Open Access
Add to BibSonomyAbstract
Electron-beam deposition of silica and alumina is used to fabricate distributed polarization rotators suitable for smoothing the intensity of large-aperture, high-peak-power lasers. Low-modulation, low-loss transmittance with a high 351-nm laser-damage threshold is achieved.
© 2014 Optical Society of America
1. Introduction
Direct-drive and polar-drive fusion experiments require highly uniform illumination of the target, and polarization smoothing has been an important tool to achieving the necessary uniformity [1–3]. One successful approach has been to use a wedge of birefringent crystal such as potassium dihydrogen phosphate (KDP) to fabricate a distributed polarization rotator (DPR) that spatially separates speckle patterns at the laser focus based on different polarization states [1,2]. The meter-scale aperture and high fluence of the National Ignition Facility (NIF) preclude, however, the use of a KDP wedge because of stimulated Raman scattering (SRS) [4]. There are currently no intrabeam polarization smoothing schemes suitable for use on the NIF. The development of a distributed polarization rotator compatible with NIF operation is a necessary enabling technology for direct-drive experiments.
Glancing-angle deposition (GLAD) is a deposition process where incident evaporant flux condenses with low surface mobility on a substrate oriented at a high incidence angle, enhancing the microstructure of the film in such a way that the film properties become highly anisotropic [5–7]. The low surface mobility is ensured by depositing the film without substrate heating or significant use of energetic assist by plasmas or ion beams in the deposition process. GLAD has been effectively demonstrated for use in fabricating high reflectors, antireflective coatings, filters, and polarization-control coatings [5–11]. A primary limitation of GLAD, however, is the area over which the film structure and associated performance is maintained since the performance of the film is dependent on the angle at which the evaporant vapor strikes the surface of the substrate. As noted by Motohiro and Taga [5], homogeneous retardation over large areas can be achieved by scanning the substrate behind an aperture and/or depositing a bilayer structure at positive and negative angular orientations. Krause et al. [12] have proposed the deposition of GLAD structures in a roll-to-roll system, but scaling to a large aperture remains necessary to demonstrate functionality for this application.
This effort describes the development of a polarization-control component consisting of sub-apertures of GLAD coatings of the correct thickness and orientation relative to the incident polarization to achieve ± π/2- or π-phase retardance for a wavelength of 351 nm, creating a spatial variation in the polarization state for a large-aperture, high-fluence laser. This approach makes polarization smoothing of the beam possible by using a single substrate, without any obstructions between deposition regions. The coated regions must be resistant to high fluences, provide the necessary optical retardance to alter the incident polarization as desired, and be incorporated within the overall component in such a manner to limit transmission losses caused by reflectance, scatter, and diffraction modulation from the edges of the deposited regions. Beam modulation can lead to damage of optical components through constructive interference and must be minimized to preserve the overall fluence capabilities of the laser system. While the current effort is focused on a very specific application, this technology may be incorporated in other large-aperture polarization-control components requiring high-laser-damage thresholds. By making use of different materials and deposition angles, different birefringent film properties can be realized. Such films may be utilized in combination to yield broader-spectrum performance, either in transmission or reflection.
