Laser Tools for Micromachining: Reducing Part Size, Time, and Cost

Medical Device & Diagnostic Industry Magazine
MDDI Article Index

An MD&DI September 1998 Column

Laser machining tools can often handle miniature devices and components better than mechanical equipment can.

Medical device manufacturers pursuing the miniaturization trend are encountering a thorny fabrication problem: new devices often require features that have sizes too small for mechanical processes. For example, a device may need holes, slots, or channels smaller than the practical range of lathes, mills, punches, or drills. Although wide variation exists among applications, most standard mechanical machining systems are difficult to use when a feature's dimensions fall below 100 µm (0.004 in.). Machining problems are often exacerbated when materials that are difficult to work are used. Soft medical plastics, ceramics, and advanced metal alloys that present machining problems on any scale are particularly challenging when the machined features are so small that a microscope is required for their examination.

Figure 1. Laser-cut stentlike pattern in miniature stainless- steel tube with 1.25-mm OD.

Figure 2. Laser-drilled hole pattern in polyurethane tube. The smallest hole diameter is 15 µm.

Laser-based tools provide fabrication alternatives that are particularly valuable at small scales. Lasers are now used in the fabrication of stents (Figure 1), catheters (Figure 2), fluid-flow-control components, lenses and haptics, miniature flex circuits (Figure 3), and wire assemblies. They are also used to mark device components, allowing even the smallest unit to be traced. As shown in Table I, laser processing can be applied to most medical device materials if the laser's characteristics are properly matched to the material's parameters.

Figure 3. Laser-excised miniature flex circuit.

In most applications, the laser light source is one of the three "workhorse" laser types—CO₂, Nd:YAG, or excimer. Together, these three laser families provide wavelengths ranging from the deep infrared to the deep ultraviolet, pulse durations from nanoseconds to milliseconds, and average power from milliwatts to kilowatts. For effective machining, the laser wavelength must be such that energy carried by the laser light pulse is strongly absorbed at the surface of the process material. The duration of the light pulse should be short enough to achieve the required degree of localization, and the pulse repetition rate or average power should be high enough to achieve the desired processing speed.

Table I. Examples of medical device materials suitable for laser micromachining.

Most laser micromachining processes remove material by a rapid vaporization process called ablation. Pulsed laser sources are normally used for laser ablation. Laser light is absorbed in a thin surface layer of the material, causing thermal or photochemical effects that release free atoms or molecules from the surface. The components of the gas jet leaving the surface may remain in a gaseous state and dissipate in the surrounding atmosphere. They may also react with atmospheric species or each other to form particulates that fall back to the surface as dust or debris.

If the laser pulse is short enough, there will be little opportunity for heat to spread into nearby regions. This allows the ablation to be highly localized. Practical optical limits dictate that the smallest dimension of an ablated region will be approximately twice the wavelength of the laser light.

COMPARING LASER AND MECHANICAL PROCESSES

With shrinking scales and new processes, it is easy for design or production engineers to lose their intuitive feel for the quality and speed of an operation. To improve potential users' intuitive understanding of laser micromachining, it is compared below to larger-scale mechanical machining processes.

Both laser-drilled and mechanically machined processes can produce a hole with relatively smooth inner walls. Depth-to-width aspect ratios as large as 10 can be achieved with both processes, but ratios in the range of 3 to 5 are preferred for production processes. Both methods may produce a small burr at the top of the hole and may leave debris that must be removed after drilling. Mechanically produced burring can be minimized with the appropriate cutting tools, coolant, and feed rates. Laser-produced burring can be suppressed with the appropriate laser wavelength, pulse duration, energy, and repetition rate, and by using assist gases and coolants. For the mechanical drill, the debris may be chips or cutting fluid; for the laser process, the debris may be loosely adherent particles or spattered molten material. For both methods, postprocessing or cleaning may be required.

Material adjacent to a mechanically machined region may melt or experience surface hardening as a by-product of the heat generated in the cutting process. Peripheral heat damage is controlled in mechanical processes by using coolant and optimizing tooling and cutting speeds. Since laser cutting involves melting and vaporization, the potential for thermal damage to surrounding material is somewhat greater. In most applications, using laser pulses of the appropriate wavelength and submicrosecond duration will reduce the size of the heat-affected zone to negligible dimensions.

There are also some qualitative differences between the two processes. A laser-cut hole will typically have a slightly tapered wall in comparison with the nearly vertical wall created by a mechanical process. Wall-taper angles of 1°—2° are often observed in laser-cut holes and channels. The angle of the taper is influenced by the energy density of the light at the work surface, the location of the beam's focus, and the cone angle of the beam as it focuses on the surface. Lasers with high beam quality tend to cut with a smaller taper. Lasers also differ from mechanical equipment because they apply very little force on the part during material removal, can machine very hard materials, and experience no tool wear. These characteristics greatly simplify fixturing.

