Laser Safety

When lasers first began appearing in laboratories, both the devices and their applications were so specialized that safe laser operation was a problem faced by a very limited group of researchers and engineers, and was not a subject of general interest. With the dramatic growth in the application of lasers in everyday activities, as well as their routine utilization in scientific laboratories and industrial environments, many more investigators must necessarily face the matter of laser safety. Lasers have become integral components of many current optical microscopy techniques, and when combined with complex optical systems, they can constitute a significant hazard if safe procedures are not strictly followed.

The two major concerns in safe laser operation are exposure to the beam and the electrical hazards associated with high voltages within the laser and its power supply. While there are no known cases of a laser beam contributing to a person's death, there have been several instances of deaths attributable to contact with high voltage laser-related components. Beams of sufficiently high power can burn the skin, or in some cases create a hazard by burning or damaging other materials, but the primary concern with regard to the laser beam is potential damage to the eyes, which are the part of the body most sensitive to light. A number of government agencies and other organizations have developed standards for laser safety, some of which are legally enforceable, while others are simply recommendations for voluntary compliance. The majority of legally required standards pertain to manufacturers of laser equipment, although the end user of the laser has the largest interest in safe operation - the prevention of personal debilitating injury or even death.

Damage to the eye can occur instantaneously, and precautions must be taken in advance to minimize the risk since avoidance at the last moment is not a possibility. Laser emission is similar to direct sunlight exposure in that the light arrives at the eye in parallel rays, which are very efficiently focused on the retina, the rear surface of the eye that senses light. General anatomy of the human eye is illustrated in Figure 1, with emphasis on the structures that are likely to be damaged by absorption of intense radiation. Potential hazards to the eye depend on laser light wavelength, beam intensity, distance from the laser, and power of the laser (both average power over long intervals and peak power produced in a pulse). The wavelength of the laser radiation is significant because only light within the wavelength range of approximately 400 to 1400 nanometers can penetrate the eye sufficiently to damage the retina. Near-ultraviolet light of certain wavelengths can damage layers of the eye near the surface, and can contribute to cataract formation in the lens, especially in younger persons, whose eye tissues have greater transparency in this wavelength range. Light in the near-infrared can produce surface damage as well, although at a higher damage threshold than for ultraviolet light.

The physical response of the human eye differs for light of different wavelengths, and this has a bearing on the potential damage that may occur for several reasons that will be discussed below. Pulsed lasers present a different hazard than those producing continuous beams. In practice, lasers operated in pulsed mode are generally of higher power, and a single microsecond-pulse of sufficient power can cause permanent damage if it enters the eye, whereas a lower-power continuous beam may only present a hazard with long-term exposure. The spectral region of greatest concern constitutes the retinal hazard region, extending from about 400 nanometers (violet color) to 1400 nanometers (near-infrared), including the entire visible portion of the electromagnetic radiation spectrum. The danger presented by these wavelengths is amplified due to the fact that ocular focus is possible, and collimated light within this range is focused by the eye on a very small spot on the retina, concentrating its power to a high density.

Classification of Lasers

Of the many laser safety standards developed by both governmental and other agencies, the one most often relied upon in the United States is the American National Standards Institute's Z136 series. The ANSI Z136 laser safety standards are the basis for the Occupational Safety and Health Administration (OSHA) technical rules used to evaluate laser hazard issues, and are also the reference for many states' occupational safety rules pertaining to laser use. All laser products sold in the USA since 1976 are required to be certified by the manufacturer as meeting specified product safety standards for their designated classification, and they must be labeled as to their class. Research results combined with an accumulated understanding of the hazards of sunlight and other light sources have led to the establishment of estimated nominal safe exposure limits for most types of laser radiation. A system of laser hazard categories, based on the known maximum permissible exposures and experience gained from years of laser use, has been developed to simplify the application of safety procedures to minimize or prevent accidents. The laser manufacturer is required to certify that a laser product falls into one of the categories, or risk classes, and to label it accordingly. The four primary laser categories are summarized in the following list. It must be emphasized that this is an abbreviated summary, and is not intended to be a complete statement of any agency's laser classification regulations.

  • Class I lasers are considered safe, based upon current knowledge, under any exposure condition inherent in the design of the product. The low powered devices (0.4 milliwatt at visible wavelengths) that use lasers of this category include laser printers, CD players, and survey equipment, and they are not permitted to emit levels of optical radiation above the exposure limits for the eye. A more hazardous laser may be incorporated within the enclosure of a Class I product, but no harmful radiation is permitted to escape during use or maintenance (this does not necessarily apply during service). No safety requirements are specified for the use of this class of laser.

  • Class IA is a special designation for lasers that are not intended for viewing, such as supermarket laser scanners. A higher power is permitted than for Class I lasers (not more than 4 milliwatts), but the Class I limit must not be exceeded for an emission duration in excess of 1000 seconds.

