SPIE Membership Get updates from SPIE Newsroom
  • Newsroom Home
  • Astronomy
  • Biomedical Optics & Medical Imaging
  • Defense & Security
  • Electronic Imaging & Signal Processing
  • Illumination & Displays
  • Lasers & Sources
  • Micro/Nano Lithography
  • Nanotechnology
  • Optical Design & Engineering
  • Optoelectronics & Communications
  • Remote Sensing
  • Sensing & Measurement
  • Solar & Alternative Energy
  • Sign up for Newsroom E-Alerts
  • Information for:


Print PageEmail PageView PDF

Lasers & Sources

Safety First

Don't fall into the traps and holes represented by poor lab practices or incomplete laser-safety training.

From oemagazine September 2005
31 September 2005, SPIE Newsroom. DOI: 10.1117/2.5200509.0008

The growth of laser development and technology during the past 40 years has been remarkable. Unfortunately, a number of traps and holes in laser safety have developed with that growth. These traps have been the cause of, or contributing factor in, many a preventable laser accident. By remaining alert to risks and by following best practices, you can increase laser safety in your lab.

The human eye is a remarkable sensor. It can readily detect photons from approximately 400 to 700 nm. Many individuals can detect wavelengths beyond 700 nm, some as far as 850 nm (see figure 1). Beyond 700 nm, to borrow a line from Rod Serling, we are now entering the iceberg zone. It is well known that only 12 to 20% of an iceberg is above the waterline, meaning it is extremely hard to visualize its actual size by visual perception alone. The same is true past 700 nm. Our eye detects less than 1% of the available photons at such wavelengths and so it can only produce a faint image. We mentally equate that faint visualization with a weak or low-power source, which can lead to dangerous underestimation of the hazard.

Figure 1. The human eye is only faintly sensitive to wavelengths above 700 nm.

Many a researcher has been injured while attempting to view, say, a 751-nm or 810-nm beam they perceived only faintly. In almost all of these cases, the user knew the actual beam power or energy was high enough to cause damage, but the fact was masked by the faint-equals-weak theorem, or iceberg effect.

The best way to avoid this trap is to remain aware; we all know not to look into a laser beam, but we get careless and forget about the many ways stray light can reach our retinas (see sidebar). If you stick with the basics, you can't go wrong. Wear protective eyewear—no ifs, ands, or buts. Fitting labs with alternative viewers such as CCD cameras and IR sensor cards can reduce the temptation to pull the goggles off "just long enough to get the job done." Just for the record, it doesn't take long to be blinded, either.

Dangerous Justifications

"How do you align if you can't see the beam?" This refrain has been a chorus from laser users since the early days of laser technology. Rather than the justification for laser alignment without protective eyewear, this mantra should be the signal to call the laser safety officer for input. The purpose of alignment eyewear is to allow the user to visualize the beam while lowering the intensity of any beam transmitted through the user's eyewear to a Class-2 level. The European Norm EN208 addresses this issue and recommends optical density for alignment eyewear versus the output of lasers used (see table).

This author often has met laser users who proudly show off their protective eyewear for visible lasers; when asked how they align with the goggles on, the response is generally silence or a lifting up of the eyewear. Given the fact that it puts their vision at risk, this approach is unacceptable.

So how do we align safely? One option to consider is the use of laser-alignment eyewear, which only partially attenuates the beam, allowing a degree of visibility. Remote viewing by CCD camera is another option, especially in the case of newer, compact, economical cameras that can be positioned to view a target, mirror, or other optics. By combining the remote cameras with motorized mounts, users can align the system without great risk.

Figure 2. An iris shutter can be used in combination with a sensor card to verify beam alignment.

A time-honored approach is to lower the output power of the laser during alignment by using a neutral-density filter or lowering the drive current. Full power is rarely needed to align a beam path; low-power coaxial beams (generally a Class 3A or 3R, depending on the ratings system used) are generally sufficient to show the beam path. A simple option for fine-tuning the alignment is to use an iris shutter or a series of irises and holes through a sensor card - if the beam goes off path, the irises attenuate it and/or the sensor card shows the offset (see figure 2).

The Hartman plate for telescope alignment provides a simple option, and fluorescent plates, or even some crosshair set ups, can aid in alignment procedures. With all of these options, there is no excuse for risky alignment practices in the lab.

Housekeeping and Safety Holes

If only our labs were like the TARDIS from the British Dr. Who television series, a place that is larger on the inside than on the outside. For those of us who have not solved the space-time dimensional problem, space is a real issue in the lab. Even as lasers become smaller, we still seem to find objects to fill all of our space.

Many laser labs look like people's garages. Unfortunately, this clutter is not confined to the space around the optical tables but extends to the tables themselves (even vertical optical table set ups are not immune to clutter). Spare optical mounts, tools, lenses, mirrors, plastic bags, plastic, and cardboard boxes all tend to find homes on optical tables. Under and beside active beam paths, they attract hands into live beams and provide reflective surfaces when lifted through the beam.

