Microfabrication of Riblets for Drag Reduction: A Novel UV Approach Utilizing Photolithographic Methods

By H. Bilinsky, CEO, MicroTau Pty Ltd.

I. Introduction

Although drag-reducing riblet microstructures have been researched for over two decades and reliably demonstrate up to 10 percent reduction in skin friction¹ attempts to apply them to aircraft as a fuel consumption reduction measure have not yet been fully successful. An economically viable implementation has so far been prevented by cost of application, maintenance and a lack of durability.²

The current paper proposes a novel method of microfabrication to overcome these issues. Drawing on photolithographic methods currently used in computer chip fabrication, the method directly “prints” riblets or other repeating microstructures onto an external aircraft surface. As a continuous and contactless method, the process is scalable to large areas, allowing reductions in time and cost of application.

II. Background

The contactless microfabrication method draws on several fields of knowledge including functional microstructures, photolithographic computer chip fabrication technology, UV-curable coatings and classical optics.³

Microstructures
Nomenclature
h = riblet height
s = riblet spacing
t = riblet thickness
y = dimensional wall units
y+ = non-dimensional wall units

Engineered microstructures such as riblet and lotus leaf structures hold great promise to reduce aircraft skin friction and provide self-cleaning properties to maintain hydraulically smooth surfaces, respectively. These microstructures have yet to be successfully implemented on aircraft due to problems with cost, maintenance, durability and application.

Riblets are small surface protrusions aligned with the direction of flow and spacing in the order of 50 to 150 μm.⁴ These microstructures have been studied for over two decades and have reliably demonstrated turbulent flow skin friction reduction of up to 10 percent.¹ Blade, sawtooth and scalloped are the most studied riblet geometries (see Figure 1). Their required dimensions are given in terms of non-dimensional wall units (y+). Optimal dimensions are spacing of s = 15±2y+; height h = 0.5-1s and thickness t = 0.02-0.04s.1 Dimensional wall units y change according to the fluid mechanics of the environment. For the wind tunnel testing, parameters used in this project optimal riblet spacing is s = 117μm. Final riblets for an aircraft at speed Mach 0.8 and altitude 30kft will have spacing s = ~50μm and increase in size going aft from the leading edge.⁵

Appliqué riblet films developed by the 3M Company demonstrated a two percent drag reduction in flight tests however failed to prove economically viable due to significant application time and cost requiring specialized workers;² issues with durability and maintaining drag reduction; and an inability to cover more than 70 percent of the aircraft without adversely impacting flight characteristics.⁶ The current microfabrication method aims to provide greater coverage, a more durable material and lower cost of application as well as riblet size optimization for drag reduction.

The proposed method may also be used to fabricate lotus leaf structures that impart self-cleaning super-hydrophobic properties.⁷ These consist of regular 2D “hierarchical” microstructures that repeat their structure on different micro- or nano-scales.⁸ Lotus leaf structures have been suggested for anti-fouling and anti-icing applications to maintain hydraulically smooth aircraft surfaces.⁹

B. Photolithography

The microfabrication method draws on a mature and commercially successful photolithographic technology from computer chip manufacturing that has achieved structural features orders of magnitude smaller than required for riblets. Photolithography utilizes a class of photocurable materials known photopolymers, or “negative photoresists” that consist of monomers, oligomers and a photoinitiator. When exposed to UV, the photoinitiator catalyzes a polymerization reaction, “curing” the monomers and oligomers into a strong network polymer.

Microstructures can thus be made by applying a thin layer of photopolymer to a substrate and exposing it to UV in the desired pattern. This is typically achieved by passing UV through a photomask and then removing the unexposed photopolymer.

The Fraunhofer Institute has attempted to repurpose this photolithographic technology to print riblets directly onto aircraft with a continuous contact process.⁴ This involves a tool consisting of a flexible UV-transparent mask with the inverse of the desired riblet structures stenciled into it. This mask contains a UV source and rolls over the photopolymer-coated aircraft surface. The riblet structures are formed out of the photopolymer with the stencil pattern and cured whilst in contact with the mask. This method places requirements on the photopolymer material used¹⁰ due to mask contact at the point of exposure and can only apply structures with the one geometry as defined by the mask. This method has not been commercially implemented to date.