  • Class II are low-power lasers that must emit a visible beam. The brightness of the beam is relied upon to prevent staring into the beam for long enough periods to cause eye damage. These lasers are limited to a radiant power less than 1 milliwatt, which is below the maximum permissible exposure for momentary exposure of 0.25 second or less. The natural aversion reaction to visible light of this brightness is expected to protect the eyes from damage, but any intentional viewing for extended periods could result in damage. Some examples of this class of laser are demonstration lasers for classroom use, laser pointers, and range-finding devices.

  • Class IIIA lasers are continuous wave intermediate power (1-5 milliwatt) devices, with similar applications as Class II lasers, including laser scanners and pointers. They are considered safe for momentary viewing (less than 0.25 second), but should not be viewed directly (intrabeam), or with any kind of magnifying optics.

  • Class IIIB lasers are of medium power (continuous wave 5-500 milliwatt, or 10 joules per square centimeter in pulsed devices), and are not safe for direct viewing or viewing of specular reflections. Specific safety measures are recommended in the standards for control of hazards with this laser class. Examples of applications of this laser type are spectroscopy, confocal microscopy, and entertainment light shows.

  • Class IV lasers emit high power, in excess of the limit for Class IIIB devices, and require stringent controls to eliminate hazards in their use. Both the direct beam and diffuse reflections from these lasers are damaging to the eyes and skin, and are potential fire hazards depending upon the materials that they strike. Most laser eye injuries involve reflections of Class IV laser light, and consequently all reflective surfaces must be kept away from the beam, and appropriate eye protection worn at all times when working with these lasers. Lasers of this category are employed for surgery, cutting, drilling, micromachining, and welding.

Although ANSI Z136 currently classifies lasers as Class I to Class IV, a new laser-hazard classification scheme will most likely be incorporated into the next revision of the ANSI standard, in an attempt to provide more harmonization with international standards such as those recognized by the International Electrotechnical Commission (IEC), and those already adopted by the United States Food and Drug Administration. The changes in the standards are primarily a response to the proliferation of laser pointers and similar devices that are likely to be used by individuals who are unfamiliar with laser safety precautions, and to the special characteristics of high-divergence sources such as laser diodes. The effects of the changes are relatively small, and in general continue a relaxation of regulations that has occurred with accumulation of data and experience since the early very conservative standards were developed in the 1970s.

The new classification scheme retains four major classes of laser, 1 to 4, but introduces relaxed versions of classes 1, 2, and 3 with less stringent requirements, and special subcategories of each: 1M, 2M, and 3R. In summary, the new categories can be generally described as follows: Class 1M includes lasers that are not capable of causing eye damage except when viewed with optical instruments. Class 2M applies to lasers emitting visible light, which are safe for viewing, provided optical instruments are not employed, for up to 0.25 seconds. Within that period of time, the natural aversion response to bright light, combined with the blink reflex, protects the eye from retinal damage. Class 3R includes lasers that are marginally safe for direct viewing, and which are allowed to have output power of up to five times that of Class 1 or Class 2 lasers. Additional measures are to be taken to prevent direct eye entry, especially for invisible wavelengths.

Eye Hazards

It is notable that a common warning for most categories of laser is to avoid viewing the beam with any magnifying optical device. A primary danger to the human eye posed by lasers results from the fact that the eye itself is a highly precise and efficient focusing optical device for light within a certain wavelength range. Utilizing lasers in conjunction with microscope optics only adds to the potential for damage to the eye. It is common for optics laboratories to contain many lasers, both as integrated components of systems such as fluorescence microscopes, and as light sources on open optical benches. The major dangers presented by these "open" lasers are potential damage to the eyes from stray horizontal beams at table height, beams reflected out of the plane of the table, and reflections from optical components and external reflective surfaces, such as belt buckles, watches, jewelry, and any reflective surfaces in the room. Even a split-second exposure to a small reflected portion of a laser beam may be sufficient to cause permanent injury and loss of vision.

The potential for laser emission causing injury to the different structures of the eye depends upon which structure absorbs the beam energy. The absorption characteristics of the different eye tissues, and the wavelength and intensity of the laser light determine whether damage occurs to the cornea, lens, or retina. The wavelengths that penetrate to the retina at the rear eye surface are determined by the overall transmission characteristics of the eye. Figure 2 illustrates eye transmission as a function of radiation wavelength over the relevant spectral range. The cornea, lens, and vitreous fluid of the eye transmit electromagnetic radiation in a wavelength range of approximately 400 to 1400 nanometers, termed the ocular focus range. Light within this range is focused onto the retina, the sensory surface that produces signals that are sent to the brain by the optic nerve. Direct viewing of a point source of light, which is the situation effectively created during intrabeam viewing of a highly collimated laser beam, produces a very small focal spot on the retina, resulting in a greatly increased power density and a high probability of damage. The dangers are similar in certain respects to those presented by direct viewing of the sun, although the potential intensity is even higher for lasers.