The comparison between laser micromachining and conventional machine-shop processes can be extended to cutting speeds and part fabrication times. To reasonably compare processing on two radically different spatial scales, it is helpful to think of a mechanical structure as an assembly of unity-aspect-ratio (UAR), or blocklike, features (Figure 4). Changing the size of the UAR features scales the overall size of the structure. Thus, a machining process like drilling a hole or cutting a channel corresponds to removal of an appropriate block of material.

Figure 4. Representation of a structure by a group of unity-aspect-ratio (UAR) features that resemble building blocks. Changing the size of the UAR blocks scales the size of the structure.

At low speeds, the cutting rate for milling and drilling machines is directly proportional to the amount of power delivered to the work surface. However, limits are imposed by cooling and chip removal. Standard machinists' reference texts provide cutting speeds and feeds that account for all of these factors. These references can be used to estimate the time required to remove a UAR feature—for example, to drill a hole with depth equal to width.

Table II. Single-pulse ablation depth for three materials for typical wavelengths and energy densities used in micromachining.

For laser cutting, the amount of material removed during a single laser pulse is a key parameter. Single-pulse ablation depth depends on the optical energy density at the work surface and is often less than 1 µm. Table II shows the ablation parameters for some actual laser processing scenarios. Under the conditions summarized in the table, the rate of feature fabrication is proportional to the laser pulse repetition rate.

Figure 5. Production rates of unity-aspect-ratio (blocklike) features in stainless steel. Production rates for large features using milling and drilling machines are compared with those for microscale features using high-pulse-repetition-rate lasers. Mechanical machining rates were derived from Machinery's Handbook, 24th ed., Green RE (ed), New York, Industrial Press, 1988; and Melin Tool Co.

Figure 5 shows the time required to remove a blocklike feature in a material like stainless steel for a range of pulse repetition rates that are typical of lasers used in micromachining. The figure also shows the feature fabrication rates for much larger features using conventional milling and drilling machines. As feature sizes decrease below 100 µm, all of the lasers produce UAR features significantly faster than mechanical machines operating on millimeter-scale sizes. On small scales, lasers can produce fundamental features at rates that are orders of magnitude faster than those of mills or drills in a standard machine shop. Miniature structures can be produced much more rapidly by lasers than larger structures of similar shape produced by mechanical means. Since time often equals money, miniature structures could cost significantly less to make than their larger counterparts.

A high-volume application of laser micromachining—the laser cutting of stents—works as an example. Laser-cut stents are typically fabricated by cutting intricate shapes in stainless-steel or nitinol tubing with a wall thickness of about 0.15 mm and an OD of about 1.5 mm. Because of the complexity of the patterns (Figure 1), the total cut length is often in the range of 500 mm for a 20-mm-long stent. Nd:YAG lasers with focal spot diameters of about 50 µm and pulse repetition rates in the kilohertz range are often used to cut stents. Single-pulse ablation depth is in the micron range, with corresponding cutting speeds that exceed 1 mm/sec. The overall time required to cut the stent pattern is therefore about 10 minutes. Now suppose the stent structure were scaled up by a factor of 100 and cut with a CNC mechanical machining center. The raw material would be a tube 2-m long, with a 15-mm wall. The 50-m cut length would be traversed by a 5-mm cutter at a recommended speed of 70 mm/min. Total cutting time would be about 12 hours.

Although this example is extreme and highly simplified, it demonstrates the relatively high speeds associated with the laser machining of microstructures. Does high speed correlate with lower part cost? Generally, yes. Depreciation and capital costs dominate the cost of owning computer-controlled laser and mechanical machining equipment. The overall capital cost of a laser micromachining center is comparable to that of a mechanical CNC center, so hourly costs of ownership are not dramatically different. With higher throughput and lower material costs, laser-cut miniature parts can be significantly less expensive than their larger counterparts.

CONCLUSION

Design, development, and production departments faced with pervasive miniaturization can use laser tools to break the 100-µm feature barrier. These laser tools can cut, drill, and mill in many ways that mimic processes used in conventional machine shops. There may still be a need for burr removal and part cleaning, but fixturing can be simplified and tool wear is negligible. The similarities between standard machine shop operations and laser micromachining allow mechanical intuition and design visualization to be readily transferred to the microscale. Perhaps the biggest surprise is that laser tools are relatively fast, potentially reducing part costs. Could it be that small is both beautiful and inexpensive?

C. Paul Christensen is chief technical officer of Potomac Photonics, Inc. (Lanham, MD).

Copyright ©1998 Medical Device & Diagnostic Industry