By using a contactless optical system, the current method avoids issues with photomask contact and eliminates associated restrictions on the photopolymer material used.

C. UV-curable Coatings

Without the restrictions associated with a contact exposure process, an extensive range of photocurable materials may be selected from to meet aircraft coating requirements. UV-curable coatings based on the same photopolymer combination of monomers, oligomers and photoinitiators are commercially used as a quick-cure and low-VOC alternative to traditional dry-to-cure coatings in wood flooring¹¹ and automotive applications. Success in the automotive industry led to development of the coatings for commercial¹² and military3 aircraft. The current project investigated whether such coatings could be used for microfabrication using the same selective exposing and developing method used in photolithographic computer chip manufacturing discussed above.

An ideal candidate material for this method is a UV-curable coating developed by the AFRL aimed to reduce the 72-hour minimum “dry to fly” time to increase production throughput and reduce the environmental burden of coating aircraft.¹³ The Air Force UV-Curable Coatings Program developed a UV-curable coating for MIL-PRF-85285 specifications that has been successfully flown for 600 hours over 14 months on a C-130H.3 By applying said UV-curable coating to an aircraft and then patterning with the proposed optical system, riblet microstructures may be fabricated from a material that already meets strict military aircraft coating specifications.

D. Oxygen Inhibition (Bottom-up Curing)

Oxygen inhibition has been a longstanding problem for the photocurable coatings industry. When the photoinitiator in the coating is exposed to curing, it produces free radicals that catalyze a polymerization reaction, curing the photopolymer. However, as oxygen from the atmosphere diffuses into the coating it consumes these free radicals before they can begin the photopolymerization reaction, thus inhibiting curing¹⁴, which can cause uncured, liquid or tacky coating on the surface rather than a complete cure. The closer to the surface of the coating, the higher the oxygen concentration and the harder it is to cure. As a result UV-curable coatings tend to cure from the bottom up.

In order to overcome this, and to cure the photopolymer despite the presence of oxygen within it, higher exposure to curing (either by increasing irradiance or exposure time) is used. This is due to the fact that with more UV, more free radicals are produced and eventually enough free radicals are present to successfully photopolymerize the coating despite oxygen inhibition.14 As the oxygen inhibition effect is more pronounced nearer to the surface (where oxygen concentration higher), greater irradiance is required to cure to that height.

The current method proposes taking advantage of this phenomenon to be able to control the heights and profiles of riblet microstructures in a single exposure. By designing an irradiance profile to expose and cure the coating, different regions can be made to cure to different heights. Therefore, a two-dimensional exposure pattern can be used to fabricate microstructures in three dimensions by varying the irradiance across said pattern.

Previous research suggests this bottom-up curing effect should reflect Fick’s law of diffusion¹⁵ as the inhibition is dependent on the concentration of oxygen in the coating at any given point. This is exhibited in Figure 2 which shows the increasing exposure time required to cure the last (topmost) coating. As the current method uses photocurable coatings that are designed to have complete through-cure, one does not expect an asymptotic approach to complete surface cure, but still a similar exposure/height profile with increasing exposure or irradiance required to cure upper regions is expected. The theoretical relationship between exposure or irradiance and cured height is shown in Figure 3, with a similar shape as the prior data.

E. Optical System

The optical system to produce the exposure irradiance profile is designed to be robust and for exposure at a distance to transition photolithographic technology from a cleanroom environment of computer chip manufacturing to a hangar for application to aircraft surfaces. The proposed method aims to achieve this by use of a diffraction grating. A diffraction grating is an opaque material with small apertures at regular intervals that allow light to pass through.

