Laser

From Scholarpedia

Author: Dr. Charles H. Townes, Astrophysics, University of California, Berkeley, CA
Author: Dr. Jeff Squier, Department of Physics, Colorado School of Mines, Golden, CO, USA

Figure 1: Femtosecond Yb:KGW laser. The lower figure is a photo of the actual system, the upper figure a schematic of the same system to clarify each of the components. Red – path followed by the laser light in the resonator that surrounds the Yb:KGW crystal. This light is transmitted from the output coupler in the form of 250 fs pulses centered at 1039 nm. GTI – Gire Tournois interferometer, SESAM(Keller et al. 1996) – semiconductor saturable absorber mirror. These are components that are used to force the production of femtosecond pulses from the laser. Blue – path followed by the pump light. The pump light is at 980 nm, and produced by a fiber-coupled, semiconductor laser.

Laser is an acronym for Light Amplification by Stimulated Emission of Radiation (Gordon et al. 1955; Schawlow et al. 1958; Maiman 1960). The basic laser system is capable of amplifying a light signal with tremendous fidelity, analogous to how a good electronic system can amplify a voltage waveform with minimal distortion. The method by which this light amplification is achieved is described by the last two words in the acronym: stimulated emission.

In addition to providing amplification, by adding feedback, the laser amplifier can be made to oscillate. Again, this is similar to many electronic circuits where a fraction of the output signal is feedback into the original circuit to create a self-sustaining waveform or oscillation. While the actual physics between optical oscillators (lasers) and electrical oscillators is quite different, the parallels are useful for establishing an intuitive model of laser operation.

History

The general idea for lasers and masers grew out of a combination of microwave spectroscopy and engineering, and originated independently from three sources: Weber of the University of Maryland, Basov and Prokhorov of the Russian Academy of Sciences, and Townes at Columbia University – all groups working then in microwave spectroscopy. By 1951, Townes had for several years been puzzling over how to amplify waves shorter than they could be amplified by electronic amplifiers and suddenly recognized that atoms and molecules, if they were in a non-thermodynamic state with more in upper than in lower energy levels, could provide the desired amplification and oscillation at short wavelengths.

The first such system was made to work by Gordon, Zeiger, and Townes at Columbia University in 1954 at a wavelength of about 1 centimeter, and they named it the maser. The field of masers became very popular, but almost no one thought such devices could be made to produce wavelengths as short as infrared or visible light. In 1957, Townes sat down to see how much shorter waves could be produced and suddenly realize the technique could even produce light waves. He and Arthur Schawlow wrote a paper, published in 1958, showing how this could be done, and many individuals then started working in this direction. Masers and lasers are really basically the same, but what are called masers are defined as producing wavelengths longer than one millimeter and lasers wavelengths shorter than one millimeter.

The first working laser was produced in 1960 by Theodore Maiman, at the Hughes Research Laboratories, using ruby a crystal, and shortly after that many others were made to work. Javan, Bennett, and Herriott at Bell Labs made the first gas discharge laser and Hall at General Electric the first solid state laser. Since then the field has grown enormously as the result of many contributors, and has had an important impact on both science and technology.

Properties

Lasers provide radiation of very pure frequencies and with simple coherent wave-fronts so that light can be focused to a small point, limited only by diffraction to a size of about one wavelength. They provide ideal amplification, limited only by quantum phenomena. And the power of lasers, especially the spatial concentration of power, can be enormous. The smallest, single-atom lasers emit only about 10^-16 watts. But the most powerful pulsed lasers emit 10¹⁶ watts (ten million billion), more than any other source. Such power lasts only a short time, but its intensity allows us to study and understand new states of matter. When focused, this can become 10²³ watts per square centimeter.

In addition to high power intensity and temperatures, lasers have also produced the coldest things we have every known – below one millionth of a degree absolute temperature.

Masers and lasers measure time very precisely, to a precision of about 10^-15, and very short times – as short as 10^-15 seconds. Lasers measure length very accurately and conveniently. We can even send a laser beam to the moon and measure its distance to one or two centimeters. Lasers and masers are great scientific tools, and have been used in research which has already produced a dozen Nobel Prizes.

