Open Access
Add to BibSonomy
Abstract
Contamination of pulse compression gratings during the manufacturing process is known to give rise to reduced laser damage performance and represents an issue that has not yet been adequately resolved. The present work demonstrates that the currently used etching methods introduce carbon contamination inside the etched region extending to a 50- to 80-nm layer below the surface. This study was executed using custom samples prepared in both, a laboratory setting and by established commercial vendors, showing results that are very similar. The laser-induced-damage performance of the etched and unetched regions in the grating-like samples suggest that contaminants introduced by etching process are contributing to the reduction of the laser-induced damage threshold.
© 2022 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Advancement of the performance of short-pulse laser systems is in part associated with improvements in the performance of the pulse-compression gratings used to achieve chirped-pulse amplification (CPA) [1]. This made it possible to develop an array of laser systems and designs that have enabled probing physical phenomena at much shorter time scales. In addition, short pulse laser systems have found important technological applications [2–7]. Of particular interest are laser systems designed to study materials at extreme conditions, where the need for maximum output tests the limits of the operation profile of the constituent optical components [8]. In such systems, but also in general for ultrashort laser systems, the gratings are the weak link presenting high risk for laser-induced damage, therefore limiting the power output of the system. As a result, efforts to improve the laser-damage resistance of gratings have been continuous since the discovery of CPA.
Current generation gratings include designs based on multilayer dielectric (MLD) coatings, which offer higher laser-induced-damage threshold (LIDT) than metal coating designs, thereby enabling higher power output [9–12]. Previous studies have demonstrated that the carbon-containing contamination on the surface generated during the fabrication process lowers the LIDT of gratings [13–17]. Absorption of the light by these contaminating compounds leads to laser energy coupling into the material followed by a localized ablation event (damage). The manufacturing process of MLD gratings gives rise to differences in the chemical environment, local to the surface of pillars (unetched) and trenches (etched, including the pillar wall regions). It has been shown that damage initiation typically occurs within the pillars [18–22] and has been attributed to the fact that they represent the regions where maximum electric-field intensification (EFI) is occurring. Despite continuous improvements in the design and manufacturing of gratings, the current generation gratings are limited by the low damage threshold that is below that after consideration of the EFI (which proportionally reduces the damage threshold of the patterned optical elements compared to the “pristine” flat coating layer) by a factor or 2 or more. It has been postulated that this behavior is due to defects introduced during the fabrication process [23,24].
The fabrication process of commercial gratings has been adopted from the semiconductor industry and involves lithographic and etching steps on fused-silica (FS) layers deposited on top of the MLD stacks. However, the targeted high periodicity, depth of the trenches and aspect ratio of the grating structures necessitates the use of specific photolithographic resists while bottom antireflective coatings (BARC) is used to improve the quality (shape and size) of the grating’s pattern of pillars and trenches. Photolithographic resists in grating manufacturing are typically applied as thick 300- to 500-nm films on top of the 40- to 70-nm BARC layers to create resist masks for reactive ion etching (RIE) or reactive ion beam etching (RIBE) of tall FS grating lines. Such thicknesses are required to ensure needed etch selectivity (e.g., a minimum amount of photoresist that will mask the pillars until the trenches are fully developed), and to ensure that the walls of the patterned photoresist mask are free of periodical standing wave features generated by the reflected exposure light. For comparison, the thickness of the resist and antireflective layers in semiconductor manufacturing rarely exceeds 100 nm. Such a significant increase in the amount of organic materials used in the grating manufacturing, combined with the prolonged etching times and porous nature of the fused-silica coating, can potentially increase the amount of contaminants that persist into the grating structures compared to the analogous cases in the semiconductor industry. Furthermore, organic contaminants can absorb the laser light, thus being more impactful for the functionality of the devise.
Previous efforts to improve the damage performance of the MLD gratings were primarily focused on advancing cleaning procedures to reduce the amount of superficial contaminants [25]. However, these efforts were only partially successful in achieving the theoretically possible LIDT values (estimated from the LIDT value for the constituent layer normalized to the electric-field enhancement), and it remains unclear why that being the case. Therefore, it is important to further investigate the presence of contamination of the different parts of the grating structures (pillars, trenches, and walls) arising from the manufacturing process and ultimately correlate with the LIDT performance.
This study examines the sources of contamination of different parts of the grating structures generated during grating manufacturing and correlates with the corresponding LIDT values. To achieve more effective removal of organic contaminants (as detected by our instrumentation) compared to previously reported cleaning methods, modifications and additional approaches were explored. Our work shows that in addition to superficial contamination, additional subsurface contamination was detected in the trenches regions. Specifically, carbon contamination inside the etched region is detected extending to a subsurface layer on the order of 50-80 nm. This study was executed using custom samples prepared in both, a laboratory setting and by established commercial vendors. Our results suggest that modification of the top grating material and/or refinement of the grating manufacturing steps are needed to produce cleaner gratings.