By illuminating the grating with a single wavelength source, an interference pattern is produced of bright and dark fringes. Grating parameters can be designed such that the irradiance peaks match the spacing’s of desired riblet microstructures and used to expose the photocurable coating even at large distances. Features of the pattern scale linearly with the distance of the diffraction grating to the exposed surface, akin to moving a projector closer or farther from a screen to shrink or grow the image (see Hecht16 for a detailed description of diffraction grating optics). The profile is then drawn out across the coated surface in a continuous exposure process that is suited to scaling for exposing large areas quickly.

III. A Novel Microfabrication Method

Fabrication consists of three key steps: (1) application of the photocurable (PC) coating; (2) exposure in the desired pattern; and (3) developing (i.e. removing unexposed material).

A. Apply Photocurable Coating

The photocurable coating is applied to the external aircraft surface using existing coating methods such as spray painting. The coating must be applied to the thickness of desired riblet height or greater (50 to 150 μm). The present proof of concept testing used a drawdown method for experimental purposes.

B. Expose Photocurable Coating

An optical system is kept at a predetermined distance from, and parallel to, the coated aircraft surface during exposure. The system projects a one- or two-dimensional interference pattern that is traversed perpendicular to said pattern to draw out the desired two-dimensional riblets. The system travels across the surface of the aircraft drawing out the riblet pattern in a continuous exposure (Figure 4).

C. Develop

The unexposed photocurable coating is then removed using an appropriate solvent or “developer.” The developer depends on the material used, e.g. mineral alcohol for unexposed UV-curable coatings.¹⁷ This may be assisted with some physical removal processes, e.g. spraying of the developer, compressed air or rubbing/wiping.

IV. Coating Investigation

A. Initial Investigation

The goal of the initial investigation was to test the patterning process of selective exposing of a photocurable coating and then removing the unexposed with a developing solvent. A commercial-off-the-shelf (COTS) UV-curable automotive primer-surfacer18 was applied to aluminum samples. Exposing with a 405 nm wavelength laser diode through a patterned photomask, the sample was then developed using mineral spirits and some light rubbing. The patterning process was successfully demonstrated (Figure 5). Methyl ethyl ketone (MEK) was used for all following developing as a more powerful solvent that required no physical rubbing or force.

B. Military Aircraft Topcoat and Reformulation

The photocurable military aircraft top coat discussed previously was then formulated using newly developed oligomers, and tested. This was a gray pigmented topcoat designed to have the same properties as the current C-130H topcoat used by the USAF.³ Applied at the required thickness for wind tunnel riblet dimensions (i.e. greater than their height) we failed to achieve adhesion with the 405 nm laser diode source. Adjustments to the formulation were made as necessary to achieve through-cure and adhesion at the required coating thickness. The resulting formulation achieved through-cure at the required coating thickness.

C. Adhesion

In order for the coating to adhere to the substrate, the bottom-most layer of coating that is in contact with the substrate must cure. The UV used to cure the coating is coming from the top (the exposed surface) and therefore has to travel through the entire uncured coating thickness first. This will cause attenuation of the irradiance that is dependent on the composition and thickness of the coating. Therefore, for a given coating and coating thickness, a minimum critical exposure is required for coating adhesion. Failing to achieve this critical exposure may result in surface curing that lifts off during developing of the sample.

D. Bottom-up Curing Investigation

The bottom-up curing effect was investigated to determine how exposure, irradiance and/or time could be utilized to control the final cure height. Experimental results supported the relationship hypothesized in Figure 3; however, time constraints and issues with the reliability of our coating process prevented determining the exact relationship between exposure parameters and cure height. Preliminary findings suggest this should depend on oxygen concentration in the atmosphere above, temperature and the coating formulation.

V. Optics Investigation

The goal of the optics investigation was to determine a method to reliably produce an irradiance profile that could be used to expose a photocurable coating in a riblet microstructure pattern. This needed to be at a distance (non-contact), robust (able to handle ambient room vibrations and small variations in distance to coated surface) and create the desired pattern spacing (117 μm for wind tunnel samples).