How does a laser achieve these remarkable characteristics? The utility, basic properties of laser media, and characteristics of laser light are perhaps nowhere better exemplified then in an ultrashort-pulsed laser. A femtosecond (1 femtosecond = 10^-15 seconds) laser system schematic is shown in figure 1, and a tour of this system will be used to illustrate some of the basic laser types and operating characteristics common to lasers in general. This system is a semiconductor diode-laser pumped, Yb:KGd(WO4)₂ ultrashort pulse laser system (Yb –Ytterbium, K – Potassium, Gd- Gadolinium, W – Tungsten, O- Oxygen, hereafter referred to as Yb:KGW for short)(Brunner et al. 2000). This is an example of an all solid-state laser system. As is typical with solid-state lasers, there is an active ion, Yb³⁺, responsible for the laser action that is doped into a host media, in this case the KGW crystal. The dopant (the laser atom) and the host material are distinguished from one another by the semicolon found in the chemical expression for the material. Other examples of lasers of this type include Nd:YAG, Nd:glass, Nd:YLF, Yb:KYW, Cr:LiSAF and Ti:Al₂O₃ (to name only a few)(Koechner 2006). While only solid-state lasers are considered here, the laser media can in fact be a gas, solid or a liquid.

The system in figure 1 actually begins with a semiconductor laser(Keyes et al. 1964) (fiber-coupled pump laser, shown lower left). This is a laser that is capable of directly converting electrical energy, into optical energy (laser light) with excellent efficiency. Perhaps the most ubiquitous of all lasers, semiconductor lasers are found in our homes and cars (CD and DVD players), at the office (in computers, laser pointers, telecommunication systems), at the store (supermarket scanners), in industry (laser marking), health care (laser surgery) and in research (laser microscopes, laser traps, laser spectroscopy, etc). These are just a few of the many possible applications of this laser source, but it really nicely illustrates how the laser has come to permeate every aspect of our lives.

Each of these applications exploits one or more of the characteristics that make laser emission so unique compared to traditional light sources: high brightness, and exquisite control over the wavelength(s) emitted by the laser. The brightness is a result of the excellent spatial mode characteristics of a laser oscillator. Spatially the emission closely approximates a plane wave: the amplitude and phase are comparatively smooth across the spatial extant of the beam. These characteristics make it possible to produce highly collimated beams (such as seen at laser light shows), and conversely, to focus the beam to tremendous intensities. Indeed, table-top size ultrashort lasers can produce focused intensities exceeding 10¹⁹ W/cm². Compare this to the solar intensity of 100 mW/cm² found at earth’s surface to truly appreciate the tremendous gains in intensity that can be achieved with a high-quality laser beam!

Figure 2: Schematic of an optical trapping system. A semiconductor diode laser bar (1 by 100 micrometer emitter) is imaged into a channel that has blood cells flowing through the system. The line emission from the laser, in turn creates an optical trap that is similarly shaped. The cells are guided along the trap as shown in the photo forming, in this case, an optical conveyer (lower left)(Squier et al. 2004).

Figure 2 is an example of how this spatial uniformity or coherence of a laser can be usefully exploited. In this case, the beam profile from a semiconductor laser is imaged into a microfluidic system, with the net result that cells flowing within the device are optically trapped and guided by the laser beam. Only a few milliwatts of laser power are required to achieve optical trapping(Ashkin et al. 1986; Ashkin 1970) in this manner. In contrast, a traditional light source does not produce a wave front that is sufficiently uniform so that it can be efficiently focused to the diffraction-limited volume needed to achieve optical trapping.

The output of the semiconductor laser for optical trapping was imaged using free-space optics. The semiconductor laser light shown in figure 1 illustrates a second common method for guiding laser light: fiber optics. Again, the excellent spatial characteristics of lasers makes it possible to couple the light into fibers with high efficiency, and, in turn, the addition of optical fibers creates a design degree of freedom that is continuously exploited in laser design, delivery and operation. Indeed, one of the most important classes of lasers is, in fact, the fiber laser.