2. Background
It has been well established that the influence of low-energy oblique-incidence deposition on the anisotropy of optical coatings leads to the formation of columnar structures exhibiting birefringence with potential application as a retardation plate [5,7]. Such a film is biaxial, with nx being the refractive index along the column, ny being the refractive index of the film perpendicular to the columns in the plane defined by the normal to the substrate surface and an individual column, and nz is the refractive index perpendicular to this plane [7]. For light normally incident on the substrate surface, the film is birefringent with refractive indices ns = nz and np given by
where ns and np are the refractive indices for s- and p- polarizations, respectively, and θ is the tilt angle of the column with respect to the substrate normal, as shown in Fig. 1 see Ref. 7. The birefringence of the film is given by Δn = ns–np, and this film birefringence can be used in the same manner as a birefringent crystal to shift the phase of one linear polarization with respect to the other. The use of a deposited film to alter the polarization state is beneficial in the ability to manufacture large-aperture components and improve laser-damage thresholds relative to other methods of polarization manipulation. By applying a film of the proper thickness, a wave plate may be fabricated, providing control of the polarization state of the incident light. The linearly polarized incident light will be oriented at 45° to the projected column orientation of the film, providing equal components encountering ns and np. The delayof one component based on the difference in Δn by π/4 or π/2 phase will alter the incident light to circular or orthogonal linearly polarized, respectively. The interaction of incident linearly polarized light with a quarter-wave plate may be expressed mathematically by using Jones matriceswhere Etx and Ety are the instantaneous components of the transmitted electric vector [13]. This equation represents a quarter-wave plate as a 2 × 2 matrix, with the ± indicating the fast axis being horizontal or vertical, respectively, multiplied by a vector representing a linearly polarized incident wave oriented at 45°. The change in the fast axis corresponds to a rotation of the substrate by 90° between depositions of different features. The resulting transmitted wave is simply left- or right-hand circularly polarized with an additional phase. Likewise, the use of a half-wave plate may be represented aswhere modification by the half-wave plate 2 × 2 matrix yields a linear polarization orthogonal to the input linear polarization incident on the film.
Fig. 1 (a) Refractive indices of the film are defined with respect to the column orientation in the film structure. The anisotropic film forms a birefringent structure, with principal refractive indices (b) defined by the incident light and the projected orientation relative to the columnar structure.
The goal of this effort is to fabricate a transmissive optical component that can provide equal areas of two complementary polarization states, either ± π/2-phase delays providing right-hand- and left-hand-circular polarizations, or alternatively the transmission of an incident linear polarization and its orthogonal polarization achieved with a π-phase retardance in the coated regions. Since any polarization state may be achieved by a combination of the two individual states, any transition region between polarization states may be treated as a superposition of the desired polarizations; provided the overall area of combined polarizations is relatively small, of the order of 10% or less, graded coating regions at the edges of patterned areas are acceptable. Such gradual transitions at the edges of the coated regions are required to limit diffraction by distinct edges, which increase the modulation in the transmitted laser intensity and risk subsequent damage to optics farther along the beamline.
The fabricated component must not only provide beam smoothing by varying polarization states, but transmission must be maximized for both laser operation and to limit potentially damaging reflections from the optic surfaces. Antireflection coatings suitable for high-fluence usage must be deposited on both surfaces of the final component, with low reflectivity not only on uncoated regions of the substrate but also over the thick GLAD-coated regions having a relatively uncontrolled film thickness relative to the interference requirements in an antireflection coating design. A means of antireflecting all regions of the substrate must be developed in order for this to be a viable beam-smoothing approach.
The reflectance from a substrate surface coated with two layers can be expressed with a characteristic matrix formalism:
where n1 and β1 refer to the refractive index and phase thickness of the single-layer antireflection coating, n2 and β2 refer to the refractive index and phase thickness of the GLAD coating, and nsub is the refractive index of the substrate [14]. The product of the characteristic matrices can be used to determine the optical admittance of the overall structure bywhere ϒ is the optical admittance. Since the optical admittance behaves like a refractive index for the combined overall structure, the goal is to achieve ϒ = 1 since this will provide zero overall reflectance from the coated surface. In order to attain an admittance match to air with n = 1, there are two primary options: If the phase thickness β2 of the GLAD structure is equal to mπ, where m is an integer, the characteristic matrix of the GLAD layer reduces to ± 1 multiplied by the identity matrix, simplifying the relationship towhich is simply the performance of the antireflection coating layer on the substrate at the use wavelength. The positive or negative sign indicates a phase change corresponding to even or odd multiples of π-phase thickness of the film, respectively. The tolerancing on such an approach, however, must also be evaluated. The phase thickness of the GLAD layer is given bywhere nGLAD and dGLAD are the refractive index and physical thickness of the GLAD layer, respectively, and λ is the wavelength of the incident light, which is 351 nm for this application. If we assume the GLAD layer thickness is 10 μm, the phase thickness of the layer isIf we consider limiting the error in the phase thickness to 5% of a quarter-wave optical thickness (β = π/2), the film-thickness errors over the GLAD region, including deposition thickness, refractive index, and film nonuniformity, must remain <0.05%—an impractical degree of performance. Given that the GLAD film is birefringent, with a reported Δn for alumina or silica on the order of 0.03, the birefringence alone exceeds this requirement [16]. Controlling the reflectivity by matching the phase thickness of the GLAD film is not viable.Another possible approach is to deposit a film with an average refractive index, defined as which is approximately equal to that of the substrate. Using a deposition material with a bulk refractive index greater than that of the substrate, the porosity of the GLAD structure can be tuned by changing the angle of the columns and the relative density of the film in order to deposit a coating with an average index of refraction equal to that of the substrate [7,14]. This is equivalent to setting n2 equal to nsub in Eq. (4), with the overall system performance simplifying to
which is simply the same as depositing the antireflection coating directly onto the substrate with a phase shift representing the additional thickness of the GLAD film, represented asThis result indicates that any thickness of GLAD film with a refractive index equal to that of the substrate will not change the reflectance of the assembly but will simply result in a phase shift of the reflected light. This result is valid for small values of Δn; for index matching to a fused-silica substrate, even Δn ~0.05 leads to a residual reflectance of <0.03%. Given Δn for alumina or silica is <0.03, the index match is sufficient for both polarizations to yield a negligible impact on the reflectance of the component [16]. Using this approach, the required antireflection coating design becomes identical for deposition on both the uncoated substrate and the GLAD-coated regions, significantly simplifying the antireflection coating requirements.The substrate to be patterned is nominally 40 × 40 cm, with the deposition to be performed on sub-apertures of 10 cm or less or continuous striped regions between 1 and 3 cm in width. Potential patterns include alternating squares or stripes of each polarization type as shown in Fig. 2, with a decreasing size of each region preferred for theoretical performance of the DPR, but increasing deposition time and transition areas between regions being detrimental. In addition, more-complex continuous patterns are also possible, such as sinusoidal paths across the substrate, or even a single continuous path that realizes a coatedarea of one-half the overall substrate aperture. The final component design may require a compromise of beam-smoothing performance and fabrication complexity.
Fig. 2 (a) Patterned deposition of right-hand-circular (blue) and left-hand-circular (green) wave plates using GLAD on a 400 × 400-mm fused-silica substrate. (b) A linear array of half-wave plates alternating with uncoated optic regions. The incident polarization would be linear, parallel to an optic edge.
GLAD coatings have been demonstrated with many different materials, but the primary interest for high laser-damage threshold use at 351 nm is limited to high-band-gap materials, with the preferred oxides being silica and alumina [17]. The porosity of a GLAD film increases as the film is deposited at higher angles of incidence, leading to a lower average index of refraction. The retardance based on Δn that may be achieved with each material varies, as does the retardance as a function of the angle of incidence of the vapor flux, which will alter the orientation angle of the film’s microstructure. The materials to be used must be selected to achieve sufficient laser-damage resistance for the application, high transmission with low scatter losses, performance stability as a function of time, and suitability for the use environment.