2. Materials and methods
2.1 Material characterization
XPS Analysis: The X-ray photoelectron spectrometer (XPS, Kratos AXIS Ultra DLD XPS) was equipped with a mono-Al x-ray source (1468.6 eV). The XPS spectra were collected using a field-of-view lens of nominally 800 × 800 µm, combined with a 2-mm aperture, resulting in an approximately 220 × 220-µm analysis area. Multiple sweeps were recorded for the survey and regional scans (typically 6 to 10 sweeps) with a pass energy of 20 eV to increase signal-to-noise ratio. The electron collection angle in all XPS measurements was zero. The x-ray gun was set at a 10-mA emission current with 15 kV at the anode. For depth profiling, the ion gun was held at an emission current of 10 mA and a source voltage of 4.86 kV, with the extractor current fluctuating ∼4 µA. The raster area for the ion beam was 1.5 × 1.5 mm to ensure that all measurements were taken near the bottom of the sputtered area. The neutralizer was kept on at all times with a 2-A filament current, charge balance of 2.6 V, and filament bias of 1.3 V in order to reduce peak shifting due to charge accumulation on the insulating surface.
The XPS signal areas were measured using Casa XPS software. The relative sensitivity factors (RSF) were provided by the Kratos XPS software. A Tougaard-style background approximation was used in all integrations. In-depth profile measurements and region energy ranges were selected based on surface regions and propagated to all subsequent measurements to ensure consistency, with occasional manual adjustments made when the background deviated too far from the average because of high noise in the data. Reported peak areas are divided by their respective RSF.
Imaging instrumentation: An optical microscope (Carl Zeiss Axio Imager.A2m) was operated with varying magnifications (objectives: ×10 to ×50) and the light sources (bright field mode: halogen 100 W) to image sample surfaces.
A scanning electron microscope (SEM, Zeiss Auriga SEM/FIB) was operated with the InLens mode, 1-4 kV and a working distance ranging from 4 mm to 8 mm. Images were recorded in the top view and at a 45° angle.
Laser-induced–damage testing: Damage testing was performed using a laser system operating at 1053 nm with a tunable pulse duration between 0.6 ps and 100 ps. The system has been described in Ashe et al[26]. Samples were tested in a vacuum environment (4 × 10–7 Torr) with s-polarized light at a 61° angle of incidence. Damage testing was performed in a single-shot (1-on-1) regime and damage was confirmed using Nomarski differential interference contract microscopy after testing.
2.1 Sample fabrication
To examine the difference in surface composition and LIDT values between the pillars, trenches, and walls of the grating lines, we used representative grating structures consisting of 5-mm-wide pillars separated by 5-mm trenches. These mm-pitch structures were otherwise prepared from the same materials and have the same wall height as the commercial pulse compression gratings (see Fig. 1). By using 5-mm structures, we were able to independently test pillars and trenches for their surface composition, roughness, and LIDT. This task would not be possible to execute using actual gratings because of the small size of the individual features.
Fig. 1. (a) Composition and manufacturing steps of the 5-mm grating-like structures. SEM images of (b) the MLD grating and (c) the 5-mm grating-like lines.
Samples were prepared using a nominal 1-µm thick monolayer coating of e-beam–deposited SiO2 on bulk fused silica using photolithography, reactive ion etching (RIE) or reactive ion beam etching (RIBE) techniques, and post etch cleaning (Fig. 1). The BARC and photoresist materials used in this work are the same to those used in the manufacturing of the commercial MLD gratings. Subsequently, the photoresist patterned substrates were etched using in-house RIE facilities as well as two established commercial vendors that used the RIBE instrumentation. Finally, the etched substrates were cleaned using previously developed and new cleaning protocols. This process is discussed in detail.
Lithography:
Etch masks were fabricated employing a standard photolithographic process using a quartz Cr photomask and contact mask aligner. We used positive PFI-88A7 photoresist and a bottom antireflective i-con 7 layer as a resist-BARC combination developed for high-aspect ratio features. Based on the resist etch selectivity and our measurements, a 400-nm layer of the resist was used to produce 800-nm wall structures in fused silica during RIE and RIBE. Both the resist and the BARC layers required prolonged bake times at 140°C and 80°C, respectively, leading to an increase in cross-linking density of the BARC and exposed resist polymers. The fabrication process is shown in Fig. 2 with the x-ray photoelectron spectroscopy (XPS) analysis of both the ARC and photoresist layers.