A. Equipment

The optical system was mounted on a breadboard held at the vertical so as to project the irradiance profile down onto the coating. The curing source used was a 405 nm, 40 mW, ∅5.6 mm, B Pin code, Sanyo laser diode, driven by a LTC100-A Laser/Tec Driver Kit. Despite photocurable coatings typically described as “UV-curable” the near-UV visible regime was tested first for a number of reasons, including better penetration ability of longer wavelengths and versatility in the photomask material (soda-lime glass is transparent to visible light but absorbs UV). The output beam shape was adjusted using various combinations of mirrors, lenses and pinholes to tidy up the beam and adjust the focal point.

The photomask design contained a matrix of different diffraction grating designs with varying slit periods and widths. The mask was made of soda-lime glass (made possible with the visible 405 nm wavelength source) and a chrome coating that had been etched with desired diffraction grating designs. Some masks included grayscale designs to slowly vary the profile (i.e. the gradient from peak to minimum irradiance). A monochrome USB CMOS Camera with pixel size 5.2 μm was used to measure the profile produced. Neutral density filters of varying optical densities were used to reduce the intensity of the exposure pattern so that the pattern could be viewed on the CMOS sensor.

B. Theory

The initial plan to produce the riblet irradiance profile using an optical component known as a diffraction grating. This produces bright and dark fringes of light due to interference of diffracted light passing through the grating in the Fraunhofer regime. After encountering issues with producing the diffraction pattern predicted by theory, it was determined this was a result of a failure to meet the optical requirement of incident normal plane waves¹⁶ hitting the diffraction grating and that such a requirement was not feasible in a laboratory context.

An unexpected result was, however, observed in which the image of the incoming light beam was projected through the photomask (as in the Fresnel regime) and replicated at each diffraction peak as predicted by the Fraunhofer diffraction grating equations. This combination of the Fraunhofer and Fresnel regimes, as well as the symmetry of the photomask pattern in the direction of the diffracted patterns, provided a means to create the desired profiles for exposing the photcurable coating (Figure 6). Whilst it is beyond the scope of the current paper to delve into the optics theory behind this and further research is required to conclusively determine the mechanism of action – we were however able to demonstrate reliable profiles with desired spacing (117 μm); sharpness (1-2 pixel or ~5-10 μm peak width); projected from distances over a range of 100 to 600 mm and demonstrated focus tolerance of 1mm to 5mm. Note that being “in focus” was determined with the CMOS – i.e. that the image remained unchanged and therefore the pattern remained the same within the tolerance of the camera’s pixel pitch (5.2 μm).

C. Results

Sharp irradiance profiles were achieved as measured on the CMOS that matched the riblet spacing of 117 μm. Peak irradiance was achieved down to the limit of the CMOS pixel size of 5.2 μm. The spacing of the peaks was able to be changed through a combination of focal adjustment and distance from mask-to-CMOS sensor to any desired spacing (within the CMOS resolution limit). Profiles were exposed onto the CMOS at distances ranging from 100 to 600 mm. This distance could be increased, as we were simply limited by the size of the optical breadboard holding the components. Depending on the optical setup, a tolerance of up to 5mm in that distance whilst maintaining the same irradiance pattern as observed on the CMOS camera (pixel size 5.2 μm). It appeared as though this could be improved upon by exposing from larger distances. As can be seen in Figure 7, different photomask designs were able to produce profiles with varying gradients from light to dark. This reflects the photomask designs and is only possible through the projection characteristic of the Fraunhofer-Fresnel phenomenon discussed previously. As the resulting pattern is essentially a projection of the photomask, arbitrarily shaped profiles can be produced and high control over microstructure profiles can theoretically be achieved.

VI. Experimental Setup

The current project aimed specifically at making riblet microstructures for testing in a Lockheed Martin Corporation (LMCO) wind tunnel facility. The goal was to produce riblets with spacing s = 117 um on aluminum panels of dimensions 24″x23″ and 0.012″ thickness. A pair of these panels were applied to LMCO’s NACA 0012 2D Airfoil (top and bottom) for wind tunnel testing.