The optical gain of the relaxed human eye for a highly collimated beam, which is the ratio of the area of the eye's pupil to the retinal (focused) image area, is on the order of 100,000. This corresponds to five orders of magnitude irradiance increase from the corneal surface to the retina. Allowing for aberrations in the lens-cornea system, and diffraction at the iris, a well-corrected eye is capable of focusing a 20-micrometer spot on the retina. The significance of this efficiency of the eye is that even a low-powered laser beam, if it strikes the eye, can be focused onto the retina and quickly burn a hole in the tissue, permanently damaging nerves responsible for vision. The seemingly low rated power of lasers can be very misleading with respect to the damage possible when the energy is concentrated to this extent. In the case of a laser beam entering the eye directly (intrabeam viewing), a 1-milliwatt beam produces a retinal irradiance value on the order of 100 watts per square centimeter. In comparison, direct viewing of the sun produces an irradiance at the retina of approximately 10 watts per square centimeter.

Figure 3 illustrates the focusing effects in the eye for an extended source, such as a conventional frosted glass lamp, in comparison to a highly collimated laser beam, which has the effective properties of a point source. Because of the differences in the nature of the sources, the power density at the retina can be 1 million times greater for a focused 1-milliwatt laser than for the standard 100-watt lamp. Assuming a perfect Gaussian laser beam, directly entering an aberration-free eye, a diffraction-limited spot size of 2 micrometers diameter at the retina is possible, compared to a focused spot of several hundred micrometers for the extended source. The corresponding irradiance (power density) values at the retina, as shown in Figure 3, are approximately 10×(E8) and 10×(E2) watts per square-meter.

It may be thought that a burned spot on the retina measuring even 20 micrometers would not be significant to vision, since the retina contains millions of cone cells. Actual retinal injuries, however, are usually larger than the primary focused spot due to secondary thermal and acoustic effects, and depending upon the location, even an extremely small injury to the retina can significantly damage vision. In a worst-case exposure, with the eye relaxed (focused at infinity) and the laser beam entering the eye directly or from a specular reflection, the beam is focused to its minimum spot size on the retina. If damage occurs to the area where the optic nerve enters the eye, the result is likely to be complete loss of vision. Retinal burns are most likely to occur in the area of central vision, the macula lutea, having dimensions of approximately 2.0 millimeters horizontally by 0.8 millimeters vertically. The central region of the macula, termed the fovea centralis, is only about 150 micrometers in diameter and provides detailed high-acuity vision and color perception. The regions of the retina outside this tiny area perceive light and detect movement, constituting peripheral vision, but do not contribute to detailed vision. Consequently, damage to the fovea, even though the structure comprises only 3 to 4 percent of the retinal area, can result in instantaneous loss of fine vision.

The band of wavelengths that pass through the outer eye structures and reach the retina includes the entire visible light spectrum from blue (400 nanometers) to red (700 nanometers), and the near-infrared range of 700 to 1400 nanometers (IR-A). Because the retina is not responsive to radiation outside the visible spectrum, no sensation results in the eye when exposed to the near-infrared wavelengths, resulting in a much greater hazard from lasers operating in this emission range. Although invisible, the beam is nevertheless focused on the retina. As discussed previously, because of the focusing efficiency of the eye, relatively small amounts of laser radiation can injure the retina, and in some instances result in serious visual consequences. Pulsed lasers that emit high intensities can cause explosive hemorrhaging when focused in the eye, and the damage can extend a considerable distance from the focused area. Retinal injuries do not heal, and cannot generally be repaired.

Absorption in the other eye components, primarily the cornea and lens, is responsible for limiting exposure of the retina to the ocular focus range of wavelengths, which can also be considered the retinal hazard region. In the absorption process, the absorbing structures become subject to damage themselves. Only the tissue that absorbs the radiation, and tissues immediately surrounding it, are subject to injury and most instances of acute damage resulting from exposure to laser radiation outside the 400 to 1400-nanometer wavelength range do not have long-lasting effects. The cornea behaves similarly to the skin in that it is constantly undergoing replenishment, and only rather severe damage results in scarring that may have some effect on vision. Most damage to the cornea results from laser radiation in the far-infrared and ultraviolet spectral regions.

Because of the high degree of focusing that occurs within the eye, exposure to a relatively weak coherent laser beam can cause permanent, instantaneous damage. Consequently, when a powerful laser is being utilized, a specular reflection (which maintains the beam coherence) of even a few percent, for a fraction of a second, is capable of inflicting eye damage. In contrast, when the laser beam is scattered by reflection from a rough surface, or even from dust in the air, the diffuse reflection enters the eye at a larger angle. With the beam energy spread over a larger area, the reflection has the characteristics of an extended source, and produces a large image on the retina, compared to the concentrated focus produced by a point source (see Figure 3). Diffusion of the beam in this fashion reduces the likelihood of eye damage, not only by increasing the source size and reducing power density, but by disrupting the beam coherence as well.