Etching:
The photolithographically patterned samples were etched in-house using reactive ion etching (RIE) and by two commercial vendors using reactive ion beam etching (RIBE) instrumentation. The RIE is a conventional method to etch microstructures in silica or silicon. Silica microstructures are usually fabricated using fluorine-based gas chemistry including molecules such as CHF3 and CF4 [27–29]. Oxidants like O2 are added to remove the horizontally accumulated fluorinated passivation, which could undermine further etching. The reactive gases are typically diluted with Ar to moderate the overall etching rate and to improve etch anisotropy. RIBE uses a highly directional energetic ion beam to chemically or/and physically remove material from a substrate. Silica layers are typically etched by RIBE using similar gas chemistries to those in the RIE processing (e.g., CHF3/CF4, O2, Ar mixture) [30,31]. RIBE is preferred to RIE in commercial grating manufacturing because it can accommodate meter-size substrates, and because it can be operated at higher temperature and lower pressure than RIE permitting more effective removal of the etched products, which minimizes re-sputtering and surface roughness in the etched areas.
The gas composition for the in-house manufactured samples using RIE was optimized to achieve the desired selectivity and feature wall morphology. The ARC layer is an organic material hardened by an extended bake and thus, different etching selectivity is required to remove the ARC first and then the inorganic fused-silica layer as shown in Fig. 3(a). To achieve both successive ARC layer and silica coating etching while the photoresist layer remained as the etch mask, several gas compositions were tested, and a two-step etching procedure was identified with different etchant gas ratios. A 10-min RIE step where 7.5% O2 and 7.5% CHF3 were applied to remove the ARC layer was followed by a 20-min RIE step where 5% O2 and 10% CHF3 were applied to create the silica trench. XPS analysis was conducted to confirm the removal of target layers during different steps. To avoid heat accumulation and possible melting of the photoresist on top of the insulating substrate, all etching was done in discrete 5-min intervals with a minimum cool-off time of 1 min between intervals.
Fig. 3. (a) Schematic depictions of the reactive ion etch process and SEM images of the etched substrate at (b) wall, (c) trench, and (d) pillar locations.
A short O2 descum step was used before the etching process to achieve a better etch profile. The tilted SEM image in Fig. 3(b) shows the side wall and surface morphologies of the trench [Fig. 3(c)] and pillar [Fig. 3(d)] regions. The pillar is still protected by a photoresist layer and the trench exhibits a rough surface while a ribbon structure is attached to the side wall, which is attributed to a fluorine passivation layer [32].
Lithography and etching protocol:
The in-house manufactured samples were fabricated using the protocol described next. The substrate was first immersed in Nanostrip (KMG Electronic Chemicals, Inc) at 75°C for 30 min, then rinsed with 0.22-µm filtered pure water and isopropanol to remove organic contaminations, and dried with N2 flow before using. Antireflective coating (ARC i-con 7) was spin coated onto the substrate at 2500 rpm and subsequently baked at 140°C for 7 h. After cooling to ambient lab conditions, positive photoresist (PFI-88A7, Sumitomo Chemical) was spin coated onto the substrate at 1500 rpm and subsequently baked at 80°C for 8 h in a N2 atmosphere. The photoresist was exposed using i-line OAI 200 Mask Aligner (200 mJ/cm2) through a Cr photomask (PhotoScience) with a pattern of 5-mm-wide lines and 5-mm spacing to allow detailed characterization of each line feature. After developing with MF-CD-26 developer (Rohm & Hass Electronic Materials LLC), the substrate was etched by reactive ion etch (RIE, South Bay Technology Reactive Ion Etcher RIE-2000). A 50-sccm O2 flow with 50-W forward power was first applied for 1 min, the time needed for the descum of the photoresist pattern to ensure a vertical wall. Then RIE was operated with a gas mixture of 42.5 sccm Ar, 3.7 sccm O2, and 3.7 sccm CHF3 with 100-W forward power for 10 min in the aim of penetrating the ARC layer. The second etching process included a 20-min RIE with 42.5 sccm Ar, 2.5 sccm O2, 5.0 sccm CHF3, 600-V bias voltage and 100-W forward power to etch the silica layer. The entire 30-min etching process was operated with 5-min/step and 1-min cooling between each step to avoid thermal accumulation of the photoresist.
The samples manufactured by two commercial vendors were prepared in a similar manner, except that instead of the RIE they used proprietary fluoride-based RIBE procedures.
Post-fabrication cleaning:
One of the goals of this study was to examine the effectiveness of an oxidative cleaning protocol in removing contaminants from the pillars and trenches of the etched structures and to determine surface composition of the cleaned interfaces.