The laser output of the fiber in figure 1 is imaged into the Yb:KGW laser gain medium. This is an example of a laser-pumped laser system, and can be used to illustrate a second important property of lasers – spectral purity. In this case, the monochromatic (single wavelength) emission of the semiconductor laser, which is at 980 nm, is tuned to the absorption band of the Yb:KGW laser to pump (optically excite) this gain media as efficiently as possible. The Yb:KGW laser, in turn, emits light at 1039 nm. In contrast, a traditional thermal light source is polychromatic, producing emissions that could range from the soft UV (350 nm) to the infrared (10,000 nm). The combination of broad wavelength emission, and the spatially incoherent wave front of a thermal light source are its distinguishing characteristics; the tightly controlled wavelength emission, and the spatially coherent, emission of a laser source are its distinguishing characteristics.

Figure 3: Upper figure – thermal image of the Yb:KGW laser as it is being pumped. Uncooled, the temperature hits 243 degrees F. Lower figure – pumped Yb:KGW crystal, with cooling. The temperature is now a more manageable 89 degrees F.

These laser properties are further illustrated through the thermal images of the Yb:KGW laser rod shown in figure . A simple two lens system images the laser output from the fiber onto the Yb:KGW laser (figure 1). The resulting focal spot size (at the crystal) is 200 µm in diameter. The monochromatic nature of the laser light simplifies the imaging design – no chromatic aberration correction is needed to create a precise focal spot. Further, as stated previously, the single wavelength design is well matched to the Yb:KGW laser absorption. This fact is illustrated by the temperature rise of the central spot (upper thermal image) to 243 F! This large temperature change indicates that the laser energy is effectively deposited and absorbed in the tight focal volume. In this instance < 25 W of laser power are used. By contrast, if a 25 W light bulb were used, we would measure no appreciable temperature change – the source would not image to the tight focal volume by our simple imaging system, only a fraction of the wattage would be captured (effects due to the spatial incoherence and spatial extension of the source), and a large percentage of the captured light would be outside the absorption band of the Yb:KGW (due to the polychromatic nature of the source).

Notably the precise energy delivery can pose a serious challenge in laser design – if the load is not properly balanced, the pump laser can damage the KGW crystal. Conversely, this can be exploited. Indeed, one of the most important applications of fiber –coupled semiconductor lasers (and lasers in general) is, in fact, laser machining and marking.

Putting it all together: the basic laser oscillator and laser amplifier

The most essential elements of the laser can be broken down into three basic areas that are common to all laser types(Siegman 1986). (1) The gain or laser media, which in this case is Yb:KGW. Once again, the Yb atoms are our laser atoms, and the KGW is the crystalline host. (2) A method for pumping or exciting the Yb atoms from the ground state, to a higher lying quantum-mechanical energy level. In this case, it is another laser – the semiconductor laser. (3) A feedback mechanism, usually a series of mirrors placed around the gain media that are optically coated and thus act as high reflectors over the wavelength range that the laser is to be operated(Kogelnik et al. 1966). For the Yb:KGW laser these are dielectric mirrors coated to be highly reflective at a central wavelength of 1039 nm. For the system in figure 1, it is seen that the mirror resonator design can be fairly complex – in this case the design is also used to influence the pulsed nature of the laser source.

All three of these basic aspects, can of course, can be generalized. As stated earlier there are many types of gain media including solid-state, semiconductor, gas and liquid. Different media will have different pumping requirements. Other methods of pumping the laser include: electrical and chemical means, as well as excitation through incoherent optical sources such as flash-lamps. The pumping can be continuous, or pulsed.

Figure 4 is a photo of another popular tunable, laser pumped-laser. In this case the gain media is Ti:Al₂O₃ or Ti:sapphire (center of picture)(Moulton 1986). The central wavelength of Ti:sapphire is 800 nm, and it absorbs at a much shorter wavelength. In this case the laser wavelength used to pump the laser is centered at 532 nm (green beam).

Figure 4: Ti:sapphire (center of photo) laser rod pumped by a frequency-doubled beam from a Nd:Vanadate laser. The pump beam can be seen in the air in the right portion of the image. The beam is no longer visible in the air after passing through the rod (left portion of the image) as the gain media has absorbed more than 90% of the incident pump light.