Biological Effects of Laser Radiation
Photobiological Spectral Domain (CIE Band) Eye
Ultraviolet C
(200-280 nm)
Photokeratitis Erythema (Sunburn)
Skin Cancer
Ultraviolet B
(280-315 nm)
Photokeratitis Erythema (Sunburn)
Accelerated Skin Aging
Increased Pigmentation
Ultraviolet A
(315-400 nm)
Photochemical UV
Pigment Darkening
Skin Burn
(400-780 nm)
Photochemical and Thermal Retinal Injury
Color and Night
Vision Degradation
Skin Burn
Photosensitive Reactions
Infrared A
(780-1400 nm)
Retinal Burns
Skin Burn
Infrared B
(1400-3000 nm)
Corneal Burn
Aqueous Flare
IR Cataract
Skin Burn
Infrared C
(3000-1 million nm)
Corneal Burn Skin Burn
Table 1

Potential damage to the eye can be categorized with respect to the laser wavelength and the eye structures affected, with the most significant injuries being to the retina and caused by radiation in the visible and near-infrared spectral region. Thermal burn, acoustic damage, or photochemical alteration is possible depending upon the energy absorbed. The biological effects on the eye tissues, manifested within various wavelength bands, are summarized as follows, and listed in Table 1.

Ultraviolet-B and C (200-315 nanometers): The surface of the cornea absorbs all ultraviolet light in this range, preventing these wavelengths from reaching the retina. A form of photokeratitis (also referred to as welder's flash) may result through a photochemical process that causes denaturation of proteins in the cornea. In addition to laser output, radiation in this range can arise from laser pump light or may be a component of blue light from a target interaction, requiring additional precautions over those specified by the ANSI standard, which only considers laser output. This type of eye damage is not usually long-lasting due to the rapid regeneration of corneal tissue.

Ultraviolet-A (315-400 nanometers): The cornea and aqueous humor transmit this wavelength range, which is then primarily absorbed by the lens of the eye. Photochemical denaturation of proteins in the lens results in the formation of cataracts.

Visible light and Infrared-A (400-1400 nanometers): This spectral region is often referred to as the retinal hazard region, due to the fact that the cornea, lens, and vitreous fluid of the eye are transparent to these wavelengths, and the light energy is absorbed in the retina. Damage to the retina is possible either through thermal or photochemical processes. Photochemical damage to photoreceptor cells of the retina can degrade overall light or color sensitivity, and the infrared wavelengths may cause cataract formation in the lens. The most likely injury when sufficient laser energy is absorbed by the eye is a thermal burn, in which absorption of light by the melanin granules of the pigmented epithelium is converted to heat. The focusing of the laser radiation by the cornea and lens within this wavelength band amplifies the irradiance by a factor of approximately 100,000 at the retina. For visible light lasers of relatively low power, the possibility of injury is reduced by the aversion reflex (taking about 0.25 second), which causes avoidance of the bright beam. If the laser energy is sufficient to produce damage in less than 0.25 second, however, this natural defense mechanism is not effective, nor does it provide any protection in the invisible near-infrared band between 700 and 1400 nanometers wavelength. Lasers operating in pulsed mode present an additional hazard from the possibility of acoustic shock wave generation in the retinal tissue. Laser pulses with duration less than 10 microseconds induce shock waves that cause tissue rupture. This type of injury is permanent and potentially more severe than thermal burn, because acoustic damage usually affects a larger area of the retina, and the required energy to produce the effect is lower. Consequently, the maximum exposure permitted in regulatory standards is reduced for short-duration pulsed lasers.

Infrared-B and Infrared-C (1400-1 million nanometers): At wavelengths longer than 1400 nanometers, the cornea absorbs energy due to water content of the tissue and the natural tear film, and the resulting temperature rise causes denaturation of proteins near the surface. Depth of penetration increases at longer wavelengths, and the thermal effects on lens proteins, at a critical temperature not much above normal body temperature, may lead to the formation of clouding, referred to as an infrared cataract. In addition to cataract formation and corneal burns, infrared radiation can produce aqueous flare, in which the normally transparent aqueous medium of the anterior chamber is compromised due to disruption of blood vessels.

In general, ultraviolet and far-infrared laser radiation is absorbed at the cornea or lens, and its effect depends upon the intensity and exposure duration. At high intensities, immediate thermal burns occur, while lower exposures may lead to the development of cataracts over a period of years. Conjunctiva tissues of the eye can also be injured by laser exposure, although conjunctival or corneal damage usually occurs at higher power levels than retinal injuries. Consequently, since retinal injuries produce more serious immediate effects, corneal hazards are generally only considered a serious concern for lasers operating at wavelengths that do not reach the retina (essentially far-infrared and ultraviolet).

Skin Hazards

Laser hazards associated with skin exposure are generally considered less important than eye hazards, although with growing utilization of higher-power laser systems, particularly ultraviolet emitters, unprotected skin may be exposed to extremely dangerous levels of radiation in systems that are not completely enclosed. Because the skin is the largest organ of the body, it is at the greatest risk of exposure to a laser beam, and at the same time effectively protects most of the other organs from exposure (with the exception of the eyes). It is important to consider that many lasers are designed for the purpose of material alteration, such as cutting or drilling of materials that are much more resistant than skin, although such lasers are not commonly employed in microscopy. The arms, hands, and head are the portions of the body most likely to be inadvertently exposed to the laser beam when alignment or other experimental manipulations are being performed, and if the beam has sufficient intensity, thermal burns, photochemical damage, and acoustic lesions may occur.