A. Experimental Rig

The experimental rig consisted of the described optical system, a programmable XY table and a black corflute enclosure with yellow lighting to protect the coating from dust and ambient light.

B. Coating method

Aluminum panels were pre-coated off site at the University of Dayton Research Institute (UDRI) with a Deft 02-Y-40 primer currently used by the USAF on C-130 aircraft. A drawdown method using wire wound rod was used for convenience, as it enabled coating application within the lab and for control over coating thickness. This method, however, caused multiple issues with coating coverage and evenness over the whole panel. Future investigations should use an alternative coating method that can be transferred to aircraft application, e.g. spray painting.

C. XY Table and Exposure Program

The continuous exposure method to cure the photocurable coating was achieved with a custom-built programmable XY translational table that allowed for automated exposures of large areas. The described profiles were traversed across the coated sample to “draw out” the extended riblet microstructures.

D. Developing

Methyl ethyl ketone (MEK) was used as a developing agent. After exposure the sample panel was placed in a tray or tub, covered in MEK and gently agitated for ~1 minute, then removed from MEK to allow drying. Post-curing in sunshine or under UV lamp was conducted to ensure completely cured structures.

VII. Results

The following results are of some of the microstructures fabricated with the above experimental setup. Each sample was produced with a different profile by the optical system and measured by the CMOS camera. Samples were printed on aluminum pre-coated with a C-130 primer.

A. Stationary Grayscale Exposures

The first attempt was made using a grayscale mask and a stationary exposure. The grayscale mask design (see top left Figure 8) consisted of bars of increasing thickness on 5.5 μm period. The smallest bar is 0.5 μm, increasing 0.5 μm to the largest at 5 μm. This was used to produce a profile with peak spacing of ~130 μm.

The resulting riblet profiles were far too short at ~2 μm in height, however they exhibited a fine structure pattern that matched the fine structured bars of the grayscale mask (see bottom Figure 8). This was an unexpected result as it appears that feature sizes down to ~1-2 μm in size are achievable with this method. The structures are also “hierarchical” in that they consist of ~1 μm structures superimposed on ~100 μm structures. This may be useful in self-cleaning superhydrophobic surfaces such as lotus leaf microstructures or perhaps a combination of riblets and lotus leaf structures to impart both drag-reducing and self-cleaning properties.

B. Stationary Slit Exposures

Given the grayscale mask riblets were far too flat, the “sharpest” irradiance profile was attempted using a slits-only design of regularly spaced apertures of 4 μm (i.e. without any slow-changing grayscale). A stationary exposure was used to produce a profile at the scale required for riblets on an aircraft (i.e. spacing s = ~50 μm). Optical microscopy revealed that regular riblet structures were produced of said spacing as well as in interesting effect at the end of the riblet segments. Where the profile faded to dark, the fabricated riblets sloped down to the substrate (Figure 9). This suggests the bottom-up curing effect is replicating the profile accurately and that adjustment of irradiance can be used to manipulate riblet heights for drag reduction optimization or variable height riblets.

C. Continuous Exposures

The process was then adjusted and aligned for a continuous exposure at the dimensions required for the wind tunnel riblets (i.e. s = 117 μm).5 Testing was completed on small 3″ size panels to allow for metrology prior to fabricating full wind tunnel panels. Exposure speeds were successfully run at 2.5 to 700 mm/min with successful adhesion.

Profilometry of riblets produced indicated a very sharp (~1-2 μm) peak radius and a reliable sawtooth profile of ~55 μm height (Figure 10).

Scanning electron microscopy (SEM) of the sample was also conducted for high resolution images of the riblet profiles produced. These images revealed far smoother and more even structures than anticipated (Figure 11). This may be a fortunate consequence of the continuous fabrication process averaging out the exposure irradiance profile any given point in the coating.