Figure 5: A Ti:sapphire laser-pumped multipass amplifier. The amplifier rod can be seen in the lower right quadrant of the image surrounded by a red fluorescence.

Feedback is not restricted to the closed system (resonator) design as shown in figure 1. Mirrors can also be used to simply multipass the gain media, to form a laser amplifier. A laser pumped, multipass Ti:sapphire amplifier is pictured in figure 5.

Creating gain: population inversions and stimulated emission

Figure 6: The optical emission occurs between two well-defined energy levels, as shown above. For amplification, the upper energy level must have a larger population then the lower level as indicated by the blue bars.

No matter the pumping or feedback mechanism, crucial to the laser process is the creation of a population inversion amongst the laser atoms. In other words, after excitation (pumping), the number of atoms residing in the higher lying quantum mechanical state must exceed those found in the lower energy level in the gain media (figure 6). Once a population inversion is achieved, there are essentially two methods whereby the atoms residing in the excited state may return to the ground state(Einstein 1917).

The first case is spontaneous emission. In the case of spontaneous emission, the atoms independently return to a lower state, in the process giving up energy that is emitted optically and/or acoustically. Spontaneous optical emission (i.e., fluorescence) produced in this manner is noisy, and since all atoms in the excited state population act independently, incoherent.

The second case is stimulated emission. In this process, an applied optical signal stimulates the return to the lower energy level. Once again, the atom gives up optical energy, however; in this case the emitted radiation is effectively in step (coherent) with the driving or applied optical signal. The applied optical field consequently sees gain- i.e., it is amplified. Notably, the atoms are acting in concert as a population (as opposed to independently) with this driving field. The resulting emission will be an amplified version of the initial applied field, tracking this signal both spatially and spectrally. The stimulated emission process therefore illustrates the necessity of surrounding the gain media with external mirrors - they are used for directing the applied optical signal through the media (for a laser amplifier) or supplying a self-generated signal through feedback (for a laser oscillator).

Crucial to having gain is the inversion. Notably if the number of atoms in the lower lying level shown in figure 6 exceed those in the excited state, the stimulated transition rate will be absorptive, and the applied signal will be attenuated as opposed to amplified. The pumping process must therefore be done in such a manner as to create this inversion (a nonequilibrium condition).

Basic laser operating characteristics

Figure 7: Power as a function of time is plotted for four different operating modes commonly used in many laser systems.

Lasers can be operated in a variety of modalities; figure 7 illustrates four of the most common. The first plot (upper left) is the most common mode – the laser power as a function of time is steady state, resulting in constant light output. This is how the semiconductor laser in figure 1 operates. This operation mode is known as continuous wave or CW. Many applications, such as laser ablation, not only require large average power output, put high instantaneous, or peak powers as well. In those applications it is advantageous to operate the laser in a pulsed mode, called Q-switching (upper right, figure 7). Pulse durations in this regime are typically 10 to 100 ns. For probing fast dynamics, or efficiently exciting optical nonlinearities, even shorter pulse durations are often desirable. In this instance, a technique called mode locking is employed (lower left, figure 7) resulting in pulsewidths from 100 ps to 10 fs. This is how the Yb:KGW laser operates. The different methods can also be mixed. The final plot shows a combination of Q-switched, mode locked operation and is useful when energetic bursts of short pulses are necessary.

Another important characteristic of many lasers, including our example of Yb:KGW, is the ability to tune the laser wavelength. Yb: KGW exhibits a tuning range of approximately 25 nm (from 1032 to 1054 nm). Ti:sapphire, the most broadly tunable of solid-state gain media, can be tuned from 700 nm to 1100 nm. A particularly nice feature of these solid-state lasers is that this tuning range is continuous – the output wavelength of the laser can be smoothly varied as a function of wavelength without sudden discontinuities in laser power or emission. Gas lasers such as Argon can also be tuned – but only along very discrete spectral lines. For instance, Argon lasers can produce light at as short as 350 nm and as long as 1090 nm, but cannot in anyway span this range continuously.

Figure 8: A frequency doubled laser beam from a semiconductor diode laser pumped Nd:Vanadate laser.

