the elastic biaxial modulus of the amorphous and metallic films is systematically smaller than that of their crystalline counterparts. This behavior stands for other films reported in the literature that were prepared by different techniques and deposition conditions. These differences were attributed to the reduction of the coordination number and to the presence of defects, such as voids and dangling bonds, in amorphous films. On the other hand, columnar structure and microcrystallinity account for the reduced elasticity of the metallic films.

Section: Introductionmentioning

confidence: 99%

Coefficient of thermal expansion and elastic modulus of thin films

Lima

Lacerda

Vilcarromero

et al. 1999

135

“…14 Second, while stress due to oxidation of the films should be minimized in this initial stage of the thermal test, some oxidation inevitably occurs which might increase the stress over values predicted solely by thermal expansion mismatch. 9 This second possibility is supported by the fact that for each film tested, the experimentally determined slope was greater than the predicted slope, as would be the case if there was an added stress component due to oxidation.…”

Section: Resultsmentioning

confidence: 95%

“…The residual growth stresses in these films were obtained using double crystal diffraction topography ͑DCDT͒, and were found to be Ϫ200Ϯ18 and ϩ416Ϯ35 MPa in the Ta and Cr films, respectively. 9 Thermal testing of these films was conducted using an apparatus consisting of a heating device/sample holder and a portable CCD x-ray imaging system, described previously by Zhao et al 10 This apparatus is shown schematically in Fig. 1.…”

Section: Methodsmentioning

confidence: 99%

Laue transmission diffraction optics for thin film stress calculation

French

Bilello

2003

Self Cite

White beam Laue transmission diffraction topography has been used to determine the curvature of a crystalline substrate, from which the stress in the overlaying coating is determined. A detailed analytical method has been developed which correlates the introduction of curvature to the substrate in two orthogonal directions with attendant changes in the dimensions of a given topographic reflection. The size of a given reflection is found to depend on the dimensions of the incident beam, the horizontal and vertical incident beam divergences, the orthogonal curvatures of the crystalline substrate, the camera length, and the indices of the selected reflection. The thermal tests of sputtered polycrystalline Ta and Cr films on Si (100) substrates are used to demonstrate the applicability of this phenomenon. Simultaneous observation of film delamination and quantification of associated stresses is shown to be possible.

“…͑1͒ deflection techniques based on determining the radius of curvature of the substrate; [3][4][5][6][7][8][9][10][11][12] and ͑2͒ strain measurement techniques based on the direct measurements of interplanar spacings in the film using x-ray diffraction. [13][14][15][16][17][18][19][20][21][22] The substrate deflection techniques include optical interferometry, 4 laser scanning, 5 and double-crystal diffraction topography ͑DCDT͒.…”

Section: Introductionmentioning

confidence: 99%

“…[13][14][15][16][17][18][19][20][21][22] The substrate deflection techniques include optical interferometry, 4 laser scanning, 5 and double-crystal diffraction topography ͑DCDT͒. 9 Optical interferometry and laser scanning determine the radius of curvature of the physical surface of the substrate R which equals the inverse of the curvature K. 4,5 The DCDT technique, on the other hand, determines the curvature of the crystal lattice planes of the substrate near the film-substrate interface by measuring the distance between successive Bragg angle contours. [6][7][8] The distance between the contours is then proportional to the curvature, K. 9,10 The resulting curvatures obtained from the above techniques are then linearly related to the average stress using standard equations.…”

Section: Introductionmentioning

confidence: 99%

Depth dependence of residual strains in polycrystalline Mo thin films using high-resolution x-ray diffraction

Malhotra

Rek

Yalisove

et al. 1996

Self Cite

The magnitude of the stress in a thin film can be obtained by measuring the curvature of the film-substrate couple. Crystal curvature techniques yield the average stress throughout the film thickness. On a microscopic level, the details of the strain distribution, as a function of depth through the thickness of the film, can have important consequences in governing film quality and ultimate morphology. A new method, using high-resolution x-ray diffraction to determine the depth dependence of strain in polycrystalline thin films, is described. The technique requires an analysis of the diffraction peak shifts of at least six independent ͕hkl͖ scattering vectors, at a variety of penetration depths from the free surface of the film. The data are then used to determine the magnitude and directions of the strain eigenvalues in a laboratory reference frame for each penetration depth from the free surface of the film. A linear elastic model was used to determine the strains in successive slabs of the film. Results are reported for two Mo films, with nominal thicknesses of 50 and 100 nm, which were deposited by planar magnetron sputtering onto Si ͑100͒ substrates. This technique can provide quantitative insight into the depth variation of residual strains ͑stresses͒ in thin films and should work with a wide variety of materials.