Initially, we used a previously developed multistep advanced cleaning protocol (ACP) [25] that relies on hot Piranha spray solutions to clean the prepared grating-like structures. We determined that after the ACP, large areas of the pillar structures were still covered by thick organic coatings left from the photoresist and ARC layers. The LIDT values measured on these areas were ∼6 × lower than those on the unmodified fused-silica substrate in 10-ps damage tests. The contaminants generated during manufacturing mainly originate from the photoresist and ARC layers, which have high cross-linking densities and can be difficult to remove by a single organic oxidation and stripping step. Moreover, purely oxidative cleaning can induce further polymerization and hardening of these materials and increase their light absorption efficiency. This motivated us to explore an alternative cleaning protocol (SCP) that relies on common commercially available reagents and presoaking and rinsing steps to soften the photoresist and ARC layers before oxidative cleaning.
SCP uses a long-time hot NMP soak as a main photoresist removing step, followed by a Nanostrip soak to oxidize and dissolve most of the organic contaminants. The substrate was first immersed in NMP at 60°C for 1 h, then rinsed with isopropanol and Millipore water, and dried with N2 flow. The substrate was then immersed in Nanostrip at 75°C for 1 h, rinsed with isopropanol and Millipore water, and dried with N2 flow. O2 plasma cleaning was applied with a pressure of 120 mTorr and 100-W forward power for 5 min and followed by 15 min Nanostrip immersion at 75°C, then rinsed with isopropanol and Millipore water, and dried with N2 flow. This was implemented to remove traces of organic residues including the fluorocarbon polymers generated during RIE. The SCP steps are depicted in Fig. 4(a). The XPS analysis of F/Si and C/Si signal ratios [Fig. 4(b)] indicates that the SCP removes most, but not all, the organic contaminants and fluorocarbons in both the trenches and pillar areas. SEM images [Fig. 4(c)] show a significant difference in surface roughness between the pillars and trenches.
Fig. 4. (a) Illustration of the simplified cleaning protocol; (b) XPS analysis of the carbon and fluorine to silicon ratio of the cleaned surface; (c) tilted SEM images of the pillar and trench regions, and (d) LIDT values of the cleaned surface with pristine coating as reference.
3. Results
To explore possible correlation of the presence of surface contaminants and roughness with the laser-damage performance of the grating devices, we have conducted measurements of the LIDT’s in the trenches and pillar areas using 10-ps and 600-fs pulses as well as on preprocessed native coatings. The LIDT experimental results are summarized in Fig. 4(d). The10-ps measurements show that LIDT values of the pillar regions are similar to those of the native unmodified coatings, while the trenches experience a significant reduction of LIDT. The 600-fs measurements demonstrate LIDT reduction in both pillar and trenches compared to the native coating. To better understand these results, we must consider that the damage initiation mechanism is different in these two pulse regimes [22,33]. Specifically, with 0.6-ps pulses, the type-I damage process (volume breakdown assisted by absorption by atomic-scale defects) is confined within the region of peak EFI. The type-II damage process is involved in10-ps damage in MLD gratings (and coatings) associated with preexisting in the coating’s nanoscale absorbing defect structures. These results suggest that the presence of contamination on the surfaces of both pillar and trench areas leads to the reduction of LIDT at 600-fs. However, this surface contamination is less impactful for the 10-ps LIDT measurements where the damage initiation mechanism is attributed to preexisting defect structures found inside the silica coating. In this case, the presence of additional absorption in the region surrounding the defect can affect the LIDT. The results show that the 10-ps LIDT of the pillar regions is closer to the LIDT of the native coating suggesting that the bulk sub-surface composition and structure of the pillar regions was not affected during the manufacturing steps. On the other hand, a significant reduction of the LIDT at 10-ps in the trenches suggests that the etching process modifies a sub-surface layer of the fused silica possibly through the infusion of sub-surface contaminants.
To further explore this hypothesis, we tested the possibility that our in-house RIE process introduces subsurface contaminants into the trench areas and that these contaminants are associated with a reduction of the LIDT performance. As such, to determine the difference in a near-surface contaminant composition of the grating areas, we examined the relative XPS concentration of carbon materials in the pillars and trenches as a function of material thickness. These measurements were compared with the control FS coating that was not exposed to the lithography and etching. In these experiments, we used the depth-profile XPS (DP-XPS) analysis with an Ar ion gun to slowly etch through the FS layer and scan for the presence of C1s and Si2p electrons.
