Current strategies for detecting and deflecting/destroying small to medium size asteroids that may possibly enter earth’s atmosphere are in their infancy. Projects such as Sentinel for detection and DE-STAR for deflecting of asteroids are gaining momentum while others are in design stages.
Single crystal silicon (SC-Si) with its unique set of mechanical and thermal properties is eminently suitable for deployment in any and all of these applications. But the superior thermal properties are particularly suited for high energy applications. We show that the exceptionally high thermal conductivity and diffusivity combined with relatively low thermal expansion enable SC-Si mirrors to handle the energy, often without active cooling, while maintaining figure. Both cooled and uncooled SC-Si mirrors have been employed successfully in high energy laser applications.
For IR detection of asteroids, SC-Si lenses, prisms and filters are invaluable in the 1-5 μm range along with SC-Si mirrors at any wavelength. This space qualified material has zero defects and is readily available in a range of sizes to 300 mm diameter from several vendors, but while boules to 460 mm have limited availability, they can be obtained. For applications requiring larger sizes, McCarter’s proprietary glass frit bonding is utilized. Components approaching one meter have been produced.
Zero defect single crystal silicon (Single-Crystal Si), with its diamond cubic crystal structure, is completely isotropic in most properties important for advanced aerospace systems. This paper will identify behavior of the three most dominant planes of the Single-Crystal Si cube (110), (100) and (111). For example, thermal and optical properties are completely isotropic for any given plane. The elastic and mechanical properties however are direction dependent. But we show through finite element analysis that in spite of this, near-isotropic behavior can be achieved with component designs that utilize the optimum elastic modulus in directions with the highest loads. Using glass frit bonding to assemble these planes is the only bonding agent that doesn’t degrade the performance of Single-Crystal Si. The most significant anisotropic property of Single-Crystal Si is the Young’s modulus of elasticity. Literature values vary substantially around a value of 145 GPa. The truth is that while the maximum modulus is 185 GPa, the most useful <110< crystallographic direction has a high 169 GPa, still higher than that of many materials such as aluminum and invar. And since Poisson’s ratio in this direction is an extremely low 0.064, distortion in the plane normal to the load is insignificant. While the minimum modulus is 130 GPa, a calculated average value is close to the optimum at approximately 160 GPa. The minimum modulus is therefore almost irrelevant. The (111) plane, referred to as the natural cleave plane survives impact that would overload the (110) and/or (100) plane due to its superior density. While mechanical properties vary from plane to plane each plane is uniform and response is predictable. Understanding the Single-Crystal Si diamond cube provides a design and manufacture path for building lightweight Single-Crystal Si systems with near-isotropic response to loads. It is clear then that near-isotropic elastic behavior is achievable in Single-Crystal Si components and will provide subsecond thermal equilibrium and sub-micron creep.
Mankind loves space and is drawn to explore its vastness. Existing space telescopes routinely
encounter data losses and delayed data collections during the constantly changing temperature and
load disruptions of space missions. The harsh environment of space thermal cycles and spacecraft
motion loads create unwanted activity such as spacecraft slew, acquisition slew, and temperature
induced blur. In order to compensate for the low performance of the materials currently used for
telescope optics, engineers and designers are using costly on-board coolers, mechanical actuators,
and deformed mirrors, for example, with limited success. However, Zero-defect Single Crystal
Silicon (SCSi) can perform in space environments without coolers, actuators, and other such devices
because SCSi is not ductile and is homogeneous and therefore is not subject to creep, and will not
jitter, or blur during operations. To take advantage of the unique advantages of Zero-defect SCSi,
we are developing and fabricating a Cryostable All-Silicon Imaging Telescope (CAIT). In this
paper, we will discuss the basis for selecting SCSi for our space telescope design, the status of the
CAIT design and fabrication progress, and compare SCSi thermal and strength properties with other
typical space optical materials.
Single crystal silicon (SCSi) is light, strong, has excellent thermal properties, is readily available and cost and delivery
are competitive with, and probably better than, either beryllium or silicon carbide. In addition, SCSi's zero-defect crystal
structure enables polishing to near-perfect surfaces.
Recent developments in direct bonding have led to simple methods of attaching SCSi, a brittle material, to enhance its
high compressive strength and avoid tensile/brittle failures. Dynamic testing of a bonded assembly has demonstrated
high resonant frequencies and damping capacity. Other recent test results have shown the excellent temporal and thermal
stability of both monolithic and bonded mirror specimens.
So why not choose silicon?
Single Crystal Silicon (SCSi) is proving to be an excellent material for the fabrication of lightweight optical components
for use in space. As part of the feasibility studies performed prior to space flight applications, it is important to
determine the mechanical properties of complex structures manufactured from individual sections of SCSi. As an
additional integral building block for future multi-component SCSi structures, the behavior of the McCarter Machine
proprietary frit-bonded metal insert technology was examined. Here we report vibration test results, the objective of
which was to measure the structural damping characteristics of a typical silicon structure and verify its structural stability
after exposure to random vibration. The tests were designed to better understand SCSi, not only as a mirror substrate,
but also as a structural material. The success of this test, combined with the already proven McCarter Machine
manufacturing techniques, give us the ability to now manufacture new lightweight and stable opto-mechanical
assemblies entirely out of SCSi. But since requirements for larger and more sophisticated SCSi structures are limited by
the practical size of available boules, the behavior of these frit-bonded SCSi structures needs to be better understood.