The solution? Keep your workspace clean. Set up a staging area on the optical table outside the active beam line or construct a second, upper-level surface on the optical table. Organize cabinets, removing seldom-used items that have been taking up space for decades. Arranging secondary storage for such items is an option all facilities should consider, as well as planning a housekeeping day once a quarter. Setting designated housekeeping days, just like preventative-maintenance days, frees users from the pressure of stopping to clean while project deadlines loom.

Safety folks like optical fiber because it contains the optical beam and the bare fiber can be jacketed to provide additional protection. So how can this frail fiber be a safety trap? For years, the rule of thumb was that the divergence of the beam from the end of a fiber was so broad it was not an eye hazard beyond about 10 cm. Here is where the trap starts. Many of today's fiber applications require a microlens at the end of the fiber, which creates a collimated beam rather than a quickly diverging one, so the hazard zone can be meters long. In addition, the amount of energy being transmitted through optical fiber has steadily increased, as witnessed by the development of high-power diode lasers and diode-laser arrays.

Most wavelengths used with optical fiber are invisible and the fiber end is far from the source, which makes it difficult to determine whether a fiber is active or inactive when one is found disconnected (warning labels are a good first attempt to combat this hazard). Handling and cutting/splicing fibers present sharps and UV hazards. One can see that fiber lasers, while a great asset to laser technology, present hazards of their own. Finally, there is the issue of broken fibers. (A number of operating room fires have been started by laser beams escaping from broken fibers.)

Individuals never wish eye damage or skin burns on themselves or their colleagues. Just as you would avoid a hole on the street, learn to look for holes in your laser-safety procedures and step safely around them. Laser accidents can be avoided - let's do so together. oe

The Risks of Reflections

If eye safety were as simple as not looking into the beam, we would have significantly fewer injuries. Unfortunately, light in a lab can propagate in unexpected ways. Consider these anecdotes and the errors that they highlight:

1. A reflection that was unknown and left unchecked.

From a September 2004 Department of Energy incident report: While aligning the diagnostics for an ultrafast titanium-doped sapphire (Ti:sapphire) Class 4 laser (800 nm), an experimenter raised his laser safety eyewear to rub his eye to alleviate an irritation due to an existing eye infection. He felt a bright flash and afterwards a light cloudiness in his left eye.

Repairs on the laser were completed earlier in the day. In his eagerness to get his experiment underway, the experimenter introduced a beam onto the table while he aligned the optics. He rotated one of the polarizing beamsplitters. The secondary beam was not considered or accounted for, therefore not blocked or contained. By doing so, an unwanted/undetected beam left the plane of the optical table at an upward 45° angle and subsequently struck his eye.

During the set-up and alignment of a laser system, it is essential to stop several times and check for reflections leaving the plane of the table. Using an IR or UV viewer/sensor card or lowering the light level are acceptable approaches.

2. A reflection that was known, but thought not to be a problem.

From another incident report: An experimental set-up had an invisible (3000 nm) reflection leaving the table at such a steep angle it struck a spot 8 feet up on an adjacent wall. Since it left the table at such a steep angle, the decision was made to disregard the reflection. Then, someone was being shown how to place an optic; during this process, the reflection traveled up and down the wall and struck the person being instructed, who was standing directly opposite the optic to get a better view of the procedure. It is noteworthy that a helium-neon (633 nm) beam ran co-axially with the 3000-nm beam. The injured person could have worn eyewear that would have allowed 633-nm visualization but blocked the 3000-nm beam; instead no eyewear was worn.

One of the most irresponsible mistakes you can make is to know of a reflection that leaves the table and not address (block or contain) it because you do not think it is worth the effort or do not think it presents a hazard.

3. An instant reflection generated by some action (e.g. moving a power meter or tools into the live beam).

The technique of moving a power detector head into an active beam is poor practice no matter how one defends it. This lapse is so common, it needs no anecdote. During inspections, laser safety officers should start by examining power-meter detector heads for burn marks and burnt-off coatings. In addition, jewelry and ID badges provide ready-made reflective sources if not removed (in the future all such valuables can be sent to this author for safe keeping). Poor housekeeping on the optical table provides a perfect set-up for reaching into the beam path or bringing reflective tools, optics, and so on into the beam path.

So how can a technician see visible reflections while wearing eyewear? This is a solid question because even with alignment eyewear, certain visible diffuse reflections may be hard to see. A few thoughts on a solution: view the room through a digital or video camera, which provides a way to image the source area; look from a known safe vantage point; finally, view the area with an IR viewer, since IR radiation may be leaking through with the visible light. -K.B.


1. www.iec.ch

Ken Barat
Ken Barat is a laser safety officer for the National Ignition Facility Programs, Lawrence Livermore National Laboratory, Livermore, CA.