The SEM images also suggest far sharper and narrower (~9 μm) riblets of a more blade-like geometry than the profilometry results indicate. It is suspected the profilometry data are in error as they are unreliable for horizontal measurements, use a 5 μm radius stylus tip and may have systematic error in scanning riblet peaks and troughs.

D. Wind Tunnel Testing

After optimization of coating formulation, thickness, exposure irradiance profile and exposure speed full 24″ x 23″ wind tunnel samples were produced. Wind tunnel testing was conducted by LMCO and a viscous drag reduction of six percent was measured.

E. Low-gloss Riblets

It was also observed that the riblet samples produced had a dull finish when compared to non-patterned cured coating, which had a high gloss finish. This may prove useful for achieving a low-gloss finish without the use of downglossing agents or pigmentation.

VIII. Conclusion

Proof of concept of a continuous and contactless microfabrication process was demonstrated with better than anticipated results. Riblet microstructures were fabricated from a durable military aircraft topcoat and achieved a six percent viscous drag reduction. This is a very encouraging first step toward a successful implementation of drag-reducing riblets microstructures on aircraft and overcoming past issues with cost of application, maintenance and durability. The six percent viscous drag reduction from wind tunnel testing has the potential to translate to a significant reduction in USAF’s $8B+ annual expenditure on aircraft fuel.

All photocurable coatings tested were able to be microfabricated, suggesting the process may work with a large range of coatings that could be selected to provide desired characteristics. High print speeds of up to 700 mm/min were achieved and should be able to be improved upon with higher irradiance exposures. The number of riblets printing at once is also easily scalable through either magnification of a scaled photomask design, or by running a horizontal array of optical systems in parallel. There are thus multiple attractive avenues for greatly reducing time of application. Exposures were conducted at distances of 100 to 700 mm from the coated surface with a robust tolerance of up to 5 mm variation in that distance.

Feature sizes down to the order of single microns were reliably produced, exceeding expectations by an order of magnitude. Bottom-up curing was demonstrated, allowing for “single exposure 3D printing” or 3D manipulation of microstructure designs in a single step. Manipulating heights and profiles of riblet microstructures was demonstrated in this way. This also enables live-manipulation of riblet heights, which may be used to optimize riblet parameters across the aircraft surface or in fabricating variable height riblets for further drag reduction. Preliminary findings suggest the process can be used to fabricate hierarchical structures in single or multiple exposures, holding promise as a possible method of fabricating self-cleaning superhydrophobic microstructures to improve maintenance and durability.

Future investigations toward the ESMC program goal of practical riblet application to the USAF legacy transport aircraft fleet should focus on building a robust, mountable optical unit with tolerances required for hangar environment; coating with existing AFRL spray painting methods and qualification testing of riblet structures at the AFRL Coatings, Corrosion & Erosion Laboratory (CCEL). Adjustments to microstructure design and formulation of the photocurable military aircraft topcoat should be made as necessary.

Acknowledgments

This work is supported by the Operational Energy Capability Improvement Fund (OECIF) from the Office of the Assistant Secretary of Defense for Operational Energy Plans and Programs, ASD (OEPP). The MicroTau Pty Ltd authors would like to acknowledge the Ohio Aerospace Institute, through which they were funded. The author would like to thank Mr. Nathan Apps and Mr. Ralf Wilson for their assistance in project management, mechanical design and technical support. This work was done in part at the OptoFab Node of the Australian National Fabrication Facility with the assistance of Ethel Ilagan, process engineer, and David O’Connor, foundry manager. The author would also like to thank the following individuals for their contributions to this project: Dr. Joe Khachan and Dr. J. Scott Brownless for their consultation on matters of experimental design and optical theory respectively; Michael J. Dvorchak, technical director of Dvorchak Enterprises, LLC for his insight into formulating UV curable coatings for this unique application; and Scott Smith, research chemist of R & D Coatings, Inc. for his formulation skills in being able to meet the parameters required.