Deposition uniformity is also a significant challenge when fabricating large-aperture coatings, particularly when traditional methods of substrate motion may not be used to spatially average the arrival rates of evaporant flux. Consider a system configuration consisting of a 1-in.-wide by 4-in.-long rectangular aperture oriented for a glancing-angle deposition at 75° incidence placed at a source-to-substrate distance of 24 in., as shown in Fig. 3(a). Deposition uniformity may be calculated as in traditional optical-coating applications, by simply using a stationary rectangular substrate [18]. Such a calculation indicates that the horizontal film nonuniformity is <0.1%, given a small angular change and a relatively negligible change in source-to-substrate distance. There is the potential for an influence in the film’s anisotropy, leading to a change in the birefringence of the structure, but given an angular range of ± 1.2°, this effect is quite small. As a result of the high deposition angles, in the vertical direction the source-to-substrate distance is changing significantly, with a change of 3.86 in. for a 4-in.-long aperture. Calculation of deposited film thickness, again ignoring the effects of small potential changes in anisotropy, leads to a thickness change over the vertical dimension of >30% as shown in Fig. 3(b). To counteract such an error, thesubstrate may be coated at both ± 75° incidence to average the two orientations; the substrate may be translated along the vertical direction of the aperture to average the deposition at different heights; or the vapor plume may be modified with a rotating mask to limit the deposition on the lower end of the aperture [5]. An additional benefit is realized by orienting the aperture vertically and translating along its length; the edge features may be quite distinct, but these may also be controlled by altering the edges of the aperture. In this manner, the shape of the transition from coated region to bare substrate may be adjusted to control diffraction and corresponding beam modulation. To fabricate large-aperture coating structures, a combination of scanning and multiple orientations has been used for this work; the use of a rotating mask may be pursued in the future, if necessary.
Fig. 3 (a) Deposition geometry for a 1-in.-wide × 4-in.-long aperture at a distance of 24 in. from the evaporant source, oriented at 75° to the incident vapor flux. (b) The calculated film nonuniformity for ± 75° indicates a significant variation in film thickness, although the film’s performance can be largely compensated by averaging the two angular orientations.
3. Experimental procedure
A 45-in.-high coating chamber equipped with an electron-beam gun, quartz-crystal monitoring, and a stationary oblique-incidence substrate fixture was used to deposit oriented-column films as shown in Fig. 4. Granular silicon dioxide or aluminum oxide was evaporated from a continuously rotating 6-in. pan-type hearth in the electron-beam gun. The chamber was operated at ambient temperature, with a stainless-steel shield in place to reduce radiant substrate heating from the deposition source. No active substrate or chamber cooling was implemented at this time, although cooling was incorporated once longer depositions were performed regularly.
Fig. 4 Coating chamber configured with a stationary substrate fixture oriented at a high incidence angle relative to the incident evaporant vapor (65° to 85°). Shields limit the coating incident on the walls of the chamber.
The silica was evaporated at a rate of 0.4 nm/s as measured on the quartz crystal with an oxygen backfill in the chamber maintained at a pressure of 1 × 10−4 Torr. Aluminum oxide was melted in thin layers to form a solid ring in the electron-beam hearth and then evaporated at a rate of 0.4 nm/s on the quartz crystal with an oxygen backfill in the chamber maintained at a pressure of 1 × 10−4 Torr. Deposition rates were different on the substrate as a function of substrate angle, temperature, and position, so all deposition control was done with respect to the fixed quartz-crystal location. Changes in deposition conditions were explored for alumina, including increasing the rate to 0.6 to 0.8 nm/s and reducing any additional oxygen added to the vacuum environment, in an effort to decrease process time and better define the columnar film structure, respectively. Deposition rate is maintained with an Inficon IC5 deposition controller and a cartridge-type quartz crystal monitor that allows one to change the monitor crystal in the event of excessive crystal noise or failure [19]. The chamber temperature was monitored, with depositions interrupted as needed to maintain a temperature of <80°C, to minimize surface mobility of the condensing vapor.
An evaporation process for antireflection coatings was pursued in a 54-in. coating chamber with dual electron-beam guns, quartz-crystal monitoring, and a planetary rotation system configured to provide a highly uniform deposited film [20]. Multilayer interference coatings were deposited using silicon dioxide as the low-refractive-index material and either hafnium dioxide or aluminum oxide as the high-refractive-index material. In addition, sol-gel antireflection coatings were deposited by dipping the GLAD-coated substrate into a silane-based sol-gel and withdrawn at 300 steps/s to yield a quarter-wave optical thickness film at a center wavelength of 351 nm. Reflection and transmission were measured with a laser-based reflectometer operating at 351 nm to evaluate the GLAD-coated region, the regions of the substrate with only an antireflection coating, and the transition between the two regions.
