Laser micromachining involves using light to remove material. Lasers can also be used in many other applications such as welding, marking, additive manufacturing and surface alteration, but these fall outside the definition. As a general rule, laser micromachining involves working on substrates that are less than 1 mm in thickness — usually much less — and feature sizes that are also less than 1 mm, with the lower end on the order of a few microns.
Why use lasers for materials processing? There are several reasons. They are non-contact, minimizing risk of damage to parts or tools. They can be used to selectively remove one material and leave another behind by varying parameters such as laser wavelength and energy density on target. They can be extremely flexible, allowing rapid prototyping and development without significant tooling costs. Finally, they can do things that other complementary technologies like mechanical methods, chemical etch and EDM (electrical discharge machining) simply cannot do.
Commercially available lasers have wavelengths spanning from the far IR (CO2 lasers at about 10 microns) through the visible and down into the vacuum UV (157 nm F2 excimer laser). For micromachining purposes, the power output is a few watts to maybe a few hundred watts on the high side with pulse lengths ranging from milliseconds to nanoseconds on most lasers, but with some available down to pico- or even femtosecond. Almost all these lasers are “pulsed” in some manner so that the combination of small spot size (frequently less than 25 microns), short pulse length and high energy per pulse produces high peak power intensity on the part — this seems to be the key to clean and efficient material removal. Smaller spot sizes can be achieved using shorter wavelength lasers. In general, the material removal is also much cleaner, but it also appears that wavelength dependence, probably because of multi-photon processes, becomes less important as the pulse length gets very short — below 1 ns.
The most important thing in material removal applications is that there is strong absorption of the incident photons — at least 50% absorption is needed and the closer to 100% the better. Absorption depth is a function of the material, the incident energy density and also the laser wavelength — as a general rule, UV photons are absorbed within fractions of microns of the material surface whereas IR photons have a penetration depth on the order of 10s of microns or more per pulse.
Lasers commonly used in micromachining applications include excimer, solid state, DPSS (diode pumped solid state), CO2 and fiber lasers. In general, these lasers all have good beam quality (measured by a ‘focus-ability factor’- called M2, with 1.0 being a perfect beam and larger numbers deviating further from “perfect”). For a Gaussian beam laser, M2 close to 1 is ideal; for imaging lasers, a very high-order multi-mode beam is used and instead of being focused, an aperture is imaged on target. The laser itself is basically a light bulb and the beam must be conditioned and directed to the work area — this is the function of the beam delivery system (BDS). The BDS consists of a series of refractive/reflective optics, sometimes enclosed and purged, to get the photons to the work piece efficiently. Beam delivery paths can be simple or quite complicated depending on the tasks, but in general it is best to use the minimum number of optical elements needed in order to do the job.
Other components of a laser micromachining system include the frame/base, electrical distribution, motion control/automation, camera/vision, and safety features. The frame/base is normally foam filled steel weldment or heavy extruded Al with some measures taken for vibration isolation and thermal stability. Power is distributed through a main AC disconnect box to the various components of the tool. Everything is enclosed and interlocked to make it into a Class I tool, meaning normal floor operators can use the tool without compromising safety. Doors and removable panels are interlocked so that any break in the interlock string causes the laser light to be shut down. Additionally, an electrical safety EMO string interrupts power back to the AC disconnect if activated. The motion/automation can be as simple as moving mirrors for the beam delivery (galvanometer based) or using x/y/z and possibly theta motorized stages, perhaps in combination with moving mirrors or with a fixed beam configuration. Automation can include roll-to-roll, robotics, cassettes, etc. and is usually key in taking small jobs and integrating for high volume manufacturing. As another general rule, the more flexible a system is designed, the less will be the throughput, and for very specialized high throughput systems, flexibility is radically compromised.
Figure 1 shows a representative plot of Etch rate (microns per pulse) vs. Fluence (J/cm2). Fluences below the ablation threshold show no material removal at all. Above the ablation threshold there is material removal, the rate of which increases until some point at which the curve plateaus. In general, a “good working area” is on the flat part of the curve where small changes in laser output do not significantly affect per pulse material removal rates. Going higher on the curve gives no payback in removal rate, so it is generally not a good idea to do so unless other factors are involved. Note also that other factors, such as taper and HAZ (Heat Affected Zone) may also be affected by the energy density on target.
Because of the wide range of available candidate lasers with differing pulse lengths, wavelengths and power outputs, almost any material can be a candidate for laser materials processing as long as the thickness and absorption are within the boundary conditions set previously. In addition, these laser tools are used in a variety of markets including medical devices (usually disposable), microelectronics, aerospace/defense, semiconductors, and the new hot field, alternative energy, specifically solar.
The medical device market is a very attractive market for laser micromachining because it can be very high volume, profitable, longterm and — very important — less sensitive to fluctuations in the economy than other markets like semiconductors and microelectronics. Plastics make up a large percentage of the materials, with materials that have numerous pi electrons being especially absorptive of many wavelengths of light. Likewise, many metals are used, but in particular there is a large ongoing market for SS stents involving hundreds of lasers worldwide. Figure 2 shows a laser generated gear for a medical implant (on a penny).
In the microelectronics market, lasers are used for wire stripping, microvia drilling, dielectric material removal, short repair, solder mask removal and patterning conductive films like ITO (Indium Tin Oxide) and thin metals to name a few applications. In some cases, the selectivity of lasers is used to preferentially remove one material and leave another behind. A good example is using a CO2 laser to remove dielectric from copper (Cu) contact pads, fingers or wire. This laser, emitting at about 10 microns wavelength, is almost totally reflective when hitting Cu, but it removes most dielectrics quite efficiently. Therefore, this removal process can be self limiting; when the photons are through the dielectric, they do not machine into the copper below. Copper on top can also be used as a conformal mask when via drilling.
A very easy application for lasers, yet not for other machining methods, is removal and patterning of photo-resist or solder mask. Solder mask, especially, gets put in undesirable places and using lasers can save otherwise scrap boards. Figure 3 shows a portion of a circuit with solder mask removed on the right side using a CO2 laser, while the left side is still untouched. Figure 4 shows solder mask removal using a hand dremel and a laser. We can also drill very small holes in ceramics with no microcracking or chip out. Figure 5 shows a wafer probe with laser 0.004" holes through 1.5 mm of alumina ceramic used for the probe tips.
The new booming application for laser technology is in the manufacture of solar panels. This is a potentially very high volume market using various types of lasers for different scoring, isolation, and deletion processes. All the commercial laser manufacturers are gearing toward this industry as well as a large number of integrators. Unfortunately, this industry could also be like the dot- com boom of the late ‘90s and early 2000s where everyone jumps on the bandwagon, but there is insufficient business to go around. Just like the dot-coms, there will be a lot of companies who lose money, but there will be a few who make lots of money. The trick is to be careful who you work with.