High Aspect Ratio 3D Microstructures

Astronomical instrumentation for measurements in the Far Ultraviolet (FUV, 90−200 nm) have historically considered aluminum (Al) thin film mirrors due to this material high reflectance over this wavelength range. However, the native aluminum oxide layer that forms on Al upon exposure to the atmosphere is strongly absorbing in this wavelength range, requiring that the films be protected with a dielectric that inhibits oxidation. Typically, magnesium fluoride (MgF₂) or lithium fluoride (LiF) coatings are used as protective layers, but each has shortcomings. For example, MgF₂ has an absorption cutoff at 115 nm that reduces performance below this wavelength, which is a critical part of the FUV spectrum for observational astrophysics. The use of LiF as a protection for Al provides a lower absorption cutoff at 100 nm, but it is hygroscopic and thus susceptible to degradation in humid conditions. Our team at GSFC has developed a new reactive Physical Vapor Deposition (rPVD) process that consists of a fluorination process with XeF₂ gas combined with our traditional PVD process. We have found that this new rPVD process produces Al+XeF₂+LiF (XeLiF) and Al+XeF₂+MgF₂ (XeMgF₂) mirror coatings with unprecedented reflectance. In addition, the rPVD process seems to produce much more environmentally stable coatings (when compared to the conventional process without the XeF₂ fluorination). We report on IR/Vis/UV reflectance of XeLiF and XeMgF₂ mirrors. The surface roughness as well as the FUV reflectance measured over a period of 8 months for a XeLiF sample with a relatively thin (≃ 30 nm) Al layer are also reported. We have also been investigating the compatibility of this rPVD coating process for potential efficiency enhancements of Si-based gratings. Since it is known that the XeF₂ vapor is a strong Si etchant, we are investigating if the native SiO₂ layer on Si is sufficient to protect the groove profile of E-beam-ruled Si gratings from degradation. Preliminary results indicate that the native SiO₂ layer is an effective barrier against etching of Si by XeF₂.

The ubiquity of particulate contamination requires dust mitigation techniques to provide low scatter surfaces and edges on sensitive optical devices in space. Poly(olefin sulfone)s have been shown to photodegrade with the assistance of a photobase generator when exposed to UV light (254 nm) and heat (120 °C). These may be applicable in minimizing dust on optical surfaces for space applications. However, their behavior in vacuum has not been fully characterized. We synthesized poly(2-methyl-1-pentene sulfone) (PMPS) and poly(1-hexene sulfone) (PHS) with and without a photobase generator. We studied the photodegradation (172 nm or 254 nm) of thin films in vacuum. Spectroscopic ellipsometry was used to quantify film thickness over time. The PMPS and PHS films both degraded when exposed to UV light in vacuum, though PHS to a lesser degree. We found that heat was not required to cause degradation, and that degradation occurred with UV irradiation even without a photobase generator. This investigation shows that poly(olefin sulfone)s could be used to protect optical surfaces until their deployment in space.

Windows for vacuum ultraviolet (VUV) sources are valuable for many applications but difficult to fabricate due to most materials being too absorptive at VUV wavelengths. We have designed, fabricated, and characterized a carbon nanotube (CNT) collimator as a window with high (VUV) transmission and significant differential pumping. The CNT collimators are arrays of square channels of various dimensions and height with sidewalls composed of vertically aligned CNT forests. The CNT collimators in this work exhibited peak intensity transmissions for VUV light (58.4 nm) of 18%–37% of that reported for the same system without a collimator present [S. Olsen, D. Allred, and R. Vanfleet, J. Vac. Sci. Technol. A (2024)]. Further analysis found that the peak intensity transmissions were lowered due to carbon deposition on the phosphor viewing screen from contaminants. The CNT collimator also had significant sidewall reflection in the VUV range (⁠R = 0.21 +/- 0.08) in the VUV range for angles 15.6 degrees and below). Pressure ratios (low pressure over high pressure) in the VUV transmission experiment were dominated by leaks in the alignment mechanism. Additional experiments demonstrated the CNT collimator’s reflection and superior differential pumping with pressure ratios less than 0.001.

Hollow cathodes are a common type of vacuum ultraviolet (VUV) light source with a wide range of design and application. We determined the VUV (58.4 nm) intensity distribution of a hollow cathode as a function of current and pressure. Our model describes the intensity distribution of a McPherson 629-like hollow cathode helium plasma within the range of 0.50–1.00 A and 0.50–1.00 Torr as a ring with a center peak. We found that for all pressures and currents considered, the ring emits more VUV light than the center peak. We also found that the center peak has a minimum VUV light emission near 0.9 Torr.

Tabletop extreme ultraviolet (EUV) sources based on high harmonic generation (HHG) have been used as a powerful tool for probing magnetism. Obtaining magnetic information via magneto-optical contrast often requires the energy of the light to be tuned to magnetic resonance energies of the magnetic element present in the material; therefore, it is essential to calibrate the HHG spectrum to well defined absorption energies of materials. We have designed and assembled a HHG based EUV source for studying transition metal magnetic materials at their resonant M-absorption edges (35-75 eV of photon energy). One material of interest is iron, for which the iron M2,3 edge is 52.7 eV (23.5 nm wavelength) according to CXRO. We prepared and characterized a thin sample of iron for absorption spectroscopy and calibration of the absorption edge with beamline 6.3.2 at the Advance Light Source (ALS) in Lawrence Berkeley National Laboratory. This well characterized sample was capped with gold to prevent oxidation. From these measurements we extracted the absorption part of the index of refraction β spectrally and confirmed that the absorption edge of iron is 52.7 eV. With this information, we can better calibrate the HHG spectrum of our tabletop EUV source. Calibration of the HHG spectrum was achieved using model fitting the HHG spectrum using the grating equation and law of cosines while taking account into the results of the ALS data. We have determined that driving wavelength of the HHG process to be 773 nm. We also conclude that the chirp of the driving laser pulse can cause an energy shift to a HHG spectrum.

Tabletop extreme ultraviolet (EUV) sources based on high harmonic generation (HHG) have been used as a powerful tool for probing magnetism. Obtaining magnetic information via magneto-optical contrast often requires the energy of the light to be tuned to magnetic resonance energies of the magnetic element present in the material; therefore, it is essential to calibrate the HHG spectrum to well defined absorption energies of materials. We have designed and assembled a HHG based EUV source for studying transition metal magnetic materials at their resonant M-absorption edges (35-75 eV of photon energy). One material of interest is iron, for which the iron M2,3 edge is 52.7 eV (23.5 nm wavelength) according to CXRO. We prepared and characterized a thin sample of iron for absorption spectroscopy and calibration of the absorption edge with beamline 6.3.2 at the Advance Light Source (ALS) in Lawrence Berkeley National Laboratory. This well characterized sample was capped with gold to prevent oxidation. From these measurements we extracted the absorption part of the index of refraction β spectrally and confirmed that the absorption edge of iron is 52.7 eV. With this information, we can better calibrate the HHG spectrum of our tabletop EUV source. Calibration of the HHG spectrum was achieved using model fitting the HHG spectrum using the grating equation and law of cosines while taking account into the results of the ALS data. We have determined that driving wavelength of the HHG process to be 773 nm. We also conclude that the chirp of the driving laser pulse can cause an energy shift to a HHG spectrum.