We first determine the etch rate of FS with the XPS Ar gun using an MLD dielectric substrate with a top FS layer of a known thickness (363 nm) and an underlying hafnia layer. By monitoring the decrease of the Si2p electron counts and the rise of the Hf4f counts, the etch rate was determined to be 0.9 nm per minute. Subsequently, the trench and pillar areas of the 5-mm grating lines and the pristine native coating were analyzed by DP-XPS. The data were collected in 3-min etch intervals (2.7 nm) for a total etch time of 30 and 18 min (80 and 50 nm). Figure 5 shows that all three regions contained similar amounts of carbon material on their surfaces before etching. This is consistent with the presence of adventitious organic molecules that are physically adsorbed on the surface of the glass from the atmosphere. Figure 5(c) shows the unmodified, e-beam deposited silica monolayer coating, while the 5-mm line structure in an e-beam deposited coating is shown both in the trench, [Fig. 5(b)] and pillar [Fig. 5(a)]. Analysis of the unmodified coating for carbon indicated that adventitious carbon does not penetrate further than 3 nm into the surface. In addition, beneath the first 2.7 nm, the amounts of carbon materials in the pillar area decreased by a factor of 10. This suggests that the adventitious organics are not absorbed by the bulk of the fused silica layer and that the lithographic, etching, and cleaning processes do not introduce subsurface contaminants into the pillar area (covered by the BARC and photoresists during RIE). Meanwhile, analysis of the trench area indicates carbon penetration of at least 80 nm below the surface. These results suggest that the RIE processing promotes carbon penetration into the bulk of the silica layer. Such substantial subsurface contamination during etching may cause spreading of the contaminants into the walls of the actual grating structures where the electric-field enhancement is highest. This contamination process, in turn, can affect the damage performance of MLD gratings.
Fig. 5. (left) XPS surface C1s scans and (right) depth profile measurements of the grating pillars, trenches, and the controlled unexposed FS substrate. (a) Pillar region, (b) trench region, (c) native unmodified coating.
We have also confirmed that this etch-induced subsurface contamination is not specific to a particular coating type or an etching method. Similar contamination profiles were observed in the trenches of 5-mm lines prepared from a much denser ion-beam–sputtered (IBS) silica coating, bulk fused silica, and in the trenches of the e-beam deposited silica formed by the reactive ion beam etching (RIBE) where the surface roughness of the trench area was much smaller than that observed with RIE. Figure 6 shows carbon contamination depth profiles of two samples prepared using RIBE instrumentation by two different commercial vendor. As evident from the XPS depth profile data, both samples show no subsurface contamination in the pillar regions, and ∼50 nm subsurface contamination in the etched trench regions. These results agree with the contamination depth profile data obtain on the in-house RIE sample [Fig. 5 (left)]. Such similarity in contamination profiles between different samples suggests that the fluoride-based ion etching in both RIE and RIBE implementations increases carbon concentration of the bulk fused silica layer and potentially contributes to the reduction of the laser damage performance of the etched structures.
Fig. 6. Depth profile measurements of carbon contamination in the 5 mm grating pillars and trenches prepared by the manufacture of the Reactive Ion Beam Etching instrumentation (left) and by the commercial grating manufacture (right)
As discussed above, we hypothesize that the difference in contamination profiles and/or surface roughness between the trenches and the pillars can affect the observed LIDT performance. The observed surface roughness, commonly referred to as “RIE grass,” is a consequence of photoresist redeposition during the dry etch [34,35]. Randomly redeposited photoresist particles can act as an etching mask in the trench area, leading to the grass-like morphology of the resulting silica interface. The RIE grass can potentially contribute to laser field intensification leading to ablation of the grass features that can be detected as laser damage. To further explore these issues, we have treated the RIE manufactured 5-mm grating-like structures with a diluted buffered oxide etch (BOE) solution. Diluted BOE can slowly dissolve silica providing a way to remove a top portion of the device interface containing subsurface contamination as well as smooth the overall roughness. As such, after application of the SCP cleaning, the substrates were soaked in 1:2800 BOE-water solution for different time interval ranging from 5 to 45 min. The etch rate of this BOE concentration was measured at 3.2 nm/min. Figure 7 shows that both 10-ps and 600-fs LIDT values increase to the levels of native FS coating only after 45 minutes of BOE exposure (removal of ∼140 nm of FS coating), while the surface roughness of the trench areas decreases significantly just after 5 minutes (removal of ∼15 nm of FS coating). Considering that the sub-surface of the trench areas is contaminated by more than 50 nm, these results suggest that the sub-surface contamination is primarily responsible for the LIDT reduction at 600-fs laser exposures. Although this contamination can be dissolved using longer BOE exposures, such long exposures will lead to a significant distortion of the wall profile of the grating features. Therefore, short exposure to diluted BOE can be beneficial as it can remove the most contaminated upper layer of the trenches, thus improving the LIDT.