References

1. Dean, B., and Bhushan, B., “Shark-skin surfaces for fluid-drag reduction in turbulent flow: a review,” Philosophical Transactions of the Royal Society of London A: Mathematical, Physical and Engineering Sciences, vol. 368, 2010, pp. 4775–4806.
2. Bushnell, D. M., “Aircraft Drag Reduction – a Review, Proceedings of the Institution of Mechanical Engineers,” Part G: Journal of Aerospace Engineering, Vol. 217, 2003, pp. 1-18
3. Williams, C.T., Dvorchak, M., and Gambino, C., “Development of UV-A Curable Coatings for Military Aircraft Topcoats,” Radtech Report [online journal], Spring 2011, URL: http://www.radtech.org/images/pdf_upload/development-of-uv-a-curable-coatings-for-military-aircraft-topcoats-spring2011.pdf [cited December 2015].
4. Malas, J., Jines, L., Ontko, N., Richey, K., and Manter, J., Introduction and Qualification of New and Improved Engineered Surfaces, Materials, and Coatings for Aircraft Drag Reduction, United Technologies Corporation, 2014.
5. Smith, B., “Wind Tunnel Validation Testing Concepts,” ESMC TIM Presentation, 20 October 2015, slide 6.
6. Lynch, F. and Klinge, M., “Some Practical Aspects of Viscous Drag Reduction Concepts,” SAE Technical Paper 912129, 1991.
7. Ensikat, H.J., Ditsche-Kuru, P., Neinhuis, C., and Barthlott, W., “Superhydrophobicity in perfection: the outstanding properties of the lotus leaf,” Beilstein J. Nanotechnol., 2011, 2, 152–161.
8. Vorobyev, A.Y., and Guo, C., “Multifunctional surfaces produced by femtosecond laser pulses,” Journal of Applied Physics, 117, 033103, 2015.
9. Malas, J., Jines, L., Richey, K., Manter, J., Ontko, N., Allport, C., and Folck, J., Engineered Surfaces, Materials and Coatings for Drag Reduction (Phase 0), United Technologies Corporation, 2014.
10. Stenzel, V., Kaune, M., and Da Silva Branco Cheta, M.R., Fraunhofer-Gesellschaft Zur Forderung Der Angewandten Forschung E.V., U.S Patent for a “Tool for generating microstructured surfaces,” Patent Number (US 7736570), filed 30 Sep 2004. URL: https://www.google.com.au/patents/US7736570
11. Dvorchak, M.J., “UV Curing of Pigmented High-Build Wood Coatings Based on Non-Air-Inhibited Unsaturated Polyesters,” Journal of Coatings Technology, 1995, URL: http://infohouse.p2ric.org/ref/25/24087.pdf
12. Baird, R.W., “Heat-Resistant UV-Curable Clearcoat for Aircraft Exteriors,” The Boeing Company, 2011, URL: https://goo.gl/spHP9Y
13. Naguy, T., and Straw, R., “Ultraviolet (UV)-Curable Coatings for Department of Defense (DoD) Applications AFRL,” AFRL, 2009, URL: https://goo.gl/YqZHvO
14. Arceneaux, J.A. “Mitigation of Oxygen Inhibition in UV LED, UVA, and Low Intensity UV Cure,” Allnex USA Inc., 2014, URL: https://goo.gl/hDSHNZ
15. Alvankarian, L., and Majlis, B.Y., “Exploiting the Oxygen Inhibitory Effect on UV Curing in Microfabrication: A Modified Lithography Technique,” PLoS ONE 10(3), URL: https://www.ncbi.nlm.nih.gov/pubmed/25747514
16. Hecht, E., “Optics,” 3rd ed, Addison Wesley, 1998, Chap 10.
17. Pfanstiehl, J., “The 2-minute Cure,” Radtech Report [online journal], November/December 2003, pp.46-50 URL: http://radtechintl.org/resources/Documents/2minutecurenovdec03.pdf
18. See product page URL: http://products.axaltacs.com/dcat/us/en/dr/product/A-3130S.html