Initial retardance was measured with a Hinds Instruments Exicor R&D Birefringence Measurement System operating at 633 nm. As this study progressed, measurement capabilities were added to allow for characterization at 355 nm; most sample retardance was determined using a Hinds Instruments Exicor 450XT Mueller Matrix Polarimeter. Reflectance, transmittance, and scatter were measured using a 351-nm laser and detector. Near-field modulation in the transmitted laser signal was also evaluated to determine the optimal configuration for joining discrete depositions in a pattern of polarization-control structures. Laser-damage testing was performed on cleaved float glass samples using nanosecond pulses at a wavelength of 351 nm [21,22].
4. Results
Silica films were used to fabricate birefringent GLAD structures that operate as quarter-wave plates at 351 nm, with a target retardance of 88 nm. By vertically aligning the center of each quadrant of a 100-mm-diam substrate in the deposition chamber, four nominally quarter-wave retarders were deposited on a single substrate, with the measured retardance on the 633-nm retardance mapper shown in Fig. 5. The arrows indicate the orientation of the retardance within each quadrant (aligned radially to the optic center), while the magnitude of the retardance is indicated by the pixel color. Deposition monitoring accuracy is limited in the development deposition system, so improved control of the nominal film retardance for each region was not pursued at this time. Variations in uniformity of retardance over each aperture, given the high angle of incidence (and corresponding variation in substrate height over the coated aperture), confirm the necessity of limiting deposition aperture based on the source-to-substrate deposition distance [23]. This sample was deposited with each quadrant stationary, without the use of positive and negative angular orientations or translation of theoptic behind an aperture. Photometric measurements of silica GLAD indicated poor performance resulting from optical scatter, with losses of 10% to 15% based on the sum of transmitted and reflected light relative to 100%. In addition, significant changes in optical retardance with respect to time have been observed; it is hypothesized that these changes are caused by hydrolysis of the silica coating, but regardless of the cause, the instability of the optical performance is a concern for component function [24]. Alumina films were also used to fabricate π/2 GLAD structures, with measured scatter losses of the film <1% at 351 nm. No discernable change in optical retardance with respect to time was observed.
Fig. 5 Retardance map of the coated 100-mm substrate, with four individual coated regions deposited with each blue arrow oriented upward in the chamber. The resulting fast axis of each coated area is aligned with the coating direction and the resulting column structure of the film.
Alumina has the added benefit that low-density alumina films may be deposited in such a way that the average refractive index of the GLAD film matches that of the fused-silica substrate, as shown in Eq. (9). As a result, thickness variations of the GLAD film do not impact substrate reflectance, and antireflection coatings based on the substrate refractive index may be used without regard to any variations in GLAD-layer thickness. Deposition at ± 75° angular orientations resulted in the uniformity variations being largely offset, forming a chevron-like columnar structure of the film, as shown in Fig. 6. Increasing the thickness of the alumina films to π retardance increased scatter losses to 5% to 6% at 351 nm, with significant spatial variations over the coated region. Depositing the alumina film over a 1 × 3-in. rectangular aperture at ± 75° angular orientations resulted in a relatively uniform retardance, as shown in Fig. 7. This provides better than ± 5% retardance over a large stationary aperture, demonstrating that the process is suitable for large-aperture coatings.
Fig. 6 Scanning electron microscope image of an alumina film deposited at ± 75° incidence, forming a chevron-like film structure on the fused-silica substrate. The film is approximately 3 μm thick, with a nominal retardance of 87 nm acting as a quarter-wave plate at 351 nm.
Fig. 7 Retardance map of an alumina film deposited through a rectangular aperture on a stationary substrate. Retardance uniformity demonstrates that the process is suitable for patterning large-aperture optics.