We have also compared the SCP cleaning method with the modified advanced cleaning protocol [see depiction in Fig. 8(a)] developed to remove contaminants induced in manufacturing of commercial MLD pulse-compressor gratings. The mACP method almost completely removes fluorocarbon contaminants from the pillar and trenches [C1s and F1s XPS measurements, Fig. 8(b)], but it still leaves a relatively rough surface in the trench areas even after the diluted BOE etch [Fig. 8(c)]. The LIDT values of the mACP samples increased for both the 10-ps and 600 fs measurements but were within 10% of the LIDT’s of the native untreated coating [Fig. 8(d)]. It must be noted that the mACP method was used (but not described) in the work reported in Ref. [22]
Fig. 8. (a) The process of mACP, (b) XPS analysis of the carbon and fluorine to silicon ratio of the cleaned surface; (c) tilted SEM images of the pillar and trench regions, and (d) LIDT values of the cleaned surface with pristine coating as reference.
4. Discussion
The results discussed above suggest that the reactive ion etch step is responsible for organic contamination that is embedded below the surface. This was observed in samples fabricated using both, the RIE and the RIBE methods using in house facilities and commercial vendors. Removal of this type of contamination requires removal of the entire contaminated layer or at least the top section of this layer. This approach was tested using BOE etching which, after sufficient material removal, led to improvement of the LIDT. Complete restoration of the LIDT was also observed with longer exposure to BOE but this has detrimental effects on the quality of the pillar walls but also the overall integrity of the grating structure.
A further refinement of the post-etching cleaning protocols was examined, but the results show incremental improvement of the grating laser damage performance. The results demonstrated that these improved cleaning protocols (SCP and mACP) achieve nearly ideal cleaning of both the trench and the pillar surface areas of the grating-like structures. The subsurface contamination embedded into the trenches during the RIE, however, cannot be removed by cleaning and must be dissolved chemically. This is of particular concern as the deep contamination is likely also penetrating the near-wall regions, which are also exposed to the etching effect. Given that actual compression gratings have very thin pillars, it is likely that at least a large portion of the pillar region along the pillar walls is also penetrated with organic contaminants. As the peak EFI in pulse compression gratings is located in these areas (where also damage is initiated at fluences near the LIDT of the gratings), interaction with the organic contaminants will arguably cause a lowering of the damage threshold.
This work demonstrates that the contamination of gratings with organic materials has at least two distinct types. First, is the well-recognized contamination by byproducts that are superficial on the surface of the etched region, believed to predominantly reside attached on the pillar walls. The second type, which is revealed for the first time through this work, is related to contamination below the surface that may be due to the interaction of organics molecules (energy or momentum transfer) by the energized ions involved in the etching process. This contamination phenomenon was observed using RIE and RIBE etching and using e-beam, IBS and plain bulk silica materials. Consequently, the manufacturing protocols based on the standard fluoride-based reactive ion etch can inherently contribute to the reduction of the grating laser damage performance necessitating the development of new etching recipes, chemistries and/or methods. In the effort by the scientific community to develop the next generation of damage-resistant compression gratings, optimizing the grating design to reduce the EFI inside the pillar materials is a well-recognized approach. Developing materials exhibiting higher LIDT that can serve as the top layer is another approach, especially if this is associated with materials that provide higher refractive index compared to the commonly used silica. Such layers can be of mixed materials (such as hafnia and silica) that offer increased refractive index but maintain the LIDT. However, in light of this work, the potential inherent limitation of reactive ion etching methods causing deep organic contamination below the surface maybe an additional problem that must be considered. Our future work will be in part dedicated to developing etching methods that reduce the surface roughness of the trenches, improve the quality of the pillars walls and reduces subsurface contaminant concentrations.
Funding
National Nuclear Security Administration (DE-NA0003856).
Acknowledgment
This material is based upon work supported by the Department of Energy National Nuclear Security Administration under Award Number DE-NA0003856, the University of Rochester, and the New York State Energy Research and Development Authority. This report was prepared as an account of work sponsored by an agency of the U.S. Government. Neither the U.S. Government nor any agency thereof, nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by the U.S. Government or any agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the U.S. Government or any agency thereof.
Disclosures
The authors declare no conflicts of interest
Data availability
Raw XPS data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon request.
References
1. D. Strickland and G. Mourou, “Compression of amplified chirped optical pulses,” Opt. Commun. 56(3), 219–221 (1985). [CrossRef]
2. H. Lubatschowski, G. Maatz, A. Heisterkamp, U. Hetzel, W. Drommer, H. Welling, and W. Ertmer, “Application of ultrashort laser pulses for intrastromal refractive surgery,” Graefe's Arch. Clin. Exp. Ophthalmol. 238(1), 33–39 (2000). [CrossRef]
3. F. Dausinger, H. Hugel, and V. Konov, “Micro-machining with ultrashort laser pulses: From basic understanding to technical applications,” International Conference on Advanced Laser Technologies, Adelboden, Switzerland, (2002), pp. 106–115.