Another method of increasing the wavelength range of the laser system is through frequency conversion. In this case, the laser is focused into an optically transparent medium such as a crystal. The laser light effectively overdrives the crystal, and a second beam is produced at one-half the wavelength (or twice the frequency) of the excitation beam. While the exact explanation of this phenomena is beyond the scope of this article, it is an example of one of the first nonlinear optical processes discovered almost immediately after the invention of the laser, and is routinely employed to extend the wavelength range of the laser(Franken et al. 1961; Kleinman 1962). Figure 8 is a picture of a frequency-doubled beam from a laser diode-pumped Nd:Vanadate laser. The outstanding spatial characteristics of the fundamental beam are clearly evident in the frequency –doubled version, as it clearly propagates from top to bottom in a tight, well-confined beam. Nonlinear frequency conversion of laser beams has enabled the production of laser light over the entire visible spectrum, pushing well into the ultraviolet and infrared regions.

Gain media that exhibit smooth, continuous tuning are also the best candidates for producing ultrashort or femtosecond (10^-15 s) lasers. A femtosecond laser pulse is actually a coherent superposition of a band of laser frequencies, and is only produced directly from those lasers that exhibit this tuning capability. With its broad tuning range, Ti:sapphire is the dominate femtosecond laser source(Spence et al. 1991). Femtosecond Ti:sapphire systems now routinely produce pulses as short as 10 fs. For a central wavelength of 800 nm, a 10 fs pulse (assuming a Gaussian pulse shape) requires approximately a 90 nm bandwidth! Our example of an Yb:KGW laser produces pulsewidths on the order of 100-300 fs due to its limited bandwidth.

Application example: nonlinear microscopy

Lasers are employed in many creative ways; one of those applications is presented here. In this example, a femtosecond laser is used as the light bulb for a microscope with several advantages following. In this instance, the time-resolution of the ultrashort laser pulse is not what is exploited, but rather its ability to efficiently excite an optical nonlinearity with extremely modest energies – in this case 100s of picojoules per laser pulse. Basically, the laser generates image contrast through its ability to drive nonlinear optical phenomena directly in the specimen that is under illumination. The most common nonlinearity used is two- photon excitation, pioneered by the Webb group at Cornell(Denk et al. 1990), but images are also obtained using second harmonic generation(Hellwart et al. 1974; Sheppard et al. 1977), third harmonic generation(Barad et al. 1997; Yelin et al. 1999; Müller et al. 1998), coherent anti-stokes raman(Zumbusch et al. 1999) processes, etc. All of these image modalities represent applications that are truly unique to the laser.

An example of laser imaging with optical nonlinearities is shown in figure 9. In this instance, the image is generated through third harmonic generation microscopy. The white light image, taken with traditional illumination is shown in the upper left for comparison. What should be immediately obvious is that the laser microscope only images a single focal plane - to capture the entire specimen it had to be moved through the focal plane of the laser. This has the advantage that a highly detailed, three-dimensional image can then be rendered from the resultant data set (figure 10).

Figure 9: Upper image – normal white light image of a spiral algae formation. Lower image sequence – third harmonic generation image series of the same algae sample. The algae specimen was moved axially (with respect to the laser beam) to create this image series. The color shown is a false color image map. The field of view in all images is approximately several 100 µm. Data taken in collaboration with Professor Michiel Müller, University of Amsterdam.

Figure 10: Cross-sectional third harmonic image data sets (figure 9) are used to render this three-dimensional image of a spiral algae formation (false color).

Conclusion

The laser represents a disruptive technology that has impacted virtually every aspect of modern life: from the way fundamental research is performed, to the way we listen to music. It is the unique characteristics of laser light that has enabled this success. The spatially uniform laser emission makes it possible to produce highly directed beams capable of traversing tremendous distances, or to be focused to microscopic volumes spanning only a mere optical wavelength. Lasers can be operated over the entire visible spectrum, and well into the UV and infrared regimes. Many lasers can produce multiple wavelengths simultaneously, a fact that has been creatively exploited to create femtosecond laser systems. The high intensities of a focused laser beam can be used for applications such as material modification or to extend the wavelength range of the laser itself through the harmonic generation process.