Silica GLAD films on cleaved float glass substrates were damage tested at 351 nm with a pulse duration of 0.7 ns. Damage thresholds as determined in a 1:1 testing protocol were measured at 10.42 ± 0.46 J/cm2, with a slight increase in N:1 damage thresholds [21,22]. Alumina GLAD films on cleaved float glass were damage tested at 351 nm in the same manner. Damage thresholds were measured at 11.31 ± 0.31 and 13.59 ± 1.11 J/cm2, with again a slight increase in N:1 damage thresholds. Based on these results, the material damage thresholds of both silica and alumina do not appear to be of significant concern; instead, the cleanliness of the surface, the fabrication of the substrate, and avoiding inclusion of contaminants in the open GLAD film structure will be of primary importance for high-damage-threshold performance in the final component. Once an antireflection coating isdeposited over the GLAD surface, the influence of this additional processing on the GLAD film must also be evaluated. There is the potential for altering the laser-damage threshold as a result of the antireflection coating process, but there is no indication at this time that sol-gel deposition would negatively impact the laser-damage threshold of a GLAD film. Given that both sol-gel and GLAD films are highly porous, they appear to behave very similarly. This will be quantitatively explored in detail in the future.
Alumina was selected for a scanned deposition process to fabricate stripes of one half-wave optical retardance of nominally 20-mm width on a 100-mm-diam fused-silica substrate. Since this is a scanning process, the configuration is identical to that needed for the final full-aperture substrate as shown in Fig. 2, requiring a larger substrate translation system and chamber and increased evaporation source capacity to enable deposition on the final optic. All other system considerations, including the aperture size and distance from the source to the substrate, will be unchanged from the 100-mm subscale component to the final optic. The chamber configuration is shown in Fig. 8 with the aperture centered above the evaporation source and the optic surface aligned with the axis of rotation of the stage assembly. The scanning system shown allows for 200 mm of linear translation in both the x and y directions. The substrate is translated horizontally to align the substrate at a position for a desired stripe in the vertical direction. By scanning vertically, aligned with the long axis of the aperture, the boundary of the coated region can be controlled by altering the profile of the aperture edges. As the substrate moves behind the aperture, the coated thickness will be directly proportional to the open region through which the substrate has passed; the edges may be controlled to minimize diffraction by transitioning the coating thickness through the use of a gradient mask edge. The substrate also incorporates large bevels to provide a means of holding it in place without the fixture extending beyond the optical surface to be coated.
Fig. 8 System for scanning deposition through a 1 × 3-in. aperture at a distance of 24 in. from the evaporant source, oriented at 70° to the incident vapor flux. Scanning in the (nearly) vertical direction creates individual stripes, while horizontal translation makes it possible to move to a new stripe location.
The coating deposition was performed with a deposition rate maintained at 0.35 nm/s on a quartz-crystal monitor without substrate heating. The chamber was evacuated to approximately 2 × 10−6 Torr, and the deposition was performed with an oxygen backfill to a pressure of 5 × 10−5 Torr. Aluminum oxide was evaporated from a continuously rotating 6-in. pan-type electron-beam gun, with a total deposition time of 11.2 h. The substrate was translated at a linear rate of 3 mm/s, moving the optic beyond the coating aperture on both ends of each linear scan, then reversing direction beyond the edge of the substrate. The coating aperture was placed 0.030 in. in front of the moving optic, minimizing the extent of the coating deposited behind the metal intended to block deposition. The retardance of the coated component is shown in Fig. 9(a). It is apparent in the photograph of a coated component between crossed polarizers in Fig. 9(b) that the GLAD region acts as a wave plate, yielding a high transmission, whereas the uncoated regions remain dark.
Fig. 9 The scanning GLAD process results in a striped substrate with (a) retardance indicating a half-wave–plate performance at 351 nm. (b) The uniformity and edge transitions are clear when viewed between crossed polarizers, with the GLAD structure oriented at ~45° to the incident polarization.