4. V. Kohli, A. Y. Elezzabi, and J. P. Acker, “Cell nanosurgery using ultrashort (femtosecond) laser pulses: Applications to membrane surgery and cell isolation,” Lasers Surg. Med. 37(3), 227–230 (2005). [CrossRef]
5. F. Calegari, D. Ayuso, A. Trabattoni, L. Belshaw, S. De Camillis, S. Anumula, F. Frassetto, L. Poletto, A. Palacios, P. Decleva, J. B. Greenwood, F. Martin, and M. Nisoli, “Ultrafast electron dynamics in phenylalanine initiated by attosecond pulses,” Science 346(6207), 336–339 (2014). [CrossRef]
6. K. Xu, Y. Chen, T. A. Okhai, and L. W. Snyman, “Micro optical sensors based on avalanching silicon light-emitting devices monolithically integrated on chips,” Opt. Mater. Express 9(10), 3985–3997 (2019). [CrossRef]
7. C. A. F. Marques, L. Bilro, L. Kahn, R. A. Oliveira, D. J. Webb, and R. N. Nogueira, “Acousto-Optic Effect in Microstructured Polymer Fiber Bragg Gratings: Simulation and Experimental Overview,” J. Lightwave Technol. 31(10), 1551–1558 (2013). [CrossRef]
8. C. N. Danson, C. Haefner, J. Bromage, T. Butcher, J. C. F. Chanteloup, E. A. Chowdhury, A. Galvanauskas, L. A. Gizzi, J. Hein, D. I. Hillier, N. W. Hopps, Y. Kato, E. A. Khazanov, R. Kodama, G. Korn, R. X. Li, Y. T. Li, J. Limpert, J. G. Ma, C. H. Nam, D. Neely, D. Papadopoulos, R. R. Penman, L. J. Qian, J. J. Rocca, A. A. Shaykin, C. W. Siders, C. Spindloe, S. Szatmari, R. Trines, J. Q. Zhu, P. Zhu, and J. D. Zuegel, “Petawatt and exawatt class lasers worldwide,” High Power Laser Sci. Eng. 7, e54 (2019). [CrossRef]
9. B. W. Shore, M. D. Perry, J. A. Britten, R. D. Boyd, M. D. Feit, H. T. Nguyen, R. Chow, G. E. Loomis, and L. Li, “Design of high-efficiency dielectric reflection gratings,” J. Opt. Soc. Am. A 14(5), 1124–1136 (1997). [CrossRef]
10. J. Britten, M. Perry, B. Shore, R. Boyd, G. Loomis, and R. Chow, “High-efficiency dielectric multilayer gratings optimized for manufacturability and laser damage threshold,” Proc. SPIE 2714, (1996).
11. B. Stuart, M. Perry, R. Boyd, J. Britten, B. Shore, M. Feit, and A. Rubenchik, “Development of high damage threshold optics for petawatt-class short-pulse lasers,” Proc. SPIE 2377, (1995).
12. F. Canova, O. Uteza, J.-P. Chambaret, M. Flury, S. Tonchev, R. Fechner, and O. Parriaux, “High-efficiency, broad band, high-damage threshold high-index gratings for femtosecond pulse compression,” Opt. Express 15(23), 15324–15334 (2007). [CrossRef]
13. B. Ashe, C. Giacofei, G. Myhre, and A. W. Schmid, “Optimizing a cleaning process for multilayer dielectric (MLD) diffraction gratings,” Proc. SPIE 6720, 67200N (2007). [CrossRef]
14. B. Ashe, K. L. Marshall, C. Giacofei, A. L. Rigatti, T. J. Kessler, A. W. Schmid, J. B. Oliver, J. Keck, and A. Kozlov, “Evaluation of cleaning methods for multilayer diffraction gratings,” Proc. SPIE 6403, 64030O (2006). [CrossRef]
15. K. Wei-Jin, S. Zi-Cai, S. Jian, S. Jian-Da, and F. Zheng-Xiu, “Investigation of Laser-Induced Damage on Multi-Layer Dielectric Gratings,” Chin. Phys. Lett. 22(7), 1757–1760 (2005). [CrossRef]
16. C. Scurlock, “A phenomenological study of the effect of trace contaminants on lifetime reduction and laser-induced damage for optics,” Proc. SPIE 5647, (2005).
17. D. A. Alessi, C. W. Carr, R. P. Hackel, R. A. Negres, K. Stanion, J. E. Fair, D. A. Cross, J. Nissen, R. Luthi, G. Guss, J. A. Britten, W. H. Gourdin, and C. Haefner, “Picosecond laser damage performance assessment of multilayer dielectric gratings in vacuum,” Opt. Express 23(12), 15532–15544 (2015). [CrossRef]
18. A. B. Jerald, A. M. William, M. K. Aleksey, and P. J. B. Christopher, “Multilayer dielectric gratings for petawatt-class laser systems,” Proc. SPIE 5273, (2004).
