UV Micromachining: Shorter Pulses or Shorter Wavelength?
PRECISION TECHNOLOGY
DPSS ultraviolet lasers are characterized by their compact platform and are now available with a choice of output wavelengths and pulse widths. |
For manufacturers of medical devices, particularly disposables, pulsed all-solid-state ultraviolet lasers are the established tools of choice in many micromachining applications. The high reliability and compact size of these diode-pumped solid-state (DPSS) lasers enable systems tool builders and end-users alike to view the focused laser beam as just another high-precision tool piece, albeit with smaller dimensions than any other physical tool type. Current applications—which involve machining plastic, metal, ceramic, and glass substrates—range from drilling holes in catheters to machining stents and other tiny implants.
Until recently, the majority of these applications have been well serviced by widely available lasers with an output wavelength of 355 nm and a typical pulse duration of 40–60 nanoseconds or more. The need for increased miniaturization and the concomitant requirement for even better edge quality are now pushing the limits for these lasers. Superior resolution and edge quality can be achieved by switching to lasers with shorter wavelengths (266 nm), but this can increase processing costs in several ways. In many materials, similar results can be obtained simply by using a 355-nm laser with shorter (~20 nanoseconds) output pulses. Manufacturers need to understand the relationship between the process and the materials involved. This article discusses materials for which 266-nm lasers may be a better option, notwithstanding the higher-per-unit machining costs.
Micromachining: Pulsed UV Lasers
The ultraviolet (UV) region of the spectrum encompasses wavelengths shorter than 400 nm. UV lasers offer two distinct advantages over longer-wavelength laser sources. The first is superior spatial resolution, i.e., tool size. The minimum spot size of a focused laser beam increases with wavelength and poor beam quality. Conversely, minimum spot size decreases with smaller wavelength and better beam quality. This spatial resolution is due to an inescapable optical effect called diffraction. Only perfect beams can be focused down to the theoretical limit. As a result, only UV lasers with high beam quality are capable of machining at the micron and submicron scale.
The second advantage of UV lasers is cold processing. Infrared and visible lasers machine materials solely by acting as an intense, highly localized spot of heat, essentially removing material by boiling it off. However, heat spreads, and this leads to unwanted peripheral thermal effects such as charring, melting, cracking, and the deposition of recast material. This is referred to as the heat-affected zone (HAZ).
In many plastics and some other nonmetals, UV light directly breaks molecular bonds. The process, called photoablation, is a relatively cold process that produces a very small HAZ, if at all. Photoablation enables the production of sharper, cleaner edges and the creation of tiny features that would be melted away by thermal processing.
Figure 1. (click to enlarge) Pulsed laser output delivers high peak power compared with the average laser power. |
In laser micromachining, it is important to use a pulsed laser rather than a continuous-wave laser, because heat dissipation takes time. If the pulse width is shorter than the time it takes for the heat to dissipate into the surrounding material, most of the pulse energy causes photoablation, avoiding destructive cumulative heating effects. In addition, pulsed output maximizes the peak power for a given average power (see Figure 1).
The laser's processing power also depends on the peak power. Ideally, pulsed output means that most of the power is delivered above the processing threshold for the target material, with just a small amount of the laser pulse at a power level at which it only causes undesirable thermal effects. This benefit is maximized in specialized micromachining lasers in which the optics and electronics have been optimized to produce pulses with very fast rise and fall times (less than 20 nanoseconds).
Shorter Wavelengths or Shorter Pulses
The key to producing smaller features and clean edges in delicate and thin materials with focused laser beams is to limit thermal effects. Thermal effects can be limited by using a laser that produces short wavelengths and short pulses with high beam quality. As already stated, this need has been met for several years by lasers with an output wavelength of 355 nm and typical pulse duration of 40–60 nanoseconds. But to produce holes and slots smaller than 10 µm, or to produce cuts with very smooth edges, as with some stents, these lasers are not always the best option. Clean edges with minimized debris can also provide the benefit of not requiring postprocessing cleaning in some cases.
Figure 2. (click to enlarge) Longer wavelength lasers process material by a thermal mechanism, essentially boiling material. In the UV, lasers remove material by a mechanism called ablation in which the laser photons directly break the interatomic bonds in a relatively cold process. |
Material removal with pulsed laser beams is caused by a combination of both thermal processing and photoablation. The latter becomes more dominant at shorter wavelengths (see Figure 2). Consequently, some applications have started to use a new generation of DPSS lasers with output in the deep UV, at 266 nm. The results can be excellent, but this approach is not without its drawbacks (see Figure 3).
Figure 3. Shorter wavelengths can deliver excellent results, but often at higher unit cost. This bioabsorbable stent has been machined on a ceramic mandrel with a 266-nm laser. |
The 266-nm lasers are only available at power levels that are a fraction of their 355-nm predecessors. Throughput is generally slower, and applications are limited to very thin materials (i.e., 250 µm in ceramics and other hard materials, and 500 µm in plastics). In addition, the cost per watt is much higher, so the laser-cost-per-unit produced increases. And equally important is that some glasses start to absorb at this wavelength, requiring the use of silica beam-delivery and focusing lenses and lowering the lifetime of the entire beam-delivery system.
