The Power of Collimation III: Moses’ Enthalpy

At InnovaQuartz, we developed a method of making very precise, tiny lenses directly upon the face of an optical fiber. In this third Chapter of The Power of Collimation, I’ll describe just how important this subtle feature really is to the performance of holmium laser URS.

In the first Chapter of this blog series, I discussed why holmium lasers launch light into fibers at a substantial fraction of their maximum propagation angle (near maximum NA). We’ll be using computer-generated ray traces to depict what happens with a typical laser focus condition (aka laser launch condition) when coupling to a series of small core fibers in common use today: Figure 1 is standard “200 μm fiber” -- all of which are actually 273 μm core fibers, Figure 2 is a widely used 242 μm core fiber, and Figure 3 is our ProFlex™ 200 – a true 200 μm core fiber. These ray traces are highly simplified versus reality, using a single mode laser beam, because reality is far too chaotic to depict in a single, simple figure, but the principles remain the same.

The actual core diameters of small fiber is far more important than is generally appreciated because the diameter of the glass cladding is directly proportional of the flexibility of the fiber and geometrically proportional to restriction to flow the fiber presents in the working channel. 273 μm core fibers have glass diameter of 300 μm and blue buffer diameters up to 475 μm. 242 μm core fibers have a glass diameter of 290 μm and a blue or green buffer diameter up to 465 μm. True 200 μm core fibers (IQ’s ProFlex™ 200) have a glass diameter of 240 μm and a blue buffer diameter up to 430 μm. Only a true 200 μm fiber permits full deflection in all flexible ureteroscopes and allow the most irrigation flow.

All three figures include a ruler and a protractor to show that the laser focal condition is typical at 10, where the maximum acceptance angle of the fiber is 13°, as discussed in the first Chapter of this series. The fiber core is light blue and the cladding is a bit darker. Laser rays are blue in transmission and black in reflection (Fresnel reflections result at refractive index transitions like cladding to core).

Figure 1: Typical Laser Coupling Condition for 273 μm core Holmium Laser Fibers

We’ll be referring back to this figures later in the Chapter, but what is immediately apparent in the above figure is that there is significant spatial overfill – the laser rays that fall on the fiber face outside of the core – and this energy is refracted and reflected to higher angles than the bulk of the beam. These rays are preserved within the fiber and interact with the plastic cladding and blue jacketing along the entire length of the fiber and are commonly called “cladding modes” by other manufacturers. The true situation is actually significantly worse than depicted due to the multimode nature of the laser, an imperfectly centered fiber that is slightly off-axis and often has a faceted or less than perfectly perpendicular polished face with chips in the cladding that scatter laser energy.

Figure 2: Typical Laser Coupling Condition for 242 μm core Holmium Laser Fibers

A 242 μm fiber is depicted in Figure 2 where the glass cladding has been overclad with doped silica; a tube of cladding material is fused to the fiber at the input end over about one centimeter length to act as a cladding mode stripper. The color scheme is the same as in Figure 1. You can see that while more energy overfills the fiber core because it is smaller than the 273 μm fiber, most of the energy that does overfill the core (and is reflected and refracted to unsafe angles) is stripped within this fiber design. This is a small, but real improvement but the fact remains that some high angle energy continues to couple to the fiber and the remaining energy still fills more than 75% of the fiber’s angular carrying capacity, as is the case for the 273 μm fiber as well.

Again, in the real world the fiber NA is usually filled a bit more than this figure represents, if for no other reason than for the reasons the laser design engineer used the lower angle focus: imperfections in the alignment of the laser port and the laser focus, etc. The glass that is added over the fiber as a mode stripper also does not function quite as perfectly as depicted due to irregularities in the fusion junction, but overall, this type of fiber is a bit more reliable than that previously depicted. Unfortunately, in spite of using a smaller core diameter, the rigid member that determines flexibility (the total glass diameter) is just 3% smaller. You may lose up to 25% of your scope’s maximum deflection when using this competitor's fiber.