The greatest hazard to the skin results from the high power density of a laser beam, and the wavelength of the radiation determines to some extent the depth of skin damage and the type of injury that results. The penetration depth of laser radiation into the skin is greatest in the wavelength range of approximately 300-3000 nanometers, reaching a maximum in the Infrared A spectral region at about 1000 nanometers. If lasers having the potential of causing skin damage are being used, adequate precautions should be taken to protect the skin, such as wearing long sleeves and gloves made of appropriate fire-resistant material. In many cases, lower laser power can be employed for alignment procedures than is required for the intended experiment.

Electrical Hazards

The hazards associated with electrical components and the supply of power to lasers are essentially the same for nearly all types, and safety precautions specific to each laser configuration or category are not necessary. Of the primary functional laser categories, gas, solid state, dye, and semiconductor, all except semiconductor lasers require high voltages, and often high current, to produce a beam. Variations exist in whether the high voltage is applied directly to the primary laser medium, or to a pump lamp or a pump laser, but it is nevertheless present at some point in the system. A particularly hazardous condition is created in lasers that may retain high voltages in capacitors or other components long after the laser is switched off. This situation is especially common in pulsed lasers, and should always be kept in mind when instrument covers are removed for any purpose. The safest approach is to always assume that a shock hazard exists until otherwise determined. Many lasers utilize high voltages only until laser emission is established, and then operate at electrical levels similar to conventional household line voltages, but this should not be taken as a justification for lack of precautions appropriate for any electrical device.

Specific Safety Considerations for Common Microscopy Lasers

Lasers and complete instrument systems that contain lasers must meet certain safety standards. Depending upon their hazard classification, lasers may be required to have beam shutters, key-controlled interlocks, or other devices to prevent injuries. Warning signs are utilized at all entry points to rooms housing lasers that present a potential hazard, and at locations near the laser where specific dangers exist (examples are illustrated in Figure 4). In devices that contain the beam so that it cannot reach the user's eye, such as laser printers and compact-disk players, additional precautions are not required.

Many laboratory lasers have properties similar to those of high-powered lasers used in industrial settings that produce the same wavelengths, and may require shielding to protect the operator from the beam. Output wavelengths for a number of commonly utilized lasers are summarized in Table 2. For working situations in which the possibility of eye exposure to the laser beam cannot be absolutely eliminated, protective goggles or glasses should be worn. It is essential that the goggles be designed to block light at the specific wavelengths emitted by the laser being employed, while transmitting light at other wavelengths to allow adequate vision. A crucial point is that filtration specific to each laser in use must be provided - there are no universal protective goggles that can be used for all lasers, or with all possible emission lines for multi-wavelength lasers. Since laser light can arrive from any angle, directly or by reflection from surfaces, the goggles must block all possible paths to the eyes.

Titanium-doped sapphire lasers (commonly referred to as Ti:sapphire lasers), are a versatile example of the tunable vibronic solid-state laser category. This type of laser requires optical pumping by an internal flash lamp or another laser, which may be internal or external to the primary laser system. Because of the varied configuration of Ti:sapphire laser systems, a standard set of safety precautions cannot be given. These lasers may be operated in either continuous-wave or pulsed modes, and depending upon the system providing the optical pumping, the electrical requirements and electrical hazards vary considerably. The tunable wavelengths of titanium-doped sapphire lasers typically range from approximately 700 to 1000 nanometers, and therefore standard safety precautions for lasers producing radiation capable of reaching the retina (shorter than 1400 nanometers) should be followed. Since the output wavelength varies, it is likely that more than one type of protective goggles will be required, and the user should be certain that any beam-blocking devices are adequate for the wavelength(s) being emitted. A single short powerful pulse emitted in pulsed-mode operation can cause permanent eye damage, and precautions must be taken to ensure that all possible paths to the eye are blocked, both direct and peripheral.

It is important to be aware that stray emission from the pump laser may be more hazardous in some Ti:sapphire laser configurations than the primary laser beam, and if there is a possibility of this light reaching the working area, eye protection that blocks the pump laser wavelength must be used. If a pump laser is employed that is separate from the vibronic laser housing, extra precautions may be required to eliminate possible exposure to stray light that might result from the coupling of the two lasers. In flash lamp-pumped systems, high-voltages are applied to the lamps that may be retained as capacitor charge in the power supply even when the unit is turned off, and precautions are necessary to avoid electrical shock when performing any laser maintenance. The near-infrared wavelengths emitted by this type of laser can be especially dangerous because, although the beam is invisible or perhaps faintly visible near the 700-nanometer end of the emission band, large amounts of infrared light will be focused on the retina.