The light incident on the GLAD-coated component will be reflected, transmitted, scattered, or absorbed. The absorption of the GLAD-coated component is very low and assumed to be negligible since large-band-gap, low-absorption materials are being used and this low absorption is confirmed by laser-damage testing. Therefore, the scatter loss can be inferred by measuring both the specular reflectance and transmittance and subtracting the sum from the total incident light (100%). Before antireflection coating, the transmitted scatter of the GLAD coating was characterized as 2.5%, based on average polarization, with transmitted scatter losses in p polarization of 2.1% and s polarization at 2.9%. No significant difference in scatter was noted over the transitions at the edges of the region. After antireflection coating with sol-gel, the reflectivity of the component is <0.1% in the regions without GLAD coating, <0.75% over the GLAD region, and <3% in the transition regions. The scatter is calculated based on transmission and reflection measurements again assuming the absorption to be negligible; scatter of the coated component varies 3% to 8%, with the highest scatter occurring in the GLAD regions. This substrate was not intended for use in a high-fluence laser use, so laser-damage testing will be pursued once an appropriate coated substrate is available.
5. Conclusions
Birefringent GLAD films fabricated using silica and alumina have been demonstrated at 351 nm with high laser-damage thresholds and optical retardance for use as quarter-wave and half-wave plates. Based on low scatter losses, alumina GLAD films have been selected for use in large-aperture distributed polarization rotators at 351 nm. The use of a bilayer structure fabricated at positive and negative deposition angles, as well as scanning the substrate behind an aperture, has been demonstrated with an improvement in retardance uniformity, as expected. A subscale DPR has been demonstrated at 100-mm aperture, using a scanning deposition process equivalent to that necessary for a full-scale substrate with an aperture of 400 mm × 400 mm. Continued development will focus on reducing scatter, determining and improving laser-damage thresholds, and modifying coating edges to limit transmitted-beam modulation. The tessellation of such depositions to fabricate a large-aperture distributed polarization rotator should make significant beam smoothing possible on high-peak-power laser systems.
Acknowledgment
The authors wish to express their appreciation to Semyon Papernov for the measurements of laser-damage thresholds. This material is based upon work supported by the Department of Energy National Nuclear Security Administration under Award Number DE-NA0001944, the University of Rochester, and the New York State Energy Research and Development Authority. The support of DOE does not constitute an endorsement by DOE of the views expressed in this article.
References and links
1. “Phase conversion using distributed polarization rotation,” LLE Review Quarterly Report45, 1−12, Laboratory for Laser Energetics, University of Rochester, Rochester, NY, LLE Document No. DOE/DP40200–149, NITS Order No. DE91010027 (1990).
2. S. Skupsky and R. S. Craxton, “Irradiation uniformity for high-compression laser-fusion experiments,” Phys. Plasmas 6(5), 2157–2163 (1999). [CrossRef]
3. T. J. B. Collins, J. A. Marozas, K. S. Anderson, R. Betti, R. S. Craxton, J. A. Delettrez, V. N. Goncharov, D. R. Harding, F. J. Marshall, R. L. McCrory, D. D. Meyerhofer, P. W. McKenty, P. B. Radha, A. Shvydky, S. Skupsky, and J. D. Zuegel, “A polar-drive–ignition design for the National Ignition Facility,” Phys. Plasmas 19(5), 056308 (2012). [CrossRef]