Ultimately, the brief synopsis as provided here is woefully inadequate for describing the many laser technologies and broad range of laser physics that now exists, which is probably, in and of itself, testimony to the tremendous success of the laser.

References

• Ashkin, A. Acceleration and trapping of particles by radiation pressure. Physical Review Letters 24, 156-9 (1970).

• Ashkin, A., Dziedzic, J. M., Bjorkholm, J. E. & Chu, S. Observation of a single-beam gradient force optical trap for dielectric particles. Optics Letters 11, 288-90 (1986).

• Barad, Y., Eisenberg, H., Horowitz, M. & Silberberg, Y. Nonlinear scanning laser microscopy by third harmonic generation. Applied Physics Letters 70, 922-4 (1997).

• Brunner, F. et al. Diode-pumped femtosecond Yb:KGd(WO4)2 laser with 1.1 W average power. Opt. Lett. 25, 1119-1121 (2000).

• Denk, W., Strickler, J. H. & Webb, W. W. Two-photon laser scanning fluorescence microscopy. Science 248, 73-76 (1990).

• Einstein, A. On the Quantum Theory of Radiation. P. Zeit. 18, 121 (1917).

• Franken, P. A., Hill, A. E., Peters, C. W. & Weinreich, G. Generation of optical harmonics. Phys. Rev. Lett. 7, 118 (1961).

• Gordon, J. P., Zeigler, H. J. & Townes, C. H. The maser-- new type of microwave amplifier, frequency standard, and spectrometer. Phys. Rev. 99, 1264-1274 (1955).

• Hellwart, R. & Christen, P. Nonlinear optical microscopic examination of structure in polycrystalline ZnSe. Optics Communications 12, 318-322 (1974).

• Keller, U. et al. Semiconductor saturable absorber mirrors (SESAMS) for femtosecond to nanosecond pulse generation in solid-state lasers. IEEE J. Select Topics Quant. Electron. 2, 435-453 (1996).

• Keyes, R. J. & Quist, T. M. Injection luminescent pumping of CaF2:U3+ with GaAs diode lasers. Appl. Phys. Lett. 4, 50-52 (1964).

• Kleinman, D. A. Theory of second harmonic generation of light. Physical Review 128, 1761-1775 (1962).

• Koechner, W. Solid-State Laser Engineering. Springer Series in Optical Sciences (2006).

• Kogelnik, H. & Li, T. Laser Beams and Resonators. Applied Optics 5, 1550-1567 (1966).

• Maiman, T. H. Stimulated optical radiation in ruby. Nature 187, 493 (1960).

• Moulton, P. F. Spectroscopic and laser characteristics of Ti:Al2O3. J. Opt. Soc. Am. B 3, 125-33 (1986).

• Müller, M., Squier, J., Wilson, K. R. & Brakenhoff, G. J. 3D microscopy of transparent objects using third-harmonic generation. Journal of Microscopy 191, 266-74 (1998).

• Schawlow, A. L. & Townes, C. H. Infrared and optical masers. Phys. Rev. 112, 1940 (1958).

• Sheppard, C. J. R., Gannaway, J. N., Kompfner, R. & Walsh, D. Scanning harmonic optical microscope. IEEE Journal of Quantum Electronics 13, D100(1977).

• Siegman, A. E. Lasers (University Science Books, Mill Valley, California, 1986).

• Spence, D. E., Kean, P. N. & Sibbett, W. 60-fsec pulse generation from a self-mode-locked Ti:sapphire laser. Optics Letters 16, 42-44 (1991).

• Jr., R. W. A., Squier, J., Vestad, T., Oakey, J. & Marr, D. W. M. Optical trapping, manipulation, and sorting of cells and colloids in microfluidic systems with diode laser bars. Opt. Express 12, 4390-4398 (2004).

• Yelin, D. & Silberberg, Y. Laser scanning third-harmonic-generation microscopy in biology. Optics Express 5 (1999).

• Zumbusch, A., Holtom, G. R. & Xie, X. S. Three dimensional vibrational imaging by coherent anti-Stokes Raman scattering. Physical Review Letters 82, 4142-4145 (1999).

External links

• The Squier Group
• U.C. Berkeley Infrared Spatial Interferometer Array

See Also