The last laser launch ray trace offered is for InnovaQuartz’ ProFlex™ 200 fiber: Figure 3. The first thing that you should notice is that the fiber’s input face is not flat like other fibers: it’s got a curve to it making it a lens. This is one of the patented features of our Pulsar™ HPC technology and it makes a world of difference. The color scheme remains the same as the prior figures, but there are no black rays in Figure 3 because there are no reflected overfill rays.

Figure 3: Typical Laser Coupling Condition for ProFlex™ 200 Holmium Laser Fibers (NOTE: Taper angle is exaggerated in the figure for illustration)

Within the Pulsar HPC, the 200 μm fiber core is grown to 290 μm (or a bit larger, depending upon the laser generator the fiber is built for) over a length of a couple of centimeters at the laser launch end. The tapered fiber is then fused into a quartz ferrule to provide a substrate larger than the fiber’s tapered cladding diameter upon which to fashion a lens that is designed to collimate the laser focus rays into the fiber core. The design works with the laser’s focusing optics to mimic a telescope, imaging the larger laser rod output onto the fiber’s core such that spatial and angular overfill are both eliminated.

The result is very low fill of the fiber NA such that far more bending stress may be applied to the fiber without causing leakage. Another benefit of collimation is lower attenuation along the length of the fiber because there are far fewer ‘bounces’ off the fiber cladding – taking some license with terminology for ease of visualization. And while eliminating fiber burn through was the motivation for developing the Pulsar HPC, we realized that a further and initially unforeseen benefit could be far more important after we started testing the fibers: significantly reduced loss from Moses attenuation in the surgical field.

For the uninitiated, the “Moses Effect” is the term used to describe how an infrared laser beam manages to function in a medium that strongly absorbs infrared energy. According to the Torah, Moses facilitated the parting of the Red Sea to provide an escape route out of Egypt. Similarly, a portion of each holmium laser pulse boils the saline irrigation between the fiber and the stone to form a steam channel for the rest of the laser pulse to pass through. It works, but a significant portion of each and every laser pulse is spent in boiling water, and that’s a fairly menial task for a surgical laser.

The historical consensus in the field is that the amount of energy lost is unavoidable and trivial, but opinions are finally changing. Some loss is unavoidable, but current losses are not trivial, especially when “dusting” at low pulse energies or when using large core fibers to fragment stones. Those without a solution to the problem will argue that there are no meaningful losses if the fiber is maintained in contact with the stone, but they are also wrong. It is impossible to maintain intimate contact with a stone for any technique.

Stones are not smooth and flat to match the surface of flat output fibers. They do not present hemispherical pits to mate with ball tips fibers either. For the sake of argument, consider a mythical flat stone surface with a fiber pressed into contact with it. How much time are you willing to spend to insure the fiber is normal to the stone surface, and how would you go about this task? There will surely be a gap. More importantly, what happens when you fire the laser? Does the fiber make a cylindrical and flat bottomed pit and follow the track with every pulse? Of course not; it’s patently ridiculous.

Considering retropulsion issues and is obvious that there is a significant average gap between the flat fiber and the stone. Clearly, the larger the gap, the slower surgery will progress.  But there is a geometric relationship between the gap and the energy lost, loss that we at IQ call “Moses Attenuation”.

Just as the focus of the laser is composed of many angles of energy rays, so it goes with the fiber’s output (Figure 5): theta (θ) in begets theta out. Again, there are those who will argue otherwise, citing the closer refractive index match of saline to glass than air to glass, but they ignore the fact that the bubble is steam, with a refractive index that is the same as air. The output may well start with a lower divergence (picoseconds), but it must reach full divergence before the beam imparts the stone.

Due to angular mode promotion in fiber bends, when it comes to flexible fibers, θ_out is larger than θ_in and may actually exceed the NA of the fiber (Figure 6), but this is beyond the scope of this blog Chapter. It is interesting though, as suggested by the figure, that the symptom of reduced performance typically prompts precisely the adjustment to the laser that causes catastrophic failure: burn through. This is not a problem seen with ProFlex™ holmium laser fibers, but I’ll likely address it sometime.