4. S. N. Dixit, D. Munro, J. R. Murray, M. Nostrand, P. J. Wegner, D. Froula, C. A. Haynam, and B. J. MacGowan, “Polarization smoothing on the National Ignition Facility,” J. Phys. IV France 133, 717–720 (2006). [CrossRef]
5. T. Motohiro and Y. Taga, “Thin film retardation plate by oblique deposition,” Appl. Opt. 28(13), 2466–2482 (1989). [CrossRef] [PubMed]
6. I. Hodgkinson and Q. H. Wu, “Serial bideposition of anisotropic thin films with enhanced linear birefringence,” Appl. Opt. 38(16), 3621–3625 (1999). [CrossRef] [PubMed]
7. I. J. Hodgkinson, “Linear and circular form birefringence of coatings fabricated by serial bideposition,” Proc. SPIE 3790, 119–132 (1999). [CrossRef]
8. K. Robbie and M. J. Brett, “Sculptured thin films and glancing angle deposition: Growth mechanics and applications,” J. Vac. Sci. Technol. A 15(3), 1460–1465 (1997). [CrossRef]
9. K. Kaminska and K. Robbie, “Birefringent omnidirectional reflector,” Appl. Opt. 43(7), 1570–1576 (2004). [CrossRef] [PubMed]
10. S. R. Kennedy and M. J. Brett, “Porous broadband antireflection coating by glancing angle deposition,” Appl. Opt. 42(22), 4573–4579 (2003). [CrossRef] [PubMed]
11. M. M. Hawkeye and M. J. Brett, “Narrow bandpass optical filters fabricated with one-dimensionally periodic inhomogeneous thin films,” J. Appl. Phys. 100(4), 044322 (2006). [CrossRef]
12. K. M. Krause, M. T. Taschuk, and M. J. Brett, “Glancing angle deposition on a roll: Towards high-throughput nanostructured thin films,” J. Vac. Sci. Technol. A 31(3), 031507 (2013). [CrossRef]
13. E. Hecht, Optics, 2nd ed. (Addison-Wesley, 1987), pp 323–326.
14. H. A. Macleod, Thin-Film Optical Filters, 3rd ed. (IOP Publishing, 2001), pp 37–50.
15. N. G. Wakefield, J. B. Sorge, M. T. Taschuk, L. W. Bezuidenhout, M. J. Brett, and J. C. Sit, “Control of the principal refractive indices in biaxial metal oxide films,” J. Opt. Soc. Am. A 28(9), 1830–1840 (2011). [CrossRef]
16. Q. Wu and I. J. Hodgkinson, “Materials for Birefringent Coatings,” Appl. Opt. 33(34), 8109–8110 (1994). [CrossRef] [PubMed]
17. B. Mangote, L. Gallais, M. Zerrad, F. Lemarchand, L. H. Gao, M. Commandré, and M. Lequime, “A high accuracy femto-/picosecond laser damage test facility dedicated to the study of optical thin films,” Rev. Sci. Instrum. 83(1), 013109 (2012). [CrossRef] [PubMed]
18. J. B. Oliver and D. Talbot, “Optimization of deposition uniformity for large-aperture National Ignition Facility substrates in a planetary rotation system,” Appl. Opt. 45(13), 3097–3105 (2006). [CrossRef] [PubMed]
19. L. L. C. Sensors Technology and C. O. Fort Collins, 80528 (http://www.sensorstec.com/).
20. J. B. Oliver, J. Bromage, C. Smith, D. Sadowski, C. Dorrer, and A. L. Rigatti, “Plasma-ion-assisted coatings for 15 femtosecond laser systems,” Appl. Opt. 53(4), A221–A228 (2014). [CrossRef] [PubMed]
21. S. Papernov, D. Zaksas, J. F. Anzellotti, D. J. Smith, A. W. Schmid, D. R. Collier, and F. A. Carbone, “One step closer to the intrinsic laser-damage threshold of HfO2 and SiO2 monolayer thin films,” Proc. SPIE 3244, 434–445 (1998). [CrossRef]
22. S. Papernov and A. W. Schmid, “Localized absorption effects during 351 nm, pulsed laser irradiation of dielectric multilayer thin films,” J. Appl. Phys. 82(11), 5422–5432 (1997). [CrossRef]
23. H. A. Macleod, Thin-Film Optical Filters, 3rd ed. (IOP Publishing, 2001), pp. 488−497.
24. H. Leplan, B. Geenen, J. Y. Robic, and Y. Pauleau, “Residual stresses in evaporated silicon dioxide thin films: Correlation with deposition parameters and aging behavior,” J. Appl. Phys. 78(2), 962–968 (1995). [CrossRef]
