The rapid pace of innovation in the medical device industry puts ever increasing pressure on manufacturers to achieve greater geometrical precision, increase device lifetime and reliability, and simultaneously reduce the cost of making diverse portfolios of products. A key step in manufacturing medical implants, such as cardiovascular stents, is laser micro machining, where the basic device geometry is cut from an extruded tube or other raw substrate. Today, most device production is performed using continuous wave (CW) lasers, which have been commercially available for decades. Nonetheless, manufacturers are switching to ultra-short pulse (USP) lasers for new device production lines at a rapid rate. In this article, we describe why this is happening and what basics device and process engineers should know to be successful with USP laser machining.

USP Laser Micro Machining Fundamentals

Medical device manufacturers typically want to immediately answer two essential questions when evaluating a new machining technology:

  1. Will it achieve the necessary quality for the device specification?
  2. Is the process fast enough for the required production run rate?

These questions are usually answered through several iterations of laser application demonstration by the medical device manufacturer and laser supplier. The closer the collaboration, the more accurate the answers.

USP laser micro machining is selected by manufacturers when they want to avoid a heat affected zone (HAZ)—the collateral damage left by machining with conventional lasers. More on this topic in the next section. Simply put, CW lasers remove material by melting it, and USP lasers remove material by vaporization. The options for industrial USP lasers today include pulse durations from tens of picoseconds (ps) down to a few hundred femtoseconds (fs). There are three ranges with the greatest number of supplier options:

  • Standard pico lasers: 5 to 10 ps
  • Long femto lasers: 700 to 900 fs
  • Short femto lasers: 300 to 500 fs

The optimal pulse duration for a given manufacturing challenge depends upon multiple factors, including the required post-machining quality. Beyond prevention of HAZ, kerf taper and sidewall average roughness are common figures of merit.

Fig. 1 – Sidewall average roughness versus laser pulse fluence for Durnico at pulse duration of 400 fs (green), 900 fs (red), and 6 ps (blue).
Kerf taper is the narrowing of the kerf width down through the material thickness owing to the Gaussian power distribution of the laser beam at focus and the distinct threshold for ablation of material to occur. Empirically, we have seen reduced taper in metals when reducing pulse duration from 6 ps to 900 fs, but no additional benefit in further reducing to 400 fs. It is not clear exactly why this happens, although it is thought to be related to the relative change in ablation threshold fluence between the pulse ranges.

The average roughness (Ra) for the sidewall surface also shows a strong dependence on pulse duration. As an example, we measured average roughness versus pulse duration for cutting Durnico, a maraging steel, at several fluences. The data, as shown in Figure 1, reveals that 900 fs pulse duration consistently produces lower roughness than either 6 ps or 400 fs. This phenomenon is not rigorously studied at this point, and the pulse duration dependence may vary with other metals or experimental conditions.

Along with machined-part quality, machining rate is a critical factor in determining the best laser source for a manufacturing process. Of course, generally more power means faster material removal, to a point. When trying to avoid any HAZ, the net heat deposit in the finite volume comprising the part will limit actual power used on the target. Even short femto lasers deposit a certain amount of heat with each pulse. This tiny amount of heat will accumulate over many pulses to result in HAZ if the process is not optimized. Nonetheless, for a given power range, the machining rate can be increased before seeing HAZ by selecting the best pulse duration.

Fig. 2 – Material removal rate versus laser pulse fluence for stainless steel at pulse duration of 400 fs (green), 900 fs (red), and 6 ps (blue).
To illustrate this effect, we measured material removal rate versus fluence for stainless steel at the same pulse durations used above. As shown in Figure 2, there is the expected trend of greater material removal with increasing fluence (more material removed with each pulse), but there is also a clear indication that 900 fs removes material faster than 400 fs and much faster than 6 ps. The reason for this is not yet obvious. Aside from the fundamental laser-material interaction, beam propagation effects inside the kerf likely play a significant role in material removal rate. However, researchers have consistently observed that shorter pulses produce faster machining until the benefit levels off, or reverses, for shorter femto pulses.

The information presented here was gathered using typical conditions for micro machining with USP lasers, e.g., 150 mm focal length lens, nitrogen purge gas, and several hundred kilohertz pulse repetition rate. A manufacturer’s results will directly depend on material type and thickness, focusing conditions, purge gas type and pressure, and tool path. These factors will be described in greater detail in the last section of this article. First, we will examine practical benefits to medical device manufacturing by using USP lasers.