19. J. Igor, G. B. Curtis, C. S. Brent, A. M. William, D. N. Norman, F. W. Benoit, A. B. Jerald, P. Deanna Marie, and P. J. B. Christopher, “Precision damage tests of multilayer dielectric gratings for high-energy petawatt lasers,” Proc. SPIE 5647, (2005).
20. J. Britten, W. Molander, A. Komashko, and C. Barty, “Multilayer dielectric gratings for petawatt-class laser systems,” Proc. SPIE 5273, (2004). [CrossRef]
21. J. Neauport, E. Lavastre, G. Razé, G. Dupuy, N. Bonod, M. Balas, G. de Villele, J. Flamand, S. Kaladgew, and F. Desserouer, “Effect of electric field on laser induced damage threshold of multilayer dielectric gratings,” Opt. Express 15(19), 12508–12522 (2007). [CrossRef]
22. B. N. Hoffman, A. A. Kozlov, N. Liu, H. Huang, J. B. Oliver, A. L. Rigatti, T. J. Kessler, A. A. Shestopalov, and S. G. Demos, “Mechanisms of picosecond laser-induced damage in common multilayer dielectric gratings,” Opt. Express 28(17), 24928–24936 (2020). [CrossRef]
23. N. Bonod and J. Neauport, “Diffraction gratings: from principles to applications in high-intensity lasers,” Adv. Opt. Photonics 8(1), 156–199 (2016). [CrossRef]
24. N. Bonod, P. Brianceau, and J. Neauport, “Full-silica metamaterial wave plate for high-intensity UV lasers,” Optica 8(11), 1372–1379 (2021). [CrossRef]
25. H. P. Howard, A. F. Aiello, J. G. Dressler, N. R. Edwards, T. J. Kessler, A. A. Kozlov, I. R. T. Manwaring, K. L. Marshall, J. B. Oliver, S. Papernov, A. L. Rigatti, A. N. Roux, A. W. Schmid, N. P. Slaney, C. C. Smith, B. N. Taylor, and S. D. Jacobs, “Improving the performance of high-laser-damage-threshold, multilayer dielectric pulse-compression gratings through low-temperature chemical cleaning,” Appl. Opt. 52(8), 1682–1692 (2013). [CrossRef]
26. A. Kozlov, S. Papernov, J. Oliver, A. Rigatti, B. Hoffman, B. Charles, and C. Smith, “Study of the picosecond laser damage in HfO2 /SiO2 -based thin-film coatings in vacuum,” (2017).
27. C. Steinbrüchel, “Mechanism of Dry Etching of Silicon Dioxide,” J. Electrochem. Soc. 132(1), 180–186 (1985). [CrossRef]
28. P. W. Leech, “Reactive ion etching of quartz and silica-based glasses in CF4/CHF3 plasmas,” Vacuum 55(3-4), 191–196 (1999). [CrossRef]
29. H. Xiong, Y. Shi, and Z. Wang, “Reactive ion etching of Ge-Sb-Se ternary chalcogenide glass films in fluorine plasma,” Microelectron. Eng. 225, 111259 (2020). [CrossRef]
30. J. Zajadacz, P. Lorenz, F. Frost, R. Fechner, C. Steinberg, H.-C. Scheer, and K. Zimmer, “Reactive ion beam etching of fused silica using vertical lamellar patterns of PS-b-PMMA diblock copolymer masks,” Microelectron. Eng. 141, 289–293 (2015). [CrossRef]
31. D. M. Brown, B. A. Heath, T. Coutumas, and G. R. Thompson, “Reactive ion beam etching of SiO2 and polycrystalline silicon,” Appl. Phys. Lett. 37(2), 159–161 (1980). [CrossRef]
32. D. Zhang and M. J. Kushner, “Investigations of surface reactions during C2F6 plasma etching of SiO2 with equipment and feature scale models,” J. Vac. Sci. Technol., A 19(2), 524–538 (2001). [CrossRef]
33. A. A. Kozlov, J. C. Lambropoulos, J. B. Oliver, B. N. Hoffman, and S. G. Demos, “Mechanisms of picosecond laser-induced damage in common multilayer dielectric coatings,” Sci. Rep. 9(1), 607 (2019). [CrossRef]
34. K. Lilienthal, M. Stubenrauch, M. Fischer, and A. Schober, “Fused silica ‘glass grass’: fabrication and utilization,” J. Micromech. Microeng. 20(2), 025017 (2010). [CrossRef]
35. H. Jansen, M. de Boer, R. Legtenberg, and M. Elwenspoek, “The black silicon method: a universal method for determining the parameter setting of a fluorine-based reactive ion etcher in deep silicon trench etching with profile control,” J. Micromech. Microeng. 5(2), 115–120 (1995). [CrossRef]