This understanding will be obtained from planned testing of larger frit bonded SCSi opto-mechanical structural
components and assemblies.
Silicon components such as mirrors and infrared lenses have been manufactured for many years, primarily from polycrystalline
silicon (poly). There are inherent advantages that Single Crystal Silicon, (SCSi), has over poly, such as strength and dimensional
stability, that make it more suitable for telescopes. However, there are challenges in the design of an all-SCSi telescope. SCSi is
brittle and has low tensile strength compared to its compressive strength. These properties therefore dictate designs that
minimize tensile stresses and eliminate direct mechanical attachments. McCarter has accepted these challenges and has designed
and is fabricating a lightweight telescope that can replace one of beryllium at substantial savings of cost and schedule.
The challenge of direct attachment has been solved with the use of bonded threaded inserts of low expansion metal. Bonding has
been studied extensively as described in a companion paper, but the proprietary frit-bonding technique developed by Frank
Anthony proved to be the most predictable, stable, and reliable. This technique is also used to fabricate complex components
from an assembly of simpler parts.
To minimize tensile stresses, the mechanical design had to be modified from the original without changing the optical
prescription. This has been successfully accomplished through a "design for manufacturing" approach teaming designers, the
stress analyst and manufacturing personnel. This approach has provided a design that is being produced at lower risk, lower cost
and with higher predicted reliability with no loss in performance.
The value of glass frit bonding to assemble silicon parts wa demonstrated by the successful evaluation of cryostability of a small multi-piece silicon mirror. This bonding technique has been extended to the assembly of a 44kg block of silicon, 113 x 400 x 400 mm. Such an assembly was considered to be a cost competitive alternative to the purchase of a custom sized silicon boule. Various types of evaluation provided the foundation upon which this accomplishment is based. Included were cryocycling of frit bonded plates, comparison of the strength of bonded silicon bend bars with that of silicon bend bars, and the fabrication and cryotest of a small concave frit bonded silicon mirror. Of the 16 bonded bars in two groups only 2 had failures in the bondline, 11 failed in the silicon, and origins could not be determined for 3 bars. The two groups of bonded silicon bars had average strengths that were 84% and 91% of the average strength of the plain silicon bars. In view of the relatively small number of bars in each group this is not surprising. The cryostability of the concave bonded silicon mirror was demonstrated by a figure error of less than 0.06 wave rms at 633 nm, cold to warm, compared to a specification of 0.1 wave rms, and 0.014 wave rms, warm to cold to warm, over an 80% clear aperture. These results are reviewed before interesting features of the large block are discussed. Finally, projections are made regarding possible future applications for this bonding process.
The excellent polishability, low density and relatively high stiffness of silicon make it an attractive candidate for optical applications that require superior performance. Assembly of silicon details by means of glass frit bonding permits significant weight reduction thus enhancing the benefit of silicon mirrors. To demonstrate the performance potential, a small lightweight glass frit bonded silicon mirror was fabricated and tested for cryostability. The test mirror was 12.5cm in diameter with a 60cm spherical radius and a maximum thickness, at the perimeter, of 2.5cm. A machined silicon core was used to stiffen the two face sheets of the silicon sandwich. These three elements were assembled, by glass frit bonding, to form the substrate that was polished. The experimental evaluation, in a liquid nitrogen cryostat, demonstrated cryostability performance significantly better than required by the mirror specification.
A grinding technique referred to as the McCarter Superfinish, for grinding large-size optical components is discussed and certain surface characterization information about flatness and the relative magnitude of the subsurface damage in silicon substrates is reported. The flatness measurements were obtained with a Wyko surface analyzer, and the substrate damage measurements were made by x-ray diffraction and acid etching. Results indicate excellent control of flatness and fine surface finish. X-ray measurements show that the diamond wheels with small particle sizes used in the final phases of the grinding operation renders surfaces with relatively small subsurface damage.
At the 1998 ASPE Meeting in Carmel, California there was much discussion of surface and subsurface damage introduced into silicon by machining operations. Many investigators have studied the problem and have defined parameters of importance. Yet there is a need for costly and time-consuming post- machining operations, such as lapping, if items of highest quality are to be produced. Significant cost reductions should be possible with machining techniques that introduce minimal damage. The combination of tool selection and treatment, speed and feed parameters, coolant choice and talent has resulted in the improved machining process. This process reduces damage without increasing machining cost. Although the McCarter Superfinish is a proprietary procedure the resulting surface condition will be compared with that of conventional machining. Data supplied by a customer is the basis for the comparison.