Chromium doping of various solid state materials has shown great promise in the development of new tunable vibronic lasers, and as these come into more common use, safety procedures specific to each type must be considered. Chromium-doped lithium strontium aluminum fluoride (Cr:LiSAF) has shown promise as a diode-pumped lasing material, and is being used in place of Ti:sapphire lasers in some multiphoton microscopy applications. With tunable emission wavelengths in the infrared, the safety precautions are similar to those required for Ti:sapphire lasers. However, because the chromium-doped lasers are relatively recently developed products, the user should be aware that protective filters and goggles may not be readily available for their specific emission wavelengths.

Argon-ion, and less-common krypton-ion, lasers produce emission at multiple wavelengths, which are widely exploited in optical techniques such as confocal microscopy. Argon lasers generally are classified as Class IIIB or Class IV under the ANSI safety code, and direct beam exposure should be avoided. The blue-green lines from a highly coherent argon-ion laser beam can penetrate the eye to the retina, causing permanent damage. Safety goggles are available that provide strong absorption of the major emission lines, and should protect the eye from damage. Krypton-ion lasers produce somewhat longer wavelengths than the argon lasers, and at lower power, partly because they emit multiple visible-wavelength lines that are widely distributed across the spectrum. The wide energy distribution presents a problem in that protective filtering goggles designed to absorb the entire laser emission would block most visible light, limiting the practicality of using them. Particular care is required, where krypton-ion lasers are concerned, to avoid exposure of the eye to the multi-line emission. Lasers using argon-krypton mixtures have become popular in fluorescence microscopy for multiple-fluorophore studies requiring stable emission at several wavelengths, and caution must be exercised to provide eye protection from the entire emission that could reach the retina. Additionally, these gas discharge lasers produce ultraviolet wavelengths that are strongly absorbed by the lens of the eye, and since the hazards of continuous-wave emission in this spectral range are poorly understood, ultraviolet-absorbing protective goggles should be worn. Krypton-ion lasers emit at several wavelengths in the near-infrared that are nearly invisible, but which can result in severe retinal damage in spite of their weak visual appearance. Electrical hazards are present due to the application of high voltages to initiate the laser discharge, and the relatively high currents required to maintain emission.

Lasers employing helium-neon mixtures are very widely used in devices such as supermarket scanners and surveying instruments, and those with powers of a few milliwatts or less present a hazard similar to direct sunlight. A momentary accidental glance at a low power beam is not likely to cause eye damage, but the highly coherent light from a He-Ne laser can be focused to a very small spot on the retina, and sustained exposure can cause permanent damage. The fundamental He-Ne emission line occurs at 632 nanometers, but variants emitting wavelengths from green light to the infrared are commonly available. Higher-powered versions of helium-neon lasers present a significantly greater danger, and should be used with great care. There is no means to predict the level of exposure that will produce a given degree of eye damage. The primary safety rule to be followed with this laser category is to avoid anything other than momentary viewing of the beam, and to observe the usual precautions with regard to the high voltages present in the power supplies.

Another gas discharge laser, based on the helium-cadmium system, is widely utilized in scanning confocal microscopy, exploiting the violet-blue and ultraviolet emission lines at 442 nanometers and 325 nanometers. The primary hazard to the eye posed by the blue line is damage to the retina, which is considered to be more susceptible to damage at lower levels of exposure at this wavelength than from longer visible wavelengths. Therefore, even at low power levels, the He-Cd laser warrants adhering carefully to safety procedures. Very little of the ultraviolet 325-nanometer radiation is able to reach the retina due to strong absorption by the eye lens, and long-term exposure may contribute to cataract development. Wearing of appropriate safety goggles can protect against this potential hazard. A difficult problem is presented by a more recent variant of He-Cd laser that simultaneously emits at red, green, and blue wavelengths. Any attempt to filter all three wavelengths with goggles blocks so much of the visible spectrum that the user does not have adequate vision to perform necessary tasks. If only two of the emission lines are filtered, then the risks remain from the third wavelength, requiring careful measures to avoid exposure.

Nitrogen lasers produce an ultraviolet line at 337.1 nanometers, and are utilized as pulsed sources for a number of microscopy and spectroscopic applications. The lasers are often utilized to pump dye molecules to produce additional longer-wavelength lines in certain imaging techniques. Nitrogen lasers are capable of producing high power at an extremely high pulse repetition rate. Corneal damage may result from exposure to the beam, and although absorption in the eye lens protects the retina from the near-ultraviolet wavelengths to some extent, it is uncertain whether this is adequate to prevent retinal damage from the high-power pulses. The safest approach is to ensure full eye protection when this type of laser is employed. Additionally, high voltages are required to operate nitrogen lasers, and precautions are required to ensure discharge of all power supply components before being exposed to contact with them.

The most common solid-state laser relies on ionized neodymium doped at impurity levels in a host crystal. The most widely used host material for neodymium doping is yttrium aluminum garnet (YAG), a synthetic crystal forming the basis of the Nd:YAG laser. Neodymium lasers in general are available in an enormous variety, producing a wide range of power ratings in both continuous-beam and pulsed types. They may be optically pumped by a semiconductor laser, pulsed flash lamp, or by arc lamp, and their characteristics vary widely depending upon the specific design and intended purpose. Because of their wide range of applications, and certain hazards they pose, neodymium lasers have probably caused more eye injuries than any other category.