Extensive studies have shown that, in many materials, comparable results can be obtained by using a laser designed to produce shorter pulse widths with high beam quality. Specifically, our studies have found that at around a 20-nanosecond pulse width or lower, and an M2 lower than 1.3, peripheral thermal effects drop off significantly in these materials. M2 is the measure for beam quality; a perfect beam has an M2 of 1. The key is to have a reliable laser with such a short pulse width and high beam quality.
Although one way to get shorter pulse widths is to lower the pulse repetition rate, this would slow the throughput for many machining processes. Alternatively, shorter pulse widths can be reached by building a physically shorter laser in which the components are pushed much closer together. In addition, shorter pulse widths can be achieved by pumping more energy into the laser crystal. To do this while maintaining stable energy output is a simple adjustment for laser manufacturers.
Choosing the Optimal Laser for Specific Materials
The type of laser selected depends on the characteristics of the material being machined and the desired results such as hole size and edge quality. As a first requirement, successful micromachining of a material necessitates that the material efficiently absorbs light at the laser output wavelength. As a rule, successful micromachining requires that 50% of the laser pulse be completely absorbed within 0.1 µm depth below the surface.
a. |
b. |
c. |
Figure 4. These Kapton samples were machined with (a) a 355-nm longer pulse laser, (b) a 266-nm laser, and (c) a 355 nm short-pulse laser. |
Plastics. Several polymer materials commonly used in medical devices absorb strongly at 355 nm. A standout example is Kapton, which is an excellent candidate for a short-pulse 355-nm laser. Figure 4 shows typical results on identical Kapton samples that have been micromachined with a short-pulse 355-nm laser, a longer-pulse 355-nm laser, and a 266-nm laser. The economic 355-nm approach delivers indistinguishable results and would therefore be the best choice. Mylar absorbs less strongly, but the short-pulse 355-nm laser can still deliver almost the same results as a 266-nm laser (see Figure 5.)
a. |
b. |
c. |
Figure 5. Mylar is a material used extensively in medical device fabrication. This figure shows the edge quality of Mylar cut using the same three lasers as Figure 4: (a) 355-nm long pulse laser, (b) 266 nm laser, and (c) 355-nm short pulse laser. Good-quality results are obtained with the 355-nm short-pulse laser, even with the lower absorption of Mylar with respect to Kapton. |
With organic polymers, UV absorption depends on the bond structure. Materials with double bonds (such as Kapton) absorb at longer wavelengths, including 355 nm. The absorption of highly saturated polymers is limited to shorter wavelengths, so these would require a 266-nm laser.
The ultimate example is Teflon. The pure material is difficult to machine even at 266 nm. Fortunately, most medical device applications use Teflon with additives, and often these are colored additives with absorption even in the visible spectrum. In this case, the short-pulse 355-nm laser can deliver great results. The bottom line with Teflon is that the optimal laser choice strongly depends on what additives are present.
Metals. Most metals absorb 355-nm wavelengths quite well, so results with the short-pulse 355-nm laser are almost as good as with a 266-nm laser (see Figure 6). The fact that the 266-nm results are slightly better is probably due to the high thermal conductivity of metal negating some of the heat-reduction benefits of a short-pulse approach. In most cases, this marginal difference is not critical. Therefore, most metal-foil applications can be well serviced with a short-pulse 355-nm laser.
a. |
b. |
c. |
Figure 6. Stainless steel is another commonly used material in medical device fabrication. Thin stainless steel and most other metals can be quickly and cleanly cut using UV lasers, the choice of which depends on the exact application at hand. Shown here is (a) 355-nm long pulse laser, (b) 266 nm laser, and (c) 355-nm short pulse laser. |
Glass. Except for colored glass, most glass materials do not absorb much light at 355 nm. Consequently, these can only be machined successfully with a shorter-wavelength laser, and this often means a laser with a wavelength even shorter than 266 nm. An even better, although costlier, solution may be to use one of the new picosecond lasers.
Ceramics. Ceramics react similarly to metals in that they are hard materials. The thermal conductivity of metals is better than that of ceramics, and the nature of the ceramic bonding is nonmetallic, so chipping and cracking are potential problems for a ceramic. In any case, ceramics respond well to 355-nm laser light, and the shorter pulse length can greatly improve processing conditions. In addition, Emulsitone is frequently used to keep the process clean and can easily be washed off in a simple postlaser process.
Conclusion
Laser manufacturers, tool builders, and contract manufacturers continue to develop new tools and techniques to support the drive for increased miniaturization in medical devices. These developments focus on processing costs as well as processing results. The advent of short-pulse 355-nm lasers enables improved results in many applications without increasing unit costs.
However, these lasers are not a panacea for micromachining needs in this industry. It is vitally important to explore all possible options before committing to a particular processing strategy. A laser or system vendor can often test samples of your material in advance to establish optimal processing parameters that help meet target costs.
Ronald Schaeffer is CEO of Photo-Machining Inc. (Pelham, NH). He can be reached at [email protected]. Tobias Pflanz is product marketing manager at Coherent Inc. (Santa Clara, CA).
Copyright ©2008 Medical Device & Diagnostic Industry
About the Authors
You May Also Like