Neodymium-YAG lasers produce 1064-nanometer near-infrared light that can cause severe retinal damage, and because it is invisible, the likelihood of injuries resulting from reflected beams is increased. Most of these lasers used in microscopy are diode-pumped and produce short intense pulses, which can produce injuries when even a single reflected pulse enters the eye. Therefore, eye protection that blocks all potential paths to the eye should be employed. Infrared-blocking protective goggles can be designed to transmit most visible light, except for applications in which the higher-order harmonics are utilized. Frequency doubling can produce the second harmonic at 532 nanometers (visible green light), which is also transmitted to the retina, and when this emission line is employed additional filtration to attenuate green light is required. Frequency tripling and quadrupling is often applied with Nd:YAG lasers to produce third and fourth harmonics at 355 and 266 nanometers, presenting different hazards, and requiring ultra-violet blocking safety goggles, and possibly skin protection to prevent burn injuries. With lasers that generate several watts of power in the infrared, outputs of hundreds of milliwatts can be achieved at the second, third, and fourth harmonic wavelengths.

Output Wavelengths of Common Lasers
Laser Type
(Spectral Region)
Argon Fluoride Excimer (UV) 193
Krypton Chloride Excimer (UV) 222
Krypton Fluoride Excimer (UV) 248
Xenon Chloride Excimer (UV) 308
Xenon Fluoride Excimer (UV) 351
Helium Cadmium (UV, Visible) 325, 442
Nitrogen (UV) 337
Krypton (Visible) 476, 528, 568, 647
Argon (Visible) 488, 514
Copper Vapor (Visible) 510, 578
Nd:YAG Frequency Doubled (Visible) 532
Helium Neon (Visible, Near IR) 543, 594, 612, 633, 1150, 3390
Gold Vapor (Visible) 628
Rhodamine 6G Dye (Visible, Tunable) 570-650
Ruby (Visible) 694
Diode Semiconductor (Visible, Near IR) 630-1600
Ti:Sapphire (Visible - Near IR) 680-1130
Nd:YAG (Near IR) 1064
Erbium (Near IR) 1540
Hydrogen Fluoride (Near IR) 2600-3000
Carbon Dioxide (Far IR) 9600, 10600
Table 2

Although some diode-pumped neodymium lasers produce relatively low power (especially at higher-order harmonics, when operated in continuous mode), most generate sufficient power to cause injuries, and eye protection should be worn when working with any laser of this type. A difficulty with any laser that generates multiple wavelengths is obtaining appropriate goggles to attenuate all dangerous emission lines. When higher-order harmonics are being employed, it cannot be assumed that light at the longer-wavelength fundamental frequency is not present, and many commercial lasers include one or more specific mechanisms to remove the unwanted radiation optically. Additional electrical hazards exist in neodymium lasers that utilize lamps for pumping instead of diodes, because of the presence of higher power supply voltages.

A considerable amount of research is being conducted to identify alternative crystalline hosts for neodymium doping, and as other materials appear in commercial lasers, special consideration is required for safe operation. As new laser types are introduced, safety devices optimized for their specific characteristics may not be initially readily available. Currently, the most widely used alternative to yttrium aluminum garnet is yttrium lithium fluoride (referred to as YLF), and both pulsed and continuous Nd:YLF lasers are available commercially. Although similar in many respects to the neodymium:YAG lasers, those employing Nd:YLF emit at a slightly different fundamental wavelength (1047 nanometers), and this should be taken into consideration in evaluating performance of safety filters such as goggles with respect to their absorption of the fundamental and higher-order harmonic wavelengths.

Semiconductor diode lasers represent a relatively recent technology that is expanding rapidly in versatility. The performance characteristics of diode lasers depend on a number of factors including electrical properties of the semiconductor, growth processes used in its manufacture, and the impurity dopants employed. The wavelength emitted by the laser medium is a function of the band gap of the material and other properties, which are dependent upon the semiconductor composition. Continued development promises to expand the range of wavelengths available in commercial diode lasers. Currently, semiconductor diode lasers with wavelengths longer than 1100 nanometers are used primarily in fiber optic communications applications. Most lasers in this category are based on active layers of indium-gallium-arsenic-phosphorus compounds (InGaAsP) of varying proportions, and emit primarily at either 1300 or 1550 nanometers. A small percentage of the 1300-nanometer emission is transmitted to the retina of the eye, while at wavelengths longer than 1400 nanometers, the cornea is subject to damage. Significant eye damage is not likely except at fairly high power levels. Most diode lasers emitting in the 1300-nanometer range are low-powered and do not present a serious hazard unless the beam is directed into the eye for long periods. Uncollimated diode laser beams and beams emerging from optical fibers diverge rapidly, providing an additional degree of safety. Safety goggles should be used with high-powered beams if the emission is not completely contained within optical fiber. For alignment of near-infrared beams in optical devices while wearing infrared-blocking goggles, fluorescent screens or other infrared viewing devices are available. Diode lasers operate at low voltages and currents, and therefore, do not generally present an electrical hazard.

Diode lasers emitting at nominal wavelengths less than 1100 nanometers are primarily based on gallium arsenide compounds, and continual development of new materials and manufacturing processes are extending the range of output emission to increasingly shorter wavelengths. With certain exceptions, diode lasers require essentially the same safety precautions as other lasers operating in the corresponding wavelength range and at the same power level. As previously stated, a factor that limits the potential hazard in some cases is the high divergence of diode laser beams, which distributes the beam power over a wide area within a short distance from the emission face of the semiconductor. When an application requires the addition of focusing optics, or some other method of collimation, this factor is negated, however. Diode lasers based on the indium-gallium-aluminum-phosphorus (InGaAlP) system are available, producing 635-nanometer radiation at the milliwatt level, and these require safety precautions similar to those for the helium-neon laser of comparable power. Other laser variants based on similar diode compositions emit at 660 or 670 nanometers, and although the natural aversion response of the eye provides some protection, the eye is not nearly as sensitive to these wavelengths as to the 635-nanometer radiation, and the use of safety goggles is advisable. Care should be taken to ensure adequate absorption at the appropriate wavelengths, as goggles designed to protect the eyes at longer wavelengths may not be effective at 660 or 670 nanometers.

Various gallium aluminum arsenide (GaAlAs) compositions are employed to produce diode lasers with emission wavelengths ranging from 750 to nearly 900 nanometers. Because of limited eye sensitivity at 750 nanometers (a weak sensation of red light is possible), and complete lack of sensitivity at longer wavelengths, these lasers present a greater eye hazard than visible light lasers. Much higher power is available in diode lasers emitting in this range (up to several watts in diode arrays), which can result in eye damage after a short exposure. Due to the invisibility of the beam, the eye's aversion response does not occur, and protective goggles should be employed, especially with high-powered lasers. Even longer-wavelength emission (980 nanometers) is produced by indium-gallium-arsenide (InGaAs) lasers, and goggles verified to attenuate 980-nanometer radiation should be utilized, again due to the hazard of invisible emission inadvertently being allowed to enter the eye.

To summarize, the principal hazards associated with laser use in any application are the risks of personal injury to the eyes and skin caused by contact with the laser beam, and electrical hazards presented by high voltages in the laser. Measures are required to avoid exposure (especially of the eyes) to the beam, and where that cannot be assured, appropriate eye protection must be worn. Four factors are significant in selection of protective goggles or other beam-blocking filtration: the laser wavelength, whether the beam is pulsed or continuous, the type of laser medium (gas, semiconductor, etc.), and the laser output power.

Additional non-beam hazards exist in laser use, some of which are relevant in microscopy applications, and others that are unlikely to be encountered. In many industrial applications, lasers are used to perform cutting and welding operations, and the heating involved can result in emission of hazardous fumes or vapors, which must be safely removed from the working environment. This type of hazard is not germane to laser use in optical microscopy, but other safety issues should be considered. In flash lamp-pumped systems, a potential explosion hazard exists from the buildup of high pressures within the flash tube. The instrument housing should be designed and maintained to contain fragments of the lamp if this type of explosive failure occurs. Cryogenic gases, such as liquid nitrogen or liquid helium, may be used to cool the laser (ruby, or neodymium crystals, for example), and exposed skin is subject to burn injuries if contacted by the cold liquids. If significant quantities of cryogenic gases are vented into a closed room or other confined space, they are capable of displacing room air and creating an oxygen-deficient atmosphere. The electrical hazards associated with laser equipment have been discussed, but cannot be overemphasized, due to the fact that instrument covers, which normally protect the user from electrical circuitry, are commonly removed during laser installation, alignment, servicing, and maintenance. Some laser systems (Class IV or 4, in particular) pose a potential fire hazard if the beam contacts flammable substances, and flame-retardant materials should be utilized wherever beam exposure is possible.

In most university and government laboratories, as well as in industrial and other corporate environments, an established framework exists for managing safety procedures to be followed in potentially hazardous activities, including the use of lasers. The general guidelines outlined in this article are not intended to supplant the more specific requirements of safety personnel in individual working situations. Typically, a local office of environmental safety will have prepared published procedures to be followed under the direction of a laser safety officer, or another person who has the responsibility of training and enforcement of proper safety procedures in the institution housing the equipment, and any potential laser user should ensure that appropriate procedures are followed. This is essential, not only to prevent possible irreversible injury to the laser user, but to protect visitors or others who may inadvertently become exposed to any of the hazards of the laser equipment.


Contributing Authors

Thomas J. Fellers, and Michael W. Davidson - National High Magnetic Field Laboratory, 1800 East Paul Dirac Dr., The Florida State University, Tallahassee, Florida, 32310.